WO2023017919A1 - 관절 상태를 정량화하기 위한 의료 영상 분석 방법, 의료 영상 분석 장치, 및 의료 영상 분석 시스템 - Google Patents
관절 상태를 정량화하기 위한 의료 영상 분석 방법, 의료 영상 분석 장치, 및 의료 영상 분석 시스템 Download PDFInfo
- Publication number
- WO2023017919A1 WO2023017919A1 PCT/KR2021/018218 KR2021018218W WO2023017919A1 WO 2023017919 A1 WO2023017919 A1 WO 2023017919A1 KR 2021018218 W KR2021018218 W KR 2021018218W WO 2023017919 A1 WO2023017919 A1 WO 2023017919A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- medical image
- joint
- region
- target
- image analysis
- 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.)
- Ceased
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H30/00—ICT specially adapted for the handling or processing of medical images
- G16H30/40—ICT specially adapted for the handling or processing of medical images for processing medical images, e.g. editing
-
- 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
- A61B5/107—Measuring physical dimensions, e.g. size of the entire body or parts thereof
- A61B5/1075—Measuring physical dimensions, e.g. size of the entire body or parts thereof for measuring dimensions by non-invasive methods, e.g. for determining thickness of tissue layer
-
- 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
- A61B5/107—Measuring physical dimensions, e.g. size of the entire body or parts thereof
- A61B5/1079—Measuring physical dimensions, e.g. size of the entire body or parts thereof using optical or photographic means
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/45—For evaluating or diagnosing the musculoskeletal system or teeth
- A61B5/4528—Joints
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/155—Segmentation; Edge detection involving morphological operators
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H30/00—ICT specially adapted for the handling or processing of medical images
- G16H30/20—ICT specially adapted for the handling or processing of medical images for handling medical images, e.g. DICOM, HL7 or PACS
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/30—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for calculating health indices; for individual health risk assessment
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10116—X-ray image
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20021—Dividing image into blocks, subimages or windows
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20036—Morphological image processing
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30008—Bone
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2210/00—Indexing scheme for image generation or computer graphics
- G06T2210/41—Medical
Definitions
- the present application relates to a medical image analysis method, a medical image analysis device, and a medical image analysis system. Specifically, the present application relates to a medical image analysis method, a medical image analysis device, and a medical image analysis system for calculating joint state information obtained by quantifying joint states.
- joint spacing In analyzing joint conditions, one of the most important factors is joint spacing, and the reduction of joint spacing is known to have an important relationship with rheumatoid arthritis, degenerative arthritis, cartilage wear, and joint conditions in various parts of the body. . In particular, many studies have proven that joint spacing values are significantly related to joint pain.
- the joint condition was estimated based on the absolute value of the joint interval.
- the absolute value of joint spacing may vary significantly depending on external factors such as gender, race, and body type.
- an error may occur in the absolute value of the joint interval depending on a recording device system and a program for capturing an image.
- a millimeter-scale error occurs, since the numerical change is large, it may cause inaccurate results in identifying the reduction rate of joint spacing compared to normal and comparing with other patients or monitoring the prognosis.
- there is a limitation in determining the joint condition based on the absolute value of the joint interval Accordingly, development of a medical image analysis method, apparatus, and system capable of obtaining objective joint condition information while minimizing the influence of external factors is required.
- One problem to be solved by the present invention is to provide a medical image analysis method for calculating joint state information, a medical image analysis device, and a medical image analysis system.
- a medical image analysis method includes acquiring a target medical image; detecting a target joint gap region from the target medical image; obtaining a first value related to a width of a joint region from the target medical image; obtaining a second value related to a joint spacing from the target joint spacing area; and calculating a target joint condition index indicating a condition of a joint based on the first value and the second value.
- An apparatus for analyzing a medical image includes an image acquiring unit acquiring a target medical image; and a controller that provides joint state information based on the target medical image, wherein the controller acquires the target medical image, detects a target joint gap region from the target medical image, and obtains the target joint space from the target medical image.
- a first value related to the width of the part is obtained
- a second value related to the joint distance is obtained from the target joint distance area
- a target joint state representing a state of a joint is obtained based on the first value and the second value. It can be configured to calculate an indicator.
- objective joint condition information can be obtained by minimizing the influence of external factors such as body shape, race, and gender.
- FIG. 1 is a schematic diagram of a medical image analysis system according to an embodiment of the present application.
- FIG. 2 is a diagram illustrating operations of a medical image analysis apparatus according to an embodiment of the present application.
- FIG. 3 is a flowchart illustrating a method of analyzing a medical image according to an embodiment of the present application.
- FIG. 4 is a flowchart embodying a step of detecting a target joint gap region according to an embodiment of the present application.
- FIG. 5 is a diagram illustrating an aspect of detecting a region of interest and a target joint gap region according to an embodiment of the present application.
- FIG. 6 is a flowchart illustrating a method of training a neural network model for obtaining a target joint gap area according to an embodiment of the present application.
- FIG. 7 is a diagram illustrating an aspect of learning a neural network model for obtaining a target joint gap area according to an embodiment of the present application.
- FIG. 8 is a flowchart embodying a step of detecting a target joint gap region according to an embodiment of the present application.
- FIG. 9 is a schematic diagram illustrating an aspect of obtaining a target joint gap area using a trained neural network model according to an embodiment of the present application.
- FIG. 10 is a flowchart embodying a step of obtaining a first value related to a width of a joint part according to an embodiment of the present application.
- FIG. 11 is a diagram illustrating an aspect of obtaining a first value according to an embodiment of the present application.
- FIG. 12 is a flowchart embodying a step of obtaining a second value related to a joint interval according to an embodiment of the present application.
- FIG. 13 is a diagram illustrating an aspect of obtaining a second value according to an embodiment of the present application.
- a medical image analysis method includes acquiring a target medical image; detecting a target joint gap region from the target medical image; obtaining a first value related to a width of a joint region from the target medical image; obtaining a second value related to a joint spacing from the target joint spacing area; and calculating a target joint condition index indicating a condition of a joint based on the first value and the second value.
- the detecting of the target joint gap region may include detecting a region of interest from the target medical image; and obtaining the target joint gap region included in the region of interest by performing segmentation on the region of interest.
- the region of interest may be acquired using a first neural network model learned to receive medical images and output regions including joint regions.
- the segmentation may be performed using a second neural network model learned to output a joint gap region by receiving a medical image including a region of interest.
- the obtaining of the first value may include: detecting a first point and a second point adjacent to a boundary between a bone region and an outer region of the bone from the target medical image; obtaining first coordinate information of the first point and second coordinate information of the second point; and calculating the first value based on the first coordinate information and the second coordinate information.
- the first point and the second point may be obtained based on a difference between brightness of the bone region included in the target medical image and brightness of an outer region of the bone.
- the first point and the second point receive a medical image including the bone region and an outer region of the bone, and the first region corresponding to the first point and the first region It can be obtained through a neural network model trained to output the second area corresponding to the two points.
- the obtaining of the second value may include obtaining a section of interest from among the joint gap region; obtaining a plurality of joint spacing values within the region of interest; and obtaining the second value based on the plurality of joint distance values.
- the second value may be a minimum value among the plurality of joint distance values or an average value of the plurality of joint distance values.
- the target joint condition index may be defined as a ratio of the second value to the first value.
- a computer-readable recording medium recording a program for executing the medical image analysis method may be provided.
- An apparatus for analyzing a medical image includes an image acquiring unit acquiring a target medical image; and a controller that provides joint state information based on the target medical image, wherein the controller acquires the target medical image, detects a target joint gap region from the target medical image, and obtains the target joint space from the target medical image.
- a first value related to the width of the part is obtained
- a second value related to the joint distance is obtained from the target joint distance area
- a target joint state representing a state of a joint is obtained based on the first value and the second value. It can be configured to calculate an indicator.
- FIGS. 1 to 13 the medical image analysis method, medical image analysis apparatus, and medical image analysis system of the present application will be described with reference to FIGS. 1 to 13 .
- FIG. 1 is a schematic diagram of a medical image analysis system 10 according to an embodiment of the present application.
- the medical image analysis system 10 may include a medical image acquisition device 100 and a medical image analysis device 1000 .
- the medical image capture device 100 may capture a medical image.
- the medical image capture device 100 may include any type of medical image including magnetic resonance imaging, computerized tomography equipment, X-ray equipment, and the like. It may mean encompassing a device that obtains.
- a medical image obtained by the medical image capture device 100 may be a 2D image. In this case, the medical image may include pixel information related to pixel coordinates, color, and intensity.
- a medical image obtained by the medical image capture device 100 may be a 3D image. In this case, the medical image may include pixel information related to the coordinates, color, and intensity of the voxel.
- the medical image analysis apparatus 1000 may obtain joint state information by analyzing a medical image. More specifically, the medical image analysis apparatus 1000 may detect a joint area from a medical image and quantify information related to a joint condition based on the joint area.
- the joint area may include an inter-articular area (or a joint space area), a bone area adjacent to the inter-articular area, and the like.
- the medical image analysis apparatus 1000 may include a transceiver 1100, a memory 1200, and a controller 1300.
- the transceiver 1100 of the medical image analysis apparatus 1000 may communicate with any external device including the medical image capture apparatus 100 .
- the medical image analysis apparatus 1000 may receive a medical image captured by the medical image capture apparatus 100 through the transceiver 1100 .
- the medical image analysis apparatus 1000 may transmit acquired joint state information to an arbitrary external device including the medical image capture apparatus 100 through the transceiver 1100 .
- the medical image analysis apparatus 1000 may transmit and receive various types of data by accessing a network through a transceiver.
- the transceiver may largely include a wired type and a wireless type. Since the wired type and the wireless type each have advantages and disadvantages, the medical image analysis apparatus 1000 may be provided with both the wired type and the wireless type in some cases.
- a wireless local area network (WLAN)-based communication method such as Wi-Fi may be mainly used.
- a wireless type a cellular communication, eg, LTE, 5G-based communication method may be used.
- the wireless communication protocol is not limited to the above example, and any suitable wireless type communication method may be used.
- LAN Local Area Network
- USB Universal Serial Bus
- the memory 1200 of the medical image analysis apparatus 1000 may store various kinds of information. Various types of data may be temporarily or semi-permanently stored in the memory 1200 . Examples of the memory 1200 include a hard disk drive (HDD), a solid state drive (SSD), a flash memory, a read-only memory (ROM), a random access memory (RAM), and the like. This can be.
- the memory 1200 may be provided in a form embedded in the medical image analysis apparatus 1000 or in a detachable form.
- the memory 1200 includes an operating system (OS) for driving the medical image analysis device 1000, a program for operating each component of the medical image analysis device 1000, and the components of the medical image analysis device 1000.
- OS operating system
- Various data required for operation may be stored.
- the controller 1300 may control overall operations of the medical image analysis apparatus 1000 .
- the controller 1300 analyzes medical images, such as an operation of detecting a region of interest or a target joint gap region from a target medical image, an operation of quantifying a width of a joint part, an operation of quantifying a joint gap, and an operation of calculating joint state information.
- Overall operation of the device 1000 may be controlled.
- the controller 1300 may load and execute a program for overall operation of the medical image analysis apparatus 1000 from the memory 1100 .
- the processor may be implemented as an application processor (AP), a central processing unit (CPU), a microcontroller unit (MCU), or similar devices according to hardware, software, or a combination thereof.
- AP application processor
- CPU central processing unit
- MCU microcontroller unit
- the medical image analysis apparatus 1000 may include any suitable input unit and/or output unit.
- the medical image analysis apparatus 1000 may receive a user's input necessary for analyzing a medical image through an input unit.
- the medical image analysis apparatus 1000 may obtain a user's input for allocating label information to each of a plurality of areas included in a medical image through an input unit.
- the medical image analysis apparatus 1000 may obtain a user's input for setting a region of interest of a joint interval region for acquiring a joint interval value through an input unit.
- the medical image analysis apparatus 1000 may output a result of comparing a target joint condition index and/or a target joint condition index and a reference joint condition index to be described later through an output unit.
- the medical image analysis apparatus 1000 may detect a joint gap region from a medical image. Also, the medical image analysis apparatus 1000 may calculate joint state information based on the joint gap area.
- FIG. 2 is a diagram illustrating operations of the medical image analysis apparatus 1000 according to an embodiment of the present application.
- the medical image analysis apparatus 1000 may include an image acquisition unit.
- the image acquisition unit of the medical image analysis apparatus 1000 may obtain a target medical image obtained from the medical image acquisition apparatus 100 .
- the image capture unit may acquire a target medical image from an external device including the medical image capture device 100 through the transceiver 1100 .
- the medical image analysis apparatus 1000 may include a region of interest detector.
- the region of interest detector of the medical image analysis apparatus 1000 may detect the region of interest included in the target medical image.
- the ROI detector may detect a ROI including a joint region from the target medical image.
- the joint area may include a joint space area and/or a bone area adjacent to the joint space area.
- the ROI detector may detect the ROI using an artificial intelligence technique.
- An operation of detecting a region of interest will be described in detail with reference to FIGS. 4 and 5 .
- an operation of detecting a region of interest from a target medical image may be omitted.
- an operation of detecting the region of interest by the region of interest detector may be omitted.
- the medical image analysis apparatus 1000 may include a joint gap area detector.
- the joint gap region detector of the medical image analysis apparatus 1000 according to an embodiment of the present application may detect a target joint gap region from a target medical image.
- the joint gap region detector may detect the target joint gap region from a region of interest including the joint region included in the target medical image.
- the target joint gap region may refer to a region between arbitrary joints included in the target medical image.
- the joint gap area detector may obtain a target joint gap area by using an artificial intelligence technique.
- the joint gap region detector may acquire the target joint gap region by using a neural network model trained on the basis of a training set in which the interjoint region (or joint gap region) is assigned (or labeled) to the medical image. Contents of acquiring the target joint gap region will be described in detail with reference to FIGS. 4 to 9 .
- the medical image analysis apparatus 1000 may include a joint condition quantification analysis unit.
- the joint state quantification analysis unit may include a joint width analysis unit and a joint spacing analysis unit.
- the joint state quantification analysis unit of the medical image analysis apparatus 1000 according to an embodiment of the present application may calculate joint state information based on a joint region and/or a target joint gap region.
- the joint width analysis unit of the medical image analysis apparatus 1000 may perform an operation of quantifying the width of a joint part based on the joint area included in the target medical image.
- the joint width analyzer of the medical image analysis apparatus 1000 detects an outer point of the joint using an arbitrary image processing technique or an artificial intelligence technique, and quantifies the width of the joint based on the coordinate information of the outer point. information can be derived. Details of quantifying the width of the joint will be described in detail in FIGS. 10 and 11 .
- the joint spacing analysis unit of the medical image analysis apparatus 1000 may perform an operation of quantifying a joint spacing based on a target joint spacing area.
- the joint gap analyzer of the medical image analysis apparatus 1000 may obtain a region of interest related to the target joint gap region and obtain at least one joint gap value within the region of interest.
- the joint spacing analysis unit of the medical image analysis apparatus 1000 may calculate quantitative information related to the joint spacing based on at least one joint spacing value. The content of quantifying the joint spacing will be described in detail in FIGS. 12 and 13 .
- the medical image analysis apparatus 1000 may calculate joint state information based on quantitative information related to the width of a joint and quantitative information related to a joint interval.
- the joint condition quantitative analysis unit of the medical image analysis apparatus 1000 may calculate a joint condition index indicating a joint condition based on quantitative information related to a width of a joint and quantitative information related to a joint interval.
- FIG. 3 is a flowchart illustrating a method of analyzing a medical image according to an embodiment of the present application.
- a medical image analysis method includes acquiring a target medical image (S1100), detecting a target joint gap region (S1200), and obtaining a first value related to the width of a joint (S1200). It may include (S1300), obtaining a second value related to the joint interval (S1400), and calculating a target joint condition index indicating the state of the joint (S1500).
- the medical image analysis apparatus 1000 may acquire a target medical image as an analysis target.
- the medical image analysis apparatus 1000 may obtain a target medical image from the medical image acquisition apparatus 100 or any external device including a database through the transceiver 1100 .
- the medical image analysis apparatus 1000 may detect the target joint gap region from the target medical image.
- the medical image analysis apparatus 1000 acquires, as a region of interest, a region including a joint part included in the target medical image from a target medical image, and precisely analyzes the region of interest to obtain a target joint gap region. It can be.
- the medical image analysis apparatus 1000 may be implemented to acquire a target joint gap region included in the target medical image from the target medical image.
- 4 is a flowchart embodying a step of detecting a target joint gap region according to an embodiment of the present application.
- 5 is a diagram illustrating an aspect of detecting a region of interest and a target joint gap region according to an embodiment of the present application.
- Obtaining the target joint gap region (S1200) may include detecting a region of interest from the target medical image (S1210) and acquiring a target joint gap region included in the region of interest (S1220).
- the medical image analysis apparatus 1000 may be implemented to detect a region including a joint region from the target medical image as the region of interest.
- the medical image analysis apparatus 1000 may acquire a region including a joint region as a region of interest by using any suitable artificial intelligence technique.
- the medical image analysis apparatus 1000 may receive a target medical image and detect a region of interest (ROI) by using a first neural network model trained to output a region including a joint region as a region of interest.
- ROI region of interest
- the first neural network model may be learned based on the medical image and label information assigned to the medical image as a joint area.
- the label information may be automatically assigned to the medical image using arbitrary software or may be manually assigned to the medical image by an arbitrary operator.
- the first neural network model may be trained to receive a medical image and minimize a difference between an output value and label information related to a joint region.
- the first neural network model may be an artificial neural network model based on deep learning.
- the artificial neural network there may be a convolutional neural network, a recurrent neural network, a deep neural network, a generative adversarial network, and the like. However, this is only an example and should be interpreted in a comprehensive sense that includes all of the above-described artificial neural networks, various other types of artificial neural networks, and artificial neural networks in a combination thereof, and does not necessarily have to be a deep learning series.
- the medical image analysis apparatus 1000 may detect the target joint gap region from the target medical image.
- the medical image analysis apparatus 1000 may obtain a target joint space region including an inter-joint region by performing segmentation on the region of interest of the target medical image. For example, segmentation of the region of interest may be performed using any suitable artificial intelligence technique.
- the medical image analysis apparatus 1000 receives a target medical image including a region of interest and detects a target joint gap region by using a second neural network model learned to output an inter-joint region. It can be.
- a neural network model may be used as a model for obtaining a target joint gap area.
- a neural network model may serve as a machine learning model.
- a typical example of a machine learning model may be an artificial neural network.
- a representative example of an artificial neural network is a deep learning artificial neural network including an input layer that receives data, an output layer that outputs results, and a hidden layer that processes data between the input layer and the output layer.
- Specific examples of artificial neural networks include a Convolution Neural Network, a Recurrent Neural Network, a Deep Neural Network, a Generative Adversarial Network, and the like.
- a neural network should be interpreted as a comprehensive meaning that includes all of the above-described artificial neural networks, various other types of artificial neural networks, and artificial neural networks in a combination thereof, and does not necessarily have to be a deep learning series.
- the machine learning model does not necessarily have to be in the form of an artificial neural network model, and in addition, nearest neighbor algorithm (KNN), random forest (Random Forest), support vector machine (SVM), principal component analysis (PCA), etc. may be included.
- KNN nearest neighbor algorithm
- Random Forest random forest
- SVM support vector machine
- PCA principal component analysis
- the techniques mentioned above may include all of the ensemble forms or even forms combined in various ways.
- centering on the artificial neural network it is disclosed in advance that the artificial neural network may be replaced with another machine learning model unless otherwise specified.
- the algorithm for obtaining the target joint space region in the present specification is not necessarily limited to a machine learning model. That is, the algorithm for obtaining the target joint gap area may include various judgment/determination algorithms other than a machine learning model. Therefore, it should be noted that the algorithm for acquiring the target joint gap area in the present specification should be understood as a comprehensive meaning including all types of algorithms for calculating the joint gap area based on medical images. However, in the following description, for convenience of description, the artificial neural network model will be mainly described.
- FIGS. 6 to 9 Details of learning a neural network model for acquiring a target joint gap region according to an embodiment of the present application will be described in detail with reference to FIGS. 6 and 7 .
- FIGS. 8 and 9 details of obtaining a target joint gap region using the learned neural network model will be described.
- FIG. 6 is a flowchart illustrating a method of training a neural network model for obtaining a target joint gap area according to an embodiment of the present application.
- a method of training a neural network model for obtaining a target joint gap region may be performed in the medical image analysis apparatus 1000 .
- the method of learning the neural network model for obtaining the target joint gap area may be performed in an external device separate from the medical image analysis apparatus 1000 .
- learning of a neural network model for acquiring a target joint gap area is performed in the medical image analysis apparatus 1000 .
- this is only an example and is not construed as being limited thereto.
- a method for learning a neural network model for obtaining a target joint gap region includes acquiring a medical image database (S2100), preparing a learning set (S2200), and learning a neural network (S2100). S2300) and obtaining a learned neural network model (S2400).
- the medical image analysis apparatus 1000 obtains a medical image database including a plurality of medical images from the medical image acquisition device 100 or an external device including an arbitrary database.
- the medical image analysis apparatus 1000 may obtain a prepared training set by allocating label information to an inter-joint region included in the medical image.
- the operation of assigning label information to the inter-articular region may be performed using any appropriate software or manually by any operator, similarly to the above description.
- the medical image analysis apparatus 1000 may learn the neural network model based on the medical image and the training set.
- FIG. 7 is a diagram illustrating an aspect of learning a neural network model for obtaining a target joint gap area according to an embodiment of the present application.
- a neural network model may include an input layer, an output layer, and a hidden layer.
- the input layer may receive a medical image
- the output layer may output an output value related to an inter-articular region.
- the hidden layer may have a plurality of nodes connecting the input layer and the output layer.
- the medical image analysis apparatus 1000 may train a neural network to output joint space area information representing an inter-joint area based on a medical image.
- the medical image analysis apparatus 1000 may input a medical image to an input layer and obtain an output value related to an inter-joint region through an output layer.
- the medical image analysis apparatus 1000 may adjust a weight (or parameter) of a node included in a hidden layer based on a difference between label information related to the joint gap region included in the training set and an output value.
- the medical image analysis apparatus 1000 inputs a first medical image acquired from a medical image database to an input layer, output values output through an output layer, and first label information allocated to a joint gap region of the first medical image.
- a weight (or parameter) of a node included in the hidden layer may be updated based on the difference between .
- the medical image analysis apparatus 1000 inputs an Nth medical image obtained from a medical image database to an input layer, output values output through an output layer, and Nth label information allocated to a joint gap region of the Nth medical image. Weights (or parameters) of nodes included in the hidden layer may be repeatedly updated based on the difference between .
- the apparatus 1000 for analyzing medical images can train a neural network model by repeatedly adjusting weights (or parameters) of nodes included in hidden layers so that a difference between label information and an output value related to the joint gap region is minimized. .
- the method of learning a neural network model for obtaining a target joint gap region may include acquiring a second learned neural network model ( S2400 ).
- the medical image analysis apparatus 1000 sets the weights or parameters of nodes included in the learned hidden layer so that the output value and label information output through the output layer are minimized. can be obtained Alternatively, in acquiring the learned second neural network model ( S1400 ), the medical image analysis apparatus 1000 may obtain a second neural network model including a hidden layer including a node having the aforementioned weights or parameters.
- the neural network model for obtaining the target joint gap region may further include verifying the neural network.
- the medical image analysis apparatus 1000 may verify a neural network model based on at least a part of a training set. Specifically, the medical image analysis apparatus 1000 may input at least some of the medical images included in the training set to an input layer of a neural network model and obtain an output value output through an output layer. In addition, the medical image analysis apparatus 1000 determines whether the weight (or parameter) of the node included in the hidden layer of the neural network model is appropriate by comparing the similarity between the output value and the label information related to the medical image included in the training set. can be verified.
- the target medical image may include a plurality of joint regions.
- the medulla bone image may include several joint regions.
- the medical image analysis apparatus 1000 may be implemented to learn a neural network model for each joint region and detect a joint gap region using at least one neural network model learned for each joint region. .
- a joint gap region included in the first joint region is detected using the learned first joint detection neural network model, and for a second joint region, the learned second joint detection neural network model is used to detect the joint interval region included in the first joint region.
- the medical image analysis apparatus 1000 may be configured to detect a joint gap region included in the second joint region.
- FIG. 8 is a flowchart embodying a step of detecting a target joint gap region according to an embodiment of the present application.
- Detecting a target joint gap area includes acquiring a learned second neural network model (S3100) and acquiring a target joint gap area using the learned second neural network model. Step S3200 may be included.
- the medical image analysis apparatus 1000 may obtain the learned second neural network model. For example, weights or parameters of nodes of the second neural network model may be obtained. As another example, the medical image analysis apparatus 1000 may obtain a neural network including a hidden layer including nodes having weights or parameters acquired through learning.
- step S3200 of acquiring the target joint gap region using the learned second neural network model the medical image analysis apparatus 1000 according to an embodiment of the present application is based on the learned second neural network model and the target medical image.
- the target joint spaced area can be obtained.
- FIG. 9 is a schematic diagram illustrating an aspect of obtaining a target joint gap area using a trained neural network model according to an embodiment of the present application.
- the medical image analysis apparatus 1000 inputs a target medical image (or a target medical image including a region of interest) to the input layer of the learned second neural network model, and obtains information related to the target joint gap region through the output layer.
- the medical image analysis apparatus 1000 outputs the target medical image (based on the learned second neural network model and the target medical image).
- a target joint gap region included in a medical image including a region of interest may be acquired.
- the method of analyzing a medical image may include obtaining a first value related to the width of a joint (S1300).
- 10 is a flowchart embodying a step (S1300) of obtaining a first value related to the width of a joint part according to an embodiment of the present application.
- 11 is a diagram illustrating an aspect of obtaining a first value according to an embodiment of the present application.
- Acquiring a first value related to the width of a joint region includes detecting a first point and a second point adjacent to a boundary between a bone region and an outer region of the bone from the target medical image.
- Step (S1310), obtaining first coordinate information of a first point and second coordinate information of a second point (S1320), and calculating a first value based on the first coordinate information and second coordinate information ( S1330) may be included.
- the medical image analysis apparatus 1000 determines the inter-joint region included in the target medical image and the second point. It is possible to detect the adjacent bone region and the outer region of the bone. In detail, the medical image analysis apparatus 1000 determines the first point P1 adjacent to the boundary between the bone region R1 adjacent to the joint and the outer region R2 of the bone from the target medical image (or the medical image including the region of interest). and the second point P2 may be detected.
- the medical image analysis apparatus 1000 may detect the first point P1 and the second point P2 using an arbitrary image processing technique.
- the medical image analysis apparatus 1000 processes a target medical image using an arbitrary image processing technique, and detects a first point P1 and a second point P2 based on brightness of the processed target medical image. can do.
- the medical image analysis apparatus 1000 determines the bone region R1 and the outer region R2 based on the difference between the first brightness of the bone region R1 and the second brightness of the outer region R2.
- the boundary of the liver may be obtained, and a first point P1 and a second point P2 adjacent to the boundary may be obtained.
- the medical image analysis apparatus 1000 may detect the first point P1 and the second point P2 using an artificial intelligence technique. For example, the medical image analysis apparatus 1000 may train a neural network model that acquires both end points of a bone based on a medical image and label information on both end points of a bone adjacent to an inter-articular region. In this case, the medical image analysis apparatus 1000 may detect both end points of the bone, eg, the first point P1 and the second point P2, by using the learned neural network model.
- the medical image analysis apparatus 1000 determines the detected first and second points P1 and P2. Coordinate information can be obtained.
- the target medical image may include a plurality of cells (eg, pixels or voxels), and the target medical image may include cell coordinate information.
- the medical image analysis apparatus 1000 may obtain first coordinate information based on the coordinates of the cell corresponding to the first point P1 and based on the coordinates of the cell corresponding to the second point P2 Second coordinate information may be obtained.
- the medical image analysis apparatus 1000 is related to the width of the joint based on the first coordinate information and the second coordinate information.
- a first value can be obtained.
- the first coordinate information e.g., e1(x,y)
- the second coordinate information e.g., e2(x,y)
- the medical image analysis apparatus 1000 calculates the first value related to the width of the joint by using a Euclidean distance calculation method.
- the medical image analysis apparatus 1000 may be implemented to calculate a first value related to the width of a joint region in consideration of an alignment direction of a target medical image.
- the method of analyzing a medical image may include acquiring a second value related to the joint spacing ( S1400 ).
- FIG. 12 is a flowchart embodying a step (S1400) of acquiring a second value related to a joint interval according to an embodiment of the present application.
- 13 is a diagram illustrating an aspect of obtaining a second value according to an embodiment of the present application.
- Obtaining a second value related to the joint distance includes obtaining a region of interest (S1410) and obtaining a plurality of joint distance values within the region of interest (S1420). ), and obtaining a second value based on a plurality of joint spacing values (S1430).
- the medical image analysis apparatus 1000 may obtain a section of interest.
- the medical image analysis apparatus 1000 may obtain, as a region of interest, an inter-joint region known to be of high clinical importance in determining a joint state.
- a region known as a region in which cartilage is present and associated with severe pain due to arthritis may be obtained as a region of interest.
- the region of interest may be the above-mentioned both end points of the bone (eg, the first point P1 and the second point P2), the width between the both end points of the bone (eg, the first value), or the above-described clinical It can be automatically obtained by utilizing statistical ratios of important intervals.
- a user may input a section of interest through an arbitrary input unit, and the medical image analysis apparatus 1000 may acquire the section of interest based on the user's input.
- a section of interest may be divided into several sub-sections, and may be continuous or discontinuous.
- the medical image analysis apparatus 1000 may obtain at least one joint distance value within the interest period. For example, the medical image analysis apparatus 1000 may obtain at least one joint distance value of an inter-joint region within the first ROI. Also, the medical image analysis apparatus 1000 may obtain at least one joint spacing value of a joint spacing area within the second ROI.
- the coordinate information of the joint spacing region may be used by the medical image analysis apparatus 1000 to acquire the joint spacing value.
- the medical image analysis apparatus 1000 may be implemented to obtain a joint spacing value based on coordinate information of a boundary defining a joint spacing region within a region of interest (eg, a first region of interest or a second region of interest).
- a region of interest e.g, a first region of interest or a second region of interest.
- this is only an example, and the medical image analysis apparatus 1000 may obtain the joint distance value within the ROI using any suitable method.
- the medical image analysis apparatus 1000 performs a joint spacing based on at least one joint spacing value obtained from a target joint spacing region within the ROI.
- a second value related to the interval may be obtained.
- the medical image analysis apparatus 1000 may acquire a minimum value among a plurality of joint spacing values as a second value related to the joint spacing.
- the medical image analysis apparatus 1000 may obtain an average value of a plurality of joint spacing values as a second value related to the joint spacing.
- the medical image analysis apparatus 1000 may obtain the second value by assigning an appropriate weight to a minimum value among a plurality of joint distance values and an average value of a plurality of joint distance values.
- a step of calculating a target joint condition index representing a condition of a joint may be included (S1500).
- the medical image analysis apparatus 1000 performs the target joint condition index based on the first value related to the width of the joint and the second value related to the joint interval. can be calculated.
- the medical image analysis apparatus 1000 may calculate a target joint condition index defined as a ratio of a second value to a first value.
- the medical image analysis apparatus 1000 may calculate the minimum joint distance value relative to the first value (width value of the joint part) as the target joint condition index.
- the medical image analysis apparatus 1000 may calculate the average joint spacing value compared to the first value (width value of the joint part) as the target joint condition index.
- the target joint condition index obtained according to an embodiment of the present application may be quantified by linking a joint spacing value affected by various external factors such as race, gender, age, and the like to the width of the joint. Accordingly, the target joint condition index obtained according to an embodiment of the present application can minimize the influence of various external factors such as race, gender, age, and the like, and can provide objective joint condition information.
- the medical image analysis apparatus 1000 may perform an operation of comparing and analyzing the target joint condition index with the reference joint condition index of the normal joint group.
- the medical image analysis apparatus 1000 may obtain a joint state data set from an arbitrary database.
- the joint state data set may include a reference medical image representing a normal joint, a joint width value calculated from the reference medical image, and a joint interval value calculated from the reference medical image.
- the medical image analysis apparatus 1000 may calculate the reference joint condition index of the normal group from the joint condition data set. For example, the medical image analysis apparatus 1000 calculates a first joint condition index (eg, a minimum joint distance value compared to a width value of a joint part) from a first reference medical image related to a normal joint, and calculates a second criterion related to a normal joint. A second joint condition index (eg, a minimum joint distance value versus a width value of a joint part) may be calculated from the medical image. Also, the medical image analysis apparatus 1000 may calculate a reference joint condition index based on the first joint condition index and the second joint condition index. For example, the reference joint condition index may be calculated as an average value of a plurality of joint condition indexes of the normal group including the first joint condition index and the second joint condition index.
- a first joint condition index eg, a minimum joint distance value compared to a width value of a joint part
- a second joint condition index eg, a minimum joint distance value versus
- the medical image analysis apparatus 1000 may quantify a target joint condition index by comparing it with a reference joint condition index.
- the medical image analysis apparatus 1000 may calculate a reduction rate of a target joint condition index compared to a reference joint condition index.
- the reduction rate may be defined as ((reference joint condition index - target joint condition index)/reference joint condition index) * 100.
- the reference joint condition index e.g., the minimum joint distance value versus the width value of the joint part
- the target joint condition index e.g., the value of the joint width part versus the minimum joint distance value
- the medical image analysis apparatus 1000 may quantify a reduction rate indicating that the joint condition of the subject to be analyzed is a state in which the joint interval is narrowed compared to the normal joint, as the reduction rate is 75% compared to the normal joint.
- the reference joint condition index (eg, the minimum joint distance value compared to the width value of the joint part), which is the average value of the knee joint condition index of the normal group, is 0.06
- the target joint condition index (eg, the value of the joint width part) If the relative minimum joint distance value) is 0.015
- the reduction rate can be calculated as about 75%.
- the medical image analysis apparatus 1000 can quantify the reduction rate of the joint state of the analysis subject's knee compared to the normal knee joint by 75%, indicating that the analysis subject's knee joint interval is narrowed compared to the normal knee joint.
- the above numerical values are only examples for convenience of explanation, and are not limited to this.
- the medical image analysis apparatus 1000 may compare conditions between joints of the same person or analyze changes in joint conditions over time.
- the medical image analysis apparatus 1000 according to the present invention may provide information on the joint condition of the subject of analysis by quantifying the joint condition between other people.
- the medical image analysis apparatus 1000 may output joint state information or comparative analysis results with a normal group through an arbitrary output unit.
- the medical image analysis apparatus 1000 according to an embodiment of the present application may transmit the acquired target joint condition index or a comparative analysis result with a normal group to an arbitrary external device including the medical image acquisition apparatus 100.
- Any external device that has received the target joint condition index or the result of comparison and analysis with the normal group may output the joint condition information or the result of comparison and analysis with the normal group through an arbitrary output unit.
- a joint state can be more accurately estimated based on objective joint state information.
- Various operations of the above-described medical image analysis apparatus 1000 may be stored in the memory 1200 of the medical image analysis apparatus 1000, and the controller 1300 of the medical image analysis apparatus 1000 may store the operations in the memory 1200. Can be provided to perform actions.
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Theoretical Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- General Physics & Mathematics (AREA)
- Public Health (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Pathology (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Primary Health Care (AREA)
- Epidemiology (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Heart & Thoracic Surgery (AREA)
- Veterinary Medicine (AREA)
- Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- Dentistry (AREA)
- Data Mining & Analysis (AREA)
- Orthopedic Medicine & Surgery (AREA)
- Rheumatology (AREA)
- Geometry (AREA)
- Evolutionary Computation (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Computational Linguistics (AREA)
- Artificial Intelligence (AREA)
- Databases & Information Systems (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
- Apparatus For Radiation Diagnosis (AREA)
Abstract
Description
Claims (12)
- 의료 영상을 획득하고, 의료 영상에 기초하여 관절의 형태학적 분석을 수행하는 장치가 의료 영상을 분석하는 방법에 있어서, 상기 방법은,대상 의료 영상을 획득하는 단계;상기 대상 의료 영상으로부터 대상 관절 간격 영역을 검출하는 단계;상기 대상 의료 영상으로부터 관절 부위의 폭과 관련된 제1 값을 획득하는 단계;상기 대상 관절 간격 영역으로부터 관절 간격과 관련된 제2 값을 획득하는 단계; 및상기 제1 값 및 상기 제2 값에 기초하여, 관절의 상태를 나타내는 대상 관절 상태 지표를 산출하는 단계;를 포함하는,의료 영상 분석 방법.
- 제1 항에 있어서,상기 대상 관절 간격 영역을 검출하는 단계는,상기 대상 의료 영상으로부터 관심 영역을 검출하는 단계; 및상기 관심 영역에 대한 세그멘테이션을 수행하여 상기 관심 영역에 포함된 상기 대상 관절 간격 영역을 획득하는 단계;를 포함하는,의료 영상 분석 방법.
- 제2 항에 있어서,상기 관심 영역은, 의료 영상을 수신하여 관절 부위를 포함한 영역을 출력하도록 학습된 제1 신경망 모델을 이용하여 획득되는,의료 영상 분석 방법.
- 제3 항에 있어서,상기 세그멘테이션은, 관심 영역을 포함하는 의료 영상을 수신하여 관절 간격 영역을 출력하도록 학습된 제2 신경망 모델을 이용하여 수행되는,의료 영상 분석 방법.
- 제1 항에 있어서,상기 제1 값을 획득하는 단계는,상기 대상 의료 영상으로부터 뼈 영역과 뼈의 외측 영역 간의 경계에 인접한 제1 지점 및 제2 지점을 검출하는 단계;상기 제1 지점의 제1 좌표 정보 및 상기 제2 지점의 제2 좌표 정보를 획득하는 단계; 및상기 제1 좌표 정보 및 상기 제2 좌표 정보에 기초하여 상기 제1 값을 계산하는 단계;를 포함하는,의료 영상 분석 방법.
- 제5 항에 있어서,상기 제1 지점 및 상기 제2 지점은,상기 대상 의료 영상에 포함된 상기 뼈 영역의 밝기와 상기 뼈의 외측 영역의 밝기의 차이에 기초하여 획득되는,의료 영상 분석 방법.
- 제5 항에 있어서,상기 제1 지점 및 상기 제2 지점은,상기 뼈 영역 및 상기 뼈의 외측 영역을 포함하는 의료 영상을 수신하여 상기 제1 지점에 대응되는 제1 영역 및 상기 제2 지점에 대응되는 제2 영역을 출력하도록 학습된 신경망 모델을 통하여 획득되는,의료 영상 분석 방법.
- 제1 항에 있어서,상기 제2 값을 획득하는 단계는,상기 관절 간격 영역 중에서 관심 구간을 획득하는 단계;상기 관심 구간 내에서의 복수의 관절 간격 값을 획득하는 단계; 및상기 복수의 관절 간격 값에 기초하여 상기 제2 값을 획득하는 단계;를 포함하는,의료 영상 분석 방법.
- 제8 항에 있어서,상기 제2 값은,상기 복수의 관절 간격 값 중 최소 값이거나 상기 복수의 관절 간격 값의 평균 값인,의료 영상 분석 방법.
- 제1 항에 있어서,상기 대상 관절 상태 지표는 상기 제2 값의 상기 제1 값에 대한 비율로 정의되는,의료 영상 분석 방법.
- 컴퓨터에 제1 항 내지 제10 항 중 어느 하나의 항에 따른 방법을 실행시키기 위한 프로그램을 기록한 컴퓨터로 읽을 수 있는 기록 매체.
- 의료 영상을 분석하여 관절의 상태와 관련된 정보를 연산하는 의료 영상 분석 장치에 있어서, 상기 의료 영상 분석 장치는,대상 의료 영상을 획득하는 영상 획득부; 및상기 대상 의료 영상에 기초하여 관절 상태 정보를 제공하는 컨트롤러;를 포함하되,상기 컨트롤러는,대상 의료 영상을 획득하고, 상기 대상 의료 영상으로부터 대상 관절 간격 영역을 검출하고, 상기 대상 의료 영상으로부터 관절 부위의 폭과 관련된 제1 값을 획득하고, 상기 대상 관절 간격 영역으로부터 관절 간격과 관련된 제2 값을 획득하고, 상기 제1 값 및 상기 제2 값에 기초하여, 관절의 상태를 나타내는 대상 관절 상태 지표를 산출하도록 구성된,의료 영상 분석 장치.
Priority Applications (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2022508473A JP7473992B2 (ja) | 2021-08-11 | 2021-12-03 | 関節状態を定量化するための医療映像分析方法、医療映像分析装置、および医療映像分析システム |
| EP21851945.2A EP4343779A4 (en) | 2021-08-11 | 2021-12-03 | Medical image analysis method, medical image analysis device, and medical image analysis system for quantifying joint condition |
| US17/634,761 US12243224B2 (en) | 2021-08-11 | 2021-12-03 | Medical image analysis method, medical image analysis apparatus, and medical image analysis system for quantifying joint condition |
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| KR1020210105754A KR102842473B1 (ko) | 2021-08-11 | 2021-08-11 | 관절 상태를 정량화하기 위한 의료 영상 분석 방법, 의료 영상 분석 장치, 및 의료 영상 분석 시스템 |
| KR10-2021-0105754 | 2021-08-11 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2023017919A1 true WO2023017919A1 (ko) | 2023-02-16 |
Family
ID=85200727
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/KR2021/018218 Ceased WO2023017919A1 (ko) | 2021-08-11 | 2021-12-03 | 관절 상태를 정량화하기 위한 의료 영상 분석 방법, 의료 영상 분석 장치, 및 의료 영상 분석 시스템 |
Country Status (5)
| Country | Link |
|---|---|
| US (1) | US12243224B2 (ko) |
| EP (1) | EP4343779A4 (ko) |
| JP (1) | JP7473992B2 (ko) |
| KR (2) | KR102842473B1 (ko) |
| WO (1) | WO2023017919A1 (ko) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118762044A (zh) * | 2024-03-05 | 2024-10-11 | 北京大学第三医院(北京大学第三临床医学院) | 医学图像处理方法、装置以及存储介质 |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20250166416A1 (en) * | 2023-11-21 | 2025-05-22 | Metatech (Ap) Inc. | System and method for hand movements recognition and analysis |
| KR102916312B1 (ko) | 2025-10-17 | 2026-01-26 | 주식회사 스포클립에이아이 | Ai 기반으로 사용자의 스포츠 일상에 관한 콘텐츠 영상을 제작하기 위한 장치 및 방법 |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2004057804A (ja) * | 2002-06-05 | 2004-02-26 | Fuji Photo Film Co Ltd | 骨関節評価方法、装置およびそのためのプログラム |
| JP2016144535A (ja) * | 2015-02-06 | 2016-08-12 | 国立大学法人名古屋大学 | 骨間距離測定装置、骨間距離測定方法、コンピュータを骨間距離測定装置として機能させるためのプログラム及び該プログラムを記憶した記録媒体 |
| KR20180092797A (ko) * | 2017-02-10 | 2018-08-20 | 연세대학교 산학협력단 | 의료 영상에 기반하여 상태를 진단하는 장치 및 방법 |
| KR101968144B1 (ko) * | 2018-10-11 | 2019-08-13 | 가톨릭대학교 산학협력단 | 척추 및 경추의 경사각 자동 진단 장치 및 방법 |
| JP2021023548A (ja) * | 2019-08-06 | 2021-02-22 | キヤノンメディカルシステムズ株式会社 | 医用画像処理装置、医用画像処理システム、医用画像処理プログラム、及び医用画像処理方法 |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6160866A (en) * | 1991-02-13 | 2000-12-12 | Lunar Corporation | Apparatus for bilateral femur measurement |
| US5509042A (en) * | 1991-02-13 | 1996-04-16 | Lunar Corporation | Automated determination and analysis of bone morphology |
| WO2012021861A2 (en) * | 2010-08-13 | 2012-02-16 | Mckinnon, Brian William | Detection of anatomical landmarks |
| US20160180520A1 (en) * | 2014-12-17 | 2016-06-23 | Carestream Health, Inc. | Quantitative method for 3-d joint characterization |
| KR102444274B1 (ko) * | 2019-07-17 | 2022-09-16 | 주식회사 크레스콤 | 관절염 심각도 정밀 분석 장치 및 방법 |
-
2021
- 2021-08-11 KR KR1020210105754A patent/KR102842473B1/ko active Active
- 2021-12-03 US US17/634,761 patent/US12243224B2/en active Active
- 2021-12-03 EP EP21851945.2A patent/EP4343779A4/en active Pending
- 2021-12-03 JP JP2022508473A patent/JP7473992B2/ja active Active
- 2021-12-03 WO PCT/KR2021/018218 patent/WO2023017919A1/ko not_active Ceased
-
2025
- 2025-07-09 KR KR1020250092393A patent/KR20250111270A/ko active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2004057804A (ja) * | 2002-06-05 | 2004-02-26 | Fuji Photo Film Co Ltd | 骨関節評価方法、装置およびそのためのプログラム |
| JP2016144535A (ja) * | 2015-02-06 | 2016-08-12 | 国立大学法人名古屋大学 | 骨間距離測定装置、骨間距離測定方法、コンピュータを骨間距離測定装置として機能させるためのプログラム及び該プログラムを記憶した記録媒体 |
| KR20180092797A (ko) * | 2017-02-10 | 2018-08-20 | 연세대학교 산학협력단 | 의료 영상에 기반하여 상태를 진단하는 장치 및 방법 |
| KR101968144B1 (ko) * | 2018-10-11 | 2019-08-13 | 가톨릭대학교 산학협력단 | 척추 및 경추의 경사각 자동 진단 장치 및 방법 |
| JP2021023548A (ja) * | 2019-08-06 | 2021-02-22 | キヤノンメディカルシステムズ株式会社 | 医用画像処理装置、医用画像処理システム、医用画像処理プログラム、及び医用画像処理方法 |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118762044A (zh) * | 2024-03-05 | 2024-10-11 | 北京大学第三医院(北京大学第三临床医学院) | 医学图像处理方法、装置以及存储介质 |
Also Published As
| Publication number | Publication date |
|---|---|
| KR20250111270A (ko) | 2025-07-22 |
| US20230360198A1 (en) | 2023-11-09 |
| EP4343779A1 (en) | 2024-03-27 |
| KR102842473B1 (ko) | 2025-08-05 |
| JP7473992B2 (ja) | 2024-04-24 |
| EP4343779A4 (en) | 2025-04-09 |
| JP2023541078A (ja) | 2023-09-28 |
| US12243224B2 (en) | 2025-03-04 |
| KR20230023899A (ko) | 2023-02-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2023017919A1 (ko) | 관절 상태를 정량화하기 위한 의료 영상 분석 방법, 의료 영상 분석 장치, 및 의료 영상 분석 시스템 | |
| WO2023153569A1 (ko) | 무릎 관절 상태의 분석 방법 및 이를 수행하기 위한 장치 | |
| WO2019132170A1 (ko) | 학습용 데이터 관리방법, 장치 및 프로그램 | |
| WO2017022882A1 (ko) | 의료 영상의 병리 진단 분류 장치 및 이를 이용한 병리 진단 시스템 | |
| WO2019235828A1 (ko) | 투 페이스 질병 진단 시스템 및 그 방법 | |
| WO2016171341A1 (ko) | 클라우드 기반 병리 분석 시스템 및 방법 | |
| WO2020196985A1 (ko) | 비디오 행동 인식 및 행동 구간 탐지 장치 및 방법 | |
| WO2016163755A1 (ko) | 품질 측정 기반의 얼굴 인식 방법 및 장치 | |
| WO2021010777A1 (ko) | 관절염 심각도 정밀 분석 장치 및 방법 | |
| WO2022131642A1 (ko) | 의료 영상 기반 질환 중증도 결정 장치 및 방법 | |
| WO2020032559A2 (ko) | 뉴럴 네트워크를 이용한 질병의 진단 시스템 및 방법 | |
| WO2025211482A1 (ko) | 위성 데이터 결측 보완 방법 및 시스템 | |
| WO2025071103A1 (ko) | 가축 개체군의 출하 정보 산출 장치 및 방법과, 상기 출하 정보 산출 장치를 포함하는 출하 정보 산출 시스템 | |
| WO2024029698A1 (ko) | 뇌혈류 데이터를 연산하는 방법 및 뇌혈류 데이터를 연산하기 위한 신경망 모델의 학습 방법 | |
| WO2019221586A1 (ko) | 의료영상 관리 시스템, 방법 및 컴퓨터 판독 가능한 기록매체 | |
| WO2024090786A1 (ko) | 레이더 데이터에 기반한 낙상 감지 모델의 학습 방법 | |
| WO2015076433A1 (ko) | 지역 미세 패턴을 이용한 얼굴 영상 분석 방법 | |
| WO2014058165A1 (ko) | 단일객체의 크기를 추정하는 영상 감시장치 및 방법 | |
| WO2025170396A1 (ko) | 다중 인공 신경망 기반 추정 선량 검사 방법 | |
| WO2025018825A1 (ko) | 영상 및 비영상 데이터를 이용한 질병 예후 통합 예측 방법, 장치 및 컴퓨터 프로그램 | |
| WO2022177069A1 (ko) | 라벨링 방법 및 이를 위한 컴퓨팅 장치 | |
| WO2023128059A1 (en) | Method and apparatus for tumor purity based on pathological slide image | |
| WO2023136695A1 (ko) | 환자의 가상 폐 모델을 생성하는 장치 및 방법 | |
| WO2023121161A1 (ko) | 모션 증폭 장치 및 이를 사용하는 방법 | |
| WO2023204342A1 (ko) | 딥러닝 기반 이미지 변환 방법 및 그 시스템 |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| ENP | Entry into the national phase |
Ref document number: 2022508473 Country of ref document: JP Kind code of ref document: A |
|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 21851945 Country of ref document: EP Kind code of ref document: A1 |
|
| WWE | Wipo information: entry into national phase |
Ref document number: 2021851945 Country of ref document: EP |
|
| ENP | Entry into the national phase |
Ref document number: 2021851945 Country of ref document: EP Effective date: 20231219 |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| WWE | Wipo information: entry into national phase |
Ref document number: 11202400677X Country of ref document: SG |
|
| WWG | Wipo information: grant in national office |
Ref document number: 17634761 Country of ref document: US |