EP4003164A1 - System zur erfassung von bewegungsabläufen und/oder vitalparametern einer person - Google Patents
System zur erfassung von bewegungsabläufen und/oder vitalparametern einer personInfo
- Publication number
- EP4003164A1 EP4003164A1 EP20764658.9A EP20764658A EP4003164A1 EP 4003164 A1 EP4003164 A1 EP 4003164A1 EP 20764658 A EP20764658 A EP 20764658A EP 4003164 A1 EP4003164 A1 EP 4003164A1
- Authority
- EP
- European Patent Office
- Prior art keywords
- person
- service robot
- detection
- module
- patient
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/117—Identification of persons
- A61B5/1171—Identification of persons based on the shapes or appearances of their bodies or parts thereof
- A61B5/1176—Recognition of faces
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B34/00—Computer-aided surgery; Manipulators or robots specially adapted for use in surgery
- A61B34/30—Surgical robots
- A61B34/32—Surgical robots operating autonomously
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/0059—Measuring for diagnostic purposes; Identification of persons using light, e.g. diagnosis by transillumination, diascopy, fluorescence
- A61B5/0077—Devices for viewing the surface of the body, e.g. camera, magnifying lens
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/01—Measuring temperature of body parts ; Diagnostic temperature sensing, e.g. for malignant or inflamed tissue
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/02—Detecting, measuring or recording for evaluating the cardiovascular system, e.g. pulse, heart rate, blood pressure or blood flow
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/103—Measuring devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
- A61B5/11—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor or mobility of a limb
- A61B5/1113—Local tracking of patients, e.g. in a hospital or private home
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/103—Measuring devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
- A61B5/11—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor or mobility of a limb
- A61B5/1116—Determining posture transitions
- A61B5/1117—Fall detection
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/103—Measuring devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
- A61B5/11—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor or mobility of a limb
- A61B5/112—Gait analysis
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/103—Measuring devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
- A61B5/11—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor or mobility of a limb
- A61B5/1121—Determining geometric values, e.g. centre of rotation or angular range of movement
- A61B5/1122—Determining geometric values, e.g. centre of rotation or angular range of movement of movement trajectories
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/103—Measuring devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
- A61B5/11—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor or mobility of a limb
- A61B5/1126—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor or mobility of a limb using a particular sensing technique
- A61B5/1128—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor or mobility of a limb using a particular sensing technique using image analysis
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/12—Audiometering
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/68—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient
- A61B5/6887—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient mounted on external non-worn devices, e.g. non-medical devices
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/70—Means for positioning the patient in relation to the detecting, measuring or recording means
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J11/00—Manipulators not otherwise provided for
- B25J11/008—Manipulators for service tasks
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/0094—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots involving pointing a payload, e.g. camera, weapon, sensor, towards a fixed or moving target
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0231—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means
- G05D1/0246—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means using a video camera in combination with image processing means
- G05D1/0248—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means using a video camera in combination with image processing means in combination with a laser
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0268—Control of position or course in two dimensions specially adapted to land vehicles using internal positioning means
- G05D1/0274—Control of position or course in two dimensions specially adapted to land vehicles using internal positioning means using mapping information stored in a memory device
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/20—Movements or behaviour, e.g. gesture recognition
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/20—Movements or behaviour, e.g. gesture recognition
- G06V40/23—Recognition of whole body movements, e.g. for sport training
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/20—Movements or behaviour, e.g. gesture recognition
- G06V40/28—Recognition of hand or arm movements, e.g. recognition of deaf sign language
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B2503/00—Evaluating a particular growth phase or type of persons or animals
- A61B2503/08—Elderly
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B2505/00—Evaluating, monitoring or diagnosing in the context of a particular type of medical care
- A61B2505/07—Home care
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/0002—Remote monitoring of patients using telemetry, e.g. transmission of vital signals via a communication network
- A61B5/0015—Remote monitoring of patients using telemetry, e.g. transmission of vital signals via a communication network characterised by features of the telemetry system
- A61B5/002—Monitoring the patient using a local or closed circuit, e.g. in a room or building
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/0059—Measuring for diagnostic purposes; Identification of persons using light, e.g. diagnosis by transillumination, diascopy, fluorescence
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/0059—Measuring for diagnostic purposes; Identification of persons using light, e.g. diagnosis by transillumination, diascopy, fluorescence
- A61B5/0075—Measuring for diagnostic purposes; Identification of persons using light, e.g. diagnosis by transillumination, diascopy, fluorescence by spectroscopy, i.e. measuring spectra, e.g. Raman spectroscopy, infrared absorption spectroscopy
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/02—Detecting, measuring or recording for evaluating the cardiovascular system, e.g. pulse, heart rate, blood pressure or blood flow
- A61B5/0205—Simultaneously evaluating both cardiovascular conditions and different types of body conditions, e.g. heart and respiratory condition
- A61B5/02055—Simultaneously evaluating both cardiovascular condition and temperature
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/05—Detecting, measuring or recording for diagnosis by means of electric currents or magnetic fields; Measuring using microwaves or radio waves
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/48—Other medical applications
- A61B5/4845—Toxicology, e.g. by detection of alcohol, drug or toxic products
Definitions
- the invention comprises a service robot for the automated implementation of geriatric tests.
- the health system suffers from a significant shortage of skilled workers.
- This shortage of skilled workers means that there is less and less time to treat patients.
- This lack of time not only leads to dissatisfaction on the part of both patients and medical staff, but can also lead to diseases not being adequately treated, which not only causes patient suffering but also reduces the value added of an economy.
- These effects go hand in hand with the fact that the need to document the patient's condition continues to increase in order to have to defend oneself from a medical point of view against claims for damages that can be attributed to inadequate therapies.
- This documentation requirement can in some cases lead to a self-reinforcing effect.
- the service robot described in this document addresses this problem in that the service robot independently carries out geriatric tests, which are currently being performed by medical personnel, using several sensors. This service robot is also able to precisely document the completed exercises, which means that the health facility of the service robot is able to comply with the relevant documentation requirements and other compliance obligations without having to separately deploy personnel. Another effect is that the use of the Service robot standardizes the assessment of the tests, because currently the assessment of a patient is subject to the experience of the medical staff, who in turn differ from other medical staff by their individual experience. This means that different evaluations are possible for the same exercise when assessments by medical staff are made, while a uniform assessment is made when the service robot is used.
- the service robot can, for example, collect the Barthel index, can carry out the so-called "Timed-up and Go" test and / or the mini-mental test with different characteristics, is the In one aspect, the service robot is also configured in such a way that the service robot can alternatively and additionally address other tasks in a clinic. This includes, for example, spectrometric examinations, with which various substances in or on a person's skin can be analyzed. These analyzes can be used, for example, to determine the delirium detection score.
- the service robot is also configured to perform delirium detection and / or delirium monitoring.
- the service robot can, in one aspect, determine possible attention disorders of the patient on the basis of the recognition of a sequence of acoustic signals.
- the service robot can evaluate cognitive abilities based on image recognition and / or cognitive abilities via the implementation in motor skills, for example by counting fingers that the patient shows in response to a primarily visual request from the service robot .
- the service robot is able to determine the pain status of a person. This can be done by means of emotion recognition, detection of the movements of the upper extremities, and / or pain vocalization of ventilated and / or non-ventilated patients.
- the service robot can, in one aspect, determine the blood pressure of a patient, their breathing frequency, and use this information, apart from original diagnostic and / or therapeutic purposes, also to control their own hardware and software components.
- the service robot is configured so that the service robot
- Service robots can also check whether users are under mental and / or physical
- the service robot can adapt its signal processing quality, but also the signal output, to environmental influences.
- the use of the service robot offers a significant relief of the medical staff, which this medical staff has to perform sometimes time-consuming and sometimes monotonous work that has no direct impact on the health of a patient and thus prevents the staff from implementing measures that immediately improve a person's health.
- CN108422427 describes a rehabilitation robot that is able to serve food on trays.
- CN206833244 in which a service robot distributes materials in the hospital, is stored in a similar way.
- the Chinese patent applications CN107518989 and CN101862245 which include a service robot that transports patients, similar to a wheelchair, are also in the hospital environment.
- CN205950753 describes a robot that recognizes patients using sensors and guides them through a hospital.
- CN203338133 describes a robot that supports the nursing staff and accompanies patients in the hospital with their daily tasks.
- CN203527474 refers to a robot that uses its arm to support the elderly.
- CN108073104 relates to a care robot that cares for infected patients by the care robot providing these patients with drugs or administering these drugs, massaging the patient, feeding, communicating with the patient, etc.
- the care robot reduces the risk of infection for medical staff by using the care robot to reduce the number of patient contacts made by staff.
- a robot for accompanying the elderly can be found in CN107598943. This robot has some monitoring functions, but most of all a floor cleaning function.
- CN106671105 is a mobile service robot for the care of the elderly.
- the service robot uses sensors to monitor body parameters such as temperature and facial expressions. The service robot also recognizes whether a person has fallen and can alert help accordingly via a network.
- the oxygen determination in the blood is also carried out in CN105082149.
- CN105078445 relates to a service robot that makes it possible to record an electrocardiogram and measure the oxygen content in the blood, especially in older people.
- CN105078450 with an electroencephalogram measurement is located similarly.
- Some of the health robots refer explicitly to the implementation of exercises with patients or tests.
- a system is described in a relatively abstract manner that carries out exercises with a patient based on stored information.
- CN108039193 describes a system for the automatic generation of health reports that is used in a robot.
- the recording of movements / fitness exercises with the help of a robot, the recording and saving of the data of the same for analysis purposes and the transmission to external systems are described in CN107544266. At the same time, this robot is able to monitor the consumption of medication via various sensors.
- CN106709254 describes a robot for the medical diagnosis of a patient, which at the same time, based on the diagnosis, also creates a treatment plan. To do this, the robot evaluates voice and image information and compares it with information stored in memories.
- a neural network is used here.
- CN106407715 describes a service robot that takes the anamnesis of a patient by means of speech processing and image recognition. In addition to querying via voice input and output devices via a touchpad, a photo of the tongue, which is taken by a camera of the robot, is used for the anamnesis.
- CN105078449 describes a service robot with a tablet computer as a communication device, via which, among other things, cognitive function training or a cognitive-psychological test takes place in order to discover Alzheimer's in patients.
- the tablet temporarily records a telephone call running according to a specific process the patient and a child and deduces from the course of the conversation whether the patient suffers from Alzheimer's.
- US10268277 describes general hand gesture recognition, as does US9372546 or US9189068.
- US9690984 for example, camera-based hand recognition is shown using a skeleton model with the aid of machine learning algorithms.
- US9423879 is dedicated to the recognition and tracking of objects in hands and proposes using a thermal sensor to differentiate hands and fingers (due to the heat given off) from other objects (which tend to be cooler).
- US9117274 describes how a depth camera is used to detect a sheet that a user is holding in his hand, while in a next step an image is projected onto this sheet, which is an example of a flat surface, with which the user can interact.
- the sheet is identified by its corners, which are compared with the squares stored in the memory, which have been rotated spatially in space.
- US 10242527 describes how gaming tables (in the casino) are monitored by automatically recognizing hand gestures, including gaming chips or playing cards that have a certain resemblance to a hand.
- gaming chips or playing cards have a certain resemblance to a hand.
- playing cards have rounded corners, which is usually not the case with a hand.
- Robots have. US20170011258 explains how based on the emotional state of a
- Facial expression of the person is assessed, which is determined by means of a histogram-of-gradients-
- the emotional state of a person can be general about
- Classification methods take place based on clustering, or by means of neural
- US2019012599 describes quite generally how a multilayer convolutionary neural network is used to generate weights based on video recordings of a face, which has at least one convolutionary and at least one hidden layer, the last level of which describes a person's emotions, which determines weights for input variables at least one level, the weights in at least one feed forward process (activated process) calculated and updated as part of a backward propagation.
- US9619613 uses a special device that works with vibrations, among other things, to evaluate a person's mental state.
- US9659150 uses, for example, acceleration sensors to carry out the timed-up and go test.
- stimuli are triggered in order to test the mental abilities by outputting a sequence of stimuli of defined length and recording the patient's reaction.
- US8475171 uses, for example, a virtual reality to show a patient various images and to diagnose, for example, Alzheimer's disease via their detection.
- US10111593 for example, uses a movement analysis to detect delirium.
- CN103956171 tries to draw conclusions about a test score of the mini-mental test based on the pronunciation of a patient.
- the service robot is configured in such a way that the service robot can collect further medical parameters by means of its sensor system, including the blood pressure in a contactless way, for example by means of a camera.
- the state of the art for determining blood pressure using a camera-based evaluation is mostly at the research stage.
- Zaunseder et al. (2016) gives an overview primarily of those methods that carry out a color evaluation of the blood flow.
- the review by Rouast et al. (2018) goes a little further. Specifically deals with evaluation algorithms for determining blood pressure on the basis of available signal data, for example. Karylyak et. al (2013) or Wang et al. (2014), while McDuff et al.
- the service robot can determine substances on the skin or within the skin, partly on the way through skin contact, partly also without contact.
- Spectrometric approaches in particular come into play here.
- Spectrometers or approaches similar to spectrometers can be found, for example, in US6172743, US6008889, US6088605, US5372135, US20190216322, US2017146455, US5533509, US5460177, US6069689, US6240306, US5222495, US8552359.
- FIG. 1 shows a schematic structure of a service robot
- Fig. 20 Data from the service robot are processed for therapy proposals
- Fig. 25 Pose recognition of the hand with a view to the numbers shown
- Fig. 58 Tampering detection based on audio signals
- Fig. 59 System for determining scores in connection with standing up / sitting down on a chair
- Fig. 61 System for recording and evaluating a folding exercise
- Fig. 65 System for cognitive analysis
- Fig. 66 System for determining the pain status
- Fig. 72 System for determining a fall risk score
- Fig. 73 System for determining the balance of a person
- Fig. 74 System for determining the position of a foot
- FIG. 85 A method for detecting the sequence of movements of a person along a line, describing the invention
- the term user is understood to mean a person who uses the service robot 17, in this case primarily evaluated by the service robot 17 using the described apparatus. These can be people of advanced age with whom the service robot 17 carries out a geriatric test, but also relatives or third parties who, for example, support people of advanced age in interacting with the service robot 17, or those who are more advanced for the people Age to perform the test.
- Fig. 1 shows the mobile service robot 17.
- the service robot 17 has a laser scanner (LID AR) 1 for scanning the environment of the service robot 17.
- LID AR laser scanner
- Other sensors are alternatively and / or in addition also possible, for example a camera (2D and / or 3D) 185, an ultrasonic and / or radar sensor 194.
- the service robot 17 has at least one display 2, which in one aspect is a touchpad. In the aspect illustrated in FIG. 1, the service robot 17 has two the touchpads. The touchpads in turn have, for example, a microphone 193 and a loudspeaker 192, which allow acoustic communication with the service robot 17. Furthermore, the service robot 17 has at least one sensor 3 for the contactless three-dimensional detection of the movement data of a patient. In one non-limiting example, the sensor is a Microsoft Kinect device. Alternatively, an Orbecc Astra 3D camera can also be used. Such 3D cameras have a stereo camera system for depth recognition, which enables the evaluation of a skeletal model of a patient, and mostly also an RGB camera for color recognition. In an alternative aspect, a conventional monaural camera can be used. Technologies that can be used in 3D cameras are time-of-flight sensors (ToF) or speckle sensors.
- ToF time-of-flight sensors
- speckle sensors speckle sensors
- a pressure-sensitive bumper 4 around the outer shell of the service robot 17, at least in the areas which are in a possible direction of travel of the service robot 17.
- the computer unit 9 is connected to the pressure-sensitive bumper 4 and detects collisions between the service robot 17 and an object. In the event of a collision, the drive unit 7 is stopped immediately.
- the service robot 17 has two drive wheels 6 which are centered and arranged parallel to one another (see FIG. 2). There are two or three further support wheels 5 around it, for example on a circular path.
- This arrangement of the support wheels 5 allows the service robot 17 to be rotated on the spot by controlling the drive wheels 6 in opposite directions.
- the axis of the two or three support wheels 5 is mounted in such a way that the axis can rotate 360 degrees around the vertical axis.
- the distance between the drive wheels is greater than shown in FIG. 2, so that the service robot 17 is prevented from tilting too easily.
- the service robot 17 also has an energy source 8 for supplying the drive and computer unit 9, the sensors (laser scanner 1, sensor 3, and bumper 4) and the input and output units 2.
- the energy source 8 is a battery or a Accumulator.
- Alternative energy sources such as a fuel cell, including a direct methanol or solid oxide fuel cell, are also conceivable.
- the computer unit 9 has at least one memory 10 and at least one
- Interface 188 for data exchange.
- These include (not shown) in an optional aspect a device to read a mobile memory (for example a transponder / RFID token).
- this mobile memory can also be written to.
- this or another interface 188 (such as WLAN) allows wireless communication with a network.
- the service robot 17 has rules, described in the further course of this document, for carrying out evaluations, which are stored in the memory 10. Alternatively and / or in addition, these rules can also be stored in the memory of a cloud 18, which the service robot 17 accesses via the at least one interface 188 (such as WLAN). This does not have to be explicitly mentioned elsewhere, but is included in the disclosure.
- the sensor 3 recognizes a person and their actions and creates a skeleton model on the basis of the person's movements.
- the sensor 3 is also able to recognize walking aids / forearm crutches (UAGS).
- UGS walking aids / forearm crutches
- the service robot 17 also optionally has one or more microphones 193, which can be implemented independently of the touchpads in order to record the person's speech and evaluate it in a computer unit.
- the navigation module 101 contains, for example, various modules. This includes a 2D or 3D environment detection module 102 which, for example, evaluates information about the environment based on various sensor data.
- the path planning module 103 allows the service robot 17 to determine its own path that it travels.
- the movement planner 104 uses, for example, the results of the path planning from the path planning module 103 and calculates an optimal route for the service robot, taking into account or optimizing various cost functions.
- the self-localization module 105 allows the service robot 17 to determine its own position on a map, for example using odometry data, the comparison of captured environmental parameters from the 2D / 3D environment detection with environment parameters stored in a map from map module 107, etc.
- the mapping module 106 allows this Service robot 17 to map its environment. Maps that have been created are stored, for example, in the map module 107, which, however, can also contain maps other than those that have been created in-house.
- the loading module 108 for automatic loading.
- there can be a database with room data 109 which, for example, contains information about the room in which an evaluation is to be carried out with a person, and so on
- Movement evaluation module 120 contains, for example, a movement sequence extraction module 121 and a movement sequence evaluation module 122. These each comprise the rules for movement evaluation, which are described in more detail in the further course of this document.
- the person identification module 110 contains, for example, a person identification module 111, which contains, for example, rules to determine from the recorded sensor data whether it is a person or something else.
- a visual person tracking module 112 for visual person tracking is based, for example, primarily on camera data as input variables; the laser-based person tracking module 113 uses LID AR 1 accordingly.
- a person re-identification module 114 allows, for example, a subsequently recorded one in the event of an interruption of the tracking process To classify a person as to whether it is one who was previously tracked.
- a seat recognition module 115 makes it possible, for example, to detect a chair.
- the service robot 17 also has a human-robot interaction module 130, comprising a graphical user interface 131, a speech synthesis unit 133 and a speech evaluation module 132.
- an odometry unit 181 for example an interface for communication with RFID transponders, a camera 185, operating elements 186, an interface 188 such as WLAN, a charging controller 190 for the energy supply, a motor controller 191, loudspeakers 192 , at least one microphone 193, for example a radar and / or ultrasonic sensor 194, a detector 195, which will be described in more detail elsewhere, also for example a spectrometer 196 and for example a projection device 920.
- LID AR 1, display 2 and Drive 7 have already been described.
- FIG. 3 illustrates that the service robot 17 is connected to the cloud 18 via an interface 188.
- a therapist has the option of accessing one in the via a terminal 13
- Medical staff can store patient data in the patient administration module 160 or, in one aspect, import these patient data from other systems via an interface 188 (such as WLAN). These other systems include above all hospital management systems (HIS) and / or patient data management systems, as they are usually used in clinics or medical practices. In addition to the name and possibly the room number of the patient, the patient data also includes information on the general state of health, etc.
- the computer 161 in the patient administration module 160 generates an ID for each person, which is stored in the memory 162 with the personal data.
- the medical staff can define the tests to be performed.
- the management system is connected to a set of rules 150 via the cloud 18, consisting of a computer unit 151 and a memory 152.
- the set of rules 150 holds rules for performing and evaluating the exercises, which can match those of the service robot 17 and, for example, centrally in the set of rules maintained and then distributed to several service robots 178.
- the classification of objects and movements is stored in the set of rules 150, but also the combination thereof in order to evaluate the observations for the purposes of the test.
- the position of the legs, upper body, arms, hands, etc. is stored on the basis of a skeleton model.
- objects that are to be evaluated as part of the test can be recognized.
- the set of rules 150 can initially be created based on a template with the help of specialists, i.e. limit values are set for individual limbs. Fuzzy algorithms can also be used for the limit values.
- medical staff can label individual images or image sequences, which can be translated into a skeleton model with a view to images of a person, and classifications are established using machine learning algorithms, including neural networks, which map the limit values.
- a cloud-based navigation module 170 with a navigation computer 171 and a navigation memory 172.
- the service robot 17 can be connected to a CI application in the cloud 18.
- the therapist can give the person performing the test a mobile Assign a storage unit such as a token.
- the token contains the patient ID and / or another token ID to which the person or his ID is assigned.
- the person can identify himself on the service robot 17 with this token or the serial number and / or the ID. Identification is also possible in other ways, e.g. by entering login data in a screen-guided menu, but also by means of biometric features such as a face scan or software on a mobile device that has a code ready that is entered or read into the service robot 17.
- the service robot 17 now downloads the test stored by the medical staff from the cloud 18, but without the personal data, via an interface 188 (such as WLAN) - the assignment is made via the person ID.
- an interface 188 such as WLAN
- the service robot 17 loads the data from the test in encrypted form into the patient administration module 160 - the assignment takes place via the person ID.
- the data is only decrypted in the patient administration module 160 (see below).
- the medical staff can then evaluate the data, as will be explained in more detail below using appropriate examples.
- the medical staff transmits the instructions for carrying out a test or a partial component thereof to a storage medium (eg transponder in the form of an RFID tag), which the person receives in order to identify himself on the service robot 17, for which purpose this has an RFID interface 183.
- the data is transferred from the storage medium to the service robot 17, including the person ID that was specified by the patient administration module 160.
- the service robot 17 transfers the data back to the storage medium so that the medical staff can transfer the data to the patient administration module 160 when reading out the storage medium.
- the data can also be transmitted in encrypted form to the patient administration module 160 via a wireless or wired interface 188 (such as WLAN).
- a LID AR 1 and radar and / or ultrasonic sensor 194 the service robot has sensors that are not only used for navigation purposes can, but also, for example, for person detection and tracking, which is why these sensors, together with corresponding software modules, form a person detection and tracking unit 4605 on the hardware side, in which case further sensors can also be used, for example in conjunction with an inertial sensor 5620, which is located on the person to be recorded and / or tracked is located.
- a person recognition module 110 can be used in a first step that recognizes a person from sensor data and can have various sub-modules. This includes, for example, a person identification module 111 that allows a person to be identified.
- the person re-identification module 114 enables the person to be recognized, for example after the interruption of person tracking, which can be done by a visual person tracking module 112 (e.g. evaluating data from a camera 185) or a laser-based person tracking module 113 (e.g. evaluating data from a LID AR 1) .
- the person can be recognized in the person re-identification module 114 by means of a pattern comparison, the patterns resulting, for example, from the stored person characteristics.
- a movement evaluation module 120 allows various movements to be evaluated. Recorded movements can be saved in
- Movement sequence extraction module 121 are first preprocessed, i.e. features of the movements are extracted that are classified and evaluated in the movement sequence evaluation module 122, for example in order to identify a specific movement.
- a skeleton model can be created in the skeleton creation module 5635, which determines articulation points on the joints of the person and direction vectors between the articulation points.
- a feature extraction based on articulation points takes place, for example, in the skeleton model-based feature extraction module 5460.
- a number of specific feature extraction modules are listed in the document as well as a number of feature classification modules which, for example, can be based on said feature extraction modules. In one aspect, this includes the gait feature extraction module 5605, which also uses data from the skeleton creation module 5635, a gait feature classification module 5610 and a gait sequence classification module 5615.
- wrist points are mentioned which can be used to represent the position of a hand, for example when it comes to holding a hand Evaluate person on objects.
- finger joint points can also be understood here as far as fingers can be evaluated over the detection distance.
- people and users are mentioned, for example.
- a person can be understood relatively broadly, while a user is usually a person who has identified himself on the service robot 17. The terms can, however, be used synonymously in many places, but the differentiation is particularly relevant for manipulation detection.
- threshold value comparisons there is sometimes talk of a threshold value being exceeded, which then leads to a specific assessment of a situation.
- Various calculations can be used, some of which could lead to a contrary interpretation of the evaluation results.
- An example is a comparison of two patterns that are used to recognize people. If, for example, a similarity coefficient is calculated here, for example a correlation, a high correlation which is above a threshold value means that, for example, there is an identity of two people. However, if there is a difference between the individual values, a high difference value means the opposite, i.e. a high dissimilarity.
- such alternative calculations are regarded as synonymous with, for example, the first calculation via the correlation.
- machine learning methods can, for example, make the establishment of explicit threshold values, for example for motion sequences, superfluous in favor of a pattern evaluation. That is, instead of threshold value comparisons, for example for dedicated distances of an articulation point from a skeleton model, pattern comparisons take place which evaluate several articulation points at the same time. If a threshold comparison is referred to below, in particular with a view to a movement sequence, a way for a pattern comparison can also be found in the case of the use of machine learning algorithms.
- a pattern comparison for example, body poses of a movement sequence, whether correct or incorrect, can be recorded over time and evaluated in a coherent manner.
- a classifier can be created which then compares it with other recorded body poses that are specified as correct or incorrect and the courses of the articulation points derived therefrom. Survey of the Barthel index
- the service robot 17 can carry out in any of the tests that the service robot 17 can carry out.
- this Barthel test an assessment of the basic abilities of independence or need for care such as eating and drinking, personal hygiene, mobility and stool / urine control is carried out on the basis of behavioral observation.
- the service robot 17 is configured in such a way that a user is asked questions about these subject areas via the communication devices.
- the user can be the person to be assessed.
- other persons, for example relatives can also be asked questions on these topics via the communication device.
- the question is asked here either via menu guidance on a display 2 of the service robot 17 or via a voice interface.
- a separate display 2 connected to the service robot 17 via an interface 188 (such as WLAN), such as a tablet computer, can be used, which the person holds in the hand or put it on a table, which makes answering and completing the exercises easier.
- an interface 188 such as WLAN
- a differentiation between the person to be assessed and, for example, relatives is made via the questionnaire.
- differentiations according to the approaches are also possible, which are described in more detail, for example, in the section on manipulation detection.
- One of the tests that can be carried out by means of the service robot 17 is the so-called “Timed Up and Go” test.
- a person to be evaluated sits in an armchair, stands up and walks ten feet to turn around and sit down again. The time used for this is recorded and, based on a table, converted into a score.
- the service robot 17 scans the room in which the service robot 17 is located with a laser scanner 1, calculates the distances to the walls and creates a virtual map as part of the mapping by the mapping module 106, which reproduces the outlines of the room, but also objects that are located between the laser scanner 1 and the walls are also noted in the XY plane.
- the created map is saved in Card module 107 deposited. If the laser scanner 1 does not have an all-round view, the service robot 17 carries out travel movements in such a way that the service robot 17 scans its surroundings by approximately 360 °.
- the service robot 17 carries out this scanning, for example, from different positions in space, in order, for example, to recognize isolated obstacles.
- the service robot 17 is able to recognize the room again by scanning a part of the room again. This further scanning is the more precise the more the space is scanned.
- the service robot 17 records, for example, the path it has covered and measures the distances, so that the service robot 17 can determine its position in space.
- the distance covered can also be measured by evaluating the rotational movement of the wheels in connection with their circumference.
- the position can be determined more easily, since characteristic dimensions are recognized not only in the XY plane, but also in the Z plane, which means that unique dimensions within space can be found more quickly let identify than this identification is only possible in a two-dimensional representation.
- More than just one sensor can also be used in the room mapping by the mapping module 106, for example the combination of the LID AR 1 and the sensor 3, the sensor 3 being an RGB camera, for example the coloring recorded in space and assigned a color value to each point in the XY plane recorded by the LID AR 1.
- the computer unit of the service robot 17 performs image processing in such a way that a Z coordinate is initially assigned to each point in the XY plane, which is represented by the inclination of the LIDAR and its height relative to the ground.
- the RGB camera in turn, has a known relative position to the LID AR as well as a known alignment angle and a known recording angle, so that this results in the distance in the image, for example a horizontal, straight line that is at a distance of 2m and 50cm above the ground.
- These parameters can be used to assign a pixel from the RGB image to each spatial coordinate that the LID AR 1 determines, and thus also the color values of the pixel.
- the LIDAR 1 By means of the LIDAR 1 it is possible to determine the position in the room where a chair is presumably located.
- the detection method is described in FIG. 4.
- Chairs typically have one to four legs, whereas one-legged chairs are Office swivel chairs, which due to their possible rotation around the Z-axis, are less suitable for people of advanced age who may have mobility impairments.
- Chairs with two or four legs are much more likely, although two-legged chairs could in most cases be so-called cantilever chairs.
- Chair legs are further characterized by the fact that the chair legs are isolated in the XY plane, with which the LID AR 1 detects isolated objects in step 405.
- the chair legs primarily have a homogeneous cross section in the XY plane with a constant Z (step 410).
- the diameter of the objects is between 0.8 cm and 15 cm, for example between 1 cm and 4 cm, and is determined in step 415.
- the distance between the objects, which may stand out as chair legs, from one another in the XY plane is typically approximately 40 cm 420.
- the legs are also primarily arranged in the form of a rectangle (step 425). This means that two objects with the same diameter indicate the existence of a free oscillator with two legs (step 430). If the front legs of the chair and the rear legs of the chair have the same cross-section, the chair is presumably a four-legged chair (step 435).
- the service robot 17 is now able, on the basis of these characteristics (two or four objects, which are isolated, which have an approximately symmetrical cross-section, which are about 40 cm apart and which may be approx. are arranged in a rectangle) to assign the attribute “chair” to these objects and to define those positions of the chairs in step 440 in which are located in the virtual map that was created using LID AR 1 and / or one or more additional sensor (s) are likely to have one or more stools.
- Each chair identified is also assigned a spatial orientation in step 445.
- the chairs are mostly located approximately parallel to a wall and usually have a distance from it that is between two and 20 cm, this distance being applicable to the back of the chair.
- the line between two of the chair legs, which is parallel to the wall, and which is typically 40-70 cm from the wall, is assigned the property “front of chair” 450, the two areas that are orthogonal to it labeled as "backs" of the chair in step 455.
- the side further away from the nearest wall can also be recognized as the front.
- a 2D or 3D camera 185 can also be used to recognize the chairs.
- the computer unit sends the recordings via an interface 188 (such as WLAN) and an API, if necessary to a web service in the cloud 18, which is set up to carry out image classifications, or the computer unit accesses those stored in the memory 10 of the service robot 17
- Image classification algorithms that are able to recognize a chair in the recordings made by the 2D or 3D camera 185, including a chair with armrests.
- There are a large number of algorithms that initially carry out such classifications and create a model that is then applied to the 2D or 3D camera 185 of the service robot 17 either in the web service in the cloud 18 or in the memory 10 of the service robot 17 can, including neural networks such as convolutionary neural networks.
- the service robot 17 is able to save the position of a chair regardless of the type of chair identification, eg in its own memory 10, which interacts 465 with the navigation module 101 of the service robot 17.
- the service robot 17 also detects the number of chairs in the Room 470 and counts these chairs in a clockwise order. Alternatively, a different order is also possible. From this sequence, the chairs receive a number that is stored as an object ID 475.
- the service robot 17 is able to map a room including existing chairs, i.e. to determine the position of the chairs in the room including their orientation. In order to carry out the Timed Up and Go test, however, it is necessary for a person to be on one of the chairs, which may also be over
- the position and the orientation of the chair in the room is identified via the preceding method in FIG. 5 in step 505.
- Service robot 17 roughly at least 20 °, for example at least 45 °, ideally at least 90 ° around the location where a chair is located 510, the LID AR 1 and / or one or more further sensors being oriented in the direction of the chair is / are (step
- the service robot 17 maintains a distance of more than 50 cm, for example more than 80 cm from a 520. This increases the accuracy of the prognosis that the chair legs will move from the chair Service robot 17 are recognized and thus allows a conclusion that a person is on the chair.
- the computer unit 9 in the service robot 17 is able to identify 550 the identified objects as walking aids. If the majority of these features are not detected, no walking aids are identified 585. Naive Bayes estimates, for example, can be used for this purpose.
- steps 525-550 are optional and / or not necessary in order to identify the person in a chair by means of the LIDAR 1.
- the legs of the person sitting on the chair are presumably located around the front legs of the chair.
- One or both legs can be positioned in front of, between the front chair legs, or behind the front chair legs. This results in an approximately funnel-shaped area that extends approximately radially forwards from approximately the center of the chair and extends up to a maximum of approximately 50 cm above the connecting line of the two chair legs 555.
- the data recorded by the LID AR 1 are evaluated to the effect that in Identify 570 two roughly round to elliptical objects 565 which have a diameter of 6-30 cm, for example 7-20 cm, 570 in this area.
- the legs can also be stored between the front chair legs or even behind these chair legs. The closer the objects are to the line between the two front chair legs, the closer the shape of the objects corresponds to a circular shape 575. If these criteria are largely met, the set of rules 150 stored in the service robot 17 recognizes a person on the basis of the LIDAR data the chair 580. Alternatively, no person is recognized 590 in the chair
- the person and walking aids can also be recognized via a general image classification, as already described, for example, a little earlier.
- the service robot 17 also improves the forecast accuracy by the Service robot 17 takes aim at the chair from several positions, so that the 2D or 3D camera 185 can record the chair, in a manner similar to, for example, explained in the previous paragraph.
- the methods (SDKs) adequately described in the prior art can also be used for person recognition via 2D or 3D cameras 185, which function on skeleton recognition, e.g. the Kinect SDK, Astra Orbbec SDK, Open Pose, PoseNet from Tensorflow, etc.
- the service robot 17 is also able to identify 605 a person in the room, for which different, alternative and / or supplementary approaches can be used Cross-sections are identified which have a diameter of at least 5 cm, for example at least 7 cm, which are not exactly round and statically have a distance of at least 3 cm, for example at least 5 cm.
- a person can be identified on the basis of an image classification by means of the 2D or 3D cameras 185, it being possible for example to use the SDKs already mentioned in the previous section. The classification as a person occurs with a higher probability in one aspect if the position in space changes over time.
- the service robot 17 also uses algorithms from the prior art which, via the SDKs of the sensors (such as cameras 185) and / or third-party software, make it possible to create a skeleton model of the person and to track it over time 610, for example by means of the visual Person tracking module 112 and / or the laser-based person tracking module 113. If a person is identified who is not on the chair 615, the service robot 17 prompts this person to sit on the chair, for example acoustically and / or visually takes place 620. The service robot 17 tracks the movement of the person in the direction of the chair 625.
- the service robot 17 If the service robot 17 does not detect any movement in the direction of the chair 630, the service robot 17 changes its position 635 Is gone or the sensors may not have correctly detected the person. If the detection was erroneously assumed by a person, this process is interrupted, alternatively continued (not shown in FIG. 6). The service robot 17 then asks the person again in step 640 to sit down. The service robot 17 tracks the movement 645 again, for example by means of the visual person tracking module 112 and / or the laser-based person tracking module 113. If the service robot 17 does not detect any movement in step 650 in the direction of the chair, the service robot 17 requests the Person again to sit down, but with increased intensity of the request, e.g.
- the service robot 17 transmits an interface 188 (such as WLAN) sends information to a computer which interacts with medical staff via a display 2 and requests the medical staff to move to the service robot 17 and to support this service robot 17 670.
- an interface 188 such as WLAN
- the service robot 17 detects the degree of person detection, ie the service robot 17 uses internal rules to determine the detection quality such as deviations from detection threshold values and, based on this, determines a number of requests that the service robot 17 directs to the person.
- the service robot 17 Since there can be more than one chair in the room on which the person can sit, and there cannot be enough space in front of each chair to cover the necessary distance of 3 m, the service robot 17 has a correction mechanism.
- This correction mechanism provides that the service robot 17 identifies from the set of identified chairs 705 those chairs in front of which there is a free area without obstacles in the orthogonal direction to the front of the chair, which is at least 3.5 m long, e.g. at least 4 m 710. If there is the corresponding free area in front of a chair that is necessary to carry out the test, this property is stored 715 as an attribute in the memory 10 of the service robot 17.
- This information is used or is used when a user navigates to the chair This information ensures that the person is on a chair that is suitable for performing the test with sufficient space in front of the chair.
- the chair can also be identified via a floor mark, which is identified below, for example, a few paragraphs.
- the service robot 17 can point this person to suitable chairs in step 720.
- the service robot 17 can also request the person to stand up again and sit in another chair 725.
- this chair is identified in more detail in step 730.
- the service robot 17 uses the object IDs and the order in which the chairs are positioned.
- the information is available within the memory 10 of the service robot 17 that, for example, the person is sitting on chair no. 6, but only chairs no. 4 and 7 are eligible for performing the test because there is enough space in front of them.
- the service robot 17 can then integrate the information in the request to the person to move that the person can position himself on a chair, for example two places to the left of the service robot 17 or one place to the right of the service robot 17.
- the service robot 17 is able, through the orientation in space that the person and service robot 17 occupies, to correct such information in such a way that the information output relates to the orientation or perspective of the person. In the example mentioned, this would be two places to the right of the person or one place to the left.
- the service robot 17 can also use the coordinates of a standing person and a suitable chair to refer this standing person to this chair, for example in the manner of “please sit down on the chair at an angle to the left in front of you”, if necessary below Mention of a distance.
- color information about the chair for example, can also flow in, which was previously collected, for example, via an RGB camera.
- the service robot 17 positions itself, for example, at a distance of more than one meter from the chair.
- the service robot 17 also has information about the direction from which a patient can approach via its navigation module 101. On the one hand, this information can be explicitly stored in the system.
- the service robot 17 is able to recognize doors or passages.
- the service robot 17 recognizes via the LID AR 1 either a plane which adjoins one of the edges in the XY direction with a length of approx. 77-97 cm 815 and an angle to the edge of 1-178 ° 820, an angle that 825 is variable over time, and / or there is a distance of at least 90 cm behind the identified opening, no further limitation of the XY plane in the map 830, which the service robot 17 records, for example via the LID AR 1.
- the service robot 17 uses its navigation module 101 for the case that the chair is unoccupied, based on the determined position of the door or the passage and the chair, the direct path between this chair and the door or the passage, which is not through any obstacle is blocked by, for example, determining the Euclidean distance.
- the service robot 17 positions itself outside of this path, for example spatially oriented in such a way that its sensors notify the chair and / or the door / the passage.
- the service robot 17 If the service robot 17 identifies a person who enters the room, the service robot 17 prompts them, as described above, to sit on the chair.
- the service robot 17 signals via an output unit, for example a loudspeaker 192, alternatively and / or additionally also via a display 2, the person to get up and to walk straight ahead three meters then return to the chair.
- an output unit for example a loudspeaker 192, alternatively and / or additionally also via a display 2, the person to get up and to walk straight ahead three meters then return to the chair.
- the service robot 17 is able to recognize a distance marking on the floor by means of a 2D or 3D camera 185, for which purpose common pattern recognition methods are used.
- the service robot 17 uses the position information of the identified chair. To ensure that it is a distance marking and, for example, not a normal pattern on the floor, the service robot 17 first uses its navigation module 101 to determine a position in space that is approximately orthogonal to the front of the chair at a distance of 3 m. Then the area on the Floor, which is located approximately in this position, scanned in order to recognize such a marking. More areas of the floor are scanned to see if the pattern is unique or repetitive. If it is unique or possibly stored in the memory 10 of the service robot 17, this pattern is used as a marking for the 3m point.
- a disadvantage of a fixed floor marking is that a chair can possibly move, be it due to cleaning work in the room, by people sitting on this chair, etc.
- the service robot 17 is in a complementary and / or alternatively equipped with a projection device 920 to project a marking at a distance of 3m in an orthogonal direction from the front of the chair.
- the XY coordinates of both the chair and the 3m point are in turn determined, for example, via the navigation module 101, this navigation module 101 having previously been updated by the service robot 17 with a view to the position of the chair.
- a light source is used for this, e.g. a laser or an LED, which in turn can be focused by lenses or functionally similar means.
- This projection device 920 is able to project a surface onto the floor as a bar, possibly with characters, which signal the person that he or she should move there.
- the projection device 920 is movably supported independently of the service robot 17, so that the service robot 17 is always positioned frontally in the direction of the person 910, for example through its own rotary movements (as in FIG Line 940 outlined), while the projected marking 915 is always at the same point orthogonal to the front of the chair 905.
- the light source of the projection device 920 itself can be movably mounted; in another aspect, mirrors, for example micromirrors or microstructured elements, ensure that the light is directed in such a way that this light appears when the service robot 17 moves, for example when it rotates remains in the same place. This changes the angles between the times a) and b) during which the person 910 moves in the direction of the marking 915, including the angle between the lines 925 and 935 and between 930 and 935. In an alternative and / or supplementary aspect, the service robot 17 also always move parallel to the direction of travel of the person.
- the light source is able to project onto a surface on the floor whose width from the perspective of the service robot 17 is over 3 m.
- This projection device 920 thus covers the route that the Person, starting from the chair, should lie back.
- the central axis of the projection direction 920 is rotated by an angle between 10 ° and 60 °, for example 20-50 °, from the central axis of the camera 185 with a view of the Z axis of rotation of the service robot 17, namely in the direction in which, seen from the service robot 17, the person should move.
- the projection mark (such as the bar) is located at the 3m point in, from the service robot 17 as seen from the right edge of the projected area.
- the projection mark moves towards the left edge of the projected area.
- a projection device 920 can be used, as is found in conventional (LCD) projectors, in which a matrix is controlled by software in such a way that different areas of the projected surface are illuminated with different levels of brightness.
- the orientations are correspondingly mirror-inverted.
- FIG. 9 c) the person 915 is sitting on the chair 905.
- the projection device 920 can illuminate an area which results from the dotted rectangle.
- the 3m mark 915 is located in its right area.
- the projection surface moves clockwise together with the service robot 17, while the 3m mark is kept on the fixed XY coordinate, which means the 3m mark migrates to the left area of the projected area (Fig. 9 d)).
- This representation described here is based on a fixed projection direction 920.
- the projection device 920 is movably mounted (effect not shown in more detail).
- the service robot 17 does not rotate and, by means of the LIDAR 1 and / or the 2D or 3D camera 185, detects more than the entire distance that the person has to cover (see FIG. 9 e), where the Sensors at least covered area is dotted).
- the 2D or 3D camera 185 is mounted so that it can be tracked, while the projection device 920 or the light source is mounted rigidly or also so that it can be tracked (not shown separately).
- a processor in the service robot 17 calculates the projection area based on the coordinates of the navigation module 101 of the service robot 17 that previously determined the position of the chair, the 3m point and the service robot 17, the inclination of the Projection device 920 and its height in order to project a largely undistorted bar from the perspective of the person completing the exercise, which bar is approximately parallel to the front of the chair.
- the shape of the marking can also be designed differently.
- the person is tracked by a procedure as has been adequately described in the prior art, for example by means of the visual person tracking module 112 and / or the laser-based person tracking module 113.
- the service robot 17 also detects the posture of the upper body to tell when the person is starting to straighten up. From this point on, the time that the person needs to complete the test is also recorded. The timekeeping ends when the person has turned and sat back in the chair after returning to the chair.
- a rotational movement is recognized by the algorithms, for example as a pattern, in that a skeleton model with articulation points of the person is generated so that the arms are approximately parallel to the plane during the rotational movement that corresponds to the distance to be covered.
- articulation points of the arms are evaluated over time and an angle change of more than 160 ° from symmetrically existing articulation points to the line that connects the starting and turning positions is determined.
- the service robot 17 is configured in such a way that the service robot 17 determines the distance that the patient is traveling while it is covering the distance. Since the starting and turning positions are 3m apart, the length of the distance to be covered is 6m, starting and ending with the chair at the starting position, which is also the end position, with the turning position at a distance of 3m from the chair. The service robot 17 does not necessarily have to detect the marking on the floor. The distance covered can be determined in various ways, including by adding up the stride length.
- the basis for this can be the distance between the ankles or ankles, which are recognized by a 2D or 3D camera 185 in connection with the evaluation frameworks used here and to which points in three-dimensional space are assigned, the distances of which are determined by the service robot 17, for example in Let the shape of vectors be determined.
- the distance covered by the patient can be determined by adding up Euclidean distances between coordinate points which the patient passes and which can be found on a map of the Let determine the environment in which the patient and service robot 17 are located, wherein the determination of the patient's coordinates can be determined on the basis of reference positions. This includes distances to recognized room boundaries or the position of the service robot 17, which can be determined by means of self-localization (self-localization module 105).
- the service robot 17 calculates the distance covered and relates this distance covered to the total distance that the patient has to cover .
- the service robot 17 can use an output device such as the display 2 and / or the speech synthesis unit 133 to give the patient feedback about how far the patient still has to go, how many steps there are still, when the patient can turn around, etc.
- the service robot 17 transmits a score based on reference data of the set of rules 150 stored in the memory 10.
- the service robot 17 is able to record the score and / or the recorded time via an interface 188 (such as e.g.
- WLAN to the patient administration module 160 in the cloud 18.
- the service robot 17 is able to use its sensor 3 to detect the movements of the person in step 1005, to record these movements as video in step 1010, to store them in step 1015 and in step 1030 via an interface 188 (such as WLAN) to a cloud storage in the cloud 18, which is located in the set of rules 150.
- the data transmission is encrypted.
- the facial features of the person to be assessed are made unrecognizable beforehand, so that the anonymity of the persons is preserved 1025.
- the video material is available within a set of rules 150 for labeling purposes in order to further improve the reference data of the set of rules 150 by means of self-learning algorithms. For these purposes, among other things, access to the stored data is possible via a terminal 1030 so that medical personnel can assess the video recordings and label them 1035.
- labels can use the Points in time are assigned. For this purpose, for example, individual start or end points of movements are marked in time and at the same time the movements such as body poses that describe an orientation of the limbs, for example over time, are classified / classified.
- the data labeled in this way are then stored in the database in which the inventory data are also located 1040.
- the set of rules 150 can then, for example, independently improve the classification rules by means of algorithms such as, for example, neural networks.
- the system comprises a computer 9, a memory 10 and at least one sensor for contactless detection of the movement of a person
- the system has a chair detection module 4540 in its memory 10, an output device such as a loudspeaker 192 and / or a display 2 for transmitting instructions, a time-distance module for determining the time taken to cover the distance 4510 and / or speed Distance module 4515 for determining the speed of the recorded person on a distance and a duration-distance evaluation module 4520 for evaluating the speed of the person on a distance and / or the time for covering the distance.
- the system can have a hearing test unit 4525 for carrying out a test of hearing skills, a vision test unit 4530 and / or a test for intellectual skills unit 4535.
- the system can be a service robot 17.
- the system has a projection device (920), for example in order to project the marking that represents the turning point and / or the starting point.
- the system has a person identification module 110, Personal identification module 111, tracking module (112, 113),
- Movement evaluation module 120 skeleton creation module 5635 and / or skeleton model-based feature extraction module 5640.
- the service robot 17 is also configured in such a way that the service robot 17 can carry out the mini-mental test.
- the aim of the mini-mental test is to identify cognitive impairments such as dementia.
- the communication devices of the service robot 17 (voice input and output, display 2) ask questions to the patient who can answer them via the communication device of the service robot 17 (e.g. as voice input, as a response to be selected on a screen, as Freehand input, e.g. for a date, place of residence, etc.).
- the display 2 of the service robot 17 on the other hand a separate display 2 connected to the service robot 17 via an interface 188 (such as WLAN), such as a tablet computer, can be used, which the person can hold in their hands or put it on a table, which makes answering and completing the exercises easier.
- an interface 188 such as WLAN
- a tablet computer such as a tablet computer
- the service robot 17 is configured in such a way that the service robot 17 can communicate with a person, as the method described in FIG. 11 shows. To this end, the service robot 17 orients itself in one aspect in space such that the display 2 of the service robot 17 is approximately parallel to the axis that runs through the two shoulders, the hips and / or knees of the user and that is drawn over the skeleton model, which are recognized by means of the 2D or 3D camera 185 and their SDKs. The service robot 17 thus aligns itself with the user from 1105.
- At least one speech sequence stored in the memory 10 is reproduced via a loudspeaker 192 and a user is prompted via a display 2 and / or via a voice output that Repeatedly reproduced sequence 1110.
- the service robot 17 records acoustic signals from the user via a microphone 193 to 1115, for example as long as the output speech sequence to be repeated was 1120. This happens, as does the following steps , through the speech evaluation module 132.
- the service robot 17 analyzes the amplitudes of the signal in the time domain 1125. If the amplitude falls to zero or almost zero (e.g. ⁇ 90% of the maximum of the amplitudes) for more than 1 second, e.g.
- the recording is ended 1130.
- Sampling continues, the sample width being defined by phases of amplitudes close to zero that are more than 1 second and that have a length of at least 70% of the sequence that the user is supposed to repeat and that is stored in the service robot 17 1135. This ensures that several attempts to speak by the user are recorded and individually evaluated.
- the service robot 17 modifies a value in a database within the memory 10 of the service robot 17 that relates to the user 1150.
- the recorded voice signals emanating from the user are stored 1155 and, together with the modified one
- the value from the database is transmitted 1160 via an interface 188 (such as WLAN) to the patient administration module 160.
- an interface 188 such as WLAN
- only the sequence recorded by the service robot 17 with the highest similarity value to the pattern sequence is stored.
- the system also counts the repeat attempts and, if the number of attempts exceeds a threshold value, stops recording the relevant repeat attempt and moves on to the next sequence to be repeated. Repeated attempts to repeat or failed attempts by the user are also noted in the database.
- the mobile service robot 17 has an optional device which contains sheets that the person to be assessed can take out for the test, for example at the request of the service robot 17.
- the mobile service robot 17 can point the person to be assessed to such a sheet, which is located in the room in which the test takes place.
- the voice output and / or the output unit of the display 2 is configured accordingly 1205.
- the service robot 17 is configured in such a way that by means of the sensor 3 in the form of a 3D camera, for example a time-of-flight camera (ToF), the hands of a user are identified, recorded and tracked, ie the Hands are recognized 1210 in the first step and tracked 1215 in the second step when this person folds a sheet.
- a ToF camera approaches are also possible in which, on the basis of a single 2D camera, hands are recognized 1210 and (hand) movements are tracked 1215 in order to recognize 1220 corresponding gestures or a sheet that is being folded.
- the weights come, for example, from a model that was classified using conventional machine learning methods such as regression methods and / or neural networks, such as, for example, convolutionary neural networks.
- skeleton models based on frameworks can also be created using the 2D camera, for example with Open Pose or PoseNet in connection with Tensorflow.
- the movements are recorded over time, for example by means of the visual person tracking module 112 and / or the laser-based laser-based person tracking module 113.
- the detection of the hands 1210 takes place, which are segmented from the total recording .
- segmentation is used to identify objects 1220 that are in the hands, for example by means of a fault-tolerant segmentation algorithm (for example RANSAC framework) that allows pattern recognition.
- RANSAC framework for example RANSAC framework
- the sheet is determined, for example, by means of a sheet classification, that is to say two- or three-dimensional data from the camera 185 which has previously been created by taking pictures of the sheet and labeling the recordings.
- the term sheet includes paper and materials that have an equivalent effect on the exercise and / or that have similar dimensions and possibly also properties as a sheet of paper.
- the service robot 17 requests the user to take a sheet 1205.
- a voice output via a loudspeaker 192 of the service robot 17 is used, for example.
- a display can also be used on a display 2, or a combination of both ways.
- the service robot 17 similarly requests the user to fold the sheet 1225, for example in the middle.
- the service robot 17 observes the folding process and, after the folding process has been completed, the service robot 17 requests the user to put down the sheet or to let it fall.
- the information on folding and / or laying down or dropping can also be provided directly after a previous request for the same exercise.
- a 3D camera such as a Kinect or Astra Orbbec is used.
- the challenge in recognizing elements of the hand, i.e. fingers, and the finger tracking 1230 derived therefrom is that, from the perspective of the camera 185, individual fingers may be hidden and therefore cannot be directly estimated. This is the case with gestures that do not require an object in hand. If, on the other hand, a sheet is folded by one or more hands, some of the fingers can also be covered, depending on the type of folding process.
- the folding process can be recognized or classified as such on the basis of finger movements 1235, for example, in that either at least one thumb and at least one, better still several fingers of the same hand touch 1240 at the level of the fingertips, ie at least two fingers are grounded, for example and tracked.
- one or more fingers of one hand can touch one or more fingers of the other hand, for example in the area of the fingertips 1245.
- the blade is in engagement 1250 with at least one finger Sheet that is recognized as described in the following paragraph.
- the system and method are alternatively and / or additionally based on the recognition of a sheet and its change in shape (step 1252), this sheet being in contact with or in engagement with at least one finger.
- the recognition focuses on the four corners of the sheet from 1255, which is in one or both hands of the user.
- each corner is tracked individually over the course of time 1260 and the distance between these corners is determined 1265.
- Successful folding is recognized, for example, by a) that the distance between two corners in each case is three-dimensional
- Space is reduced by more than 90%, e.g. reduced by more than 98% 1270.
- the distance between two opposite Edges of the sheet are tracked and a folding process is detected if the distance falls below these values.
- the surface of the sheet is tracked with a view to its curvature 1275.
- the folding module determines the center between two corners 1277 and monitors, for example, the curvature of sheet 1279 in these areas successful folding in this case 1280 is recognized by the fact that the curvature in this area increases 1282 over time, while the sheet edges / margins near the corners behave approximately parallel 1284 (i.e.
- the spacing of the sheet edges is greatly reduced 1285, for example to a distance of less than 2mm, which means that individual detection of the two approximately equal sheet parts is no longer possible as a rule, since the depth resolution of the camera 185 cannot record two sheets lying on top of one another due to the thinness of the sheets .
- the area of the sheet in three-dimensional space is also recorded over time, with a depth of the sheet of less than 2mm being detected only poorly or not at all.
- a folding of the sheet is determined by the fact that the area of the sheet is reduced by more than 40%, for example by approx. 50%, over time. This approach can also be implemented, for example, without explicitly analyzing and tracking the fingers.
- the distance between the ends of a sheet edge parallel to one another is recorded and evaluated 1293 and, if the distance between the sheet ends is less than 20mm, a fold is detected 1294.
- the overall detection accuracy increases. If this exercise has been completed successfully, ie the sheet has been folded and it then moves towards the center of the earth 1295, alternatively comes to rest on a plane 1297, this is noted in a database 1299, in particular in the database in which the test results are stored .
- FIG. 61 shows the system, for example a service robot 17, for recognizing a folding exercise:
- the system comprises a computer 9, a memory 10 and a sensor for contactless detection of the movement of a person such as a 2D and / or 3D camera 185, a LID AR 1, a radar and / or ultrasonic sensor 194 as well as several modules in its memory 10.
- the memory 10 comprises a fingertip Ab Stands module 4750 for detecting the distance between fingertips of at least one hand and a sheet detection module 4705 for detecting a sheet, for example consisting of a sheet segmentation module 4755 for detecting a sheet and / or a module for the Sheet classification 4760.
- the system also has an output device such as a loudspeaker 192 and / or a display 2 for transmitting instructions and an interface 188 to a terminal 13.
- the system has a person recognition module 110, person identification module 111, tracking module ( 112, 113),
- Movement evaluation module 120, skeleton creation module 5635 and / or skeleton model-based feature extraction module 5640 The sequence involves recording, identifying and tracking at least one hand of a person; a detection, identification and tracking of a sheet and a common classification of dimensions, shapes and / or movements of the detected sheet and elements of a hand as a folding process.
- the sheet is identified using a fault-tolerant segmentation algorithm and, for example, sheet classification and / or classification of a folding process based on comparison with two-dimensional or three-dimensional patterns, including shape patterns and / or movement patterns.
- the service robot 17 can also request the user to spontaneously think up a sentence whose evaluation spelling and grammar are not relevant, but which must contain at least one subject and one predicate.
- the service robot 17 requests the person to be assessed via the communication devices (display 2; loudspeaker 192) to think up a spontaneous sentence 1305 and to write it 1320 with his fingers on the touchpad of the service robot 17 Voice output 1315.
- a pen or pen-like object is kept ready 1320 by the service robot 17 for this purpose.
- a pen and a sheet of paper are kept ready that the person can use to write down the sentence 1325 and the service robot 17 prompts the person to do so via the communication device to hold 1330 the written sheet in front of a camera 185 of the service robot 17 in order to record it and store it in the memory 10 of the Service robot 17 to store.
- the sensor system (2D, 3D camera 185) tracks the user movements 1335, for example by means of the visual person tracking module 112 and / or the laser-based person tracking module 113, uses the internal object recognition of a sheet (see previous approaches) and recognizes that the user holds the sheet 1340 in front of the 2D camera of the service robot 17 and the service robot 17 recognizes the sheet 1345, which the service robot 17 photographs 1350 with the 2D camera.
- OCR processing of the sentence contained in the photo takes place 1355.
- the processor of the service robot 17 accesses corresponding, established libraries for image processing, for example word processing, which enable OCR processing.
- a Natural Language Parser 1360 is used to determine the existence of subject and predicate in the sentence.
- the recorded sentence is broken down into individual words in the first step (tokenization) 1365.
- the root form of the words is then formed (stemming and / or lemmatization) 1370.
- the POS tagging (part-of-speech) tagging takes place, via which the words are classified into subject, predicate, object, etc.
- step 1375 In this context, an approach based on neural networks can also be used. Toolkits such as NLTK or SpaCy can be used for this.
- the results are stored in a memory in step 1380 and it is compared in the next step 1385 whether a subject and a predicate occur in the sentence specified by the user. In this case, it is stored in a database that the exercise was successfully completed (step 1390).
- Another element of the test involves drawing two pentagons that intersect.
- the person to be assessed is given the opportunity to produce the drawings on a display 2 which is located on the service robot 17.
- the display 2 is freely movable within the room in which the service robot 17 and user are located, but is wirelessly connected to the service robot 17 via an interface 188 (such as WLAN).
- the drawing itself can be done either with the fingers or with a pen suitable for tablets.
- the drawing can be made on a sheet of paper, the person drawing being requested by the service robot 17 by means of the communication devices to hold the finished drawing in front of a camera 185 of the service robot 17.
- the captured images are compared by the computer with those that are stored in a database, a set of rules 150 being used that compares the features of an image with features of classified images and assigns them based on probabilities.
- a set of rules 150 being used that compares the features of an image with features of classified images and assigns them based on probabilities.
- methods are used that are described in the prior art and that were previously created on the basis of automated training, primarily using methods of neural networks.
- classification mechanisms can also come into play that were created without training and whose classification features were determined on the basis of characteristic features of a pentagon, of overlapping pentagons in the form of defined rules (such as the number of angles and lines). This takes into account, for example, rounded edges, uneven lines and possibly lines that do not form a closed pentagon.
- smoothing approaches can be used in order, for example, to simplify the classification. If a threshold value is reached in the similarity comparison (e.g. correlation) between the comparison patterns recorded by the service robot 17 and stored in the set of rules 150 or the recognition rules for two overlapping pentagons, it is stored in a database that the exercise was successfully completed.
- a threshold value is reached in the similarity comparison (e.g. correlation) between the comparison patterns recorded by the service robot 17 and stored in the set of rules 150 or the recognition rules for two overlapping pentagons, it is stored in a database that the exercise was successfully completed.
- the service robot 17 has a functionality to recognize manipulations by third parties when completing the exercises.
- the sensor system with which the user and his activities are also analyzed, records the presence of other people in room 1405. It is analyzed here whether the person (s) (including users) position themselves spatially during the test in such a way that they can Manipulating the service robot 17, ie whether they are in a so-called critical distance from the service robot 17 1410.
- the manipulations include, in one aspect, entering the data on a display 2 of the service robot.
- the distance of the person from the service robot 17 is determined and then determined in at least one of the three following ways whether the person is positioned sufficiently close to the service robot 17 to be able to make inputs (especially on the display 2) if necessary: a) it will assumed a blanket distance value, e.g. 75cm. If this is not reached, the service robot 17 assumes that the display 2 can be operated (step 1415). Alternatively and / or in addition, the distance between the hand and / or fingers of the person and the service robot 17 can also be detected, the distance from when manipulation is assumed to be shorter than that of the person per se.
- a blanket distance value e.g. 75cm.
- the arm length of the person is determined 1420, in particular by determining the distances between a shoulder joint point and a wrist joint point or the finger joint points. If this is not reached, the service robot 17 assumes that it can be operated 1425.
- An average arm length is deduced 1435 from the size of the person who is determined by the service robot 17 (which is stored in the memory 10, for example). and, as soon as this distance is undershot, it is assumed that it can be operated / manipulated 1425.
- the service robot 17 can calculate the positioning of the person in space relative to the position of the display 2 (step 1440). If, for example, the alignment of the shoulders, hip joints, etc.
- the frontal plane of the person derived therefrom is approximately parallel to the display 2 or at an angle of less than 45 °, and is the person oriented in the direction of the display 2, as is e.g. .from the primary direction of movement of the person, the posture of the arms, head, knees, feet, facial features, etc., this increases the likelihood of an interaction with them.
- this approach can also be implemented for other elements of the service robot 17 instead of a display 2, for example a switch-off button.
- a virtual plane is taken into account which is oriented orthogonally to the axis of symmetry of the operating element 186 towards the center of the service robot 17.
- the sensor system analyzes whether the input or manipulation of the service robot 17 is made 1450 by the user or by a third person.
- the service robot 17 tracks the people in its environment on the basis of characteristic features 1445, as in the prior art Technology is described in general (for example on the basis of the size, dimensions of the limbs, gait features, color and texture of its surface such as clothing, etc.), for example by means of the visual person tracking module 112 and / or the laser-based person tracking module 113.
- the differentiation into users and third parties is made by identification on the service robot 17, it being assumed that the person who is identifying himself is the user. This is done with a view to inputs via the display 2 via the optical sensors of the service robot 17.
- an orientation of the person relative to the person can be determined here Service robot 17 by determining the angle between the frontal plane of the person and the axis standing perpendicular to the operating elements 186 of the service robot 17, each projected into a horizontal plane, and a comparison of the determined angle with a threshold value, with falling below the threshold value a detection of a increased likelihood of manipulation.
- Identification features of the person take place, followed, for example, by the detection and tracking of the person, detection of identification features of the person, comparison of the recorded identification features with the identification features of the person stored during registration and comparison with a threshold value, with similarities being compared here and a threshold value Minimum similarity implies.
- a detection of an increased manipulation probability takes place if the threshold value is undershot and a detection of a lower one
- the determined manipulation probabilities can be multiplied to determine a manipulation score, which, for example, is stored together with the evaluation results during or after evaluations are carried out by the robot with the recorded person. Depending on the type of comparison, other interpretations can also be made, as was presented in the introduction, for example.
- the system for example a service robot, comprises a computer 9, a memory 10 and a sensor for contactless detection of the movement of at least one person, for example a 2D and / or 3D camera 185, a LID AR 1, a radar and / or ultrasonic sensor 194.
- the system includes modules with rules in its memory 10.
- a manipulation attempt detection module 4770 by at least one person who is in the vicinity of the system is detected, a person identification module 111, a person-robot distance determination module 4775 to determine the distance of the at least one person to the service robot 17, a size-arm-length orientation module 4780 to determine the size, arm length and / or orientation of the at least a person, and / or an input registration comparison module 4785 for comparing whether a person identified by the system makes inputs to the system, for example via di e control elements 186.
- the system has, for example, an output device such as a loudspeaker 192, a display 2 for transmitting instructions and / or a Interface 188 to a terminal 13.
- the system has a person recognition module 110, tracking module (112, 113), movement evaluation module 120, skeleton creation module 5635, skeleton model-based feature extraction module 5640 and / or a movement planner 104.
- the verbal communication between the persons is evaluated 1455 (in FIG. 14) using microphones 193.
- speech signals are recorded 1560 in the vicinity of the service robot 17 via at least one integrated microphone 193.
- the speech source is identified in two alternative or supplementary ways, for example also in the speech evaluation module 132 identified 1565 and tracked 1570 and then synchronized in time 1575 with the voice signals recorded by the service robot 17.
- Image recognition and tracking methods from the prior art are used to recognize the speech movements of the lips.
- the service robot 17 can thus assign who the registered language is coming from and whether it corresponds to the user who is supposed to perform the exercise, the language of the user being recorded, for example, when the user identifies himself on the service robot 17.
- the second way which circumvents this problem, consists of the sound analysis of several microphones 193 (step 1480), which are attached to different positions of the service robot 17 and which are recorded over several channels in the course of time and frequency, the processor of the service robot 17 carries out a runtime analysis 1485 and from the time offset when the signals arrive, calculates the person from whom they originate 1490.
- a microphone 193 can also be used, in which case a triangulation is carried out by changing the position of the service robot 17 can.
- the time courses are correlated by means of the calculated time offset via triangulation and the origin in space is determined (which can happen in two or three dimensions). This is then compared with the positioning of the people in the room, which is determined by the 2D or 3D camera (s)
- the service robot 17 can determine in this way which person spoke 1495. If it is the third party The person (and not the user) to whom the speech signals are assigned could have been manipulated 1498. A value can then be adapted in a memory and, in one aspect, a note or an error message can be generated in the user dialog.
- the further person only assists the user with the input, ie does not take any of his own inputs, but only inputs what has been spoken, recorded, etc. into the service robot 17 via the display 2 or the microphones 193.
- the recorded word sequences are analyzed to the extent that they are assigned 1505 to the individual persons via at least one of the routes shown, for example, in the previous sections.
- FIG. 15 illustrates the cornerstones of this procedure.
- the language can be recorded 1510 in the Elmfeld of the service robot 17 and the speakers can be differentiated 1515 on the basis of different language features / language characteristics, including above all the language frequencies (in particular fundamental frequencies), different language intensities and / or different language speeds, in particular within the speech evaluation module 132.
- This method in combination with that from FIG. 14, which use speech signals from people either by lip tracking or location based on speech signal flow behavior, makes it possible to assign speech signals that have been identified to the people without having to repeat the lip movements and / or spatial ones every time To have to determine the position of the speaker and, if necessary, to compare it with the 2D / 3D person tracking results.
- This matching of the people with the language characteristics 1520 allows voice to be recorded and, at the same time, to be tracked in a user-related manner 1525 it is checked whether the same text fragments or speech fragments / speech patterns occur several times in succession 1530 and originate from different people 1535, which is done by tagging the patterns and the speech characteristics assigned to the people such as the fundamental frequencies (alternatively and / or in addition, the ones shown in Fig . 14 are used).
- Text fragments, speech fragments or speech patterns are understood to mean, for example, identical words and / or word sequences. For an evaluation of these sequences with a view to the assistance of the user or the manipulation of the service robot 17, it is relevant from which person a relevant sequence was mentioned for the first time.
- the first step it is checked whether a speech fragment is first from recorded by a person who is not the user before the user repeats that speech segment. For this purpose, correlations are made, especially in the time domain, in order to search for identical words. In particular, it is examined whether more than one single word that occurs one after the other is repeated.
- a lexical analysis using natural language processing is also possible 1545, in which words such as in previous paragraphs are analyzed and the sequence of the tagged words is compared, based on, for example, tokenization, lemmatization and part-of- Speech tagging, e.g. using spaCy or NLTK in Python.
- This procedure also makes it possible to check whether a “suggestion” is not repeated acoustically by the user for the recording by the service robot 17, but rather a corresponding input is made directly by the user in the service robot 17.
- Manipulation probability the detection and tracking of at least one person in the vicinity of the robot by a contactless sensor, the determination of the position of the person in the vicinity of the robot, the recording and evaluation of audio signals, the position determination of the source of the audio signals, a comparison of the determined position of the person and the position of the source of the audio signals and comparing the position difference with a threshold value and determining a probability of manipulation of the Robot based on the comparison of the position difference with the threshold value.
- the position of the source of the audio signals can be determined by detecting the direction of the audio signals using at least one microphone and triangulating the determined directions, for example also by changing the position of the service robot 17 or using a second microphone.
- the determination of the position of the source of the audio signals includes the detection of the direction of the audio signal by a microphone, the determination of the position of at least one person by the contactless sensor, the triangulation of the direction of the audio signal and the determined position of the person. Furthermore, there is, for example, the evaluation of the person's face, the recording of the lip movements over time, a temporal comparison of the recorded audio signals (e.g. by means of correlation evaluations) with the recorded lip movements in relation to a threshold value and, if the threshold value is exceeded, the recorded audio signals are assigned to recorded person.
- the method can also include the registration of the person on the robot (as a user) and the acquisition and storage of identification features of the person (as a user), with identification features including frequency, intensity and / or spectrum of the audio signals from the person, for example also a Detection and tracking of the person, a detection of identification features of the person, a comparison of the recorded identification features with the identification features of the person stored as part of the registration of the person on the robot and a comparison with a threshold value (i.e. showing a minimum similarity), the registration of inputs from the Person on the operating elements (186) and an assignment as to whether a registered person (a user) makes entries on the operating elements (186).
- identification features including frequency, intensity and / or spectrum of the audio signals from the person, for example also a Detection and tracking of the person, a detection of identification features of the person, a comparison of the recorded identification features with the identification features of the person stored as part of the registration of the person on the robot and a comparison with a threshold value (i.e
- an increased probability of manipulation of the robot can be determined if an unregistered person makes inputs on the operating elements (186) of the robot.
- the method can further include, for example: a determination of words and / or word sequences in the recorded audio signals or audio sequences, an assignment of the determined words and / or word sequences to recorded persons, a determination of an increased probability of manipulation of the robot, if a comparison of the determined word sequences results in a word and / or word sequence difference that is above a threshold value, ie that a minimum correlation is not achieved.
- the method can, for example, determine words or word sequences that are entered by the person via an operating element (186), determine words and / or word sequences in the recorded audio signals, assign the determined words and / or word sequences from the captured audio signals to captured people who Detection of identification features of the person, the determination of an increased likelihood of manipulation of the robot, if a comparison of the word sequences entered via the operating elements (186) with word sequences determined from the recorded audio signals a word and / or word sequence minimum similarity and at the same time a minimum similarity of the recorded identification features of the person with the identification features recorded and stored during registration.
- FIG. 58 shows the architectural view for the system for manipulation detection based on audio signals.
- This includes a computer 9, a memory 10 and a sensor for contactless detection of the movement of a person who is detected in the vicinity of the system, at least one Microphones 193, a person position determination module for determining the position of a person in the room 4415, an audio source position determination module for determining the spatial origin of an audio signal 4420, a module for correlating two audio signals 4025, an audio signal person Module 4430 for assigning audio signals to a person, and / or a speech evaluation module 132.
- the system also has an audio sequence input module 4435 for comparing an audio sequence (i.e. a sequence of tones which, for example, reproduces words) with a sequence of tactilely input letters.
- an output device such as a loudspeaker 192 and / or a display 2 for transmitting instructions.
- a connection to a terminal can be established via an interface 188 (e.g. WLAN).
- the sensor for contactless detection of the movement of a person is a 2D and / or 3D camera 185, a LID AR 1, a radar and / or ultrasonic sensor 194.
- the system has a person recognition module 110, a person identification module 111, tracking module (112, 113), movement evaluation module 120, skeleton creation module 5635 and / or skeleton model-based feature extraction module 5640.
- the service robot 17 is configured in one aspect in such a way that the service robot 17 carries out a short hearing test, in addition or as an alternative to this, a short eye test with the user before the start of the exercises.
- the method steps taking place here are essentially shown in FIG.
- the service robot 17 initially informs the user, optionally via a screen output and / or an acoustic output, of the problem that understanding problems could possibly arise and therefore a calibration of the service robot 17 to the user is necessary.
- the vision and / or hearing test represents such a calibration.
- the service robot 17 requests the user to take part in the calibration 1605.
- the service robot 17 requests the user to press appropriate fields in the menu on the display 2 if the user has heard certain tones.
- a voice input for the user is also possible, which in turn is evaluated by means of natural language processing, as described in the prior art, for example within the language evaluation module 132.
- the service robot 17 then plays a sequence of tones different frequency and different volume, but individually essentially have a constant frequency and volume 1610, and each time "asks" whether the user heard the sound, which can happen, for example, because the service robot 17 has a display 2 Input options presented to the user, via which the user can specify the extent to which he has heard the tone 1615.
- the tones become quieter and more frequent 1620 in one aspect over time. However, a different sequence is also possible in this regard. conceivable.
- the user's responses are recorded.
- a score is then determined 1625 which indicates the extent to which the user has heard the tones. If, for example, the user does not reach certain threshold values in the hearing behavior, ie the service robot 17 does not reach predefined limit values in the case of positive answers via its either screen menu-guided or voice menu-guided and correspondingly evaluated answers (e.g. only three out of seven tones recognized) a corresponding score value can be determined.
- this score is stored 1630 in a database in the service robot 17, for example together with user information that characterizes the medical condition of the person.
- the service robot 17 can also use the volume of the responses given by the user, for example relative to the ambient noise level 1635, which is recorded via at least one additional microphone 193, to determine whether the user needs a higher volume of the Service robot 17 has output signals.
- the volume of the output of acoustic signals of the service robot 17 is adjusted accordingly, for example increased if it is found in at least one of these described ways that the user is hard of hearing.
- the service robot 17 prompts the user to press appropriate fields in the menu on the display 2 if the user can recognize certain letters or other symbols 1650.
- voice input is also possible for the user possible, which in turn is carried out by means of natural language processing, as described in the prior art.
- the service robot 17 then plays a sequence of characters or images on the display 2 from 1655.
- the user signals whether the user has recognized the character or not or which character the user has recognized.
- the characters or images become smaller in one aspect over time (step 1665).
- different color patterns are also possible in order to identify any color blindness of the user.
- the user's responses are recorded.
- the results of the test are mapped 1670 in the form of a score. If, for example, the user does not reach certain threshold values for visual acuity or color blindness is determined, ie a certain number of objects / patterns is not recognized (such as three out of seven), If this influences the score, one aspect is stored 1675 in a database in the service robot 17.
- the size of the letters is adjusted accordingly when text elements are output on the display 2 of the service robot 17, possibly also the Menu design in order to be able to display corresponding menu items with larger letters 1680.
- the coloring of the display 2 can also be adapted in a supplementary aspect in order to enable better recognition of the display menu in the event of color blindness.
- the service robot 17 can vary the distance to the user, for example, for users with visual difficulties, to move closer to the user 1695.
- a parameter value is temporarily modified in the navigation module 101, which is the usual distance between user and service robot 17 is defined 1690.
- the contrasts and / or the brightness of the display 2 of the service robot 17 can also be adapted to the environmental conditions, taking into account the visual capabilities of the user 1685. Improvement of the signal processing quality through adaptation to environmental influences
- the service robot 17 is able in a further aspect to adapt the input and output units to the environment in such a way that operation is possible with different degrees of brightness and / or background noise is.
- the service robot 17 has a commercially available brightness sensor in the vicinity of the display in order to determine how much light falls on the display 2.
- the brightness value of the display 2 is adapted to the environment, i.e. especially when there is intense incidence of light, the brightness of the display 2 is increased and the brightness of the display 2 is reduced when the brightness values are low.
- the service robot 17 is able to use one or more microphones 193 to determine the background noise.
- At least one further microphone 193 records the background noise and uses noise cancellation (ie noise suppression) methods (phase shifts of the input signal around the recorded background noise) in order to improve the signal quality of the acoustic input signal and thus improve speech processing enable, in order to avoid e.g. detection errors, the renewed asking of a question or request by the service robot 17, etc.
- noise cancellation ie noise suppression
- the service robot 17 also inquires as a measure for
- the interaction between the service robot 17 and the person to be evaluated takes place via the communication device already described elsewhere. Such information is stored in the user's database record.
- the service robot 17 also receives as a further measure for
- Accuracy improvement of the test results from the patient administration module 160 the information when the patient was admitted to the clinic in which the test is taking place, and calculates the length of the previous stay in order to take into account the decline in cognitive performance due to inpatient clinic purchases. At the same time, it is recorded in the patient administration module 160 whether the patient has already made a diagnosis for a Disease received. This information is also taken into account when displaying the result of the mini-mental test and is stored in the user's database record.
- the service robot 17 transmits the results of the stored test tasks described above via an interface 188 (such as WLAN) to the patient administration module 160 and in this way makes them available to the medical staff, with the results being documented at the same time .
- an interface 188 such as WLAN
- the service robot 17 is configured in such a way that the service robot 17 is able to determine whether a patient has certain excretions over the skin, which in one aspect indicate certain diseases that can be diagnosed therewith. For example, the service robot 17 can determine whether and, if applicable, how much a patient is sweating in bed.
- a spectrometer 196 such as a near-infrared spectrometer can be used here, for example, and in another aspect also a Raman spectrometer.
- the processes for measuring precipitates 2100 are shown in FIG. Measurements at different points on the body are possible. The procedure for three locations is described as an example: on the hands, on the forehead, and a measurement on the torso, especially the bedding. Detection of sweat at these points is, for example, part of the Delirium Detection Score, another test to detect delirium in patients.
- the service robot 17 is configured in such a way that the service robot 17 can record a patient in a bed by means of a 3D sensor, for example a 3D camera.
- this sensor is positioned on the service robot 17, for example, in such a way that the service robot 17 is at a height of at least 80 cm, for example at least 1.2 m, and is, for example, rotatable and / or tiltable.
- the service robot 17 is able to identify 2105 beds based on object recognition.
- the service robot 17 can, in one aspect, scan the room via the 2D or 3D sensor, for example the LID AR 1 which is known a priori that I have beds in this room.
- dimensions can also be determined using a map stored in memory 10, which contains, for example, spatial information such as the width and depth of a room. In this way, the spatial dimensions are evaluated 2110.
- the service robot 17 can also determine the dimensions determine measured objects 2115, for example by triangulation in connection with an implemented odometry unit 181 (step 2120), which can determine the positional deviations of the service robot 17.
- the dimensions of the measured objects in space relative to the space information are determined 2122, for which no recourse to the odometry function is necessary.
- the determined dimensions, especially the outer dimensions of the bed are classified on the basis of rules stored in the memory 10 in order to determine whether this is a bed 2124. In one aspect, this includes the dimensions of a bed can take.
- the objects recognized by the LID AR 1 and / or the 2D and / or 3D camera 185 can also be classified on the basis of characteristic features that uniquely identify 2125 a bed Act configuration of the wheels of the bed and / or the lifting device for adjusting the height of the bed.
- classification rules can also be used that have been created by learning typical bed features on the basis of machine learning methods and / or neural networks.
- the beds can also be equipped with sensors and / or barcodes 2130, for example RFID or Bluetooth transmitters, which allow bed identification.
- the positioning of the sensors on the bed can be used to determine the orientation of the bed in the room 2140, for example by using backscatter signals, which are reflected differently on the bed frame and are based on the transit time and / or phase differences determine the orientation of the bed in the room.
- Barcodes can also be attached to the service robot 17 in such a way that reading them out allows the spatial orientation of the bed to be determined.
- the codes stored in the sensor system and / or the barcodes are read out by the service robot 17 and compared with those that are stored in the memory 10 of the service robot 17 and are assigned to the beds, whereby the service robot 17 can determine that the sensor read out and / or barcode is assigned to a bed.
- the bed can be oriented in the room using the bed dimensions, in one aspect also by comparing the position to the next wall 2135: That is to say, the service robot 17 determines the orientation of the bed, in particular the head end, on the basis of a priori information, the a priori information aiming at the fact that a bed has an essentially rectangular shape and its shorter sides represent either the head or foot end. The shorter side is recognized as the head end, which is, for example, closer to a wall of the room.
- the service robot 17 identifies a person in bed, in particular their head and arms, which can be evaluated, for example, in the context of a skeleton model.
- the service robot 17 determines where the service robot 17 can navigate relatively close to the patient's head. For this purpose, the service robot 17 next determines on one side of the bed how far the service robot 17 can travel to the head end. If the distance to the wall at the head end on one side of the bed is less than 1 m, the service robot 17 drives along this side of the bed. If the distance is more than 1 m, the service robot 17 on the other side of the bed determines the distance to the wall and then moves as far forward as possible to the wall at the head end, namely on the side where the service robot 17 is as far as possible to the wall can drive at the head end. In an alternative aspect, the service robot 17 first checks both sides for their depth, as described, and then moves in the direction of the head end on the side on which the service robot 17 can reach the farthest in the direction of the wall at the head end.
- the service robot 17 next determines a candidate region of the head 2150. For this purpose, the service robot 17 positions itself in such a way that its front points in the direction of the presumed position of the head. This can be done, for example, by rotating the service robot 17 on the spot, the service robot 17 having an angle of rotation between 25 and 90 ° measured to the long side of the bed. By means of a 2D or 3D sensor, the service robot 17 detects the surface of the bed, in particular in the area towards the head end.
- the service robot 17 calculates a candidate region in which the head is usually located and which lies in an area that is at least 25 cm from each long side of the bed and at least 10 cm from the head end of the bed, up to a distance of 60 cm from the head end .
- intervals can also be stored for the width.
- the bed stands lengthways to the wall and the service robot 17 moves along the long side of the bed on which there is sufficient space.
- the service robot 17 uses the already described determination of the candidate region for the Head, alternatively the service robot 17 scans the entire bed by means of the camera 185, the images of which are evaluated using a framework available on the market that has implemented head detection.
- the service robot 17 is able to determine the forehead 2152 on the basis of features of the head.
- an area is defined for this which is limited by the following facial features: approx. 4 cm above the line connecting the center of the eye , Hairline on the sides, recognizable by a color contrasting with the patient's skin.
- the shape of the head can also be used here, the end face limitation being restricted by the rounding of the head.
- approaches such as histogram-of-gradients can be used, which are implemented in frameworks such as OpenCV or Scikit-image.
- the angle can be used as a limitation, one leg of which consists of a light beam from the sensor of the head and the perpendicular at the point at which the light beam hits the surface of the head.
- the service robot 17 can, in one aspect, change the side of the bed in order to ensure that the patient has not turned the back of the head towards him.
- the service robot 17 can use its output units such as display 2 and / or the speech synthesis unit 133 to request the patient to move his head 2154, for example to look at it. After such a request, an attempt to identify the head or forehead is made again.
- the service robot 17 uses further classification algorithms to classify the hands of a
- Patient i.e. there are two candidate regions for the hand 2157 roughly in the middle of the long side of the bed and with a depth of approx. 30 cm parallel to the short
- RGB or RGB-D camera 185 can be used to create a (partial) skeleton model of the
- the service robot 17 can, via its output units such as display 2 and / or speech synthesis unit 133, request the patient to move his hands or arms 2159, for example to put them out from under the blanket. After such a request there is another attempt to identify the arms or hands. Similar to the forehead, the service robot 17 also identifies the surfaces of the hands, either the back of the hand and / or the palm. For improved localization, joint points from the skeleton model can alternatively and / or additionally also be included, with the hand area of interest lying between the wrist and the finger joints. Alternatively and / or in addition, the palm recognition can take place via an image classification, the classification algorithms being carried out by training using images that represent the palms of the hands.
- Another body region of interest is the patient's upper torso, which is pre-defined over a candidate region as a head-down area that extends for a length of approximately 1.5 head-heights beginning half a head-height below the head and which has a width of approx. 2 head widths.
- the area is defined over a width of approx. 45 cm and a height of approx. 50 cm, which begins approx. 10 cm below the patient's head, alternatively is positioned roughly in the middle of the bed at a distance of approx. 50 cm from the patient Head end.
- a classification using the three-dimensional shape can also take place.
- the width of the bed is scanned with a view of the height and, in the area of the axis parallel to the long side, the area that is located in half of the bed that is oriented towards the head end.
- the part that is below the candidate region for the head is selected along the ridge line that results in this zone.
- a candidate region for the upper body 2160 can be determined.
- the elevation is then scanned in relation to the level of the mattress, which is detected by the 3-D sensor system of the service robot 17 and, if an elevation is detected in the candidate region, this area is identified 2162 as the upper body.
- the service robot 7 is thus able to detect and identify three target regions of the patient: the forehead, the surface of the hand / the back of the hand and the upper part of the torso. These can be identified in space by means of the sensor system, for example by means of the RGB-D camera 185, so that their surface can be represented accordingly in a three-dimensional coordinate system.
- the images created by the sensor system are segmented in order to determine body regions by means of a classification, so that the spectrometer (196) can be aimed at these body regions.
- a corresponding classification can be stored in memory 10.
- the service robot 17 can, for example, also have stored regions in the application for controlling the spectrometer on which the measurement is to be carried out.
- the service robot 17 tracks the movements of the hand or the head (optionally also of the upper body) in step 2170 over a defined period of time. If no movement is detected for a period of time that is above a defined threshold value (e.g. 5 seconds), or only a movement is detected in which the hand / head does not exceed a defined threshold value (e.g. 5 mm) 2180, a measurement and evaluation of the recorded data takes place 2185.
- a defined threshold value e.g. 5 seconds
- a defined threshold value e.g. 5 mm
- the head of the patient or the hand on which the measurement is being made is continuously tracked by the RGB-D camera 185 as part of a safety check 2178. If movements are detected, for example a turning movement of the head, a lowering or lifting of the head, which exceeds a defined threshold value, the measurement is immediately interrupted.
- the service robot 17 continues to track the regions on which a measurement is to take place and starts a new measurement attempt if the movements of the head are less than a defined threshold value.
- the service robot 17 also has, for example, a near-infrared spectrometer for a substance analysis 2186, which is rotatably and pivotably mounted and in this regard. is electronically adjustable.
- the service robot 17 is able to align the spectrometer 196 via this bearing so that the path of the radiation emitted by the spectrometer 196 reaches the coordinates of the target region in three-dimensional space and the reflected radiation is also detected again by the spectrometer 196
- Infrared diode with focusing optics come into question, an infrared laser is used in one aspect.
- the measurement takes place, ie the signals of the spectrometer 196 are evaluated and classified 2189 using a reference database which contains reference spectra and thus allows to determine what is in or on the target region 2188, namely qualitatively or quantitatively 2190.
- classification rules for determining the substances from the measured spectra can also be stored directly, based on correlation analyzes, for example work.
- characteristic signals, ie above all spectral curves of sweat can be determined 2191, which are composed of individual spectra of water, sodium and / or chloride and occur, for example, on the patient's skin such as the forehead or the hand. With a view to the target region trunk, it is recorded to what extent the bedspread of the patient is damp, ie the classification used here for signal evaluation takes into account the material of the bedding accordingly.
- the amount of water excreted as sweat can also be estimated by means of classification using the reference database.
- the database with reference spectra has those that can determine the concentration of other substances, including various drugs 2192 such as heroin, opiates (such as morphine), amphetamine, methamphetamine,
- Cocaine including benzoylecgonine, if applicable
- 9-tetrahydrocannabinol THC
- other substances 2193 such as glucose, lactate, uric acid, urea, creatinine, cortisol, etc.
- the service robot 17 has a further reference database which, on the basis of the combination of different substances and / or their concentration (s), allows a classification of measured values determined to that effect, which different diseases are possibly associated with the measured spectra 2194.
- Both threshold values are here of concentrations or the measured amount of substance, the relationship between the amounts of substance and / or concentrations as well as combinations thereof are part of this classification.
- An example is the combination of urea, uric acid and creatinine, in which the concentration of uric acid is greater than 0.02 mmol / l, creatinine 0.04 mmol / l (higher at lower temperatures), and urea> 15 mmol / l (at low temperatures) or> 100 mmol / 1 (at higher temperatures).
- the service robot 17 takes into account the ambient temperature by means of a thermometer located in the service robot 17, the season or the outside temperature and, in the latter case, is equipped with an interface 188 (such as WLAN) to display the outside temperature for its own via the cloud 18 To determine the location, ie the service robot 17 is able to acquire further data for improved evaluation, either via further sensors 2195 and / or via interfaces 188 (such as WLAN) to further databases 2196.
- a thermometer located in the service robot 17
- the season or the outside temperature is equipped with an interface 188 (such as WLAN) to display the outside temperature for its own via the cloud 18
- an interface 188 such as WLAN
- the measurement results are stored 2197 in a database that is located in the service robot 17 and / or they can be transmitted via an interface 188 (such as WLAN) to a server in the cloud 18 and stored there 2198 a display 2 and / or a voice output can be output 2199, for example via the service robot 17 and / or a terminal to which medical personnel have access for evaluation purposes.
- an interface 188 such as WLAN
- a display 2 and / or a voice output can be output 2199, for example via the service robot 17 and / or a terminal to which medical personnel have access for evaluation purposes.
- the spectrometry system (for example the service robot 17) is shown in FIG.
- the spectrometry system comprises a computer 9, a memory 10 and a sensor for contactless detection of a person (e.g. a 2D and / or 3D camera 185, a LID AR 1, a radar and / or ultrasonic sensor 194), a spectrometer 196 and a spectrometer alignment unit 4805 for aligning the spectrometer 196 with a body region of a person, similar to a tilting unit.
- the system can have a thermometer 4850 for measuring the ambient temperature and / or an interface 188 to a terminal 13.
- a body region detection module 4810 for detecting body regions
- a body region tracking module 4815 for tracking body regions before and / or during a spectroscopic measurement on this body region
- a spectrometer measuring module 4820 for monitoring, incl.
- a visual person tracking module 112 and / or a laser-based person tracking module 113 The system accesses a reference spectrum database 4825 and / or a clinical picture database 4830 with stored clinical pictures and associated spectra for comparing the measured spectra and determining the measured substances that are located in the cloud 18 and / or in the memory 10.
- a perspiration module 4835 for determining the amount of exuded moisture
- a delirium detection score determination module 4840 for determining a delirium detection score and / or a cognitive skills assessment -Module 4845 for determining cognitive abilities.
- the system has a person recognition module 110, person identification module 111, tracking module (112,
- the service robot 17 is configured (see FIG. 22) in such a way that the service robot 17 outputs a pulsed sequence of different tone signals 2205 (for example a tone sequence) via a loudspeaker 192, for example with a pulse frequency of 0 , 3-3 Hz, e.g. about one Hertz.
- the service robot 17 can detect signals from at least one tactile sensor 4905 (step 2210) and synchronize 2220 with the output signals.
- a value in the memory 10 can also be assigned to each sound signal.
- There is a time delay 2215 between the output of the audio signals and the detection of the signals from the tactile sensor 4905 i.e. for example a phase shift of a maximum of half a pulse length, which lags behind the pulsed signal.
- the signals of the at least one tactile sensor 4905 registered with the possible phase shift are evaluated as to whether they occur 2225 in a defined tone spectrum stored in memory 10, ie a comparison takes place as to whether the detected signals occur after a defined tone sequence. If this is the case, a counter in the memory 10 is increased by an integer value 2230; alternatively, there is no increase 2235. The determined counter value is then classified in such a way that the determined counter values are assigned a diagnosis associated with counter values 2240 The output of the tones is directed to a patient in order, for example, to check his or her cognitive abilities. The higher the value that is incremented, the less the patient's cognitive abilities are impaired.
- the diagnosis is stored in the memory 10 of the service robot 17 2245, optionally transferred to a cloud-based cloud storage in the cloud 18 and optionally made available to medical staff via a terminal.
- the tactile sensor 4905 is a piezoelectric, piezoresistive, capacitive or resistive sensor. However, other types of sensors can also be used, as described in Kappassov et al. 2015 (DOI: 10.1016 / j.robot.2015.07.015).
- the tactile sensor 4905 is located on an actuator 4920 of the service robot 17, which has at least one joint and can be positioned such that the service robot 17 reaches the patient's hand, ie the tactile sensor 4905 is at a distance from Hand positioned that is below a threshold value that is stored, for example, in a memory.
- the sensor is integrated into a robotic hand.
- the sensor is attached to the surface of the service robot 17. For this purpose, the service robot 17 can identify the patient with at least one camera 185, tack it, and determine the position of his hands, for example at least those of the right hand.
- the service robot 17 outputs a sequence of letters via a loudspeaker 192 which corresponds to a word. Each letter is output every second.
- the patient is ordered by the service robot 17 to perform a pressure movement with his hand when certain letters are recognized. These pressure movements are evaluated by the tactile sensor 4905 described and it is counted how often the specific letters were recognized. The higher the detection rate, the less the patient is affected.
- the attention analysis system is summarized as follows:
- the system for example a service robot 17, comprises a computer 9, a memory 10, an output unit for acoustic signals such as a loudspeaker 192, for example.
- the system can also include an actuator 4920, for example a robot arm, and a camera 185.
- the tactile sensor 4905 is, for example, on the Actuator 4920 positioned.
- an actuator positioning unit 4925 which positions the tactile sensor 4905 by means of the actuator 4925 next to a person's hand, and via a person identification module 111 and / or a hand identification module 4930 and a cognitive skills evaluation module 4845 for determination cognitive abilities of the person.
- the system has a person recognition module 110, tracking module (112, 113),
- Movement evaluation module 120 skeleton creation module 5635 and / or skeleton model-based feature extraction module 5640.
- the latter is configured to evaluate and classify the recognition of images by a patient in order to evaluate the cognitive abilities of the patient, in particular whose attention.
- 23 shows the sequence used here by way of example.
- the service robot 17 uses a speech synthesis unit 133 to indicate to a patient that the service robot 17 should memorize several images 2305. Following this voice output, a sequence of images is displayed on the monitor of the service robot 17 displayed 2310, for example five three seconds apart. The patient is then informed via a speech synthesis unit 133 that the service robot 17 should signal by moving his head whether these images then shown are known to him, ie the service robot 17 should have a corresponding Perform classification in step 2315.
- a shake of the head is seen as a rejection, a nod as a confirmation.
- This is followed by a display of ten images on the screen of the service robot 17 (step 2320) at intervals of three seconds. Five of these are repeated compared to the previous sequence from the five images, for example each image only once.
- the sequence of the images and / or the differentiation into new image vs. already shown image can take place 2325.
- the service robot 17 stores the shown sequence of images, whether they have already been shown or not 2330, and records during of pointing (or up to one second afterwards) the head movements of the patient.
- the service robot 17 has at least one sensor, for example an RGB-D camera 185, which can recognize and track the head of a patient 2335, the evaluation e.g. using the visual person tracking module 112 and / or the laser-based person tracking module 113 takes place. This includes turning your head and / or nodding your head.
- the service robot 17 is in the Able to use classification methods to identify prominent points on the face, including the eyes, eye sockets, mouth and / or nose. Solutions for this are known in the prior art (eg DOL 10.1007 / 978-3-642-39402-7_16; 10.1007 / sl 1263-017-0988-8) which, among other things, use histograms of gradients.
- the head movements of the patient are next classified in order to be able to detect head shaking and / or nodding 2340. Frameworks from the prior art are also used for this.
- the movements of nodding or shaking the head recognized above are synchronized 2345 accordingly with the images shown.
- the image sequence shown is then coded to determine whether the patient has correctly recognized a repeated pointing or first-time pointing 2350.
- the service robot 17 optionally stores the comparison of the values, e.g. with the date on which it was carried out, in a database which, for example, has also stored the image sequences shown.
- a counter is incremented 2355.
- the score that results from the incrementation serves as a measure of whether the patient suffers from cognitive impairments.
- the determined score is classified and assigned a medical interpretation 2360.
- the score and its medical interpretation are stored 2365 in a database, possibly stored 2370 in the cloud memory in the cloud 18, and are available to medical personnel via a terminal for evaluation purposes Order 2375.
- the service robot 17 is able to determine the position of the patient's eyes in three-dimensional space 2410 and that of the display 2 (step 2405). In one aspect, the service robot 17 uses this data to check the visual connection between the eyes and the display 2 for the presence of obstacles. Thus, for example, in the case of a patient in bed, the fall out protection can possibly represent such an obstacle. To this end, the service robot 17 first calculates the coordinates that are on the line of sight 2415 and checks, for example by means of a 3D camera, whether these coordinates of the visual connection are associated with detected obstacles 2420.
- the display inclination is repositioned 2450, alternatively and / or additionally a repositioning of the service robot 17 in the XY plane 2455.
- the service robot 17 is configured so that the service robot 17 with the aid of the spatial coordinates of the display 2, for example the display corners , and determining the angle between the patient's eyes and the display surface (step 2425) ensures that at least one angle lies in an interval 2430, which is e.g. can be device-specific.
- the service robot 17 is able to adjust the display inclination accordingly 2450 and / or to reposition the service robot 17 in the room.
- the font size and / or other symbols on the display 2 can also be adapted as a function of the distance between the patient and the display 2.
- the service robot 17 first calculates the Euclidean distance between the eyes and the display 2, compares this with reference values stored in the memory 10 of the service robot 17 as to whether this distance is usually acceptable for recognition, and includes, in a supplementary aspect, patient data on vision, to adjust the reference values if necessary.
- the service robot 17 can adjust the size of the display 2 (i.e. the size of the displayed objects and characters) and / or the service robot 17 is positioned in the space within the XZ level (i.e. the floor level) that the distances for recognizing the display contents are sufficient.
- the service robot 17 With a view to the repositioning of the service robot 17 in the XZ plane, the inclination of the display 2 and / or the size adjustment of the display, the service robot 17 is able, by scanning its surroundings and a possibly expanded or alternative viewing corridor, which is free of obstacles, the inclination of the display 2 and / or the display dimensions of the display 2 to determine what a position in the XZ plane, a display inclination and / or a display size would be, so that the patient does not have any obstacles between his eyes and the Display 2, the display 2 is spatially positioned in such a way that it is largely free of reflections and / or that the display size is sufficient for the patient's visual capabilities.
- the service robot 17 has a control for the display inclination and a dialog function in the display 2 or configured as a voice interface.
- this dialog function the patient is asked for feedback on the extent to which the display is sufficiently recognizable for him.
- the service robot 17 can change the orientation of the display 2. This can be done in one aspect by repositioning the Service robot 17 vs. the patient. This can be done on the one hand by turning on the spot, on the other hand also by taking a different position (for example defined by the area that the service robot 17 covers on the floor). In an alternative and / or supplementary aspect, the inclination of the display 2 can be adjusted, with the tilt axes being able to be aligned horizontally and / or vertically.
- this described process is run through again in order to check that the patient can see the display 2 well.
- the service robot 17 is configured in such a way that the camera 185, for example an RGB-D camera 185, can identify and track the fingers in order to identify the poses shown by the hand with a view to Evaluate numbers, for example by means of the visual person tracking module 112 and / or the laser-based person tracking module 113.
- FIG. 25 illustrates this process.
- the depth image 2505 generated by the 3D depth camera is transformed into a 3D point cloud in which each pixel of the camera is assigned a spatial coordinate 2510, from which the skeleton recognition 2515 takes place via camera SDKs or software from third-party providers such as NUITrack.
- the points of articulation are recognized accordingly, including the wrist and finger joints.
- a joint selection 2520 takes place, that is to say joint points that are only necessary for the calculations to be carried out subsequently are processed further.
- Angle calculations 2525 are then carried out, for example for the angle between the third and second phalanx, the second and first phalanx, and the first phalanx and the metacarpal bone.
- the third phalanx is generally referred to as the one with the fingertip). Since the thumb does not have a second phalanx, it is the angle between the third and first phalanx, the first phalanx and the metacarpal bone, and, in one aspect, between the metacarpal bone and the carpal bones.
- Each phalanx or each hand bone is represented as a direction vector, in each case from the observed joint point.
- a feature extraction 2530 takes place next, in which, for example, the angles of said joint points per finger are jointly evaluated.
- the A feature classification 2535 that is implemented on the basis of defined rules, for example, defines an extended index finger as an angle of 180 ° between the first and second and second and third phalanx.
- threshold values can be defined that soften the condition 180 ° somewhat and, for example, for the angle between the third and second phalanx between 150 ° and 180 °, and the angle between the first and second phalanx between 120 ° and 180 ° and between the metacarpal bones and the first phalanx between 90 ° and 180 °.
- the angle between the third and first phalanx of the thumb is less than 120 °, for example less than 100 °, in order to ultimately be recognized as 2.
- the angle between the hand bones is less than 145 °, for example less than 120 °.
- Feature extraction, feature classification, and hand pose classification can be carried out on the one hand by predefined rules such as angle definitions of individual articulation points and their combination, or they can be trained using machine learning approaches 2550 such as support vector models in which certain angle combinations are labeled accordingly , ie the combination of angles between the individual phalanges of the fingers can indicate that, for example, two fingers corresponding to the value 2 are shown.
- predefined rules such as angle definitions of individual articulation points and their combination
- machine learning approaches 2550 such as support vector models in which certain angle combinations are labeled accordingly , ie the combination of angles between the individual phalanges of the fingers can indicate that, for example, two fingers corresponding to the value 2 are shown.
- an output of the service robot 17 is first triggered, which is output via the speech synthesis unit 133 via loudspeaker 192 and / or via text and / or images on the screen of the service robot 17.
- This speech prompts the patient to two Pointing fingers 2605.
- the camera 185 identifies the patient's hands, their fingers, and tracks the finger movements.
- the service robot 17 evaluates this as part of the pose classification in order to determine how many fingers are displayed 2610. In an optional aspect, as will be described below, it can be taken into account whether the finger pose displayed by the patient is associated with a code 2665.
- the service robot 17 then saves a value which indicates whether the patient has shown two fingers 2615, ie an evaluation of the comparison of the evaluated finger poses with the visually and / or acoustically output numerical values takes place.
- the service robot 17 has at least one actuator 4920, for example a robot arm with at least one joint, which also has at least one robot hand 2620 that has at least two human-like fingers, for example at least over five fingers, one of which corresponds to a thumb in terms of its arrangement, and which, for example, have as many finger joints as the human hand.
- the service robot 17 is able to display numbers by means of these fingers, with stretched fingers and finger poses resulting from the angles of the phalanges that have already been classified above with a view to the recognition of phalanges by the camera 185.
- the service robot 17 is thus able to also display the value 2 2670, for example by stretching the thumb and forefinger on the robot hand, ie the angles between the first three phalanxes are approximately 180 °, while the angles of the others Phalanges and phalanges and hand bones are less than 120 °.
- the service robot 17 is configured in such a way that the service robot 17 can synchronize the regulation of the poses of the robot hand via the voice output and / or display 2 in such a way that the value 2 is displayed by the robot hand while the request is made via display 2 and / or speech synthesis unit 133 occurs to the patient to display as many of the fingers as the robot hand displays 2675.
- the hand tracking and the pose classification for recognizing two of the fingers 2610 on the patient then take place in order to determine that the patient has the number two has indicated to store a value 2615.
- the hand poses shown by the patient are evaluated, for example, within a time window of 3 seconds after the service robot 17 prompts the patient to display a numerical value via its output units such as loudspeaker 192 and / or display 2 has or gives him the corresponding numbers through the robot hand were shown.
- the findings obtained in the course of this test allow an assessment of the extent to which the patient is impaired by disorganized thinking and thus represent a test method with a view to recognizing and monitoring delirium.
- the service robot 17 is configured in an optional alternative and / or supplementary aspect in such a way that it is possible to display the numbers with the fingers based on cultural and / or national differences. Alternatively and / or in addition, a recognition of the numbers shown can also be facilitated by the fact that the service robot 17 takes these differences into account when evaluating the hand poses. As a result, for example, the number 2 is more likely to be displayed by the thumb and forefinger in patients from Germany, while US Americans use the forefinger and middle finger to display the number 2.
- the service robot 17 contains in the memory 10 codes for different poses which display the same number 2650 and which are accordingly country-specific / culturally determined.
- the patient data that the service robot 17 holds in its memory 10 can also contain one of these codes 2652, which accordingly refers to the national / cultural background of the patient.
- Several poses are thus stored in memory 10 for each number, in particular several combinations of fingers.
- the next step is a comparison of the codes in order to determine the patient's preferred poses 2655.
- the patient the numbers in the familiar hand and / or finger pose.
- the service robot 17 is thus able to show the patient the hand and / or finger pose, for example the number 2, which corresponds to his cultural / national background 2660, which is implemented by the robot hand of the actuator 4920 2670.
- this information can be used for the appropriately coded cultural / national background of the patient to better recognize the two fingers shown by the patient.
- This finger output and / or recognition taking into account such codes is an optional implementation.
- the service robot 17 is further configured in such a way that the robot hand can be spatially aligned via the actuator 4920 with at least one joint in such a way that the robot hand can be recognized by the patient in step 2638.
- the service robot 17 detects the patient's head and its orientation in space by using methods of facial pose recognition established in the prior art 2625 uses such as are contained in the OpenPose framework, for example.
- approaches such as histograms-of-gradients can also be used, which are implemented in frameworks such as OpenCV or Scikit-image.
- the service robot 17 uses these frameworks to determine the orientation of the head in space and calculates a field of vision for the eyes.
- the service robot 17 thus has a cone-of-view recognition 2630.
- the service robot 17 detects its position in space and the position of the actuator 4920, in particular the position of the robot hand, and determines whether this position is within the cone 2632 If it does not position itself within the cone, the service robot 17 calculates which angular settings of the joints of the actuator 4920 are necessary in order to position the robot hand within the cone.
- the service robot 17 calculates a three-dimensional area in space which has a minimum distance from the patient and which varies, for example, depending on the body region of the patient. Such minimum distances are stored in the memory 10 of the service robot 17 in step 2636.
- an “allowed zone” is calculated in which the robot hand is allowed to move, whereby the distance to the patient's head is further than, for example, to the torso or arms.
- the distance to the head is 50 cm and that to the rest of the patient's body is 30 cm.
- the service robot 17 determines in step 2638 where in which part of the permitted zone the service robot 17 can position the robot hand so that this robot hand is located within the two cones. Then, in step 2638, the service robot 17 aligns the robot hand by means of the actuator 4920 in such a way that the hand can be “easily recognized” by the patient. If no such positioning is possible, in step 2640 the service robot 17 requests the patient to look at the service robot 17 via the output units such as display 2 and / or speech synthesis unit 133 via loudspeaker 192. A new test then takes place in step 2642, ie, steps 2630-2640 are run through.
- the service robot 17 aborts the test in step 2644 and transmits information to medical personnel in step 2646, including, for example, information to the server and / or a mobile device via an interface 188 (such as WLAN) Terminal.
- the service robot 17 can prompt the patient again and / or wait a little longer. If the robot hand is aligned in such a way, so that the patient can easily recognize this, the display of two fingers with this hand takes place in step 2670 and the process is continued as described above. This alignment of the robot hand on the basis of face pose detection and the consideration of the “allowed zone” is an optional aspect.
- the cognitive analysis system is illustrated in FIG. 65.
- the system for example a service robot 17, comprises a computer 9, a memory 10, an output unit and in the memory via a numerical value output module 4940 for outputting numerical values, a person detection and tracking unit (4605) with a camera (185) and a person recognition module (110).
- the output unit is a sound generator such as a loudspeaker 192, a display 2 and / or an actuator 4920, for example a robot arm, in one aspect with a robot hand 4950.
- the system has a hand pose in the memory 10.
- Detection module 4960 for detecting hand poses of the person
- finger pose generation module 4955 for generating finger poses of the robot hand (4950), the finger poses representing numerical values, for example.
- the system also has a cognitive skills assessment module 4845 for assessing the cognitive abilities of the recorded person.
- the system is connected to a patient administration module 160.
- the system has rules to determine cognitive abilities of the recorded person, which have been described elsewhere.
- the system has a personal identification module 111, tracking module (112, 113),
- Movement evaluation module 120 skeleton creation module 5635 and / or skeleton model-based feature extraction module 5640.
- the service robot 17 is configured in such a way that the service robot 17 is able to carry out a test of the patient's sensation of pain, which test is implemented by observing the patient's behavior by the service robot 17. The procedure is based on the Behavioral Pain Scale, a pain rating scale established in medicine. Such tests are also carried out as part of Delir monitoring.
- the facial expression of the patient who is in a bed is analyzed.
- approaches were presented as to how the service robot 17 can identify patients in a bed and possibly track their faces, including those associated therewith Navigation of the service robot 17.
- a bed is detected by sensors and the image generated in the process is evaluated using a pattern comparison in order to assess whether it is a bed. In one aspect, these approaches can also be used here.
- the service robot 17 evaluates the patient's emotions, which result from his facial expression.
- the service robot 17 can use a face classification database in one aspect, in which classification rules for classifications within a candidate region of the face and across several candidate regions of the face are stored, which allow conclusions to be drawn about the emotional state of the patient on the basis of facial features and which are described in more detail below.
- This two-stage process differs from the state of the art, which was described, for example, in US20170011258 or US2019012599 as a one-stage process.
- histograms-of-gradients are used, which are implemented, for example, in frameworks such as OpenCV or Scikit-image. Emotion recognition focuses primarily on those emotions that measure the patient's tension, from relaxed to great tension, which is expressed in grimaces.
- the head of the patient is first identified in step 2705, for example by means of frameworks such as OpenPose. These evaluations can take place by means of the 2D or 3D camera 185.
- candidate regions within the face are first identified 2710 before the feature extraction required for the evaluation of the emotional state based on histogram-of-gradient algorithms takes place in at least one candidate region in step 2715, with which, for example, the movements of the mouth can be evaluated or that of the eyebrows.
- a feature classification takes place, which falls back on an existing feature classification, which is carried out using established clustering methods such as K-Means, support vector machines and / or based on weights that are generated by training a neural Network were collected, for example a multilayer convolutionary neural network with backward propagation with the help of labeling of facial expressions in step 2720.
- the classifications made at the candidate region level are classified across several candidate regions, which is also done by means of classification based on established clustering methods of machine learning such as, for example, K-Means, Support Vector Machines, and / or the convolutionary neural networks already mentioned in step 2725. With this, for example, movements of the mouth and eyebrows are evaluated together.
- the recognition algorithms can be filtered in various aspects in step 2730, ie for example corrected for the age of the patient in step 2735, which the service robot 17 can receive from a database via an interface 188 (such as WLAN), provided that the Evaluation of the emotions takes place directly on the service robot 17.
- the recordings that the camera 185 creates of the patient's head for the purpose of recognizing the emotions can also be transmitted via an interface 188 (such as WLAN) to the cloud 18 and analyzed there. In that case, any age information would be transmitted from the cloud storage in the cloud 18 to the module carrying out the emotions.
- Another filter is whether the patient has an endotracheal cannula (step 2740) which ensures artificial ventilation of the patient through the mouth.
- classification algorithms that are used for the described evaluation of the emotions have, for example, been created in one aspect using training data with images of corresponding endotracheal ventilated patients. Further details on the detection of cannulas can be found below and can also, in one aspect, be used within the scope of the procedure described here.
- the emotions are rated on a scale from 1-4, for which purpose determined emotions are compared with those that are stored in memory 10 and to which scale values are assigned.
- a value of “1” means a facial expression classified as regular, while the tension increases on the scale up to a value of 4, which implies grimaces.
- the values are recorded over the course of several hours or days, which possibly simplifies an evaluation of the emotional state of the patient, for example if, at the beginning of the series of emotion measurements made by the service robot 17, a patient is in is in a relaxed state, which is, for example, about a terminal and a menu configuration can be stored in the memory 10, for example by medical personnel, to which the service robot 17 has access.
- This dynamic classification (step 2750) of the facial expressions increases the quality of the classification, since the classifications are possible on the basis of different observations of the face at several points in time.
- a retrospective classification can also be carried out, in which case, for example, only the extracted features, together with a time stamp that characterizes the recording time, are stored and reclassified. In order to do this, the recordings of the face are saved.
- Pain status detection of the movement of the upper extremities.
- a second part of the test focuses on movements of the upper extremities such as the upper arm, forearm, hand and fingers.
- the service robot 17 tracks these, which the service robot 17 has recognized as described above, over the course of time, which happens either by means of the 2D camera and frameworks such as OpenPose or 3D camera (possibly an RGB-D camera 185), with For example, the evaluation takes place by means of the visual person tracking module 112 and / or the laser-based person tracking module 113.
- each point is assigned a spatial coordinate in step 2810, by camera frameworks or other software tools in the prior art Technique a skeleton model recognition takes place in step 2015, in which the articulation points of the skeleton are recognized.
- a joint selection is then carried out in step 2820, that is to say the detection is aimed in particular at joint points such as the shoulder joint, the elbow joint, the wrist and the finger joints.
- An angle calculation of these articulation points takes place in step 2825, in which the angle, for example, via the Direction vectors is defined, which take the hinge point as a starting point.
- the angles of these limbs are recorded over time.
- the limbs are then classified in such a way that the number of angular changes per unit of time, the speed, ie for example the angular speed, etc., is used as a measure of the intensity of movement.
- the service robot 17 classifies these movements in step 2835 on a scale from 1 to 4, and saves this value.
- a value of 1 corresponds to no movement within the tracked time.
- the value 2 corresponds to few and / or slow movements of the arms, 3 corresponds to movements of the fingers, and 4 corresponds to a high movement intensity of the fingers, which are defined, for example, by the number of finger movements per unit of time and / or their speed and which are threshold-dependent.
- a third part of the test is based on the pain vocalization of the patient and can take place in two principally different processes which are based on two different scenarios and are expressed in two variants of the assessment.
- the first scenario involves an artificially ventilated patient whose vocalization is assessed on the basis of a cough.
- the second scenario the patient is not ventilated and typical pain sounds are assessed. The procedure here is described in more detail in FIG.
- Pain status pain vocalization of ventilated patients
- a first variant involves ventilated patients. These can either have a tracheostomy tube, which ensures ventilation through an opening in the neck, or an endotracheal tube, which enables ventilation through the mouth. Means
- the service robot 17 is able to use image recognition algorithms to identify patients ventilated in this way in step 2901.
- the head and neck region of the patient are recorded, either as a 2D or 3D image, which serve as candidate regions in the first step, the neck in the case of a tracheostomy tube and the mouth in the case of a
- Endotracheal cannula serves as a candidate region.
- the candidate regions are recognized, for example, in connection with histogram-of-gradient (HoG) -based face recognition in step 2905 and candidate regions derived therefrom, such as the mouth and neck, in the step
- HoG histogram-of-gradient
- Model assumptions can be used as a basis, i.e. the shape that such cannulas typically have (step
- the pixels recorded by the camera 185 are then evaluated by an optionally real-time capable and fault-tolerant segmentation algorithm 2920, to recognize such a cannula.
- the service robot 17 is thus able to detect the cannula.
- a database-based detection can be carried out in step 2902, the service robot 17 providing information on the ventilation of the patient in step 2925 via an interface 188 (such as WLAN) from a cloud-based database with patient information in the cloud 18
- Step 2927 queries and / or the information is located, together with further patient data, in the memory 10 of the service robot 17 (step 2929).
- the service robot 17 determines the extent to which the patient is breathing normally or even coughing 2930. This determination can take place in various ways. In one scenario, the service robot is using in step 2935
- step 2936 the service robot engages
- Curves are compared in step 2941 either with threshold values that are typical for various ventilation scenarios such as pressure- or volume-controlled ventilation and those that occur in these ventilation scenarios in the event of a cough.
- labeling can also be carried out in the cases, for example by medical staff, in order to recognize atypical ventilation patterns such as coughing, which is then classified as a cough in step 2942 by the algorithms of machine learning and / or neural networks .
- the curve progressions pressure, volume, flow
- an anomaly can not only be compared directly with the cycle before and after in step 2942, but also chains of several cycles in order, for example, to
- the corresponding modes of the ventilator alternatively the ventilation curves derived therefrom (pressure / flow over time) can be recognized by the service robot 17 and accordingly in the classification of the ventilation of the Patient in step 2944 must be taken into account.
- the service robot 17 can also receive the information from the ventilator that it is in the cough support mode or that it triggers a cough, with which the system would detect a coughing event.
- an adapter can also be accessed in step 2937, which uses the pressure and / or flow in the supply tube between the cannula and the ventilator a pressure and / or flow sensor and transmits the signals, for example wirelessly via an interface 188, to the service robot 17, which then creates corresponding evaluations of the ventilation profile, which can be evaluated as described above.
- coughing can also be detected via at least one sensor, which is located, for example, on the patient's body 2950.
- sensors 2952 for example with magnetometer
- strain sensors 2954 eg, strain gauges that are applied to the patient's skin at precisely these locations
- contact microphones 193 step 2956
- thermistor 2958 located on or in the nose, for example, and which are each wirelessly connected to the service robot 17 via an interface 188 (e.g.
- WLAN Wireless Local Area Network
- At least one microphone 193 (step 2960) is used that is on the patient and / or on another Position is located in the patient's room and that is directly (or indirectly with data storage (in one variant also storage of data already evaluated with a view to cough signals) in memory 10, to which the service robot 17 has access) connected to the service robot 17.
- this at least one microphone 193 is integrated 2962 in the service robot 17, records the noises in the patient's environment in the memory 10 of the service robot 17 and classifies the sound signals as to whether a cough is occurring.
- machine learning algorithms and / or neural networks that have been trained using recorded coughing noises are used here.
- a system is trained in one aspect that has at least one processor for processing the audio data and at least one audio data memory in which audio data are located, in one aspect also present as spectral data, and which are labeled accordingly.
- the 3D sensor of the service robot 17, e.g. the 3D camera detects movements around the mouth, but also of the chest and / or neck, i.e. by means of a fault-tolerant
- Segmentation algorithm 2976 said candidate regions 2974 are evaluated. How the mouth can be detected has already been described elsewhere. With a view to the detection of the chest and stomach area, the candidate region 2974 is, for example, via the
- the trunk under the blanket which consists of the chest and abdomen, both of which can be evaluated while breathing.
- the blanket which consists of the chest and abdomen, both of which can be evaluated while breathing.
- Candidate region 2974 can also be determined, for example, from twice the head height, which extends downward from the chin, and approximately 2.5 times the head width. In both cases, the identification of the head can be used as an initial step and thus as reference 2972 in order to identify the candidate regions from there. As an alternative and / or in addition, the dimensions of the bed can also be used as reference 2972 for this purpose
- Camera capture the elevations of the duvet surface, in one aspect a
- Histogram-of-gradients evaluation takes place, which are evaluated on the basis of a classification that has been trained by a system that uses 2D or 3D as input variables
- Beds are or are not and which are removed from the area using classification procedures machine learning and / or neural networks were evaluated, while the evaluation results form the classification, in particular to detect the upper body of a patient.
- step 2978 the movements of the mouth, cheeks, neck and upper body recorded by the 3D camera are evaluated over time.
- an interference reduction 2980 made, as far as the detection of movements is concerned, which are covered by fabric or the duvet, i.e. primarily of the upper body / stomach. Interferences that arise from different phases of the recorded movements of the duvet and make the actual breathing movements more difficult to detect are eliminated by determining the power density spectrum, which makes it easier to record the movements of the chest.
- the power density spectrum is determined, for example, by means of a fast Fourier transformation (FFT) in step 2982, then the power density spectra for all pixels is aggregated in step 2984 and the maximum is determined via quadratic interpolation in step 2986, the position of the maximum indicating the respiratory rate in step 2988. This is then monitored with a view to frequency changes that reflect the cough in step 2990. A frequency determination of the detected body parts is then undertaken 2990 in step 2990. For example, histogram-of-gradient calculations are used in step 2990.
- FFT fast Fourier transformation
- the subsequent feature classification in step 2992 is based on a classification that was generated by recording coughing movements and non-coughing movements, for which purpose standard classification methods and / or neural networks can be used, as already described elsewhere in this document. If no cough is detected by the approaches described, this criterion is rated with a score of 1. If a cough is detected, the score is given a 2.
- the sequence can be described as follows: detection of the person, face and neck recognition of the person, evaluation of the face and neck area of the person according to patterns that describe a device for artificial ventilation, storage of a value when a pattern is detected, the one device for artificial Describes ventilation, the device for artificial ventilation describing a pain status.
- Pain status pain vocalization in non-ventilated patients
- the service robot 17 If the service robot 17 has not recognized a cannula in the patient by means of the implemented image recognition and / or no information on artificial ventilation is stored in the database with patient information, another variant of the third part of the test takes place, the service robot 17 analyzes using Microphone 193 sounds emitted by the patient. These noises are classified on the basis of algorithms that have been trained on the basis of labeled noise data using machine learning algorithms and / or neural networks in such a way that the noises can recognize pain vocalizations in different forms. If no pain vocalization is found, this criterion is given the value 1.
- a pain vocalization is recorded for a duration of less than 3 seconds and with a frequency of less than three pain vocalizations per minute, this criterion is rated 2.
- a higher frequency or longer duration is rated 3, while for example verbal pain vocalizations, which in one aspect can also be determined via a dialogue with the patient, is rated 4.
- the scores over the three parts of the test are finally summed up.
- the results are stored in a database, in one variant transmitted via an interface 188 (such as WLAN) to a server in the cloud 18 and stored there.
- the medical staff has access to the evaluation, which also enables a detailed examination of the test results (partial results and overall result), and this data can be displayed visually via a terminal.
- the individual parts of the test can also be carried out independently of one another.
- the pain status determination can be based on a
- Assignment of scale values to the recorded acoustic signals includes and the Scale values represent a pain status.
- the position of the source of acoustic signals is also determined, the person whose pain status is determined is determined and the determined position is compared by comparison with a threshold value (ie with regard to a minimum similarity of the position values), and a value is stored when the value falls below the threshold value in relation to the determined pain status.
- a threshold value ie with regard to a minimum similarity of the position values
- a system for determining the pain status of a person is summarized as follows in FIG. 66:
- the system for example a service robot 17, comprises a computer 9, a memory 10 and a sensor for contactless detection of the person, for example a 2D and / or 3D camera 185, a LID AR 1, a radar and / or ultrasound sensor 194.
- a sensor for contactless detection of the person for example a 2D and / or 3D camera 185, a LID AR 1, a radar and / or ultrasound sensor 194.
- it can have different modules in its memory 10.
- the system has a person recognition module (110), a visual person tracking module (112), face recognition module 5005 for recognizing the face of the person, a face candidate region module 5010 for selecting candidate regions within the face, an emotion -Classification module 5015 for classifying the surface curvatures of the candidate regions in emotions, and an emotion evaluation module for determining a scale value for the emotion 5020.
- the system has, for example, a bed recognition module 5025 for recognizing a bed and / or a Upper extremities evaluation module 5035 for detecting the upper extremities of the person, tracking them over time, and evaluating the angles between the trunk and the upper arm, the upper arm and the forearm and / or the phalanges and the hand bones with regard to the intensity of the angle changes , their speed and / or the number of angle changes per unit of time is evaluated, as well as, for example, via a pain status calculation module 5040 to determine a scale value for the pain status.
- a bed recognition module 5025 for recognizing a bed and / or a
- a Upper extremities evaluation module 5035 for detecting the upper extremities of the person, tracking them over time, and evaluating the angles between the trunk and the upper arm, the upper arm and the forearm and / or the phalanges and the hand bones with regard to the intensity of the angle changes , their speed and / or the number of angle changes per unit of time is evaluated, as well as,
- the system comprises a microphone 193 for recording acoustic signals, for example an audio source position determination module (4420) for evaluating the position of the source of acoustic signals and an audio signal person module (4430) for assigning audio signals to a person .
- the system can, for example, via a pain vocalization module (5055) to classify the intensity and frequency of the acoustic signals and to determine a scale value representing a pain vocalization.
- a ventilation device identification module 5065 for identification of a device for artificial ventilation, ie selection of candidate regions of the artificial ventilation, expansion of the candidate regions for artificial ventilation by means of object recognition, and object classification for the identification of cannulas, around tracheostomy tubes or endotracheal tubes.
- a pain perception evaluation module 5085 for evaluating sensors attached to a person such as an inertial sensor, strain sensor, contact microphone and / or thermistor that detect movements, air currents and / or sounds that are classified with regard to an expression of pain.
- the system has a person identification module 111, movement evaluation module 120, skeleton creation module 5635 and / or skeleton model-based feature extraction module 5640.
- the robot 17 is further equipped with a system which detects repetitive movements of the human body that correlate with the expulsion of blood from the heart into the large blood vessels with each heartbeat. Changes are recorded that arise, on the one hand, through movements of the large blood vessels and, for example, propagate in waves in the body, and / or that result from movements of arteries in the skin. The latter are more robust against fluctuations in the lighting of the body part and / or different color tones of the skin. Alternatively and / or in addition, changes in the blood volume or the blood flow in the skin, e.g. over time, are recorded which correlate with the heartbeat. 30 below shows the process of data acquisition and evaluation.
- a body region and multiple subregions are identified in step 3005.
- the body region is the face, the evaluation taking place, for example, by means of the camera 185.
- the system uses algorithms from the prior art to record the face (alternatively other body regions), for example the OpenCV, OpenPose or dlib framework, and to track it, with the evaluation using the visual person tracking module 112 and / or the laser-based person tracking module 113 takes place.
- the forehead, the cheeks or the chin are recorded as partial regions, for example several body regions together, which are then evaluated individually and / or separately according to the steps described below.
- a selection of candidate regions can be made, ie partial regions of the face that are relevant for the evaluation, for which purpose segmentation methods known in the prior art (such as, for example, RANSAC) can be used.
- These sub-regions like the Body region as such, tracked over time in step 3010 by the frameworks mentioned.
- the camera 185 is aligned as parallel as possible to the region that is to be tracked.
- the detection angle of the face which results from an axis that is perpendicular to the camera detecting the face, and an axis that is perpendicular to the sagittal plane of the face can be minimized.
- the system determines, for example, a plane which runs through the face and which is essentially parallel to the top view of the face and which corresponds, for example, to the sagittal plane. Based, for example, on histogram-of-gradients (HoG), the system has a classification that describes the deviations from this top view in order to record the inclination of the face in three-dimensional space.
- HoG histogram-of-gradients
- the system uses this procedure to assess a face looking into the system's camera 185 according to the extent to which the face is aligned parallel to the camera lens.
- the system can, in one aspect, adjust the inclination of the camera 185 in three-dimensional space using a corresponding mechanism such as a tilting unit 5130, for example by controlling two tilting axes arranged orthogonally to one another, which are driven by servomotors.
- the term tilting unit is therefore understood to mean a tilting unit with at least two axes which, on the one hand, enables the horizontal plane to be tilted and, on the other hand, enables rotation about the vertical axis.
- the wheels of the service robot 17 are controlled in such a way that the service robot 17 rotates in the direction of the person in order to reduce the determined deviation of the camera plane from the face plane.
- such a deviation triggers the voice output of the service robot 17 in order to give the patient instructions to align his face accordingly.
- rules are stored that require an alignment in the XY plane if a deviation in the XY plane has been determined. Adjustment of the face inclination and control of the tilting mechanisms, the orientation of the service robot 17 and / or the voice output with a view of the patient are carried out, for example, until the angle between the camera plane and the face plane have reached a minimum.
- the detection angle of the camera 185 can be minimized in comparison to the sagittal plane and possibly also the transverse plane, which also includes driving maneuvers of the robot 17, for example.
- the system is configured, in an optional aspect, to illuminate the body part such as the face in step 3020.
- the face or another body part
- At least one lamp is used, which is located close to the camera 185, for example. This is ideally located below and / or above the camera 185, that is to say is offset vertically around this camera 185.
- the emitted light is scattered in order to ensure the most homogeneous possible illumination of the area to be recorded.
- a lateral arrangement of the camera 185 can, depending on the position of the face and its dimensions, possibly lead to a shadow cast by the nose, which affects the cheeks, which are located to the side of the nose and whose recording provide an above-average signal-to-noise ratio, and thus possibly worsen the quality of the evaluation.
- the camera 185 used provides at least one color channel for the evaluation, including, for example, at least the green color channel, since the light emitted here is particularly well absorbed by hemoglobin.
- the camera 185 also provides a color channel for the shade of orange and / or cyan.
- the color depth per channel is, for example, at least 8 bits, the image frequency 30 images per second.
- the camera 185 can also be an RGB-D camera 185, which provides depth detection in addition to color detection, for example on the basis of time-of-flight sensors or speckle patterns to record rhythmic, vascular blood flows and rhythmic, vascular expansions.
- a signal extraction takes place.
- the input signal is first selected on the basis of the video signals of the tracked regions, which can either be a movement caused by the pumping rhythms of the heart and / or a color change with a view to the flow of blood, in particular hemoglobin, whereby said rhythmic , vascular blood flow and / or rhythmic vascular expansion.
- a color channel evaluation takes place, including known information, which features are mapped by which color channel, provided that the blood flow is recorded.
- This is understood to mean, in particular, a weighting of the channels in step 3030. It is primarily the green color channel that can be evaluated, the green and red color channels (e.g. as a difference consideration of the green and red channel), the combination of green, cyan and orange, etc.
- the spatial resolution is determined alternatively or in addition. That is to say, the vertical and / or horizontal movements of recorded properties of the face are tracked, for example the position of the face and its subregions over time and are evaluated. This includes both the movements of the head and the individual parts of the face.
- the subsequent signal determination uses, for example, at least one filter in the first sub-step (preprocessing 3035).
- This includes trend adjustment (e.g. with scaling and / or normalization); a moving average consideration, a high-pass filtering; a bandpass filtering, possibly designed as an adaptive bandpass filtering; an amplitude-selective filtering; a Kalman filter; and / or a continuous wavelet transform.
- trend adjustment e.g. with scaling and / or normalization
- a moving average consideration e.g. with scaling and / or normalization
- a bandpass filtering possibly designed as an adaptive bandpass filtering
- an amplitude-selective filtering e.g., a Kalman filter
- / or a continuous wavelet transform e.g., a linear polynomial approximation of the least squares.
- a signal separation method 3040 in order to improve the signal-to-noise behavior and to reduce the number of feature dimensions to be considered.
- a main component analysis or a one-dependency analysis can be used, in one aspect also machine learning methods.
- the signal processing 3045 comprises the determination of the pulse rate and, if necessary, further variables in the context of a Fourier transformation (fast, for example, discrete Fourier transformation, in particular to determine the maximum power spectrum density), autoregressive models (for example using Burg's method), the use of bandpass filters in Connection with the detection of maxima, for example a peak detection, a continuous wavelet transformation, and / or models of machine learning, in particular of non-awakened learning.
- a discrete cosine transformation can also be used.
- methods can also be used to compensate for errors caused by movements of the head, etc., for example, including Kalman filters, (adaptive) bandpass filters, outlier detection, sliding Averages, Bayesian fusion and / or machine learning methods can be used.
- step 3055 takes place on the basis of various prior art approaches described, with which, for example, the systolic and diastolic blood pressure can be determined, for which purpose linear or non-linear prognosis methods can be used.
- the mentioned machine learning methods for example neural networks such as convolutionary neural networks, are able to recognize hidden and partly unknown features in the data and to take them into account in the evaluation, for example in the context of performed cluster analyzes .
- weights for the classifications or linear or non-linear forecast models are generated, which are then used in productive operation as part of the process described.
- the values determined for the pulse rate, pulse rate variability and, if applicable, further values, possibly after post-processing has taken place, are compared in one aspect with values stored in the memory 10 in step and the pulse rate, pulse rate variability, or other variables are determined based therefrom, especially the systolic and diastolic blood pressure.
- the filtering in the context of preprocessing and postprocessing depends on the variables to be detected.
- a bandpass filter can be used that covers the spectrum 0-6 Herz, for example at least 0.7-4.5 Hz.
- the signal of the pulse can also be sampled more closely in this frequency range , for example with a window of 0.1 Hz. Smoothing can then be carried out using a low-pass filter.
- the pulse rate, heart rate or pulse rate or heart rate can be processed, for example, by means of a bandpass filter which has a width between 0.7 and 4 Hz.
- a bandpass filter with a window between 0 and 0.4 Hz can again be used, in one aspect sampled at intervals of 0.02 Hz
- Bandwidth between 0.5 and, for example, 6 Hz can be evaluated, in one aspect sampled in intervals of 0.1 Hz.
- the pulse wave transit time can thus be determined for values from several regions.
- the pulse shape results from an unsampled course in the spectral window of approx. 0-6 Hz, for example characterized by the area under the curve, the height and / or width.
- the pulse energy results from the first derivative of these values.
- the blood pressure can thus be determined, for example, by means of a linear model from the pulse wave transit time and the pulse or heart rate as well as the previous blood pressure value, and linear regression models or neural networks can be used.
- a previous blood pressure value for example, the shape of the measured pulses can also be evaluated, e.g. by determining the differences between the pulse curve and the vertical running through the maximum value.
- the system for determining blood pressure is illustrated as follows, as set out in FIG. 67:
- the system for determining cardiovascular parameters of a person in one aspect a service robot 17, comprises a computer 9, a memory 10 and a camera 185 (e.g. a 2D and / or 3D camera) and also a body region detection module 4810 for the detection of body regions, a body region tracking module 4815, a face recognition module 5005, a face candidate region module 5010 and a cardiovascular movement Module 5110 for recording movements that can be traced back to cardiovascular activities.
- the camera 185 provides at least the green color channel with 8 bits.
- the system also has a light 5120 in order to illuminate the face during the recording by the camera 185, which light is located above and / or below the camera 185, for example.
- the system has a blood pressure determination module 5125 for determining the systolic or diastolic blood pressure 5125 and / or a tilting unit 5130 in order to minimize the detection angle of the camera 185 compared to the sagittal plane.
- the system has rules, for example, to place a vertical line between the eyes of the person being captured, so that the head is divided into two halves. The face is segmented, with histograms-of-gradients being placed over the individual segments. If these have a (mirror-inverted) similarity that is below a threshold value, the face is perceived as being recorded vertically.
- the camera 185 can now be controlled via a tilting unit 5130 in such a way that this comparison of the mirror-inverted halves of the face is made via the histograms of gradients during the control and the camera is positioned in such a way that these threshold values of the histograms of gradients are undercut.
- the system has a person recognition module 110, person identification module 111, a visual person tracking module 112, movement evaluation module 120, skeleton creation module 5635, skeleton model-based feature extraction module 5640 and / or movement planner (104).
- the service robot 17 can also be equipped with a detector 195, which is located, for example, on the side of the service robot 17 facing the patient. In one aspect, this detector 195 can be permanently integrated in or on the surface of the service robot 17. In an alternative and / or supplementary aspect, the detector 195 is mounted on an actuator 4920, for example a robot arm, and can, as described by way of example for the alignment of the spectrometer 196 on the skin of a patient, on surfaces of the body identified by the service robot 17 Align the patient and, in one aspect, touch the patient's skin in this way. As an alternative and / or in addition, the service robot 17 can also request the patient to touch the detector 195, for example with a finger.
- the service robot 17 is able to verify that the patient actually touches the detector 195. In one aspect, this verification can take place via a test measurement, the recorded values being compared with those from a measurement interval stored in the memory 10 in order to assess whether the patient has actually placed a finger on the detector 195. With this approach, however, it cannot be ruled out that the measurement results may be influenced by the orientation of the finger on the sensor.
- camera-based tracking of the finger is therefore carried out, the evaluation being carried out, for example, by means of the visual person tracking module 112 and / or the laser-based person tracking module 113. Such tracking has already been described elsewhere in this document.
- a dialog-based method can be used in which the patient is asked by the service robot 17 whether the service robot 17 has put his finger down correctly, which can take place on the way via a display 2 and / or a voice output.
- the surface of the detector 195 consists of a crystal, for example a crystal with a cubic lattice structure, for example in a diamond structure, with a hexagonal lattice structure or with a tetragonal lattice structure.
- the crystal has a refractive index of 1-4, e.g. 1.3-1.4, 2, 2-2, 4 or e.g. 3.4-4.1.
- the spectral width of the crystal is in the interval lOOnm - 20,000 nm, e.g. in the interval 900nm - 12,000nm.
- the measuring method of the detector 195 uses deflections of an evaluation laser 5205 on the crystal surface based on laser-induced excitation of substances which are excited by a further laser 5210 on and / or within the skin of a patient.
- the surface that was excited by the further laser 5210 is in engagement, for example, with the detector 195 at the point at which the evaluation laser 5205 is deflected on the crystal surface.
- a feature extraction takes place in which the variations in the wavelength of the further laser 5210 and the deflection of the evaluation laser 5205 caused by this and detected by a sensor based on the photoelectric effect are included as features.
- the steps shown in FIG. 30, in particular 3025-3050, which have been described in more detail elsewhere, can be used here.
- a camera-based system which is aimed at the surface of the skin of a patient and can carry out measurements.
- the system can either be rigidly attached to the service robot 17 or in such a way that it can, for example, be aligned three-dimensionally so that it can capture the skin surface of a patient without the patient moving.
- the service robot 17 detects, as described, for example, for the detection of emotions, areas of the patient in which the skin surface of the patient is to be recorded.
- the skin surface to be recorded is illuminated by the at least one camera 185, for which purpose LEDs are used in one aspect, which together map a light spectrum in the range of 550-1600 nm, for example at least 900-1200 nm and thus lies in the infrared range.
- the sensor of the at least one camera 185 is a sensor based on indium gallium arsenide or lead sulfide, which is supplemented in one aspect with a sensor based on silicon, which is optionally integrated in a further camera 185 is.
- a laser is used in place of the LEDs.
- the light sources can be controlled in such a way that the wavelength of the light sources varies over time.
- the at least one camera 185 detects the emissions of substances that are on or within the skin, stimulated by the light.
- a feature extraction takes place during the measurement, which determines the phase and frequency of the substances emitted by substances on and within the skin, and in a further aspect also takes into account the frequencies of the emitted light.
- Pre- and / or post-processing can take place here, in which different filters are used, for example band and / or low-pass filters.
- steps 3025 to 3050 shown in FIG. 30 and described in more detail elsewhere can also be run through here. Concentrations of the substances are then determined on the basis of a feature classification.
- the system for substance measurement is illustrated in FIG. 68 as follows:
- the system for measurement of substances on and / or within the skin of a person, in one aspect a service robot 17, comprises a detector 195 with an evaluation laser 5205 and another Laser 5210, where the evaluation laser is deflected from entering a medium 5215 such as a crystal surface and the further laser 5210 excites a substance while varying the wavelength, the area of the excited substance at the point with the medium 5215 such as the crystal is engaged, at which the evaluation laser 5205 is deflected, further comprising a laser variation module 5225 for feature extraction and feature classification of the wavelength variation of the further laser 5210 and a laser deflection evaluation module 5220 for evaluating the deflection of the evaluation laser.
- the system has, for example, a sensor for contactless detection of a person, a movement evaluation module (120) for evaluating recorded movements of the person over time and / or a finger positioning detection module 5230 for automated detection of the positioning of a finger on the medium 5215 and Carry out the measurement after placing the finger on the medium.
- the system for measuring substances on and / or within the skin of a person for example a service robot 17, comprises a detector 195 with a medium 5215 consisting of a crystal with a cubic, hexagonal or tetragonal lattice structure, a refractive index of 1- 4 and a spectral width that lies in the interval between 100nm-20,000nm.
- the system can further comprise an evaluation laser 5205 and a further laser 5210, wherein the Evaluation laser 5205 is deflected from the crystal surface and the further laser 5210 excites a substance while varying the wavelength, the area of the excited substance engaging with the medium 5215 at the point at which the evaluation laser 5210 is deflected.
- the system can further include a laser variation module 5225 for feature extraction and feature classification of the wavelength variation of the further laser 5210 and laser deflection evaluation module 5220 for evaluating the deflection of the evaluation laser 5205.
- the evaluation laser is evaluated by means of a sensor on the basis of the photo effect 5250.
- the system can furthermore comprise an interface for the transmission of data into a patient administration system 160.
- the detector 195 can be positioned on an actuator 4920 and the system can have a module with rules to position the detector 195 on the skin of a person, for example by comparing the positions of the actuator 4920 and the position on which the actuator is positioned is to be, and control of the actuator in such a way that the distance between the actuator 4920 and the position on which the actuator 4920 is to be positioned is reduced to at least close to zero.
- the system can have a sensor for contactless detection of a person such as a 2D or 3D camera 185, a LID AR 1, a radar and / or an ultrasonic sensor 194.
- the system can include a body region detection module 4810 and a body region tracking module 4815 for tracking the measurement region.
- the system for measuring substances on and / or within the skin of a person is, in one aspect, equipped with a camera 185 and a tilting unit (5130) for horizontal and / or vertical alignment of the camera 185, with a body region detection module (4810) and a body region tracking module (4815) for the identification and tracking of a person over time (identical in one aspect to the person identification module 111 and the tracking modules 112 and 113), with at least one light source 5270 to detect the To illuminate the skin of the person, the system having a wavelength variation unit 5275 for varying the wavelength of the light emitted by the at least one light source, and a wavelength variation evaluation unit 5280 for evaluating the variation in the wavelength of the detected signals.
- the at least one light source 5270 can be a laser (identical in one aspect to lasers 5205 and / or 5210) and / or a plurality of LEDs with different spectra, which can be controlled accordingly.
- the wavelength of the emitted light is between 550 and 1600 nm, for example 900 and 1200 nm.
- the camera 185 can, for example, have a photodetector made of indium gallium arsenide or lead sulfite.
- the system has one Another camera 185 for the detection of light in the 400-800 nm spectrum.
- the system can, for example, via a substance classification module 5295 for feature extraction and feature classification of recorded data and comparison of the classified data with a substance classification Light by comparing evaluated features with stored features.
- the system has a person recognition module 110, person identification module 111, a tracking module (112, 113),
- Movement evaluation module 120 skeleton creation module 5635 and / or skeleton model-based feature extraction module 5640.
- the service robot 17 In the environment in which the service robot 17 moves, it can happen that the floors on which the service robot 17 and a person tracked by the service robot 17 move are damp, for example due to cleaning work or the spillage of liquids . With a view to people who are instructed to train by the service robot 17, such moist surfaces can represent a danger that is associated with an increased risk of falling.
- the service robot 17 has, in one aspect, a corresponding sensor system to detect moisture on the floor.
- various sensor technologies are described that can be used here:
- Yamada et al 2001 "Discrimination of the Road Condition toward Understanding of Vehicle Driving Environments", IEEE Transactions on Intelligent Transportation Systems, vol 2 (1), March 2001, 26-31 (DOI: 10.1109 / 6979.911083) describe a Method for the detection of moisture on the ground by means of polarization of the incident light. They use the Brewster angle (53.1 °) as the angle of inclination in order to set the reflection to 0 in the horizontal polarization plane, while the vertical polarization shows a strong reflection. On the basis of the ratio of the measured intensities of the horizontal and vertical polarization, it is determined to what extent there is moisture on the measured surface.
- 10.1109 / IVS.2008.4621205 suggest an approach that does not involve a polarization filter and is based on image parameters such as contrast, brightness, sharpness, hues and saturation, which are extracted from the images as features.
- the brightness is taken into account by the Koschmieder model established in image processing, with the brightness primarily depending on the attenuation and scattering of the light.
- the contrast is determined from the difference between local extremes of brightness, with the lightest and darkest pixels in the observed region being compared with one another.
- the approach is based on the tenacity criterion established in image processing. Color tones and saturation are determined using defined groups of pixels.
- a histogram with 10 areas is generated for each feature and a vector is derived from this, which contains the results of the features.
- These vectors can be classified using machine learning / artificial intelligence methods, including k-NN, neural networks, decision trees, or support vector machines. Initially, pre-labeled data is available for training the algorithms, and the classifications determined in the process allow future recordings of the floor to be assessed in terms of the extent to which the floors are moist.
- US patent application No. 2015/0363651 A1 analyzes the texture of the surface recorded by a camera in order to check it for moisture, comparing two images recorded at different times and extracting features.
- the features include the spatial proximity of pixels within a region (i.e. a search is made for recurring patterns), edge detection and their spatial orientation, the similarity of gray levels between the images, the established Laws' texture energy measure, autocorrelation and power density Models, as well as texture segmentations (both region-based and border-based, ie edges that lie between pixels with different textures).
- McGunnicle 2010, "Detecting Wet Surfaces using near-Infrared Lighting, Journal of the Optical Society of America A, vol 27 (5), 1137-1144 (DOI: 10.1364 / JOSAA.27.001137) uses infrared diodes with a spectrum approx. 960 nm and records the light emitted by these diodes using an RGB (CCD) camera in order to evaluate the light spectrum accordingly.
- McGunnicle can show that damp surfaces emit a characteristic spectrum, which allows moisture to be detected on the surface.
- Another approach uses radar waves instead of light in the visible or invisible range, in particular ultra-wideband radar waves, which are used for substance analysis. The reflected signals can be evaluated (ie classified) as described in the prior art, with characteristic features being recognized when measuring moisture on surfaces, so that the type of moisture is detected.
- the sensors on the service robot 17 are arranged in such a way that the sensors detect at least the surface in front of or below the service robot 17, in one aspect also facing laterally or backwards.
- the algorithms for determining moisture are stored in the memory 10 of the service robot 17, for example as values in a database that allow the detected light spectrum in the infrared range to be spectrally or the radar waves emitted by the service robot 17 and reflected from the surfaces to evaluate.
- the service robot 17 has a self-learning system 3100 (see FIG. 31) in order to differentiate between moist and dry soils. This self-learning system is particularly helpful, for example, for the optical methods that determine the texture and / or reflections of the surface.
- the service robot 17 drives the surfaces from 3110 over which the service robot 17 usually moves while these surfaces are in the dry state.
- the service robot 17 draws the surfaces on 3120 by means of at least one integrated sensor and performs a feature extraction 3130, for example according to the approaches of Roger and Mossmann or according to the teaching of US patent application No. 2015/0363651 A1. This is preferably done at different times of the day in order to be able to take different lighting conditions into account (daylight, artificial lighting and / or combinations thereof).
- a feature extraction 3130 for example according to the approaches of Roger and Mossmann or according to the teaching of US patent application No. 2015/0363651 A1. This is preferably done at different times of the day in order to be able to take different lighting conditions into account (daylight, artificial lighting and / or combinations thereof).
- an input device which in one aspect is connected to the service robot 17 via an interface 188 (such as WLAN), the recorded measured values are assigned a value that characterizes the recorded surfaces as dry (labeling 3140). This value, together with the recorded measured values, is stored in the memory 10 in step 3145.
- the service robot 17 at least partially traverses the previously traversed surfaces again, but these previously traversed surfaces are moist.
- an input device which in one aspect is connected to the service robot 17 via an interface 188 (such as WLAN), is used to assign the recorded measured values a value which the recorded surfaces as characterized wet (Labeling 3140).
- a feature classification 3160 of the features recorded by the sensors is then carried out, as shown, for example, in Roger and Mossmann. As a result, surfaces are detected as wet or dry in step 3170.
- the results of the feature classification ie whether the surfaces are moist or dry, are stored in the memory of the service robot 17 in step 3180.
- the service robot 17 can access the stored classifications (e.g. for radar reflections, infrared spectral absorptions, light reflections or textures via camera recordings) and, by means of these stored classifications, evaluate whether the measured values recorded by its sensors Surfaces are damp.
- stored classifications e.g. for radar reflections, infrared spectral absorptions, light reflections or textures via camera recordings
- the service robot 17 shows the navigation of the service robot 17 with the detection of moisture on surfaces 3200. If the service robot 17 navigates accompanied by a person 3210, for example a patient during training, the service robot 17 draws the surface properties of the floors on 3120, over which the service robot 17 moves. A feature extraction 3130, a feature classification 3160 and associated therewith a moisture detection 3170 take place.
- the service robot 17 can, in one aspect and depending on the type of sensors used or the implemented evaluation algorithms, in an optional step by a rotary movement 3220, for example around a vertical axis of the service robot 17, the width of the moisture or the dry area 3230 capture. This embodiment can, in one aspect, be stored in the movement planner 104.
- a tilting unit 5130 can be used instead of a rotary movement of the service robot 17 or the detection angle of the sensor is sufficiently wide to detect the area in the direction of travel even without movement.
- the width is determined here, for example, orthogonally to the direction of travel.
- the width of the dry (alternatively the humid) area is compared 3240 with a value stored in a memory.
- the width of the humid area is determined, for example in relation to the width of the room in which the service robot 17 is moving. If the detected width of the dry area is less than the width stored in the memory, the service robot 17 does not drive onto the damp surface, but instead stops and / or turns in step 3250, as is stored in movement planner 104.
- an output takes place via the output unit (display 2, loudspeaker 192, possibly also projection device 920 / warning lights), which indicates the surface identified as moist.
- the service robot 17 sends a message to a server and / or a terminal in step 3260 via an interface 188 (such as WLAN). If, on the other hand, the detected dry area is wider than the threshold value, the service robot 17 navigates through the dry area in step 3270, as is stored in the movement planner 104. The service robot 17 maintains a minimum distance, stored in the motion planner 104, to the surface detected as being wet, a 3280.
- the service robot 17 can point the accompanied person to the person being accompanied via an output unit (display 2, loudspeaker 192, possibly also projection device 920 / warning lights) indicate damp surface 3290.
- the classification of humidity also includes the degree of humidity. For example, even on surfaces that are perceived as dry, there can be a very thin film of moisture, which, however, has almost no effect on the friction that an object would experience on the surface.
- the detection and evaluation of moisture on surfaces can be summarized as follows: Detection of a surface such as the floor, classification of the surface structure for the detection of moisture on the surface, segmentation of the recorded surface into moist and non-moist areas, Determination of the width of the recorded areas, evaluation of the width of the recorded areas by comparison with at least one stored value.
- FIG. 81 An alternative sequence can be summarized as follows, as illustrated in FIG. 81: Detection of a surface 6005, surface classification for moisture detection 6010, surface segmentation into moist and non-moist areas 6015, storage of moist areas in a map 6020, determination of area with minimum dimensions 6025 and based thereon an output via an output unit 6030, transmission of a message 6035 and / or modification of a value in memory 10 (step 6040). Alternatively and / or in addition, a modification of a path planning and / or movement planning 6045 can take place. 82 also illustrates part of the process. The service robot 17 moves in a corridor 6071 along an initial planned path 6072 (see FIG.
- the service robot 17 plans a new path 6073 in the path planning module 103, which is calculated on the basis of the moisture as an obstacle.
- the service robot 17 compares the width 6074 between surface segments that are stored as an obstacle and determined to be moist, for example resulting from a map stored in the card module (107), maintains, for example, safety distances to these surface segments recorded as moist and follows the newly calculated path ( see Fig. 81 c).
- an area segment detected as wet is so wide that the service robot cannot navigate around it because the width between the area classified as wet and the walls of the aisle is less than the width of the service robot 17 , which is why the service robot 17 stops before this.
- a system for moisture detection is described as follows, as illustrated in FIG. 69: It comprises a sensor for contactless detection of a surface (e.g. camera 185 or a radar sensor 194), a segmentation module 5705 for segmenting the detected surface, a moisture determination module 5305 for Classification of the segments in terms of moisture on the surface and a
- Moisture evaluation module 5310 for evaluating the dimensions of the classified segments of the surface. Furthermore, it can comprise a map module 107 that contains obstacles in the area of the system and the segments classified with regard to moisture.
- the system comprises a movement planner 104 and / or a path planning module 103, for example an output unit (2 or 192) and outputs stored in the memory 10 for indicating the area detected as wet.
- the system can be a service robot 17, e.g. accompanied by a person.
- the system for detecting the position of moisture on a surface comprises a unit for
- Determination for example a camera 185) and a moisture determination module 5305 for
- Moisture on the surface and a moisture evaluation module 5310 for evaluating the dimensions of the classified segments of the surface. For example, classified areas are evaluated in such a way that the width of the moisture is evaluated approximately perpendicular to the direction of movement of the system and the width of dry and / or moist areas is determined.
- the system can, for example, have a movement planner 104 with rules in order, when a dry area is determined, the width of which is one in the memory 10 exceeds the stored value to move through this area.
- the movement planner 104 can, for example, have rules for determining a minimum distance from the damp area, for example by entering the areas classified as damp in a map and comparing its own position with the map.
- the system has an output unit (2 or 192) and rules stored in the memory 10 for indicating the area detected as moist and / or warnings.
- the movement planner 104 can, for example, have stored rules that instruct the system to abort its movement in a predefined destination direction if the determined width of the moist area exceeds a threshold value or the width of the determined dry area falls below a threshold value, similar to rules from the state technology, in which a mobile system moves towards an obstacle.
- the system can have a unit for sending a message to a server and / or terminal 13.
- the service robot 17 has fall detection, i.e. the service robot 17 is configured in such a way that the service robot 17 can detect falls of people directly or indirectly.
- This evaluation of fall events 3300 is shown in FIG. 33.
- Indirect means that the service robot 17 uses external sensors, direct means an evaluation using its own sensors.
- a person is equipped with a sensor unit for fall detection, i.e. the service robot 17 is connected via an interface 188 (such as WLAN) to an external fall sensor which is located on the person to be monitored
- a sensor unit for fall detection i.e. the service robot 17 is connected via an interface 188 (such as WLAN) to an external fall sensor which is located on the person to be monitored
- This sensor unit contains at least one control unit, a power source, possibly a memory, an interface 188 (such as WLAN), and at least one inertial sensor for detecting the movements of the person 3315, for example an acceleration sensor.
- the signals of the inertial sensor are within the sensor unit in step
- the signals are sent to the in step 3320
- Service robot 17 is transmitted so that an evaluation of the signals in the service robot 17 is implemented 3330.
- the recorded measured values are classified in step 3335 as to whether the person has fallen. This classification can for example.
- Acceleration take place that is above a defined threshold value. Based on a
- Fall detection then takes place in step 3345 via an interface 188 (such as WLAN) a
- Notification ie an alarm system is notified and / or an alarm, for example (for example an alarm sound) triggered, etc. If the classification of the detected movements takes place within the sensor unit, the notification and / or alarm is carried out by the sensor unit (via an interface 188 (such as WLAN)). If the service robot 17 performs the classification of the movements, it triggers the notification and / or initiates the alarm.
- an interface 188 such as WLAN
- the sensor unit is designed such that the sensor unit detects the person's movements for the purpose of detecting the severity of the fall, collects measured values and classifies the measured values in step 3340 directly within the sensor unit and / or via the service robot 17 . Specifically, this means that the extent to which the person who is equipped with the acceleration sensor continues to move is recorded. For this purpose, acceleration and / or orientation data of the sensor unit can be evaluated. For this purpose, rules are stored in the memory of the sensor unit and / or of the service robot 17, which, based on the measured movement data, trigger different notifications.
- the notification and / or the alarm can be modified, for example in such a way that the Notification priority is reduced.
- the notification and / or the alarm can be modified, for example the priority of the notification can be increased.
- the notification or the alarm is only issued after the person's movement behavior has been analyzed after the fall, i.e. several seconds after the actual fall, which may reduce the notifications associated with the fall event.
- the service robot 17 is equipped with a wireless sensor unit in order to determine fall events of a person in step 3350.
- This sensor unit can be a camera 185, for example a 3D camera, a radar sensor and / or an ultrasonic sensor 194 or a combination of at least two sensors.
- the sensor unit is used, for example, to identify a person and / or to track the person over time in step 3355, which is implemented, for example, by means of the visual person tracking module 112 and / or the laser-based person tracking module 113.
- the service robot 17 with a Kinect or Astra Orbbec be equipped, ie an RGB-D camera 185, which is able to create a skeleton model of detected persons in step 3360 by means of the method described in the prior art (e.g. by means of camera SDK, NUITrack, OpenPose, etc.), in which body joints are represented as articulation points and the body parts connecting the articulation points, for example as directional vectors.
- a feature extraction 3365 different orientations of the direction vectors as well as distances from articulation points to the detected surface on which the detected person is moving are determined.
- the service robot 17 evaluates whether the detected person is standing or walking, sitting or, if applicable, has fallen.
- the rules of the feature classification can be fixed in one aspect; in an alternative and / or supplementary aspect, the rules can be learned by the service robot 17 itself. During this learning process, recordings of people who have fallen, as well as recordings of people who have not fallen, are evaluated, with labeling stating which case is involved. Based on this, the service robot 17 can use machine learning / artificial intelligence methods to make classifications that allow future recordings of people to be classified as to whether they have fallen or not.
- a fall is detected, for example, on the basis of the extraction of the following features in step 3367, with the body parts being evaluated, ie classified, with a view to the fall: Distances and changes in distance towards the floor or accelerations derived from the distances or changes in distance (for example, over defined minimum periods of time) of articulation points in a direction, the vertical directional component of which is greater than a horizontal directional component, the vertical directional component preferably pointing in the direction of the center of the earth.
- a detected distance from the hip joint point to the floor of less than 20 cm can be classified as a fall event, a change in distance from, for example, greater than 70 cm to less than 20 cm, or an acceleration of the hip joint point towards the floor, this acceleration being e.g.
- the orientation of at least one direction vector (as a connection between two articulation points) in space or the change in the direction vector in space can also be classified as a fall event.
- the height of the person is determined by the 3D camera. If the height of the person falls below a defined height, a fall event is detected.
- the area that the person occupies on the floor can also be determined. For this purpose, in one aspect, the area can be determined by a vertical projection of the tracked person on the ground.
- the service robot 17 triggers a notification and / or an alarm in step 3345, for example via an interface 188 (such as WLAN), as an acoustic alarm via a loudspeaker 192, etc.
- a notification and / or an alarm for example via an interface 188 (such as WLAN), as an acoustic alarm via a loudspeaker 192, etc.
- the service robot 17 detects people by means of radar and / or ultrasound 194, primarily the outer dimensions of the person are recorded, including their height. If a height reduction is detected, possibly an acceleration of the height reduction, this is classified as a fall event, possibly in connection with falling below threshold values. Alternatively and / or in addition, the service robot 17 can also classify the area of the person that this person (projected vertically in one example) occupies on the floor.
- the service robot 17 also detects the position of the person's head in step 3369. This position is tracked with a view of the ground and / or (the position) of detected obstacles. This means, for example, that walls detected by the sensor unit (camera 185, radar and / or ultrasonic sensor 194) are recorded. Alternatively and / or in addition, the position of the walls can also be determined by means of the LIDAR 1. The service robot 17 compares the horizontal position of the head of the person with the (horizontal) position of walls and / or other obstacles in the room.
- the vertical position is also taken into account.
- the camera 185 can also evaluate the distances between the tracked head and objects in three-dimensional space such as tables.
- the (two-dimensional, essentially horizontally oriented) LID AR 1 would, for example, recognize the table legs, but not necessarily the position of a table top in space, which the person's head could touch in the event of a fall.
- the camera-based evaluation allows the three-dimensional recording of the head and other obstacles in the room and the determination of the distance between these other obstacles and the head, which is evaluated as part of the classification in step 3374. If the service robot 17 detects that the distance between the head and one of the other obstacles falls below a value, a value in the memory of the service robot 17 is modified and, if necessary, a separate notification or a separate alarm is triggered.
- the service robot 17 also tracks the person after their fall and detects the extent to which they stand up or try to stand up again, ie there is a post-fall movement detection and classification in step 3340.
- extent means, for example, that distances to Ground, accelerations in vertical directions opposite to the ground, the orientation of the body part or limb vectors, the height and / or the (projected) area of the person are evaluated.
- the degree of changes in the position of articulation points is also evaluated.
- There is a classification based on the extent to which the person moves or even tries to get up. In this way, values are adapted in a memory of the service robot 17 and, in one aspect, the degree of notification via the interface 188 (such as WLAN) / the alarm is modified.
- the process of evaluating a fall event is as follows: detection and tracking of the movements of a person, detection of a fall event by means of feature extraction and classification of the orientation of the limbs and / or the trunk of the person, detection and classification of the movements of the person after they have occurred Fall and assessment of the severity of the fall event.
- the service robot 17 is also able to use its sensor system to record vital parameters of the fallen person in step 3380.
- An integrated radar sensor such as an ultra wideband radar, for example, can be used for this purpose, as has been explained elsewhere.
- parts of the person's body that are not covered by clothing can be detected using radar and / or camera-based methods, and the person's pulse can be measured in these areas, for example using radar. This information can be used in the classification of the notification and / or the alarm can be taken into account; in one aspect, the vital parameters such as the pulse can also be transmitted with the notification.
- the system for detecting a fall of a person for example a service robot 17, comprises a memory 10, at least one sensor for contactless detection of the person's movements over time, a person identification module 111 and a person tracking module 112 or 113, a fall detection module 5405 for extracting features from the sensor data and classifying the extracted features as a fall event, a fall event evaluation module 5410 for classifying the severity of the fall event.
- the system can also have an interface 188 to a server and / or terminal 13 for the purpose of transmitting messages.
- the fall detection module 5405 can, for example, have a skeleton creation module 5635 for creating a skeleton model of a person.
- the fall detection module 5405 can perform classifications for determining distances or changes in distance from points of articulation to the ground originating from the skeleton model; Accelerations of points of articulation in the vertical direction; the orientation of direction vectors resulting from the connection of at least two points of articulation; the change in orientation of the direction vectors; the height and / or height change in the height of the person, for example via the person size evaluation module 5655, which determines the height of the person, for example by vector subtraction of two directional vectors that extend from a common origin to at least one foot and at least the head of the person ; include the area that a person projected in the vertical direction occupies on the floor, and / or the position of the head of the person relative to the floor and / or relative to detected obstacles.
- the system can also include a vital parameter acquisition unit 5415 for acquiring vital parameters of the person (eg a camera 185, a LID AR 1, a radar and / or an ultrasonic sensor 194) and a vital parameter evaluation module (5420) for evaluation include recorded by vital parameters of the person.
- a vital parameter acquisition unit 5415 for acquiring vital parameters of the person (eg a camera 185, a LID AR 1, a radar and / or an ultrasonic sensor 194) and a vital parameter evaluation module (5420) for evaluation include recorded by vital parameters of the person.
- the system has a person recognition module 110, movement evaluation module 120 and / or skeleton model-based feature extraction module 5640.
- the service robot 17 records vital parameters of the person while a test and / or an exercise is being carried out, as shown in step 3400 in FIG. 34.
- the service robot 17 identifies and tracks the person, for example by means of the visual person tracking module 112 and / or the laser-based person tracking module 113 in conjunction with a camera 185 or a LID AR 1.
- person identification and person tracking are found in step 3355 instead, for which purpose the personal identification module 111 is used.
- the system is (optionally) in front of the person (step 3420) and moves (optionally) in front of the person in step 3430.
- the body region of the person is identified and tracked in step 3440, with which the exercise and / or the test is carried out in order to record vital parameters in step 3450 by measuring this body region.
- the person's face, hands, chest area, etc. come into question as body regions.
- the procedure for detecting such a body region was described elsewhere in this document and / or in the prior art.
- the measured vital parameters are the pulse rate, pulse rate variability, the systolic and / or diastolic blood pressure, or the breathing of the person (such as the breathing rate). How these vital parameters, cited by way of example, can be determined by the service robot 17, for example, has been described elsewhere in this document. However, other methods are also possible in order to determine the vital parameters.
- the vital parameters are recorded with at least one sensor such as the camera 185 and / or the radar sensor (e.g. a microwave pulse radar, a range-controlled radar, a Doppler radar, a continuous wave radar, an ultra-wideband radar) and / or combinations thereof, which records the named body regions of the person and the vital parameters, preferably over time.
- the radar sensor e.g. a microwave pulse radar, a range-controlled radar, a Doppler radar, a continuous wave radar, an ultra-wideband radar
- the senor used here determines movements on and / or under the skin and / or clothing of the person.
- the movements of the skin surface and / or the person's clothing are evaluated relative to the movement of the person with which the person moves to the service robot 17, ie the detected signals of the body region are corrected for the movement of the person 3460.
- the service robot 17 detects the body region that is evaluated for the purpose of determining the vital parameters, at least one other body region and determines the distance of this further body region to the service robot 17.
- the recording of the movements of the body region evaluated for the purpose of evaluating the vital parameters the body region recorded for the purpose of determining the relative movement of the person synchronously with one another
- using the camera 185, for example the RGB-D camera 185, ultrasonic and / or radar sensors 194 can take place.
- the measurement that the service robot 17 takes can be a continuous or discontinuous measurement, i.e. for example at intervals of 10 seconds.
- the measured vital parameters are stored in the memory of the service robot 17 in step 3470 and can be transmitted to other systems via an interface 188 (such as WLAN).
- the service robot 17 compares the determined vital parameters with threshold values stored in a memory in step 3480. These values stored in the memory can be permanently stored and / or result dynamically from past values of the recorded vital parameters, for example as average values of previously recorded values over a time interval be evaluated. If exceeding or falling below the recorded vital parameters is recognized by the threshold values in step 3490, the service robot 17 modifies, for example, a value in a memory. This modification can trigger at least one of the following events:
- An output unit display 2
- Loudspeaker 192, projection device 920, etc. is triggered in step 3492, i.e., for example, a voice output is initiated.
- the service robot 17 can request the person to reduce their speed.
- the service robot 17 can request the person to sit down.
- the service robot 17 can move to a defined position 3498. This can be at least one seat to which coordinates in the map of the service robot 17 are assigned. The service robot 17 can then steer towards this seat.
- the seat can be a chair.
- the service robot 17 can identify the chair directly in its surroundings via its implemented sensor system (as described elsewhere in this document), the service robot 17 can alternatively and / or additionally also be stored on a card of the service robot 17 in the card module 107.
- the service robot 17 can trigger a notification in step 3494, i.e., for example, send a notification via an interface 188 (such as WLAN) and / or trigger an alarm. Furthermore, in one aspect, it can reduce its speed in step 3496.
- the service robot 17 carries out a gait exercise with a person, for example gait training on forearm rests.
- the service robot 17 detects the person by means of at least one sensor and carries out a feature extraction, a feature classification and a gait classification in order to evaluate the gait of the person.
- the camera 185 mounted on the service robot 17 detects the person's face and determines the systolic and diastolic blood pressure over time, which is each stored in a blood pressure memory in the service robot 17. The determined measured values of the blood pressure are evaluated over time, optionally stored and compared with values stored in the blood pressure memory.
- the service robot 17 reduces its speed and sends a notification via an interface 188 (such as WLAN) such as WLAN to a server, which in turn alerts personnel in the vicinity of the service robot 17 and calls for support.
- the service robot 17 optionally tries to detect a chair within its surroundings, ie within a defined distance from its position. If the service robot 17 detects a chair, the service robot 17 slowly navigates to this chair and prompts the person to sit down via an output.
- the service robot 17 detects the breathing rate over time when completing a gait exercise by evaluating the movements of the chest and / or the abdominal area of the person, which is done by means of an ultra-wideband radar sensor that is mounted on the service robot 17.
- the system for recording vital parameters can be described according to FIG.
- the system for recording vital parameters of a person for example a service robot 17, comprises a computer 9, a memory 10 and at least one sensor for contactless recording of the Movements of the person over time (such as a camera 185, a LID AR 1, an ultrasound and / or radar sensor 194, e.g. a person identification module 111 and a person tracking module (112, 113) for detecting and tracking the person Vital parameter evaluation module 5420. It further comprises a body region detection module 4810 and a body region tracking module 4815 for tracking the recording region of the vital parameters and a vital parameter recording unit 5415 for recording vital parameters of the person, for example over time , contactless and / or contact-based.
- the vital parameter evaluation module 5420 can, for example, compare the recorded vital parameters with mi Carry out at least one stored threshold value and, based on the comparison, send a notification to a system via an interface 188, an output via an output unit (2 or 192), a change in the speed of the system (e.g. initiate a speed reduction) and / or a control of a target position of the system.
- the latter are implemented, for example, by a navigation module (110), for example by adapting a path plan to a seat such as a chair that is located, for example, within a defined minimum distance from the system.
- the threshold value used in the vital parameter evaluation module 5420 can be determined dynamically from previously recorded vital parameters, for example based on an averaging of recorded vital parameters over a defined time interval.
- the vital parameter evaluation module 5420 can also record body movements of the person and an evaluation of the recorded vital parameters while comparing the recorded body movements.
- the recorded vital parameters can be the pulse rate, pulse rate variability, the systolic and / or diastolic blood pressure and / or the respiratory rate.
- the system can acquire data from a vital parameter sensor 5425 attached to a person via an interface 188 and evaluate it in the vital parameter evaluation module 5420.
- An application module 125 has rules for performing at least one exercise, for example the exercises stored as examples in this document.
- the recorded and evaluated vital parameters can be used to determine the risk of falling, for example the acute risk of falling, in which a fall is to be expected within a time interval of only several minutes.
- the system has a person recognition module 110, person identification module 111, tracking module (112, 113), Movement evaluation module 120, skeleton creation module 5635 and / or skeleton model-based feature extraction module 5640.
- the service robot 17 is configured in such a way that the service robot 17 evaluates the gait of a person for their risk of falling, as is explained in FIG. 35.
- the person reports to the service robot 17 to 3510, which can be done by means of an input unit, an RFID transponder, a barcode, etc.
- the service robot 17 then carries out a person identification by means of its person identification module and then tracks the person 3355, for example by means of the visual person tracking module 112 and / or the laser-based person tracking module 113.
- a sensor is used for this tracking that enables contactless detection enables the person, for example a camera 185, an ultrasound sensor and / or a radar sensor 194.
- the service robot 17 requests the person whose risk of falling is to be assessed to follow the service robot 17 via an output 3520 of an output unit.
- This service robot 17 is (optionally) in front of the person in step 3420, moves in front of the person (optional) in step 3525 and (optionally) detects their speed in step 3530. In one aspect, this is done by determining the speed of the service robot 17 with synchronous detection of the distance of the identified person, whereby the relative speed of the person to the service robot 17 and the speed of the person is determined via the speed of the service robot 17 itself.
- the speed of the service robot 17 itself is determined via its odometry unit 181 and / or via the tracking of obstacles stored in the stored map of the service robot 17 and the relative movement of the service robot to these obstacles.
- the service robot 17 performs a feature extraction in step 3365 in order to extract features from the skeleton model in step 3360, such as the position of articulation points 3541, direction vectors which connect articulation points 3542, the perpendicular through a person, etc.
- features can also be extracted from inertial sensors that are attached to at least one limb of the person, etc., such as the instantaneous acceleration 3543, the direction of an acceleration 3544, etc.
- the feature extraction 3365 is followed by a feature classification 3370, in which several characteristics are evaluated in combination.
- the speed of the person which can be determined as a feature from the recorded data of the service robot 17 as an alternative and / or in addition to the method described above, the individual classified features, for example, the step length 3551 and / or double step length 3552 of the Can be a person who the service robot 17 records over time and determines the speed over the step length per recorded time unit, with several steps being evaluated in one aspect.
- the step length is extracted in the context of feature extraction 3365 in one aspect via the position of the ankle points in the skeleton model, the skeleton model being carried out in step 3360 the evaluation of camera recordings of the person is created.
- the points in time and the times between these points in time are recorded / extracted at which a circular movement is started by the sensor, the radius of which points towards the ground, ie the direction vectors of the acceleration 3544 are evaluated for this purpose.
- the instantaneous acceleration is determined or extracted in step 3543, preferably in the sagittal plane, and the distance covered is determined via the instantaneous speeds in step 3543 and the time period between the mentioned points in time as part of the feature classification, which then represents the step length in step 3551 . Viewed individually, these are extracted features that are classified in this combination.
- the ankle joint is detected by means of a radar sensor and / or ultrasonic sensor 194.
- the ankle joint point is determined via the position of the knee joint point, a direction vector that is oriented from the knee joint point parallel to the lower leg, and the height of the knee joint point above the floor in the case of the direction vector passing through the perpendicular, the height of the knee joint point above the ground in the case of the passage of the direction vector through the perpendicular indicates the distance at which the point of the ankle is located from the point of the knee.
- Said double step length 3552 is determined via the distances between the recorded ankle points, with the single step lengths 3551 being added successively in one aspect.
- the service robot 17 evaluates the respective length and / or duration for the individual steps within a double step for each double step and relates the length or the duration in step 3553 to one another.
- the service robot 17 adds the recorded values from more than one double step in order to determine an average value over more than one double step.
- the service robot 17 evaluates flexion and / or extension 3554, i.e. the angle of the thigh compared to the perpendicular.
- the service robot 17 then evaluates the recorded speed of the person, the recorded stride length and the double stride length as well as the cadence.
- the standing time of at least one foot is also evaluated in step 3555.
- the track width can also be evaluated in step 3556, the distance between the ankles being evaluated.
- the service robot 17 records further joint points from the skeleton model of the person, such as the head, the shoulder joint points, the pelvis / hip joint points, etc. and records their position in space, for example three-dimensionally, and evaluates these parameters over time.
- this evaluation includes the height of these parameters above the ground, but also their movement in the sagittal plane (both vertically and horizontally). In one aspect, the acceleration of at least one of the named points from the skeleton model is also evaluated 3557 for this purpose.
- the service robot 17 stores the acquired values in step 3570, classifies the gait sequence using the classified features in step 3580 and compares it with a gait sequence classification stored in its memory (or in a memory available via an interface 188 (such as WLAN)) in step 3585.
- a gait sequence classification stored in its memory (or in a memory available via an interface 188 (such as WLAN)) in step 3585.
- at least one of the named classified features preferably several, is (jointly) evaluated and compared with those from the memory.
- the service robot 17 determines a score which reflects the risk of falling in step 3590, for example a probability that the recorded person will fall within a defined period of time.
- the classification includes values for a determined speed, cadence (steps per minute) and step length as a function of person parameters such as person size.
- people are associated with an increased risk of falling if they have an average height of approx. 1 m / s, a cadence of less than 103 steps / min and a step length of less than 60 cm.
- detected accelerations are evaluated in three-dimensional space and the harmonics are formed by means of discrete Fourier transformation. Then the ratio of the summed up amplitudes of the even harmonics is formed by the summed up amplitudes of the odd harmonics.
- values of the lateral acceleration in the frontal plane that are below 1.8 indicate an increased Risk of falling.
- the corresponding evaluations are evaluated in the gait characteristic classification module 5610. For example, several parameters such as acceleration, step length, cadence, etc. are also evaluated at the same time.
- the system for determining a score that describes the risk of a person falling such as a service robot 17, as illustrated in FIG. 72, can be described.
- the system for determining a score describing a person's risk of falling comprises a computer 9, a memory 10 and a sensor for recording the movements of a person over time (including a gait), such as a camera 185, a LID AR 1, an ultrasound and / or radar sensor 194, a movement sequence extraction module 121 and a movement sequence evaluation module 122, configured in one aspect to determine a fall risk score within a fall risk determination module 5430, for example by evaluating accelerations in the horizontal and / or vertical plane, step size, Speed and / or variables derived therefrom, etc.
- the movement sequence extraction module 121 can include a gait feature extraction module 5605 for extracting features of a gait sequence, the movement sequence evaluation module 122 a gait feature classification module 5610 for the feature classification of a gait sequence on the basis of the extracted features (e.g.
- a gait classification module 5615 for a gait classification including, for example, the step length, the length of the double step, the walking speed, the ratio of the step lengths in the double step, the flexion and / or extension, the standing time, the track width, and / or the course (position) and / or the distance between joint points and / or acceleration of articulation points
- the classification comprises a comparison of recorded gaits with gaits stored in the memory as well as the determination of the fall risk score.
- the gait sequence classification module 5615 can comprise a person speed module 5625 for determining the speed of the person, wherein the determination of the speed of the person via the number and step size of the steps taken by the person per unit of time, relative to the speed of a detection and evaluation unit or of the system, with the inclusion of an odometry unit 181, with the inclusion of obstacles recorded in a map and / or relative to the position of obstacles recorded in a map.
- the system comprises a person identification module 111 and a person tracking module (112 or 113) and components (e.g. 2, 186) for logging the person on to the system, with, for example, visual characteristics of the person being stored and within the scope of the Person re-identification module (114) can be used.
- the system can receive sensor data from an inertial sensor 5620 via an interface 188 and evaluate these sensor data in the movement sequence extraction module 121.
- the sensor can, for example, be attached to the person, for example, to the lower limbs, or to a walking aid used by a person, such as an armpit or forearm support, and can detect movements of the walking aid.
- the system has a person recognition module 110, movement evaluation module 120, skeleton creation module 5635 and / or skeleton model-based feature extraction module 5640
- Detection of the gait of a person e.g. by the mentioned sensor for contactless detection of the person, extraction of features of the recorded gait, classification of the extracted features of the gait, comparison of at least two of the classified features of the gait sequence with a gait sequence classifications stored in a memory and determination of a fall risk score.
- the service robot 17 is configured in one aspect in such a way that the service robot 17 can evaluate different body positions and / or movements of a person, while sitting, standing, walking, as shown in FIGS. 36-52, around a get a holistic view of a person's mobility. A number of the method steps can be found in most of the steps, which is why these have been summarized in FIG. 36 by way of example. In step 3525, the service robot 17 can move in front of the person in one aspect while walking and behind the person in an alternative aspect.
- the person can be logged on to the service robot 17 in step 3510, as well as personal identification and person tracking by the service robot 17 in step 3355, for example by means of the visual person tracking module 112 and / or of the laser-based person tracking module 113 in connection with a LID AR 1 and / or a camera 185.
- the service robot 17 can request the person to take certain actions in step 3521 via an output unit, for example to get up, run, etc. , whereby the outputs can be made via the display 2, the voice output, etc.
- This step 3521 is optional or dependent on the respective evaluation.
- the evaluation is preferably carried out in Time lapse, for which defined time intervals are used.
- the service robot 17 uses information from the skeleton model in one aspect, which is created by recording the person with the at least one 3D sensor and / or the camera 185 in step 3360 and can be implemented using SDKs from the prior art.
- a feature extraction takes place in step 3365, which includes, for example, articulation points in step 3541 and direction vectors between articulation points in step 3542.
- a feature classification then takes place in step 3370, which in detail depends in particular on the task of the evaluation.
- the results of the feature classification in step 3370 are (optionally) stored and coherently classified, which in turn is task-dependent, which is why this method step is referred to as a further classification in step 3700 in FIG. 36.
- a threshold value comparison can take place.
- a score is then determined for each task. Recorded data such as data that result from complete or partial evaluation and / or classification can be saved (even in the meantime).
- a plurality of articulation points from the feature classification can be evaluated simultaneously in step 3365 without explicitly specifying angles resulting from the connections of the articulation points.
- the position in three-dimensional space (alternatively, a position estimate for the three-dimensional space based on two-dimensional data) can be made based on classifiers in which body poses were recorded that describe poses identified as correct or incorrect and the classifiers were then determined.
- body poses can be specified which describe a correct sequence and for which the positions and the course of the joint points are evaluated over time.
- the course of articulation points can be evaluated, for example, on the basis of a demonstration of a body pose and a classifier can be created based on this, which is then compared with other recorded, correctly specified body poses and the resulting courses of the articulation points in space, after which again a classifier is created that takes into account all existing joint point course data.
- a classifier For example, the Dagger algorithm in Python can be used for this purpose.
- a neural network creates a classifier that recognizes a correct movement and consequently also movements that do not proceed correctly.
- Body poses that are evaluated and classified are (not limited to) those in the following paragraphs, including sitting balance, standing up, trying to get up, standing balance in different contexts, the beginning of the aisle, gait symmetry, step continuity, path deviation, torso stability, rotation through 360 °, sitting down or taking down, the use of forearm supports , etc.
- the service robot 17 detects the person and evaluates the extent to which a seated person leans to one side, slides on a chair or sits securely or stably.
- Features of the skeleton model are extracted, e.g. the joint points of the knees, pelvis, shoulder, head, etc. and direction vectors between the individual joint points are used to record and evaluate the orientation of the body parts / limbs of the person.
- the direction vector between at least one shoulder and at least one hip joint point is evaluated (preferably on one half of the body in each case; and / or parallel to the spine) and its deviation from the vertical / perpendicular 3601 in FIG. 37.
- the orientation of the person is evaluated, i.e. in this case, at least one direction vector between the shoulder points, the hip points, for example also the knees, etc. is recorded in step 3603. More than one direction vector is preferably recorded. This direction vector is used, for example, to determine the frontal plane of the person in step 3602 which runs parallel to this direction vector.
- the position of the hip in space is recorded and deviations over time in the transverse plane are evaluated in step 3604. This is used to determine the extent to which the person is sliding back and forth on his or her seat, for example.
- the deviation and / or inclination of the direction vector between at least one shoulder and at least one hip joint point from the vertical / perpendicular in the frontal plane is evaluated in step 3711. Furthermore, the change (amplitude, frequency, etc.) in the position of the shoulder joint points in the transverse plane is determined in step 3712.
- a threshold value comparison takes place in step 3713 via these two steps 3711 and 3712 and / or a comparison with patterns such as, for example, movement patterns. If at least one of the determined values is greater than a threshold value (such as 1.3m), the measurement result is considered to be a low one Seat balance classified in step 3714, otherwise as a high seat balance in step 3715. For this, a score is assigned in each case in step 3716, which is stored in a seat value memory.
- the service robot 17 evaluates the extent to which the person is able to get up (see also FIG. 38). As part of the feature extraction, the service robot 17 identifies objects and / or obstacles in step 3545, as is described in the prior art. As an example, the service robot 17 extracts the point cloud near the tracked wrist points and carries out a segmentation of the point cloud, which allows the hands to be differentiated from the objects. This segmentation takes place preferably in real time (with, for example, 30 fps).
- the recorded point cloud can also be compared with point clouds stored in a memory, to which objects are assigned, for example, in order to establish an association between sensor-recorded objects and their semantic meaning, which in turn makes it possible to classify certain objects as more relevant than others , for example a chair with armrests or a walking aid compared to, for example, a vase.
- the standing is determined in step 3610.
- a distance measurement between the head and the floor is carried out in step 3611, for example on the basis of the position of the head joint point and at least one ankle joint point.
- these values are compared, if necessary, with values stored in a memory and / or with a threshold value and / or pattern. If the determined height is greater than the threshold value (e.g. 1.4m), the person is classified as standing in step 3616, otherwise as sitting in step 3617.
- the threshold value e.g. 1.4m
- the orientation of the Direction vectors between the at least one foot and at least one knee joint point, the at least one knee and at least one hip joint point and at least one hip and at least one shoulder joint point are evaluated in step 3612, in the event that these three direction vectors are essentially parallel to one another, such as For example, a threshold value comparison 3615 and / or pattern comparison can show, the threshold value being calculated, for example, on the basis of the deviation from the parallel.
- the orientation of a direction vector between at least one knee and at least one hip joint point is evaluated to what extent this point Direction vector is substantially perpendicular. If this deviation from the parallel and / or from the vertical rake is classified as smaller than the threshold value, the service robot 17 detects these features as a stand in step 3616, otherwise as sitting 3 in step 617.
- step 3620 it is detected whether a hand is using an aid, an aid being broadly understood here as a walking aid, an armrest of a chair, a wall, etc., ie everything that a person can use to help yourself to get up.
- the distance between at least one wrist and at least one of the extracted objects is determined in step 3621. If the distances between the at least one hand and the object or objects or obstacles fall below a threshold value 3622 (e.g. 8 cm), this is classified as using aids in step 3623, otherwise as not using aids in step 3624.
- a minimum distance is used to the body of the observed person, ie to the articulation points and / or direction vectors that connect the articulation points with one another.
- the service robot 17 classifies as follows: is the person in the stand 3721 after a defined time or an input is made by a person 3722, in particular an input that the person is unable (alone ) to get up, the situation is classified as a situation in step 3723 in which the person needs help. If the stand occurs within a defined time in step 3724 and the person uses tools in step 3623, the person is classified as a person in step 3725 who needs tools to get up 3725.
- the third case, which is classified here is the case that the person does not need any aids in step 3624 and reaches the level in step 3724 within a defined period of time, whereby the person can get up without aids in step 3726.
- a get-up score 3727 is determined based on steps 3723, 3725 and 3726.
- Feature classification is therefore, in addition to the feature classification in FIG. 38, the
- the knee-hip direction vector is evaluated for its horizontal position, ie to what extent the service robot 17 is parallel to the transverse plane.
- the following steps take place: If, on the basis of the information from the attempt to stand up feature classification 3370, no stand is detected within a defined time in step 3731, or if the person makes an entry in step 3732 and (in comparison to steps 3731 and 3732) if no aid is detected in step 3624, the person is classified in step 3733 in such a way that it is not possible to get up without help. If no aids are detected in step 3624 and the local maxima are not equal to the global maximum and the number of local maxima is greater than 1, several attempts to stand are detected in step 3735.
- the course of the joint points that define the stance is evaluated over time and / or the angle or the change in angle of the direction vector between the hip and the knees with reference to the horizontal (alternatively: vertical), with the horizontal being described via the transverse plane. If, for example, it is detected twice that the angle changes from approx. 0 ° (transverse plane) to approx. 30 ° (change in one direction of rotation), but there are then changes in another direction of rotation (for example again 30 °) , and only then an angle change of »30 ° is detected, for example 90 °, three attempts to stand are detected (the last of which was successful).
- step 3624 If, on the other hand, no aids are detected in step 3624, status 3616 takes place, the situation is classified in step 3736 as a situation in which the person does not need any aids. Overall, based on steps 3733, 3735 and 3736, a standing attempt score 3737 is assigned.
- the service robot 17 evaluates the standing balance of a person, as FIG. 40 shows.
- a balance determination 3630 takes place in the feature classification 3370.
- the amplitude, orientation and / or frequency of the change in position of at least one of the shoulder joint points, at least one of the hip joint points or at least one of the ankle joint points in the transverse plane 3631 is evaluated over time (e.g. for 5 seconds) and a threshold value comparison takes place in step 3632 and / or a comparison with patterns such as movement patterns.
- the step length and / or the existence of steps can also be evaluated in one aspect.
- the amplitude, orientation and / or frequency of the position change are less than the threshold value 3632 (such as a lateral fluctuation of 10 cm) and / or they do not correspond to a pattern, stability 3635 is assumed, otherwise of instability 3636.
- the deviation (amplitude, orientation and / or frequency of at least one direction vector can be out of perpendicular and / or in the sagittal and / or The frontal plane can be evaluated over time (for example, for 5 seconds) in steps 3633 3631.
- the joint points above include, for example, at least one shoulder joint point and a head joint.
- the standing balance classification in step 3740 classifies the person with an unsafe stance 3741 if the person is standing 3616 and is unstable 3636.
- the person is assigned to the class " Safe stand with aids "assigned if the person is standing 3616, uses aids 3623 and is in balance, s stable 3635 is.
- a secure stance without aids 3743 is assumed if the person stands 3616, does not use any aids 3624, and stands stable 3635.
- a standing balance score of 3744 is assigned based on this classification.
- the ankle points and / or the knee joint points are used from the position of the extracted joint points 3541, in one aspect also the orientation of the direction vectors 3542 between hip joint point and knee joint point and / or knee and ankle joint point.
- the distance between the ankle points 3641 is determined on the basis of this data, in one aspect within the frontal plane.
- a threshold comparison 3642 and / or a pattern comparison is then used to classify whether the feet are wide 3643 or close together (i.e. a small distance 3643), whereby the threshold value can be, for example, 12 cm (from the center of the joint to the center of the joint).
- the stance is classified into three classes: In the first class (unsafe stance 3746), the people who are standing 3616 and are unstable 3636 are grouped. In the second class, the standing people 3616, who stand steadily 3635, with aids 3623 or wide foot distance 3644 are classified. Third grade will be the people assigned who are standing 3616, with a stable balance 3635, and that without aids 3624 and with a small foot spacing 3643. This classification results in a standing balance foot spacing score 3749.
- first class unsafe stance 3746
- the standing people 3616, who stand steadily 3635, with aids 3623 or wide foot distance 3644 are classified.
- Third grade will be the people assigned who are standing 3616, with a stable balance 3635, and that without aids 3624 and with a small foot spacing 3643. This classification results in a standing balance foot spacing score 3749.
- the ankle points may not be obtained directly from the data of the SDK, which extracts the skeletal model, but alternatively via the knee joint point.
- the position of the knee joint point, the direction vector, which is oriented from the knee joint point parallel to the lower leg, and the height of the knee joint point above the floor in the case of the passage of the direction vector are determined by the perpendicular determined, wherein the height of the knee joint point above the ground in the case of the passage of the direction vector through the perpendicular indicates the distance at which the ankle joint point is seen from the knee joint point.
- the service robot 17 detects the person who has at least one impact on the standing balance
- Hip forward evaluated in step 3651 i.e. within the sagittal plane.
- the standing balance at impact is then evaluated using the standing balance-impact classification 3750.
- the standing balance is safe / stable 3753, which is characterized by the standing 3616, the stability of the balance 3635, no aids 3624 and a small foot distance 3643 after a push 3655 b) the person shows one
- the standing balance is recorded and evaluated with the eyes closed.
- the service robot 17 can capture the face of the person, whose eyes, and through changes in color, color contrast, etc., which are captured by an RGB camera, differentiate between closed and open eyes.
- the service robot 17 outputs an output, for example acoustically, which prompts the person to close his eyes.
- the movements are recorded after detection of the closed eyes and / or the output.
- the standing balance is determined in the same way as in FIG. 42, except that no impact is evaluated, and a stable or unstable stance is classified as a result, which leads to a standing balance eye score.
- the service robot 17 records the gait behavior of the tracked person, preferably after an output that contains a request to walk, and determines the time until the gait begins, as shown in FIG. 43.
- Aisle determination 3660 takes place within the framework of the feature classification 3370.
- the change in position of the shoulder joint points / hip joint points, ankle joint points in the transverse plane and / or the distances between the ankle joint points 3661 is determined over time.
- a threshold value comparison takes place in step 3662 and / or a pattern comparison, and if the threshold value is exceeded (e.g. 10 cm) a walking and / or attempted walk is assumed 3666, otherwise not 3665.
- the curve profile of articulation points in the sagittal plane 3663 can be evaluated, wherein threshold values and / or curve comparisons 3664 or pattern comparisons can be used. Based on this, the movement is classified into walking and / or walking attempts 3666 or no walking 3665. In one aspect, attempts to walk are detected by the fact that the movement in the sagittal or transverse plane is relatively slow and / or discontinuous, with relatively slow implication that the value falls below a threshold value. As part of the start of aisle classification 3755, the length of time between the request and walking motion 3756 evaluated.
- this walking movement is above a threshold value (such as 2 seconds) and / or different walking attempts are detected in step 3666, this is classified as dragging / different attempts in step 3757. If this walking movement takes place within a time interval that is below the threshold value, this is classified as no hesitation in step 3758. The result is assessed with a start-of-walk score 3759.
- a threshold value such as 2 seconds
- the service robot 17 evaluates the walking movement of a person (as described, for example, in the previous section), as also shown in more detail in FIG. 44, by the step lengths of the left and / or right To determine the leg.
- the service robot 17 records the distance between the ankle points relative to one another over time, the maxima occurring in the sagittal plane corresponding to the step length 3672.
- the service robot 17 alternately evaluates the position of the ankle points in relation to one another in the sagittal plane.
- the foot length is taken into account in the step position classification 3760 that takes place below, for which purpose the foot length is determined in step 3675. In one aspect, this is interpolated over the height or size of the person, with different foot lengths being stored in a memory for different sizes of a person, i.e. reference values from the memory are used for this in step 3676.
- the step length is related to the foot length in step 3761.
- the position of the respective ankle points in the sagittal plane is evaluated when running through the stance phase, and in step 3762 it is compared in which position the ankle points are relative to one another 3762, the position data originating from step 3661.
- leg under consideration is placed in front of the foot of the other leg 3763 or not 3764. It is placed in front of the foot of the other leg if the comparison of stride length and foot length in step 3761 shows that the
- the stride length is shorter than the foot length and / or if the ankle point position of the leg under consideration in the gait direction in the sagittal plane is not in front of the foot of the other leg is set as the position of the ankle points results in 3762 when running through the stance phase.
- a step position score 3765 is assigned on the basis of this classification. Such an evaluation can, in one aspect, take place separately for each leg.
- Stand 3616 is to be understood here (and also in the further (for example, following) evaluations in which it is a question of walking that the person is essentially in an upright position in which the person is on a Otherwise, the described procedures could record movements of the person that would not generally be referred to as walking.
- the service robot 17 follows the person during the aisle or drives in front of the person 3525, the service robot 17 adapting its speed to the speed of the person 3530 in one aspect, with a possibly discontinuous speed of the person turning into a continuous one Speed of the service robot 17 is implemented, for example.
- Speed of the service robot 17 is implemented, for example.
- the service robot 17, as shown in FIG. 45 detects the walking movement of a person and classifies the extracted features so that a foot height determination 3680 (above the ground) takes place.
- the amplitude height of the ankle joint points and / or knee joint points plus directional vectors are evaluated over time in step 3681 and, for example, in the sagittal plane, with these being used for deriving the ankle joint points with a view to the knee joint points, as already described above.
- the curve profile of the ankle joint points and / or knee joint points plus direction vectors is evaluated in step 3682.
- the increases / decreases in the amplitudes which serve as a proxy for the step height, are evaluated and a comparison with threshold values and / or reference data takes place in step 3683
- a higher probability implies that the leg will be lifted off the ground as much as possible, with a movement that is more like a trapezoidal movement with a higher probability of a grinding movement in which the foot does not move off properly Floor is lifted off.
- the recorded step heights are evaluated using a threshold value comparison 3771 and / or a pattern comparison. If the step height falls below the threshold value (e.g.
- Curve courses can, in one aspect, also be inferred directly from this that the foot is lifted or not lifted.
- the results of the classification flow into the step height score 3774. Such an evaluation can, in one aspect, take place separately for each leg.
- the service robot 17 evaluates the symmetry of the course of the aisle when it detects the aisle, as has been described, for example, in the preceding sections (see also FIG. 46), which is within the scope of aisle symmetry.
- Classification 3775 takes place.
- This gait symmetry classification 3775 uses in particular data from the step length determination 3760, ie the step lengths 3762 and evaluates this in one aspect when the person is standing 3616 or walking 3666.
- the step length ratio is compared to a threshold value 3776 and / or movement patterns are evaluated over time.
- the symmetry of the stride lengths per double step is evaluated, with a double step taking place by adding a step of the left and right leg (or vice versa).
- the step length ratio can, in one aspect, be formed as the ratio of the individual step lengths to one another, or, in another aspect, as the ratio of a single step length to the double step length. If the respective ratio is below a threshold value or if a comparison with patterns shows, for example, a high pattern similarity, the gait is classified as symmetrical 3777, otherwise as asymmetrical 3778.
- a step length ratio of 1: 1.1 or less (or 60:66 cm when referring to the single step or 60: 126 cm when referring to the double step) classified as symmetrical, larger proportions as asymmetrical.
- the classifications are then converted to a gait symmetry score 3779.
- the service robot 17 evaluates when detecting the aisle, as for example in the previous sections has been described, the step continuity (see Fig. 47).
- the position of the ankle points can also be determined in the stance phase 3673 in one aspect.
- the curve profile of the articulation points in the sagittal plane 3663 is evaluated with a view to the symmetry of the curve profiles 3781.
- a high degree of symmetry is classified as a continuous gait 3784, otherwise as a discontinuous gait 3785.
- the step lengths 3672 are evaluated with simultaneous detection / extraction of the points at which the feet touch the ground, i.e. the position of the ankle points is determined in the stance phase 3673. If the service robot 17 detects, for example, that at the times at which the left and right feet (or vice versa) touch the ground, the distance between the ankle points falls below a threshold value (e.g.
- this step continuity is also classified as a discontinuous gait 3785. This is the case, for example, if the person always puts one foot in front and pulls the second foot behind so that both feet are approximately parallel when standing. Alternatively and / or in addition, such a case can also be detected by the service robot 17 if both legs are in parallel (in the sagittal plane) beyond a defined time threshold 3783. The classifications are then converted into a step continuity score 3786.
- the service robot 17 evaluates the deviations of the aisle from a line when it detects the aisle, as was described in the previous sections, as shown in FIG is virtual or real. Via the output 3521, the person is requested to move along a line that is at least 2 m long, preferably 3 m.
- a line determination 3690 is used.
- a projection of a line and / or at least one marking on the floor 3691 can be detected; in an alternative and / or supplementary aspect, at least one marking and / or a line on the floor is detected in step 3692.
- the marking and / or line is projected onto the floor by the service robot 17.
- the line can also be virtual and, for example, consist of the direct connection of the person to a marking and / or the line in which the sagittal plane of the person intersects the floor, the line being determined at the beginning of the evaluation and / or after the Issue 3521 of the request to cover the distance.
- a distance determination takes place 3910 in order to check whether the person has covered the distance along the line.
- outputs from the service robot 17 can occur in order, if necessary, to instruct the person to take more steps to reach the target distance (e.g. 3 m) or to stop when the target distance has been reached .
- the distance can be determined in various ways.
- the service robot 17 records the distance that the service robot 17 covers 3911, for example using the odometry data 3912 and / or using position data 3913, the distance in the latter case being determined by the difference between at least two positions.
- the distance to identified obstacles and / or objects can also be evaluated.
- the distance to the person is evaluated over time in step 3914 and the distance covered is calculated from this.
- the distance can be determined by adding the step lengths in step 3915, which were recorded in step 3672.
- the position can also take place by evaluating the position of the person in space in step 3916 (see also step 3695 below), i.e. in particular by evaluating the distance between the coordinates that change when the position changes.
- the position of the person is evaluated in step 3920 by determining the position of the head joint point, for example in the transverse plane and / or the center of the direction vector between the shoulder joint points or the hip joint points and / or the center between the direction vector between the knee joint points (e.g. . Projected into the frontal plane), the direction vector between the ankle points (e.g. projected into the frontal plane) and / or a direction vector between at least two similar arm joint points (e.g. projected into the frontal plane) is evaluated.
- an evaluation is carried out to determine whether the person is using aids, as already described above in step 3620.
- the service robot 17 determines the distance of the center of the body to the line in the course of time 3791 and / or the distance of the ankle points to the line within the frontal plane in the course of time 3792.
- a deviation calculation including threshold value 3793 or a pattern comparison then takes place, ie for the determined distances, the maximum of the deviation, the least squares of the individual deviations per step, etc. are calculated, whereby other approaches described in the prior art for distance evaluation can also be used.
- a classification then takes place as follows: The result is classified as a significant deviation in step 3793, if the person is standing 3616, 3666 goes, and the value of the line deviation in the deviation calculation including threshold value 3793 is above a threshold value or in has a minimum pattern similarity to a pattern comparison. The result is classified as a slight deviation and / or use of aids 3794, if the person stands 3616, walks 3666, and the value of the line deviation in the deviation calculation including threshold value 3793 and / or pattern comparison lies in an interval whose upper value is the threshold value for the Classification according to 3793 represents. Furthermore, as an alternative and / or in addition to the deviation from the line, the use of aids is detected in step 3620.
- the result is classified as no deviation without the use of aids 3795 if the person is standing 3616, goes 3666, and the value of the line deviation in the deviation calculation including threshold 3793 is below a threshold (or a pattern similarity is not achieved in a pattern comparison) and no use of aids is detected in step 3620.
- the path deviation score is calculated in step 3796.
- the service robot 17 evaluates the torso stability while walking, which happens analogously or similarly to the standing balance determination, with the difference that the person also walks (cf.
- the result of the feature classifications 3370 of the various aspects is evaluated as follows.
- the person is classified as swaying or using aids in step 3951 if the person is standing 3616, walking 3666, using aids 3623, and is unstable 3636.
- the person is classified as not swaying, but bent or balancing in step 3952, if the person is standing 3616, walking 3666 and either (as a partial aspect of the balance determination) is leaning forward (e.g.
- the arm joint points themselves are at a distance from the body which is above a threshold value or which, for example, has a pattern dissimilarity in a pattern comparison (for example, evaluated via the periodic movements of the arm joint points in the transverse plane in 3631).
- the person is classified as torso stable 3953 if the person is standing 3616, walking 3666, has no aids 3624 and is stable 3635.
- the path deviation score 3954 is calculated based on this classification.
- the amplitude and / or frequency of the periodic or aperiodic movements that are detected essentially parallel to the frontal plane can also be subjected to a threshold value comparison and / or a pattern comparison.
- the service robot 17 evaluates the lane width / step width when detecting the aisle, as was described, for example, in the previous sections (see FIG. 50).
- the track width is evaluated 3695 within the feature classification 3370, which is implemented, for example, as a distance measurement of the ankle points over time in the frontal plane 3696.
- a threshold value comparison 3956 and / or a pattern comparison is carried out using the gauge data. It is also taken into account whether the person is standing 3616 and walking 3666. If the track width is below the threshold value (e.g.
- the track width is classified as narrow 3958, otherwise as wide 3957.
- the result is transformed into a 3959 track width score.
- the track width can be corrected by the width of the hip joint point, which is approximated over the length of the direction vector between the hip joint points. 360 'rotation
- the service robot 17 evaluates the rotational movement of the recorded person (see FIG. 51), preferably a rotational movement of 360 °, when the aisle is detected, as was described in the previous sections, for example .
- the step length 3930 is determined within the feature classification 3370, but differently at this point than in step 3670 because it is not the distance in the sagittal plane that is evaluated, but the absolute distance, since the person's rotational movement can also step at an angle.
- it is evaluated, for example with data from step 3661, whether and to what extent the person turns, i.e. a rotation detection takes place in step 3925.
- the rotation in the transverse plane of the direction vector between the shoulder joint points, the hip joint points and / or the knee or arm joint points or the head is evaluated in step 3926.
- the angles of rotation are recorded, added and the added values are evaluated to determine whether the addition has reached the value of 360 ° (angle of rotation addition up to the threshold value 360 ° in step 3927 or pattern comparison). In one aspect, this takes place after an output in step 3521 by the service robot 17.
- step 3925 In the case of the rotation through 360 ° (as a result of step 3925), a detected walking movement in step 3666 and the standing in step 3616, an evaluation of the previously recorded step lengths is carried out in which the distances between the steps are compared in step 3961 become.
- the symmetry of the double steps 3962 is evaluated, i.e. the double step lengths, the ratio of the single step lengths to each other and / or in comparison to the double step are evaluated.
- the step frequency 3963 can also be evaluated, in particular with a view to its periodicity, i.e. the rise and fall of the curves and, above that, the amplitude symmetry.
- Periodicity and / or the symmetry of the double steps is evaluated via a threshold value comparison in step 3964 and / or a pattern comparison, with a high degree of symmetry leading to steps 3965 classified as continuous, otherwise to discontinuous steps 3966.
- the results are converted into a turning step score 3967 converted.
- a rotation stability score 3970 is also recorded.
- a rotational movement detected in 3925 is evaluated with regard to the balance in step 3630.
- such rotational movements in which the balance is stable 3635 is where the person stands 3616 and walks 3666, classified as 3968 stable rotary motion.
- movements in which the person is standing 3616 and walking 3666, but the balance is unstable in step 3636 are classified as unstable rotary movement 3969. The results are transferred to the 3970 Rotation Stability Score.
- the service robot 17 detects the sitting down of the person as a result of an output 3521 via at least one output device of the service robot 17 and uses at least some of the feature classifications as already described above (cf. FIG. 52).
- the service robot 17 evaluates the transition from standing 3616 to sitting 3617 in the sitting down classification 3980, this step being described with 3981. In particular, the speed of transition 3982 is evaluated.
- the service robot 17 determines the continuity of the transition 3983, for example by sequential comparison of the instantaneous speeds while sitting down and / or by comparison with values stored in a memory. Based on these two steps 3982 and 3983, a threshold value evaluation 3984 and / or a pattern comparison takes place.
- results of the classification 3620 are used, which evaluate whether the hand uses an aid, which also includes supporting oneself. The results are then classified as follows:
- step 3984 If the speed of sitting down exceeds a threshold value 3984, then sitting down is classified as an insecure sitting down 3987. If the value of the discontinuity in the course of sitting down exceeds a threshold value 3984 and / or an auxiliary means 3623 is used, the movement is classified as an arduous sitting down 3985. If no aid is detected in step 3624 and the speed of sitting down falls below a threshold value in step 3984 and / or the value of the discontinuity of the course of sitting down falls below a threshold value in step 3984, sitting down is considered safe in step 3986 classified. The results of this evaluation are converted into a 3988 Sit Down Score.
- the service robot 17 creates two- or three-dimensional recordings of the person and compares these recordings with recordings stored in a memory of people who also adopt the postures or make movements and who were classified accordingly to what extent the recorded person leans to the side, slips or sits securely or stably; the extent to which this person stands up and uses aids, tries to stand up, takes steps, has gait symmetry, shows trunk stability, rotates 360 °, etc.
- This classification can be carried out using classification methods described in the prior art, e.g. methods of machine learning / artificial intelligence.
- the service robot 17 classifies the recorded images of the person. The score is assigned to the individual exercises in the same way as described above.
- the service robot 17 can also transmit the recorded data via an interface 188 (such as WLAN) to other systems in which the data are then evaluated, for example in the manner described for the evaluation within the service robot 17 .
- an interface 188 such as WLAN
- the various aspects of the mobility evaluation are described below with the aid of several figures.
- the system for determining the balance of a person comprises a sensor for contactless detection of a person over time, a skeleton creation module 5635 for creating a skeleton model of the person, a skeleton model-based feature extraction module 5640 for feature extraction based on articulation points and / or Direction vectors between the articulation points of the person, a transverse articulation point evaluation module 5645 for evaluating changes in position of the articulation points within the transverse plane with a view to the amplitude, orientation and / or frequency of the change in position and for comparing recorded values with threshold values stored in memory 10 and / or patterns.
- the system for determining the balance of a person comprises a sensor for contactless detection of a person over time, a skeleton creation module 5635 for creating a skeleton model of the person, a skeleton model-based feature extraction module 5640 for feature extraction on the basis of articulation points and / or direction vectors between the articulation points of the person, a perpendicular articulation point evaluation module 5650 for determining the deviation of a direction vector from the perpendicular of the person, the direction vector being formed as a connection of at least one articulation point of the foot, knee or hip with at least one vertically above articulation point of an upright person.
- the system includes, for example, a perpendicular joint point evaluation module 5650 for determining the deviation of a direction vector from the person's perpendicular with a threshold value and / or pattern stored in memory 10, a track width increment module 5675 for determining the track width and / or increment of a Person on the distance between the ankle points in the frontal plane over time, when the track width has fallen below a threshold value, a person size evaluation module 5655 for evaluating the height of the person, the size e.g. over the distance between the ground and / or at least one ankle point and at least one point in the head area is determined, for example is determined by vector subtraction of two direction vectors which extend from a common origin to at least one foot and at least the head of the person.
- a common origin is understood to mean, for example, a sensor such as a 3D camera from which depth information is recorded.
- the system comprises a hand distance evaluation module 5660 for determining the distance between the at least one recorded wrist point of the person and at least one recorded object in the vicinity of the person and modification of a value in the memory (10) when the distance falls below a threshold value.
- the system comprises a sagittal plane-based joint point course evaluation module 5665 for evaluating the course of the joint points within the sagittal plane and for comparing the determined values with values stored in the memory 10.
- the sensor for contactless detection of the person can be a camera 185, a LID AR 1, an ultrasound and / or radar sensor 194.
- the system has a person recognition module 110, person identification module 111 and / or movement evaluation module 120.
- Example 1 Delirium prevention and delirium monitoring
- the service robot 17 can be used to reduce the length of stay of patients in clinics if the patients are of advanced age and require an operation, which is usually carried out under general anesthesia. There is a high risk of developing dementia from the anesthesia. Risk patients are those who have previously suffered from cognitive impairments.
- the service robot 17 can now be used to monitor the cognitive abilities of the patient at least once, for example automatically over the course of time, to provide medical staff with diagnoses that enable improved and targeted prophylaxis and treatment of patients.
- FIG. 17 shows the sequence of how the service robot 17 automatically moves towards the patient for this purpose.
- a hospital management system HIS
- patient data are stored that an operation is to take place for a certain patient, including the type and date of the OP 1705.
- the patient management system which is linked to the hospital management via an interface 188 (such as WLAN) System (HIS) accesses, the room information about where the patient is located can be obtained from the HIS.
- further information can be transmitted to the patient administration module 160 in step 1710, including the type of operation, the appointment, disease-related information, etc.
- the service robot 17 accesses the patient administration module 160, receives the room information in step 1715 and the same in step 1720 the room information with information stored in his navigation module 101 in order to then move in the direction of the patient's room in step 1725.
- the information comparison between the patient administration module 160 and the navigation module in the cloud 170 takes place in step 1730, and in step 1735 the navigation module in the cloud 170 synchronizes with the navigation module 101 of the service robot 17. The service robot 17 then moves in the direction of the room Patient 1725.
- the service robot 17 If the service robot 17 is in front of the room door to the patient 1805, the service robot 17 must pass this so that the patient can carry out the test on the service robot 17.
- the service robot 17 is also configured in such a way that the service robot 17 is able to recognize a door through its sensor system, as described, for example, further above or in FIG. 8 1810. If the door is open, the service robot 17 navigates in the Step 1815 straight to the patient's room. If the door is closed 1820, the service robot 17 uses an integrated communication module 1825 which is connected to the call system of the clinic 1840 via an interface 188 (such as WLAN) (step 1835).
- an interface 188 such as WLAN
- the service robot 17 uses this to send a signal which allows the medical staff to draw conclusions about its position and the instruction to please open the door to the patient's room.
- the service robot 17 has, in one aspect, a database with assignment of location data in its memory to room numbers 1830, which can be part of the database with room data 109 as part of the navigation module 101, which can be connected to a hospital information system via an interface, for example.
- this database can also be available in a cloud 18. If the room door is opened by hospital staff in step 1845, the service robot 17 moves into the patient's room in order to carry out the test with the patient in step 1850.
- the service robot 17 is configured in such a way that the service robot 17 has direct access to the door control 1855 via an interface 188 (such as WLAN) (step 1835) and this has a code for the Opening the door in step 1860 sends.
- an interface 188 such as WLAN
- the service robot 17 in front of the door observes the surroundings 1865 through its at least one sensor 3 and, if the service robot 17 detects people, the service robot 17 tracks these detected people in step 1870 and, in one aspect , predicts the movements of the recognized persons in the optional step 1875 and, if the persons orientate themselves in its direction 1880, aligns itself so that its display 2 is orientated in the direction of the persons (step 1890).
- the tracking takes place, for example, by means of the visual person tracking module 112 and / or the laser-based person tracking module 113.
- the service robot 17 waits until people are at a minimum distance from the service robot 17 1885 before the service robot 17 aligns the display 2 in the direction of the people 1885. At the same time, the service robot 17 visually and / or acoustically signals its wish that the person open the door to the patient's room in step 1892. The addressed person opens the door in step 1894.
- the service robot 17, as already described above or FIG. 8, is able to detect the opening process of the door 1896. As soon as the door is opened, the service robot 17 navigates into the patient's room to carry out the test in step To be carried out in 1850. If there is no door between the service robot 17 and the patient, as is sometimes the case in intensive care units, these steps are not necessary.
- the service robot 17 carries out a test in step 1905, in particular the mini-mental test, and uses this to determine a score in step 1910 which, in the case of the mini-mental test, determines the degree reflects the patient's cognitive impairment.
- Test procedures are used. These data are transmitted via an interface 188 (eg WLAN) (step 1915) to the patient administration module 1920 and are made available there to the medical staff, who can access them via a display 2 (step 1925). If necessary, the data is also transferred to the HIS via an interface 188 (e.g. WLAN) 1930.
- the patient administration module 160 is able to receive further data on the patient's medical history from the HIS, including, for example, drugs that the patient is taking. Based on this information, the patient administration module 160 determines a risk value which indicates the probability with which the patient can expect the dementia to progress through the planned operation 1935. This risk value can in turn be made available to the medical staff via a display 2 and / or be transferred to the HIS. On the basis of such information, the medical staff can initiate suitable preventive measures in order to prevent possible postoperative dementia or at least to reduce the likelihood of it.
- the service robot 17 and the patient administration module 160 are also configured so that the service robot 17 after the operation 1955 (the information comes from the HIS), as already described above 1950, moves back to the patient and reconnects with him Test conducted in 1960, in particular, a geriatric test, such as the mini-mental test.
- a geriatric test such as the mini-mental test.
- step 1970 this procedure is repeated after a few days in step 1970 in order to record, evaluate and document the progress of the patient's recovery.
- FIG. 20 it is shown how data from the service robot 17 are processed for therapy proposals.
- the set of rules 150 is connected to the patient administration module 160 in the cloud 18 in such a way that it is configured such that the patient information can be transmitted in anonymized form to the set of rules 150 in step 2025.
- the patient administration module 160 can previously be accessed from the Hospital Information System 2015 received further relevant data via an interface 188 (e.g. WLAN), including type of operation, type of anesthesia, comorbidities, medication taken, measures to prevent delirium, postoperative measures to alleviate delirium or its treatment, etc. as well as the results of the Exercises that have been carried out by means of the service robot 17 in step 2020.
- the latter can also come from the patient administration module 160.
- These data which are each available as a time series, are anonymized and encrypted in the patient administration module 160, transmitted to the set of rules 150 and stored there 2030.
- step 2035 established machine learning methods and neural networks are used to provide patient-specific data such as age, The co-morbidities, the type of anesthesia, the operation, the medication taken etc. to make a prognosis of the extent to which the patient is expected to have postoperative delirium in 2040.
- Determinants of delirium are in turn the degree of cognitive impairment shortly after the operation, when the service robot 17 usually carries out the first test, with which the parameters that are collected as part of a CAM-ICO test, the delirium detection score, the behavioral pain scale, the critical care point observation tool, the Richmont agitation sedation scale, the motor activity assessment Scale, in different forms and / or from the service robot such as Data described and collected above (e.g.
- section delirium detection for example the data collected in example 11, 12 or 17 etc.
- Another determinant is the improvement in the degree of delirium that occurs over a specified time, over which the service robot 17 determines the cognitive abilities 2045.
- Another , alternative or supplementary determinant is the time that is required to (again) reach a certain level of cognitive skills in 2050. This data can be used in the form of a training data set.
- the effect of interventions on the determinants described in the previous paragraph is estimated using established machine learning methods and / or neural networks.
- These interventions include the use of gentle anesthesia, accompanying measures such as the provision of a caregiver, the use of medication, etc. 2055
- weights are determined 2060, which are transmitted from the set of rules 150 to the patient administration module 160 in step 2065 and are used to submit recommendations to the medical staff, such as according to specific test results determined by the service robot 17 the creation of therapy plans for the respective patient should be proceeded in step 2070.
- recommendations can be made available to the medical staff by the patient administration module 160 preoperatively, postoperatively and / or over the course of time.
- the updates of these recommendations are optionally based on entries in the HIS to which the patient administration module 160 has access, on entries in the patient administration module 160 as well as on the results of the mini-mental test, based on the test for the delirium detection score with, for example, a sweat detection, the Confusion Assessment Method with, for example, the evaluation of cognitive abilities in connection with the detection of acoustic signal sequences, image recognition or finger recognition and / or the determination of the pain status on the basis of an evaluation of emotions, movements of the upper extremities, any coughing activities and / or acoustic pain articulations that the service robot 17 completes or performs with or on the patient, as described.
- the situation can be represented in such a way that it is a system for predicting postoperative dementia or delirium, consisting of a computer, a memory and at least one interface 188 (eg WLAN) via which the system exchanges data with a mobile data acquisition unit, which has at least one camera 185, in one aspect also a spectrometer 196.
- the mobile acquisition unit which records and evaluates data by machine, is, for example, a service robot 17.
- the system as such can also be mapped in the service robot 17.
- the system has an interface 188 (eg WLAN) via which it receives data on the state of health of a person, on treatments, medication status, measures for delirium prevention, measures for postoperative delirium treatment and / or evaluations of measurements that the mobile data acquisition unit such as For example, the service robot 17 has performed.
- the system evaluates the data assigned to the persons over time.
- historical data is evaluated, which depicts the pre- and postoperative course of illness and treatment in patients.
- Predictions are made, ie in particular the probability of the occurrence of postoperative data such as, for example, the probability of occurrence of postoperative dementia, its course, etc. are forecast, for which, for example, machine learning methods are used.
- the prognoses also take into account medical interventions such as the initiation of certain treatments pre- and post-operatively and their influence on the postoperative course of the disease is predicted.
- rules are determined and stored, ie in particular weights (or regression coefficients, if regression models are used). These can then be used for forecasts in a second step on the basis of data obtained via the interface 188 (eg WLAN), including data that have been collected by the mobile data acquisition unit and, in one aspect, processed.
- These data are primarily newly recorded data from patients for whom the future course of the disease is still unclear at the time of the survey or who has not yet been conclusively recorded.
- This second step can take place in a separate system which receives the rules or weights from the system mentioned.
- the process of the empirical-based prognosis of postoperative dementia / delirium is as follows: recording of persons over time, determination of health status data of persons based on recording of the person over time, receipt of preoperative data of persons, receipt of intervention data on the Persons, determination of the influence of the preoperative data and the intervention data on the health status data of the persons by calculating a weight estimate for parameters of the preoperative data and the intervention data as well as, for example, prognosis of the health status of a recorded person based on the weight estimate and newly collected preoperative data and intervention data of a person .
- the service robot 17 can also carry out such an evaluation on patients who are not in bed.
- the service robot 17 is able to identify the postures / poses of a patient on the basis of the skeleton recognition using frameworks from the prior art.
- the service robot 17 can use the RGB camera to make recordings of the patient's surface and evaluate them to determine whether the patient is clothed. The patient picks up on this Classification algorithms designed to recognize skin by its color.
- a cross-validation can take place in such a way that the target region on the skin on which the measurement is to take place is compared in terms of color and / or texture with that of the person's face, which is determined via frameworks can be recognized according to the state of the art, for example by means of approaches such as histogram-of-gradients, which are implemented in frameworks such as OpenCV or Scikit-image.
- a filter can be used here that carries out color corrections to the extent that the detected colors on the face can be darker than on the area on the skin on which the measurement is to take place. Regions of the face that are eligible for determining the comparison value are the cheeks or the forehead (the identification of the latter region has already been described elsewhere).
- Such a correction factor can, in one aspect, also be dependent on the season. If the comparison taking place in this case results in a similarity value that is above a defined threshold value, the detected point is recognized as skin for the measurement.
- An alternative and / or supplementary filter can also be used, which excludes color tones that are not typical for the skin (certain red tones, blue, green, yellow) etc.
- the service robot 17 detects a target region of the skin for the measurement, determines whether this is skin, and if this is not the case, the service robot 17 detects further target regions and / or asks the patient via the speech synthesis unit 133, to clear a corresponding region. To this end, the service robot 17 tracks the corresponding region, for example the patient's arm, and, for example, after the patient moves less than defined by a threshold value, the evaluation of the location starts again in order to identify whether the skin is exposed there or, if necessary, still through Clothing is covered. If the service robot 17 has detected this concealed skin, the service robot 17 carries out the measurements described elsewhere.
- Example 4 Triggering elements of embodiment based on cardiovascular indicators
- the service robot 17 calculates the pulse rate or pulse frequency of a person with whom the service robot 17 interacts by the service robot 17 the pulse rate and pulse frequency with the system described by recording and evaluating the face and the reflected cardiovascular movements of the face surface and /or of the head and / or vascular blood flows under the skin.
- the determined pulse frequency is used in the event that the service robot 17 has elements of embodiment, ie those elements that at least partially reproduce a person such as a stylized head or parts of it such as stylized eyes, mouth, etc., by imitation the frequency of unconsciously interacting with the person. This interaction can mean, for example, adapting the blinking of stylized eyes to the pulse rate.
- the service robot 17 can also be adapted to the pulse frequency.
- the service robot 17 can, if the person has an above-average pulse rate, which, for example, also indicates a high level of nervousness (alternatively, other determined parameters indicate high level of nervousness), by selecting a movement frequency of, for example stylized eyes, chest or other elements that are smaller than the frequency identified in the patient try to calm the patient down.
- the service robot 17 determines the patient's pulse frequency over time and, if necessary, reduces its movement frequency until the patient also has a normal pulse frequency. The difference between the recorded pulse frequency and the movement frequency of the service robot 17 can remain approximately constant.
- Stylized means, for example, that the eyes can be implemented in hardware, for example spheres with printed circles and hemispheres that can mechanically cover the printed circles on the spheres.
- the eyes can also be shown on a display, for example in the form of circles, etc.
- When looking at a mouth it can be defined, for example, as in the case of a smilie, by a line that can adopt different orientations and / or curvatures.
- the service robot 17 can identify and also track a person's chest using the camera 185, which can be done using a framework such as OpenPose, OpenCV, etc., for example using the visual person tracking module 112 and / or the laser-based person tracking module 113.
- Camera 185 and the two person tracking modules 112 and 113, but possibly also other sensors such as LID AR 1, are also referred to as person detection and tracking unit 4605.
- the service robot 17 can detect movements of a person over time, which, for example, reflect breathing. This detection includes in the event that the patient is roughly in front of the service robot movements in the horizontal direction as well as in depth.
- These movements can be determined, for example, by means of a bandpass filter with a window of 0.005 to 0.125 Hz, at least 0.05 to 0.08 Hz, and subsequent rapid Fourier transformation. This can be used to determine the breathing rate, which can be used instead of the pulse rate to mirror the patient's movements and, if necessary, to calm the patient.
- the pulse rate or pulse frequency and / or breathing or breathing frequency is detected by means of a movement frequency determination unit 4606, the computer-implemented method for determining the pulse rate or pulse frequency and / or described, for example, from camera 185 and elsewhere in this document Breathing or breathing rate, but other movements of the person are also conceivable.
- the parameters for pulse and / or breath are specifically recorded and evaluated by a pulse-breath evaluation unit 4615. In particular, breathing is about a
- Movement signal detection and processing unit 4620 detects and evaluates which background signals of the body differ from those of clothing. For this purpose, details are given elsewhere in this document with a view to signal processing.
- the mentioned stylized faces or facial elements, heads, torso or rib cages, are also referred to as stylized embodiment elements 4625. These are moved by a movement unit 4607 at a certain frequency. This can take place in different ways, depending on the type of stylized embodiment element. Eye movements on a display can be implemented purely software-based, for example, while physical embodiment elements, for example, require servomotors that move eyelids or move a stylized chest.
- the system also has a person recognition module 110, person identification module 111, tracking module (112, 113) and / or movement evaluation module 120. An overview of the components of the system is shown in FIG. 60.
- the synchronization of movements of a person with a service robot 17 is characterized by the following aspects ASBPS1 to ASBPS19:
- ASBPS1 System for synchronizing the movements of a person and a system, comprising a person detection and tracking unit (4605), a movement frequency determination unit (4606) for determining the frequency of the movements of the person, and a movement unit (4607) for movement stylized Embodiment elements (4625) of the system with a frequency which lies within a defined bandwidth around the determined frequency of the movements of the person.
- ASBPS2 System according to ASBPS1, further comprising a pulse-breathing evaluation unit (4615) for measuring the pulse rate and / or the breathing rate of the person.
- ASBPS3. System according to ASBPS1, further comprising a motion signal detection and
- ASBPS4 System according to ASBPS1, wherein the stylized embodiment elements (4625) are implemented in hardware and / or software.
- stylized embodiment elements (4625) implemented in software comprise the display of at least one stylized face or face element on a display 2.
- ASBPS6 System according to ASBPS5, wherein hardware-implemented, stylized embodiment elements (4625) comprises at least one stylized face, facial element or a torso or rib cage.
- ASBPS7 The system of ASBPS5, wherein the movement of stylized embodiment elements (4625) comprises movement of a stylized face, facial element, torso or chest by the movement unit (4607).
- ASBPS8 System according to ASBPS4, wherein the stylistic embodiment elements are stimulated to imitate the breathing movement by the movement unit (4607).
- ASBPS9 System according to ASBPS1, whereby the system is used to calm the person.
- ASBPS12 The computer-implemented method of ASBPS10, wherein the movement of stylized embodiment elements (4625) comprises movement of a stylized face, facial element, torso or chest and / or mimicking breathing.
- ASBPS13 Computer-implemented method according to ASBPS10, wherein the movement of the stylized embodiment elements (4625) by the movement unit (4607) is kept lower than the detected frequency of the movements of the person.
- ASBPS14 Computer-implemented method according to ASBPS10, the frequency difference between the movements of the stylized embodiment elements (4625) and the person being kept approximately constant over time by the movement unit (4607).
- ASBPS15 Computer-implemented method according to ASBPS10, the recorded movements of the person being the pulse rate and / or the breathing rate.
- ASBPS16 Computer-implemented method according to ASBPS10, wherein the movements of the stylized embodiment elements (4625) are set by the movement unit (4607) to a frequency which is lower than the frequency of the detected movements of the person.
- ASBPS17 Computer-implemented method according to ASBPS10, wherein the movements of the stylized embodiment elements (4625) are controlled by the movement unit (4607) in such a way that they become slower over time.
- ASBPS18 Computer-implemented method according to ASBPS10, wherein the frequency difference between the stylized embodiment elements (4625) and the person is kept approximately constant by the movement unit (4607) over time.
- ASBPS19 Computer-implemented method according to ASBPS10, whereby the
- Movement unit (4607) initiated bandwidth of the frequency of the determined frequency of the person moving in an interval of 50% downward and / or upward or in an interval of less than 15% downward and / or upward.
- Example 5 Method, device and / or system for carrying out a get up and go test Determination of a score in connection with getting up and sitting on a chair is characterized here by the following aspects ASASS1 to ASASS20:
- ASASS Computer-implemented method for recording and evaluating the
- Covering a distance by a person comprising: outputting an instruction via an output unit;
- ASASS2 Computer-implemented method according to ASASS1, whereby the distance covered by the person is determined by
- ASASS3 Computer-implemented method according to ASASS1, whereby the distance covered by the person is determined by tracking the person's movements between the starting position and the turning position.
- ASASS4 Computer-implemented method according to ASASS1, comprising a detection of a rotational movement of the person and comparison with patterns.
- ASASS5 Computer-implemented method according to ASASS4, whereby the rotational movement of the person at the turning position is recorded.
- ASASS6 Computer-implemented method according to ASASS1, whereby the rotational movement of the person is recorded and the position of the rotational movement defines the turning position.
- ASASS7 Computer-implemented method according to ASASS4-6, whereby the detection of the rotary movement is carried out
- ASASS8 Computer-implemented method according to ASASS1, whereby the turning position is determined by a recorded marking on the ground.
- ASASS9 Computer-implemented method according to ASASS1, further comprising a determination of the person getting up from and / or the person sitting on a chair.
- ASASS10 Computer-implemented method according to ASASS9, whereby the determination of the person getting up from and / or the person sitting on a chair is carried out by evaluating the inclination of the upper body over time.
- ASASS11 Computer-implemented method according to ASASS10, with evaluation of the inclination of the upper body over time over the
- ASASS12 Computer-implemented method according to ASASS10, whereby the determination of the person getting up from and / or the person sitting on a chair by evaluating the height of the person and / or a change in height of the person compared to a threshold value and / or pattern he follows. ASASS13.
- Computer-implemented method according to ASASS10 whereby the determination of the person standing up from one and / or the person sitting on a chair via the detection, tracking and evaluation of the movements of the person's head over time and one recognized, at least partially circular movement of the head takes place within the sagittal plane.
- ASASS14 Computer-implemented method according to ASASS1, further comprising the
- ASAS15 Computer-implemented method according to ASASS14, further comprising the creation of a score for the determined time.
- ASASS16 Computer-implemented method according to ASASS1, further comprising the performance of a test on the hearing, vision and / or mental abilities of the person.
- ASASS17 Device for carrying out a method according to ASASS1 - ASASS16.
- ASASS18. System comprising a computer (9), a memory (10) and at least one sensor for contactless detection of the movement of a person, with a chair detection module (4540), an output device such as a loudspeaker (192) and / or a display (2) for transmission of instructions, a duration-distance module (4510) for determining the duration for covering the distance and / or a speed-distance module (4515) for determining the speed of the recorded person on a distance and a time-distance evaluation module (4520) to evaluate the time to cover the distance.
- ASASS19 System according to ASASS18, further comprising a hearing test unit (4525), a vision test unit (4530) and / or a test-for-intellectual-ability-unit (4535).
- ASASS20 System according to ASASS18, further comprising a projection device (920) for projecting a turning position onto the ground.
- Example 6 Method, device and system for evaluating a mini-mental test folding exercise
- AMMTF1 to AMMTF25 The determination of a score in the evaluation of a folding exercise is characterized here by the following aspects AMMTF1 to AMMTF25:
- AMMTF Computer-implemented method for detecting and evaluating a folding process, comprising
- AMMTF2 Computer-implemented method according to AMMTFG1, comprising folding the sheet approximately in the middle.
- AMMTF3 Computer-implemented method according to AMMTFG1, wherein the tracking of the at least one hand of the person comprises the creation of a skeleton model of the at least one hand of the person.
- AMMTF4 Computer-implemented method according to AMMTFG1, comprising an identification of the leaf via a fault-tolerant segmentation algorithm.
- AMMTF5. Computer-implemented method according to AMMTFG4, further comprising a sheet classification and / or classification of a folding process based on comparison with two-dimensional or three-dimensional patterns.
- AMMTF6 Computer-implemented method according to AMMTFG1, wherein the classification of the folding process comprises touching at least one thumb and at least one other finger on the fingertips or fingertips as hand movements.
- AMMTF7 Computer-implemented method according to AMMTFG1, wherein the classification of the folding process comprises a detection of a change in shape of a sheet which is in engagement with at least one element of a hand.
- AMMTF8 Computer-implemented method according to AMMTFG1, comprising identification and tracking of at least one corner and / or edge of a sheet.
- AMMTF10 Computer-implemented method according to AMMTFG9, further comprising a classification of the folding process by comparing determined distances with a threshold value and / or pattern and detection of a folding process if the determined distance is below the threshold value and / or a minimum pattern similarity is detected.
- AMMTF11 Computer-implemented method according to AMMTFG1, wherein the classification of the folding process comprises a detection of a curvature of the sheet which is above a threshold value and / or which has a minimum pattern similarity.
- AMMTF12 Computer-implemented method according to AMMTFG1, wherein the classification of the folding process comprises a distance reduction between at least two sheet edges.
- AMMTF13 Computer-implemented method according to AMMTFG1, wherein the classification of the folding process comprises an approximately parallel alignment of the ends of a sheet edge and / or a distance between the ends of a sheet edge which is smaller than 20 mm.
- AMMTF14 Computer-implemented method according to AMMTFG1, whereby the classification of the folding process comprises a size reduction of the recorded and tracked sheet by more than 40% over time.
- AMMTF15 Computer-implemented method according to AMMTFG1, further comprising output on a display 2 and / or a voice output for folding and depositing a sheet or folding and dropping a sheet.
- AMMTF16 Computer-implemented method according to AMMTFG15, further comprising the detection of the sheet over time and the adjustment of a value in a memory after a folding process has been detected and the sheet is deposited and / or dropped.
- AMMTF17 Device for carrying out a method according to AMMTFG1-AMMTFG16.
- AMMTF18 System comprising a computer (9), a memory (10) and a sensor for contactless detection of the movement of a person, comprising a memory (10)
- AMMTF20 System according to AMMTFG18, wherein the folding movement detection module (4710) via a sheet-distance-corner-edge module (4720) for detecting the distance of edges and / or corners of a sheet, a sheet shape change module (4725), a sheet curvature module ( 4730), a sheet dimension module (4740) and / or a sheet edge orientation module (4745).
- AMMTF21 System according to AMMTFG18, further comprising a fingertip Ab Stands module (4750) for detecting the distance between fingertips of at least one hand.
- leaf detection module (4705) includes a leaf segmentation module (4755) and / or a leaf classification module (4760).
- AMMTF23 System according to AMMTFG18, comprising an output device such as a loudspeaker (192) and / or a display (2) for transmitting instructions.
- an output device such as a loudspeaker (192) and / or a display (2) for transmitting instructions.
- AMMTF24 System according to AMMTFG18, comprising an interface (188) to a terminal (13).
- AMMTF25 System according to AMMTFG18, wherein the at least one sensor for contactless detection of the movement of a person is a 2D and / or 3D camera (185), a LID AR (1), a radar and / or ultrasonic sensor (194).
- the at least one sensor for contactless detection of the movement of a person is a 2D and / or 3D camera (185), a LID AR (1), a radar and / or ultrasonic sensor (194).
- the manipulation detection is characterized here by the following aspects AMI to AM18:
- AMI Computer-implemented method for determining a manipulation probability on a robot, comprising
- AT 4 Computer-implemented method according to AMI, comprising a skeleton model creation of the recorded and tracked person and an extraction and classification of articulation points.
- AM5 Computer-implemented method according to AM4, further comprising a determination of the orientation of the person relative to the robot.
- Computer-implemented method according to AM5 further comprising the determination of an orientation of the person relative to the robot by determining the angle between the frontal plane of the person and the axis perpendicular to the operating elements 186 of the robot, each projected in a horizontal plane, and a Comparison of the determined angle with a threshold value, falling below the threshold value includes a detection of an increased probability of manipulation.
- AM9 Computer-implemented method according to AM3, AM6 and / or AM8, comprising a multiplication of the manipulation probabilities to determine a manipulation score.
- AM10 Computer-implemented method according to AM9, comprising the execution of evaluations by the robot with the person and the storage of the manipulation score together with the evaluation results.
- AMI 1. Device for carrying out a method according to AM1-AM10.
- System comprising a computer (9), a memory (10) and a sensor for contactless detection of the movement of at least one person, comprising an attempted manipulation detection module (4770) for detecting an attempted manipulation by at least one person.
- System according to AMI 2 further comprising a person-robot distance determination module (4775) for determining the distance between the at least one person and the robot.
- AM15 System according to AMI 4, wherein the person-robot distance determination module (4775) has a height-arm-length orientation module (4780) for estimating the height, arm length and / or orientation of the at least one person to the robot.
- the person-robot distance determination module (4775) has a height-arm-length orientation module (4780) for estimating the height, arm length and / or orientation of the at least one person to the robot.
- System according to AM13 further comprising an input registration comparison module (4785) for comparing whether a person registered with the system is recorded by the system or makes inputs in the system via the operating elements (186).
- the at least one sensor for contactless detection of the movement of the at least one person being a 2D and / or 3D camera (185), a LID AR (1), a radar and / or Ultrasonic sensor (194) acts.
- the manipulation detection is characterized here by the following aspects AMM1 to AMM17:
- AMM1 Computer-implemented method for determining a manipulation probability on a robot, comprising
- AMM2 Computer-implemented method according to AMM1, the position of the source of the audio signals being determined by detecting the direction of the audio signals by at least one microphone and triangulating the determined directions.
- AMM4 Computer-implemented method according to AMM1, further comprising
- AMM5. Computer-implemented method according to AMI, further comprising
- identification features of the person including frequency, intensity and / or spectrum of the audio signals from the person.
- AMM8 Computer-implemented method according to AMM1, further comprising
- AMM10 Computer-implemented method according to AMM1, where the position of the source of the audio signals is determined by repositioning the microphone and capturing the audio signals from two microphone positions with subsequent triangulation.
- AMM1 Computer-implemented method according to AMM1, comprising the determination of a manipulation score by multiplying determined manipulation probabilities.
- AMM13 System for manipulation evaluation on the basis of audio signals, comprising a computer (9), a memory (10), operating elements (186), a sensor for contactless detection of the movement of a person, a manipulation attempt detection module (4770), at least one microphone (193), a person position determination module (4415) for determining the position of a person, an audio source position determination module (4420) for Determination of the spatial origin of an audio signal, an audio signal comparison module for comparing two audio signals (4425) and an audio signal person module (4430) for assigning audio signals to a person.
- AMM14 System according to AMM13, comprising a speech evaluation module (132).
- AMM15 System according to AMM13, comprising an input registration comparison module (4785) for comparing whether a person identified by the system makes inputs to the system.
- the sensor for contactless detection of the movement of a person being a 2D and / or 3D camera (185), a LID AR (1), a radar and / or ultrasonic sensor (194).
- AMM17 System according to AMM13, comprising an audio sequence input module (4435) for comparing an audio sequence with a sequence of tactilely input letter sequences.
- ASP1 to ASP20 The spectroscopy is characterized here by the following aspects ASP1 to ASP20:
- Computer-implemented method for spectrometri see analysis of at least one body region of a person, comprising
- ASP2 Computer-implemented method according to ASP1, the body regions being the person's forehead, the surface of the hand and / or the upper body.
- ASP7 Computer-implemented method according to ASP1, comprising the evaluation of the ambient temperature.
- ASP8 Computer-implemented method according to ASP5, comprising the quantitative evaluation of a person's perspiration.
- ASP9 Computer-implemented method according to ASP1, comprising the determination of a delirium detection score.
- ASP10 Computer-implemented method according to ASP1, comprising the determination of the person's cognitive abilities.
- ASP12. System comprising a computer (9), a memory (10) and a sensor for contactless detection of a person, further comprising a spectrometer (196), a visual person tracking module (112), a body region detection module (4810) for Detection of body regions, a spectrometer alignment unit (4805) for Alignment of the spectrometer (196) to a body region of a person, and with access to a reference spectra database (4825) with reference spectra for comparing measured spectra for the purpose of determining measured substances.
- ASP13 System according to ASP12, comprising a spectrometer measuring module (4820) for monitoring a measuring process of the spectrometer (196).
- ASP14 The system of ASP12, wherein the visual person tracking module (112) includes a body region tracking module (4815).
- ASP15 System according to ASP12, comprehensive access to a clinical picture database (4830) with stored clinical pictures.
- ASP17 System according to ASP12, comprising a delirium detection score determination module (4840) for determining a delirium detection score.
- ASP18 System according to ASP12, comprising a cognitive skills assessment module (4845) for assessing cognitive abilities of the person.
- thermometer 4850
- ASP20 System according to ASP12, wherein the at least one sensor for contactless detection of the movement of a person is a 2D and / or 3D camera (185), a LID AR (1), a radar and / or ultrasonic sensor (194).
- the at least one sensor for contactless detection of the movement of a person is a 2D and / or 3D camera (185), a LID AR (1), a radar and / or ultrasonic sensor (194).
- AAA1 to AAA18 The attention analysis is characterized here by the following aspects AAA1 to AAA18:
- AAA1 A computer-implemented method for comparing detected signals of a tactile sensor (4905) with a sequence of output sound signals, comprising
- AAA3 Computer-implemented method according to AAA1, with a delay or phase shift between the output, pulsed sound signals and the signals detected by the tactile sensor (4905).
- AAA4 Computer-implemented method according to AAA3, the delay or phase shift being approximately half a pulse length.
- AAA5 Computer-implemented method according to AAA3, wherein the detected signal of the tactile sensor (4905) tracks the pulsed tone sequence.
- AAA6 Computer-implemented method according to AAA1, comprising the assignment of a value to each output sound signal.
- AAA7 Computer-implemented method according to AAA6, further comprising an adjustment of a value upon detection of a signal according to a defined value.
- AAA8 Computer-implemented method according to AAA7, wherein the adjustment of the value represents an incrementation of the value.
- AAA9 Computer-implemented method according to AAA7, further comprising the creation of a diagnosis on the basis of the adjusted value.
- AAA10 Computer-implemented method according to AAA9, whereby the diagnosis represents an assessment of cognitive abilities.
- AAA1 1. Computer-implemented method according to AAA1, comprising the detection of a person and the detection and position determination of a hand of the person.
- AAA12 Computer-implemented method according to AAA11, comprising positioning the tactile sensor (4905) at a distance from the hand which is below a threshold value.
- AAA13 Device for carrying out a method according to AAA1-AAA12.
- AAA14 System comprising a computer (9), a memory (10), an output unit for acoustic signals (192), a tactile sensor (4905) and a tactile sensor evaluation unit (4910) for evaluating signals from the tactile sensor (4905) and a tactile sensor output comparison module (4915) for carrying out a comparison as to whether the detected signals occur after an output of acoustic signals.
- AAA15 System according to AAA14, comprising an actuator (4920) on which the tactile sensor (4905) is positioned.
- AAA16 System according to AAA14, comprising an actuator positioning unit (4925) for positioning the tactile sensor (4905) within a defined distance to the hand.
- AAA17 System according to AAA14, comprising a camera (185), a personal identification module (111) and a hand identification module (4930).
- AAA18 System according to AAA14, comprising a cognitive skills assessment module (4845) for assessing cognitive abilities of the person.
- the cognitive analysis is characterized here by the following aspects AKA1 to AKA16:
- AKAT Computer-implemented method for comparing finger poses of a person determined on the basis of video signals with visually and / or acoustically output numbers, comprising
- AKA2 Computer-implemented method according to AKA1, whereby the finger poses represent numerical values.
- AKA3 Computer-implemented method according to AKA2, whereby a numerical value can represent several finger poses.
- AKA4 Computer-implemented method according to AKA1, the optical output of numerical values representing output of finger poses by an actuator (4920).
- AKA5 Computer-implemented method according to AKA1, further comprising a detection and tracking of the head of the person and a determination of the field of vision of the person.
- AKA6 Computer-implemented method according to AKA5, further comprising a positioning of the actuator (4920) and / or a display (2) in the field of vision of the person.
- AKA7 Computer-implemented method according to AKA1, further comprising a determination of cognitive abilities of the person by evaluating the comparison of the evaluated finger poses with the visually and / or acoustically output numerical values.
- AKA8 Device for carrying out a method according to AKA1-AKA7.
- AKA9. System comprising a computer (9), a memory (10), an output unit and a numerical value output module (4940) for outputting numerical values, a person detection and tracking unit (4605) with a camera (185) and a person recognition module (110 ).
- AKA10 System according to AKA9, the output unit being a tone generator such as a loudspeaker (192), a display (2) or an actuator (4920).
- a tone generator such as a loudspeaker (192), a display (2) or an actuator (4920).
- AKA1 1. System according to AKA10, whereby the actuator (4920) is a robot arm.
- AKA12. System according to AKA10, whereby the actuator (4920) has a robot hand (4950).
- AKA13 System according to AKA9, further comprising a hand pose detection module (4960) for the detection and of hand poses of the person.
- AKA14 System according to AKA12, wherein the system comprises a finger pose generation module (4955) for generating finger poses of the robot hand (4950).
- AKA15 System according to AKA9, the system being connected to a patient administration module (160).
- AKA16 System according to AKA9, further comprising a cognitive abilities assessment module (4845) for assessing cognitive abilities of the recorded person.
- a computer-implemented method for determining the pain status of a person comprising
- ASB2 Computer-implemented method according to ASB1, the individual and / or coherent classification of the surface curvature of the candidate regions representing a determination of emotions.
- ASB4 Computer-implemented method according to ASB2, further comprising the evaluation of emotions over time.
- a computer-implemented method for determining the pain status of a person comprising
- a computer-implemented method for determining the pain status of a person comprising
- the pain intensity classification includes the assignment of scale values to the recorded acoustic signals and the scale values represent a pain status.
- ASB1E Computer-implemented method for determining the pain status of a person, comprising
- ASB12 Computer-implemented method according to ASB1, ASB6, ASB9 or ASB11, with at least two of the methods being carried out in parallel or sequentially.
- ASB 13 Computer-implemented method according to ASB1, ASB6, ASB9 or ASB11, further comprising the evaluation of the determined scale values or stored values within a delir detection.
- ASB14 Device for carrying out a method according to ASB1-ASB13.
- ASB 15. System for determining the pain status of a person, comprising a computer (9), a memory (10), a sensor for contactless detection of the person and a pain status calculation module (5040).
- ASB 16 System according to ASB 15, comprising a face recognition module (5005) for recognizing the face of the person, a face candidate region module (5010) for selecting candidate regions within the face, an emotion classification module (5015) for Classification of the surface curvatures of the candidate regions of the face in emotions, and an emotion evaluation module (5020) for determining a scale value for the emotion.
- ASB17 System according to ASB15, comprising a bed recognition module (5025) for recognizing a bed.
- ASB18 System according to ASB15, comprising a person recognition module (110), a visual person tracking module (112), an upper extremity evaluation module (5035) for detecting and tracking the upper extremities of the person and evaluating the angles of the upper extremities.
- ASB19 System according to ASB15, comprising a microphone (193) for recording acoustic signals, pain vocalization module (5055) for classifying the intensity and frequency of the acoustic signals and determining a scale value representing pain vocalization.
- ASB20 System according to ASB 19, further comprising an audio source position determination module (4420) for evaluating the position of the source of acoustic signals and a
- ASB21 System according to ASB 15, comprising a ventilation device identification module (5065) for identification of a device for artificial ventilation.
- ASB22 System according to ASB 15, comprising a pain perception evaluation module (5085) for evaluating sensors attached to a person.
- ASB23 System according to ASB 15, the sensor for contactless detection of the person being a 2D and / or 3D camera (185), a LID AR (1), a radar and / or ultrasonic sensor (194).
- the determination of the blood pressure is characterized here by the following aspects ABI to AB16:
- a computer-implemented method for determining cardiovascular parameters of a person comprising
- AB 10 System for determining cardiovascular parameters of a person, comprising a computer (9), a memory (10) and a camera (185), further comprising a body region detection module (4810) for detecting body regions, body region tracking Module (4815), and a cardiovascular movement module (5110) for recording movements that can be traced back to cardiovascular activities.
- ABI 1 System according to AB 10, further comprising a face recognition module (5005) and a face candidate region module (5010).
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Surgery (AREA)
- Medical Informatics (AREA)
- Public Health (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- Molecular Biology (AREA)
- Animal Behavior & Ethology (AREA)
- Veterinary Medicine (AREA)
- Pathology (AREA)
- Biophysics (AREA)
- Physiology (AREA)
- General Physics & Mathematics (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Dentistry (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Human Computer Interaction (AREA)
- Robotics (AREA)
- Psychiatry (AREA)
- Social Psychology (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radar, Positioning & Navigation (AREA)
- Cardiology (AREA)
- Mechanical Engineering (AREA)
- Radiology & Medical Imaging (AREA)
- Aviation & Aerospace Engineering (AREA)
- Remote Sensing (AREA)
- Automation & Control Theory (AREA)
- Geometry (AREA)
- Evolutionary Computation (AREA)
- Optics & Photonics (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
Abstract
Description
Claims
Applications Claiming Priority (4)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| DE102019123304 | 2019-08-30 | ||
| DE102020102315 | 2020-01-30 | ||
| DE102020112853 | 2020-05-12 | ||
| PCT/EP2020/074240 WO2021038109A1 (de) | 2019-08-30 | 2020-08-31 | System zur erfassung von bewegungsabläufen und/oder vitalparametern einer person |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| EP4003164A1 true EP4003164A1 (de) | 2022-06-01 |
Family
ID=72292546
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| EP20764658.9A Pending EP4003164A1 (de) | 2019-08-30 | 2020-08-31 | System zur erfassung von bewegungsabläufen und/oder vitalparametern einer person |
Country Status (4)
| Country | Link |
|---|---|
| US (1) | US20220331028A1 (de) |
| EP (1) | EP4003164A1 (de) |
| CN (2) | CN114980810B (de) |
| WO (1) | WO2021038109A1 (de) |
Families Citing this family (71)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019222641A1 (en) * | 2018-05-18 | 2019-11-21 | Corindus, Inc. | Remote communications and control system for robotic interventional procedures |
| KR102715879B1 (ko) * | 2019-03-07 | 2024-10-14 | 삼성전자주식회사 | 전자 장치 및 그 제어 방법 |
| US11129032B2 (en) * | 2019-11-26 | 2021-09-21 | Motorola Mobility Llc | Optimal device position for wireless communication |
| KR102777562B1 (ko) * | 2020-01-06 | 2025-03-11 | 주식회사 비햅틱스 | 촉각자극 제공 시스템 |
| US12162516B2 (en) * | 2020-02-18 | 2024-12-10 | Toyota Motor North America, Inc. | Determining transport operation level for gesture control |
| US12430889B2 (en) | 2020-02-18 | 2025-09-30 | Toyota Motor North America, Inc. | Distinguishing gesture actions among transport occupants |
| US20210275807A1 (en) * | 2020-03-06 | 2021-09-09 | Northwell Health, Inc. | System and method for determining user intention from limb or body motion or trajectory to control neuromuscular stimuation or prosthetic device operation |
| SG10202002676SA (en) * | 2020-03-23 | 2021-10-28 | Nec Asia Pacific Pte Ltd | Method and apparatus for determining a contextual threshold |
| KR20210130325A (ko) * | 2020-04-21 | 2021-11-01 | 현대자동차주식회사 | 노이즈 제어 장치, 그를 가지는 차량 및 그 제어 방법 |
| US12125137B2 (en) * | 2020-05-13 | 2024-10-22 | Electronic Caregiver, Inc. | Room labeling drawing interface for activity tracking and detection |
| US11796637B1 (en) * | 2020-09-10 | 2023-10-24 | Amazon Technologies, Inc. | Fall detection on uneven surfaces using radar |
| JP2022052345A (ja) * | 2020-09-23 | 2022-04-04 | キヤノンメディカルシステムズ株式会社 | 超音波診断装置、撮像方法、及び撮像プログラム |
| WO2022086161A1 (ko) * | 2020-10-20 | 2022-04-28 | 삼성전자주식회사 | 로봇 및 그 제어 방법 |
| KR20230091149A (ko) * | 2020-11-06 | 2023-06-22 | 힌지 헬스 인크. | 컴퓨터 비전을 통해 객체의 움직임에 대한 인사이트 도출 |
| JP7591727B2 (ja) * | 2020-12-21 | 2024-11-29 | パナソニックIpマネジメント株式会社 | 在席情報管理システムおよび在席情報管理方法 |
| US20220222853A1 (en) * | 2020-12-22 | 2022-07-14 | Industrial Technology Research Institute | Method and device for calculating the distance between eyes and the eyes? target |
| EP4278328A1 (de) * | 2021-01-13 | 2023-11-22 | Signify Holding B.V. | Erkennung menschlicher gesichtsrichtungen unter verwendung von wärmebildern aus eingebetteten overhead-sensoren |
| EP4043914B1 (de) * | 2021-02-15 | 2025-09-03 | Leica Geosystems AG | Six-dof-messhilfsmittel |
| KR102862315B1 (ko) * | 2021-02-25 | 2025-09-18 | 체리쉬 헬스, 인크. | 정의된 영역 내에서 객체를 추적하기 위한 기술 |
| CN112926514A (zh) * | 2021-03-26 | 2021-06-08 | 哈尔滨工业大学(威海) | 一种多目标检测及跟踪方法、系统、存储介质及应用 |
| US12433496B2 (en) * | 2021-04-29 | 2025-10-07 | National University Of Singapore | Vital sign detection apparatus and system and data processing method |
| US20240242842A1 (en) * | 2021-05-17 | 2024-07-18 | Sony Group Corporation | Information processing apparatus, information processing method, and program |
| CN113536901B (zh) * | 2021-05-31 | 2025-09-05 | 浙江大华技术股份有限公司 | 一种目标捕获方法以及相关装置 |
| CN113456042A (zh) * | 2021-06-30 | 2021-10-01 | 浙江师范大学 | 一种基于3d cnn的无接触面部血压测量方法 |
| US20230024435A1 (en) * | 2021-07-22 | 2023-01-26 | Kar-Han Tan | Autonomous mobile robot |
| CN113743239A (zh) * | 2021-08-12 | 2021-12-03 | 青岛图灵科技有限公司 | 行人重识别方法、装置及电子设备 |
| BR102021017369A2 (pt) * | 2021-09-01 | 2023-03-14 | Foxconn Brasil Industria E Comercio Ltda. | Estação de auto-atendimento e sistema de gerenciamento de informações médicas para triagem de consultas médicas |
| US20230082016A1 (en) * | 2021-09-15 | 2023-03-16 | Covidien Lp | Mask for non-contact respiratory monitoring |
| CN113925476A (zh) * | 2021-10-18 | 2022-01-14 | 谢俊 | 一种检测人体心率的智能系统及其检测方法 |
| CN114147702B (zh) * | 2021-11-04 | 2024-07-19 | 合肥欣奕华智能机器股份有限公司 | 一种机器人及手指端的位置调整方法 |
| CN114330384B (zh) * | 2021-11-24 | 2024-06-25 | 南京邮电大学 | 一种基于无源rfid相位信息的手臂运动轨迹追踪方法 |
| US12548151B2 (en) * | 2022-02-18 | 2026-02-10 | The Mitre Corporation | Systems and methods for remote optical screening of individuals in entryways |
| DE102022110175A1 (de) * | 2022-04-27 | 2023-11-02 | Bearcover GmbH | Überwachungsvorrichtung und Verfahren zum Betrieb einer Überwachungsvorrichtung |
| EP4276770B1 (de) * | 2022-05-09 | 2024-10-09 | Axis AB | Objektneuidentifizierung in videodatenströmen |
| US20250208698A1 (en) * | 2022-07-03 | 2025-06-26 | Chian Chiu Li | Methods to Use Robot and Buttons in Surveys |
| CN114898471B (zh) * | 2022-07-12 | 2022-09-30 | 华中科技大学 | 一种基于人体骨架特征的行为检测方法及存储介质 |
| CN115588234A (zh) * | 2022-09-29 | 2023-01-10 | 北京电影学院 | 一种基于人工智能的步态识别声音制作方法 |
| DE102023123138A1 (de) | 2022-10-31 | 2024-05-02 | tediro GmbH | Objektlokalisierung durch einen mobilen Roboter |
| DE102022128862A1 (de) | 2022-10-31 | 2024-05-02 | tediro GmbH | Objektlokalisierung durch einen mobilen Roboter |
| DE102022128864A1 (de) | 2022-10-31 | 2024-05-02 | tediro GmbH | Objektlokalisierung durch einen mobilen Roboter |
| WO2024097300A1 (en) * | 2022-11-03 | 2024-05-10 | Tellus You Care, Inc. | Mapping a living area using lidar |
| TWI880130B (zh) * | 2022-11-21 | 2025-04-11 | 立積電子股份有限公司 | 雷達裝置、訊號處理電路及用於雷達裝置的訊號處理方法 |
| US12430949B2 (en) * | 2023-01-02 | 2025-09-30 | Valeo Schalter Und Sensoren Gmbh | Systems and methods for person classification and gesture recognition |
| US12608970B2 (en) * | 2023-03-27 | 2026-04-21 | Taiwan Semiconductor Manufacturing Company, Ltd. | Graph algorithm and motion capture for improving manufacturing processes |
| US12454358B2 (en) * | 2023-05-08 | 2025-10-28 | B/E Aerospace, Inc. | Aircraft seat motion planner and seat architecture |
| CN121079698A (zh) * | 2023-05-10 | 2025-12-05 | 柯惠有限合伙公司 | 使用机器学习对对象生理信号进行上下文化 |
| JP2025001757A (ja) * | 2023-06-21 | 2025-01-09 | キヤノン株式会社 | 映像処理装置、映像処理方法、および、プログラム |
| CN116935016B (zh) * | 2023-07-28 | 2024-03-08 | 上海济目科技有限公司 | 一种集成隧道施工现场风险识别与激光定位方法 |
| WO2025028199A1 (ja) * | 2023-07-31 | 2025-02-06 | 株式会社Orgo | 情報処理装置、情報処理方法、及び情報処理プログラム |
| CN116740846B (zh) * | 2023-08-02 | 2024-08-27 | 深圳零和壹物联科技有限公司 | 一种rfid智能顶装式门禁终端控制方法 |
| US12487329B2 (en) * | 2024-02-09 | 2025-12-02 | Ford Global Technologies, Llc | In-vehicle biometrics with ultrawideband radar |
| CN117806356B (zh) * | 2024-02-28 | 2024-07-19 | 南京大学 | 一种基于物联网的飞艇数据传输管理系统及方法 |
| CN118051841B (zh) * | 2024-04-12 | 2024-08-23 | 上海迎智正能健康科技有限公司 | 一种基于生理参数提取的情绪识别方法及系统 |
| CN118351595A (zh) * | 2024-04-28 | 2024-07-16 | 北京小米机器人技术有限公司 | 步态识别方法、装置、电子设备及存储介质 |
| KR20250165155A (ko) * | 2024-05-17 | 2025-11-25 | 주식회사 메디칼솔루션시스템 | 이동형 보행 분석 로봇 |
| CN118398850B (zh) * | 2024-07-01 | 2024-09-17 | 同济大学 | 一种燃料电池闭环状态管理方法、装置、介质及产品 |
| DE102024120233A1 (de) | 2024-07-17 | 2026-01-22 | TEDIRO Healthcare Robotics GmbH | Wendepositionsermittlung eines Serviceroboters |
| WO2026017691A1 (de) | 2024-07-17 | 2026-01-22 | TEDIRO Healthcare Robotics GmbH | Verfahren zum betrieb einer personenerfassungsvorrichtung |
| DE102024120226A1 (de) | 2024-07-17 | 2026-01-22 | TEDIRO Healthcare Robotics GmbH | Vorrichtung und Verfahren für die Bewegungsanalyse einer Person und die Feedbackgenerierung |
| DE102024120232A1 (de) * | 2024-07-17 | 2026-01-22 | TEDIRO Healthcare Robotics GmbH | Sicherheitsbewegungssteuerung für einen mobilen Roboter |
| DE102024120228A1 (de) | 2024-07-17 | 2026-01-22 | TEDIRO Healthcare Robotics GmbH | Vorrichtung und Verfahren für Bewegungsfeedback beim Laufen |
| KR20260021890A (ko) * | 2024-08-07 | 2026-02-19 | 주식회사 메디칼솔루션시스템 | 회전 라이다를 적용한 이동형 보행 분석 로봇 |
| GB2643435A (en) * | 2024-08-15 | 2026-02-18 | Bae Systems Plc | Vehicle movement control |
| CN118839937B (zh) * | 2024-08-30 | 2025-01-24 | 国网甘肃省电力公司经济技术研究院 | 基于曲线纹理相似度匹配的分布式光伏本地平衡消纳方法 |
| US20260062025A1 (en) * | 2024-09-05 | 2026-03-05 | GM Global Technology Operations LLC | Transparent operator impairment detection for a motor vehicle |
| CN119188800B (zh) * | 2024-09-23 | 2025-05-06 | 深圳市视壮科技有限公司 | 一种智慧医疗养老的智能陪护机器人 |
| CN119622496B (zh) * | 2024-11-25 | 2025-08-01 | 哈尔滨理工大学 | 一种基于坐垫的坐姿分类方法 |
| CN119888784B (zh) * | 2024-12-20 | 2025-12-09 | 北京师范大学珠海校区 | 一种基于多特征循环确认的在线行人重识别方法及其系统 |
| TWI896488B (zh) * | 2025-02-05 | 2025-09-01 | 國立成功大學 | 輕度認知障礙的檢測方法 |
| CN119871434B (zh) * | 2025-02-28 | 2026-03-24 | 北京银河通用机器人股份有限公司 | 防摔倒方法、装置及机器人 |
| CN120370991B (zh) * | 2025-04-18 | 2025-11-04 | 广州泛海智能科技有限公司 | 四足机器人多场景智能巡检与资源分配优化方法及系统 |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2011110644A (ja) * | 2009-11-26 | 2011-06-09 | Fujitsu Ltd | 自律移動ロボット |
| DE102010063818A1 (de) * | 2010-12-21 | 2012-06-21 | Offis E.V. | Verfahren und Vorrichtung zur automatischen Erfassung der Mobilität einer Person |
| WO2019061126A1 (zh) * | 2017-09-28 | 2019-04-04 | 法玛科技顾问股份有限公司 | 自动随行系统 |
Family Cites Families (54)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5222495A (en) | 1990-02-02 | 1993-06-29 | Angiomedics Ii, Inc. | Non-invasive blood analysis by near infrared absorption measurements using two closely spaced wavelengths |
| AU2245092A (en) | 1991-12-31 | 1993-07-28 | Vivascan Corporation | Blood constituent determination based on differential spectral analysis |
| US6172743B1 (en) | 1992-10-07 | 2001-01-09 | Chemtrix, Inc. | Technique for measuring a blood analyte by non-invasive spectrometry in living tissue |
| US5460177A (en) | 1993-05-07 | 1995-10-24 | Diasense, Inc. | Method for non-invasive measurement of concentration of analytes in blood using continuous spectrum radiation |
| US5533509A (en) | 1993-08-12 | 1996-07-09 | Kurashiki Boseki Kabushiki Kaisha | Method and apparatus for non-invasive measurement of blood sugar level |
| US6240306B1 (en) | 1995-08-09 | 2001-05-29 | Rio Grande Medical Technologies, Inc. | Method and apparatus for non-invasive blood analyte measurement with fluid compartment equilibration |
| US6088605A (en) | 1996-02-23 | 2000-07-11 | Diasense, Inc. | Method and apparatus for non-invasive blood glucose sensing |
| US6008889A (en) | 1997-04-16 | 1999-12-28 | Zeng; Haishan | Spectrometer system for diagnosis of skin disease |
| US7702420B2 (en) * | 2004-07-07 | 2010-04-20 | Panasonic Corporation | Method for making mobile unit accompany objective person |
| US20090298025A1 (en) | 2007-05-09 | 2009-12-03 | Oregon Health & Science University | Object recognition testing tools and techniques for measuring cognitive ability and cognitive impairment |
| JP4371153B2 (ja) * | 2007-06-15 | 2009-11-25 | トヨタ自動車株式会社 | 自律移動装置 |
| WO2010114736A1 (en) | 2009-04-01 | 2010-10-07 | The Curators Of The University Of Missouri | Optical spectroscopy device for non-invasive blood glucose detection and associated method of use |
| GB0917600D0 (en) | 2009-10-07 | 2009-11-25 | Univ Edinburgh | Testing apparatus and method |
| US8698888B2 (en) * | 2009-10-30 | 2014-04-15 | Medical Motion, Llc | Systems and methods for comprehensive human movement analysis |
| CN101862245A (zh) | 2010-05-28 | 2010-10-20 | 上海市古美高级中学 | 医院服务机器人 |
| US20170011258A1 (en) | 2010-06-07 | 2017-01-12 | Affectiva, Inc. | Image analysis in support of robotic manipulation |
| US10628741B2 (en) | 2010-06-07 | 2020-04-21 | Affectiva, Inc. | Multimodal machine learning for emotion metrics |
| US9189068B2 (en) | 2011-03-14 | 2015-11-17 | Lg Electronics Inc. | Apparatus and a method for gesture recognition |
| US9117274B2 (en) | 2011-08-01 | 2015-08-25 | Fuji Xerox Co., Ltd. | System and method for interactive markerless paper documents in 3D space with mobile cameras and projectors |
| US8971572B1 (en) | 2011-08-12 | 2015-03-03 | The Research Foundation For The State University Of New York | Hand pointing estimation for human computer interaction |
| CN103930027B (zh) | 2011-10-07 | 2016-03-02 | 皇家飞利浦有限公司 | 用于监测患者并检测该患者的谵妄的监测系统 |
| US9171131B2 (en) | 2012-06-22 | 2015-10-27 | Integrated Deficit Examinations, LLC | Device and methods for mobile monitoring and assessment of clinical function through sensors and interactive patient responses |
| US9659150B2 (en) | 2012-07-20 | 2017-05-23 | Care Innovations, Llc | Method for assessing cognitive function and predicting cognitive decline through quantitative assessment of the TUG test |
| CN203338133U (zh) | 2012-11-12 | 2013-12-11 | 常熟理工学院 | 一种智能医疗服务机器人 |
| US9423879B2 (en) | 2013-06-28 | 2016-08-23 | Chia Ming Chen | Systems and methods for controlling device operation according to hand gestures |
| CN203527474U (zh) | 2013-07-09 | 2014-04-09 | 常熟理工学院 | 一种助老服务机器人 |
| CN103956171B (zh) | 2014-04-01 | 2017-06-13 | 中国科学院软件研究所 | 一种多通道简易精神状态检测系统 |
| US9594964B2 (en) | 2014-06-12 | 2017-03-14 | GM Global Technology Operations LLC | Vision-based wet road surface detection using texture analysis |
| DE102014108424B3 (de) | 2014-06-16 | 2015-06-11 | Johann Wolfgang Goethe-Universität | Nicht-invasive Stoffanalyse |
| US10268277B2 (en) | 2014-09-30 | 2019-04-23 | Hewlett-Packard Development Company, L.P. | Gesture based manipulation of three-dimensional images |
| WO2016058085A1 (en) | 2014-10-16 | 2016-04-21 | Arb Labs Inc. | Systems, methods and devices for monitoring game activities |
| US20180361581A1 (en) * | 2015-01-06 | 2018-12-20 | Discovery Robotics | Robotic platform with following mode |
| US9690984B2 (en) | 2015-04-14 | 2017-06-27 | Microsoft Technology Licensing, Llc | Two-dimensional infrared depth sensing |
| CN204772554U (zh) | 2015-06-30 | 2015-11-18 | 广州绿松生物科技有限公司 | 智能健康服务机器人 |
| CN104889994A (zh) | 2015-06-30 | 2015-09-09 | 广州绿松生物科技有限公司 | 智能健康服务机器人 |
| CN105082149B (zh) | 2015-08-24 | 2017-10-20 | 华南理工大学 | 一种可实现血氧饱和度检测的健康服务机器人 |
| CN105078450B (zh) | 2015-08-24 | 2018-02-27 | 华南理工大学 | 一种可实现脑电检测的健康服务机器人 |
| CN105078445B (zh) | 2015-08-24 | 2018-11-02 | 华南理工大学 | 基于健康服务机器人的老年人健康服务系统 |
| CN105078449B (zh) | 2015-08-24 | 2018-07-20 | 华南理工大学 | 基于健康服务机器人的老年痴呆症监护系统 |
| EP3205269B1 (de) * | 2016-02-12 | 2024-01-03 | Tata Consultancy Services Limited | System und verfahren zur analyse des haltungsgleichgewichts einer person |
| CN205950753U (zh) | 2016-05-14 | 2017-02-15 | 深圳市华科安测信息技术有限公司 | 医院导诊服务机器人 |
| CN107544266A (zh) | 2016-06-28 | 2018-01-05 | 广州零号软件科技有限公司 | 家庭健康服务机器人 |
| CN106407715A (zh) | 2016-10-19 | 2017-02-15 | 上海派毅智能科技有限公司 | 一种智能服务机器人健康辨识系统及方法 |
| CN108073104A (zh) | 2016-11-10 | 2018-05-25 | 贺州学院 | 基于stm32嵌入式多用途护理服务机器人 |
| RU2649048C1 (ru) | 2016-11-25 | 2018-03-29 | Самсунг Электроникс Ко., Лтд. | Система компактного спектрометра, предназначенного для неинвазивного измерения спектров поглощения и пропускания образцов биологической ткани |
| CN106709254B (zh) | 2016-12-29 | 2019-06-21 | 天津中科智能识别产业技术研究院有限公司 | 一种医疗诊断机器人系统 |
| CN106671105A (zh) | 2017-01-17 | 2017-05-17 | 五邑大学 | 面向老人的智能陪护机器人 |
| CN206833244U (zh) | 2017-04-21 | 2018-01-02 | 山东大学 | 一种基于云平台的医院服务机器人 |
| US11161236B2 (en) * | 2017-09-14 | 2021-11-02 | Sony Interactive Entertainment Inc. | Robot as personal trainer |
| CN107518989A (zh) | 2017-10-21 | 2017-12-29 | 长沙展朔轩兴信息科技有限公司 | 医院服务机器人 |
| CN107598943A (zh) | 2017-10-30 | 2018-01-19 | 文杨 | 一种陪伴老人的机器人 |
| CN108039193A (zh) | 2017-11-17 | 2018-05-15 | 哈尔滨工大服务机器人有限公司 | 一种自动生成体检报告的方法及装置 |
| CN108053889A (zh) | 2017-12-20 | 2018-05-18 | 中国科学院合肥物质科学研究院 | 一种基于Agent技术的健康促进服务机器人 |
| CN108422427A (zh) | 2018-03-21 | 2018-08-21 | 南通市巨久新材料科技有限公司 | 一种康复服务机器人 |
-
2020
- 2020-08-31 EP EP20764658.9A patent/EP4003164A1/de active Pending
- 2020-08-31 WO PCT/EP2020/074240 patent/WO2021038109A1/de not_active Ceased
- 2020-08-31 CN CN202080076764.2A patent/CN114980810B/zh active Active
- 2020-08-31 CN CN202511318678.9A patent/CN121337261A/zh active Pending
- 2020-08-31 US US17/639,254 patent/US20220331028A1/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2011110644A (ja) * | 2009-11-26 | 2011-06-09 | Fujitsu Ltd | 自律移動ロボット |
| DE102010063818A1 (de) * | 2010-12-21 | 2012-06-21 | Offis E.V. | Verfahren und Vorrichtung zur automatischen Erfassung der Mobilität einer Person |
| WO2019061126A1 (zh) * | 2017-09-28 | 2019-04-04 | 法玛科技顾问股份有限公司 | 自动随行系统 |
| US20200238527A1 (en) * | 2017-09-28 | 2020-07-30 | Farmer Licensing Co., Ltd. | An automatic accompanying system |
Non-Patent Citations (2)
| Title |
|---|
| POPOV VASIL L ET AL: "Detection and Following of Moving Targets by an Indoor Mobile Robot using Microsoft Kinect and 2D Lidar Data", 2018 15TH INTERNATIONAL CONFERENCE ON CONTROL, AUTOMATION, ROBOTICS AND VISION (ICARCV), IEEE, 18 November 2018 (2018-11-18), pages 280 - 285, XP033480795, [retrieved on 20181218], DOI: 10.1109/ICARCV.2018.8581231 * |
| See also references of WO2021038109A1 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114980810B (zh) | 2025-10-10 |
| WO2021038109A1 (de) | 2021-03-04 |
| US20220331028A1 (en) | 2022-10-20 |
| WO2021038109A9 (de) | 2021-05-27 |
| CN114980810A (zh) | 2022-08-30 |
| CN121337261A (zh) | 2026-01-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2021038109A1 (de) | System zur erfassung von bewegungsabläufen und/oder vitalparametern einer person | |
| US11659350B1 (en) | Metaverse dating | |
| US11602855B2 (en) | Autonomous companion mobile robot and system | |
| US11989340B2 (en) | Systems, methods, apparatuses and devices for detecting facial expression and for tracking movement and location in at least one of a virtual and augmented reality system | |
| Xu et al. | Continuous hybrid BCI control for robotic arm using noninvasive electroencephalogram, computer vision, and eye tracking | |
| EP3908969A1 (de) | Verfahren und system zur erfassung des bewegungsablaufs einer person | |
| JP7625673B2 (ja) | コミュニケーション方法及びシステム | |
| US20220044821A1 (en) | Systems and methods for diagnosing a stroke condition | |
| CN116945156B (zh) | 一种基于计算机视觉技术的老年人智能陪护系统 | |
| US20150320343A1 (en) | Motion information processing apparatus and method | |
| Romdhane et al. | Automatic video monitoring system for assessment of Alzheimer's disease symptoms | |
| EP3846678A1 (de) | Therapeutische raumbeurteilung | |
| Buchs et al. | Waist-up protection for blind individuals using the EyeCane as a primary and secondary mobility aid | |
| US20250095292A1 (en) | Human-machine interaction collection method, apparatus, and system for wearable extended reality device | |
| CN108714020A (zh) | 便携自适应心理调节仪、实时眼动延时求取方法及存储有计算机程序的计算机可读存储介质 | |
| DE102022104499A1 (de) | Synchronisierung von Bewegungen eines Serviceroboters mit denen einer Person | |
| Lanata et al. | Robust head mounted wearable eye tracking system for dynamical calibration | |
| Penčić et al. | Anthropomorphic robotic eyes: Structural design and non-verbal communication effectiveness | |
| Atallah et al. | An ear-worn sensor for the detection of gait impairment after abdominal surgery | |
| Mazzei et al. | Robotic social therapy on children with autism: preliminary evaluation through multi-parametric analysis | |
| Cui et al. | Research on multimodal control method for prosthetic hands based on visuo-tactile and arm motion measurement | |
| KR102743861B1 (ko) | 디지털 콘텐츠를 이용한 환자 치료 시스템 | |
| Akshay et al. | ialert: An alert system based on eye gaze for human assistance | |
| Sinyukov | Semi-autonomous robotic wheelchair controlled with low throughput human-machine interfaces | |
| de Almeida | Intellwheels-Controlling an Intelligent Wheelchair using a Multimodal Interface |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: UNKNOWN |
|
| STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: THE INTERNATIONAL PUBLICATION HAS BEEN MADE |
|
| PUAI | Public reference made under article 153(3) epc to a published international application that has entered the european phase |
Free format text: ORIGINAL CODE: 0009012 |
|
| STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: REQUEST FOR EXAMINATION WAS MADE |
|
| 17P | Request for examination filed |
Effective date: 20220224 |
|
| AK | Designated contracting states |
Kind code of ref document: A1 Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR |
|
| DAV | Request for validation of the european patent (deleted) | ||
| DAX | Request for extension of the european patent (deleted) | ||
| RAP1 | Party data changed (applicant data changed or rights of an application transferred) |
Owner name: TEDIRO GMBH |
|
| STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: EXAMINATION IS IN PROGRESS |
|
| 17Q | First examination report despatched |
Effective date: 20230717 |
|
| 19U | Interruption of proceedings before grant |
Effective date: 20231228 |
|
| 19W | Proceedings resumed before grant after interruption of proceedings |
Effective date: 20250401 |
|
| RAP3 | Party data changed (applicant data changed or rights of an application transferred) |
Owner name: TEDIRO HEALTHCARE ROBOTICS GMBH |