WO2022021990A1 - 信息判断方法和装置 - Google Patents

信息判断方法和装置 Download PDF

Info

Publication number
WO2022021990A1
WO2022021990A1 PCT/CN2021/092019 CN2021092019W WO2022021990A1 WO 2022021990 A1 WO2022021990 A1 WO 2022021990A1 CN 2021092019 W CN2021092019 W CN 2021092019W WO 2022021990 A1 WO2022021990 A1 WO 2022021990A1
Authority
WO
WIPO (PCT)
Prior art keywords
information
review
prescription
training
data set
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
Application number
PCT/CN2021/092019
Other languages
English (en)
French (fr)
Inventor
赵俊
匡哲祥
刘潇龙
黄思晋
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing Jingdong Tuoxian Technology Co Ltd
Original Assignee
Beijing Jingdong Tuoxian Technology Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Beijing Jingdong Tuoxian Technology Co Ltd filed Critical Beijing Jingdong Tuoxian Technology Co Ltd
Priority to EP21851572.4A priority Critical patent/EP4191599A4/en
Priority to US18/014,073 priority patent/US20240296926A1/en
Priority to JP2023512260A priority patent/JP7518971B2/ja
Publication of WO2022021990A1 publication Critical patent/WO2022021990A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • 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
    • G16H20/00—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance
    • G16H20/10—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance relating to drugs or medications, e.g. for ensuring correct administration to patients
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0637—Strategic management or analysis, e.g. setting a goal or target of an organisation; Planning actions based on goals; Analysis or evaluation of effectiveness of goals
    • 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
    • G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
    • 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
    • G16H50/70—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients
    • 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
    • G16H70/00—ICT specially adapted for the handling or processing of medical references
    • G16H70/40—ICT specially adapted for the handling or processing of medical references relating to drugs, e.g. their side effects or intended usage

Definitions

  • the present application relates to the field of computer technology, in particular to the technical field of smart medical care and information processing, and in particular to a method and apparatus for judging information.
  • the present application provides an information judgment method, apparatus, device, and storage medium.
  • an information judgment method includes: obtaining prescription review information, wherein the prescription review information includes: prescription information issued by a doctor and review information given by a pharmacist according to the prescription information; Perform feature extraction with review information to generate a feature data set corresponding to the prescription review information; according to the current review rules, judge the feature data set to obtain review results corresponding to the prescription review information, wherein the review rules are used to characterize the feature data set and reviews.
  • the correspondence between the results, and the audit rules are updated based on the training results of the classification decision model obtained by training.
  • the method further includes: training a classification decision model according to a preset condition; the classification decision model is obtained by training based on the following steps: acquiring a training sample set, wherein the training samples in the training sample set include: different extracted Various feature datasets of hospitals and review results corresponding to various feature datasets of different hospitals; using machine learning methods, various feature datasets of different hospitals included in the training samples in the training sample set are used as the input of the detection network, and the The review results corresponding to various feature data sets of different hospitals are used as the expected output of the detection network, and the classification decision model is obtained by training.
  • the review rule is updated based on the training result of the classification decision model obtained by training, including: judging whether the classification decision model is trained; in response to the completion of the classification decision model training, according to the input parameters and output parameters of the classification decision model, Update the audit rules.
  • the method before the feature extraction is performed on the prescription information and the review information, the feature data set corresponding to the prescription review information is generated, and before the training sample set is obtained, the method further includes: cleaning the prescription information and the review information, and after generating the cleaning The prescription review information, where cleaning is based on data structuring of prescription information and review information.
  • cleaning the prescription information and review information to generate cleaned prescription review information includes: extracting information from the prescription information to obtain personal information corresponding to the prescription information, diagnostic information corresponding to the prescription information, and corresponding prescription information According to personal information, determine the group label corresponding to the prescription information; standardize the diagnosis information to obtain the diagnosis name corresponding to the prescription information; standardize the drug information and unit normalize, obtain the drug standard information corresponding to the prescription information; According to the text classification technology, the review information is classified to determine the category of the review information; the cleaned prescription review information is generated according to the population label, diagnosis name, drug standard information and the category of the review information.
  • the method further includes: encoding the feature data set, and obtaining the processed feature dataset.
  • the method further includes: according to the preset pharmacological knowledge, performing the feature data set on the feature data set. Filter to generate a filtered feature dataset.
  • the method further includes: optimizing the structure of the product and/or the product strategy based on the relevance of the review results to the product.
  • the feature data set includes: discrete features with relatively independent features and associated features with associated relationships between features.
  • an information judging device includes: an obtaining unit configured to obtain prescription review information, wherein the prescription review information includes: prescription information issued by a doctor and comments given by a pharmacist according to the prescription information information; the feature extraction unit is configured to perform feature extraction on the prescription information and the review information, and generate a feature data set corresponding to the prescription review information; the judgment unit is configured to judge the feature data set according to the current review rules, and obtain the prescription
  • the review result corresponding to the review information wherein the review rule is used to represent the correspondence between the feature data set and the review result, and the review rule is updated based on the training result of the classification decision model obtained by training.
  • the apparatus further includes: a training unit configured to train the classification decision model according to preset conditions; the classification decision model in the training unit is obtained by training based on the following steps: acquiring a training sample set, wherein the training samples
  • the centralized training samples include: various feature datasets of different hospitals obtained by extraction and review results corresponding to various feature datasets of different hospitals; using machine learning methods, the training samples are collected in the training samples.
  • the feature data set is used as the input of the detection network, and the review results corresponding to various feature data sets of different hospitals are used as the expected output of the detection network, and the classification decision model is obtained by training.
  • the judging unit includes: a judging module, configured to judge whether the training of the classification decision model is completed; an update module, configured to respond to the completion of the training of the classification decision model, according to the input parameters and output parameters of the classification decision model, Update the audit rules.
  • the apparatus further includes: a cleaning unit configured to clean the prescription information and the review information to generate cleaned prescription review information, wherein the cleaning is based on data structuring of the prescription information and the review information.
  • the cleaning unit includes: an extraction module configured to perform information extraction on prescription information to obtain personal information corresponding to the prescription information, diagnostic information corresponding to the prescription information, and drug information corresponding to the prescription information; a determination module, which is It is configured to determine the group label corresponding to the prescription information according to personal information; standardize the diagnosis information to obtain the diagnosis name corresponding to the prescription information; standardize the drug information and unit to obtain the drug standard information corresponding to the prescription information; classification module , is configured to classify the review information according to the text classification technology, and determine the category of the review information; the generation module is configured to generate the cleaned prescription according to the category of the crowd label, diagnosis name, drug standard information and review information. Review information.
  • an extraction module configured to perform information extraction on prescription information to obtain personal information corresponding to the prescription information, diagnostic information corresponding to the prescription information, and drug information corresponding to the prescription information
  • a determination module which is It is configured to determine the group label corresponding to the prescription information according to personal information; standardize the diagnosis information to obtain the diagnosis name corresponding to the prescription information
  • the apparatus further includes: a processing unit configured to perform encoding processing on the feature data set to obtain a processed feature data set.
  • the device further includes: a screening unit configured to screen the characteristic data set according to preset pharmacological knowledge, and generate a screened characteristic data set.
  • a screening unit configured to screen the characteristic data set according to preset pharmacological knowledge, and generate a screened characteristic data set.
  • the apparatus further includes: an optimization unit configured to optimize the structure of the product and/or the product strategy based on the relevance of the review result to the product.
  • an electronic device comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor.
  • the at least one processor executes to enable the at least one processor to perform a method as described in any implementation of the first aspect.
  • the present application provides a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to cause a computer to execute the method described in any implementation manner of the first aspect .
  • the prescription review information is obtained, the feature extraction is performed on the prescription information and the review information, the feature data set corresponding to the prescription review information is generated, and the feature data set corresponding to the prescription review information is judged according to the current review rules, and the corresponding prescription review information is obtained.
  • the acquired rules are static rules, it is impossible to dynamically learn the auditing rules of prescriptions.
  • a method for judging prescription review information by using the auditing rules obtained by learning is implemented, which improves the auditing efficiency and efficiency of the prescription auditing system. Accuracy.
  • FIG. 1 is a schematic diagram of a first embodiment of an information judgment method according to the present application.
  • FIG. 2 is a scene diagram in which the information judgment method according to the embodiment of the present application can be realized
  • FIG. 3 is a schematic diagram of a second embodiment of the information judgment method according to the present application.
  • Figure 4a is an example diagram of prescription review information after cleaning
  • Figure 4b is an example diagram of an audit rule
  • FIG. 5 is a schematic structural diagram of an embodiment of an information judging device according to the present application.
  • FIG. 6 is a block diagram of an electronic device used to implement the information judgment method of the embodiment of the present application.
  • FIG. 1 shows a schematic diagram 100 of a first embodiment of an information judgment method according to the present application.
  • the information judgment method includes the following steps:
  • Step 101 obtaining prescription review information.
  • the execution subject may obtain prescription review information from other electronic devices or locally through wired connection or wireless connection, and may also obtain prescription review information by parsing the prescription review request.
  • the prescription review information includes: prescription information and review information. Prescribing information means that it is issued by a doctor and includes the patient's personal information (such as age, gender, etc.), the patient's diagnosis information (including the disease, allergy history, physical examination indicators, etc.) dosage, route of administration, frequency of administration, etc.). Comment information means that it is given by the pharmacist based on the prescription information, including the pharmacist's judgment information on whether the prescription is reasonable.
  • the comment information can include: the drug does not match the diagnosis, the drug does not match the gender, the single dose is too high, etc.
  • wireless connection methods may include but are not limited to 3G, 4G, 5G connection, WiFi connection, Bluetooth connection, WiMAX connection, Zigbee connection, UWB (ultra wideband) connection, and other wireless connection currently known or developed in the future connection method.
  • step 102 feature extraction is performed on the prescription information and the review information, and a feature data set corresponding to the prescription review information is generated.
  • the execution body may perform feature extraction on the prescription information and the review information based on the feature extraction method, and generate a feature data set corresponding to the prescription review information.
  • the feature dataset includes: discrete features with relatively independent features and associated features with associated relationships between features.
  • the features in the feature dataset may include: gender, age, population characteristics of patients, diagnostic characteristics of patients, generic names of drugs, administration frequency, single dose, and administration route characteristics. Among them, gender, population, diagnosis, common name, and route of administration are discrete features, while age, frequency of administration, and single dose are associated features.
  • Step 103 according to the current review rule, judge the feature data set, and obtain a review result corresponding to the prescription review information.
  • the execution subject may judge the feature data set according to the current review rule, and obtain the review result corresponding to the prescription review information.
  • the audit rules are used to represent the correspondence between the feature data set and the review results.
  • the audit rules are updated based on the training results of the classification decision model obtained by training, so that the audit rules are dynamically learned to obtain the optimal audit rules.
  • the information determination method 200 of this embodiment runs in the electronic device 201 .
  • the electronic device 201 After the electronic device 201 receives the review request, the electronic device 201 first obtains the prescription review information 202, and then the electronic device 201 performs feature extraction on the prescription information in the prescription review information and the review information in the prescription review information, and generates a corresponding prescription review information.
  • Feature data set 203 Next, the electronic device 201 judges the feature data set according to the current review rules, and obtains a review result 204 corresponding to the prescription review information.
  • the information judging method provided by the above-mentioned embodiments of the present application adopts the method of obtaining prescription review information, performing feature extraction on the prescription information and the review information, generating a feature data set corresponding to the prescription review information, and judging the feature data set according to the current review rules, Obtain the review results corresponding to the prescription review information, in which the review rules are updated based on the training results of the classification decision model obtained by training, which solves the problem of configuration errors caused by manual formulation of rules, and frees pharmacists from complicated rule configuration work.
  • FIG. 3 a schematic diagram 300 of a second embodiment of the information determination method is shown.
  • the flow of the method includes the following steps:
  • Step 301 obtaining prescription review information.
  • Step 302 cleaning the prescription information and the review information, and generating the cleaned prescription review information.
  • the execution body may clean the prescription information and the review information respectively, generate the cleaned prescription information and review information, and summarize the cleaned prescription information and review information to obtain the cleaned prescription review information, wherein the cleaned prescription information and review information are obtained. Based on data structure processing of prescription information and review information.
  • cleaning the prescription information and review information to generate cleaned prescription review information includes: extracting information from the prescription information to obtain personal information corresponding to the prescription information and corresponding prescription information. According to personal information, determine the group label corresponding to the prescription information; standardize the diagnosis information to obtain the diagnosis name corresponding to the prescription information; standardize the drug information and unit to obtain the prescription The drug standard information corresponding to the information; according to the text classification technology, the review information is classified to determine the category of the review information; according to the population label, diagnosis name, drug standard information and the category of the review information, the cleaned prescription review information is generated. A specific example is shown in Figure 4a.
  • the crowd labels may include: newborns, children, adults, the elderly, and the like.
  • Common diagnostic name standardization methods are divided into two types: the first uses the international ICD (International Classification of Diseases, International Classification of Diseases) 10 code table for standardization, trains word vectors and uses cosine similarity to map diagnostic names to ICD10 codes The standard disease name in the table; the second is to use a syntactic analysis algorithm to extract the hyponymous relationship of the phrase in the diagnosis name, and use the basic diagnosis name as the standardized disease name of the diagnosis.
  • the categories of review information can include: reasonable prescription, inconsistency between diagnosis and drug use, drug use and gender inconsistency, repeated drug use, high single dose, high dose frequency and inconsistent route of administration.
  • Standardization and unit normalization of drug information may include: extracting drug usage and dosage information from prescriptions, standardizing the description of administration route and drug administration rate in usage and dosage, and performing unit of single dose in usage and dosage. Normalized. By unifying the description of each feature in the prescription review information, the complexity of prescription review rules is reduced.
  • step 303 feature extraction is performed on the prescription information and the review information, and a feature data set corresponding to the prescription review information is generated.
  • Step 304 encoding the feature data set to obtain a processed feature data set.
  • the execution subject may classify and encode the relevant features of the diagnosis description in the feature data set generated in step 303, and perform dimension reduction processing on the coding in combination with the drug instructions to obtain the processed feature data set. Due to the complexity and diversity of the description of the diagnosis name, if the one-hot encoding ONE-HOT is used directly, the dimension of the feature will be too high, which is not conducive to the training of the classification decision model. Dimension, to solve the sparse problem of diagnostic names.
  • the drug insert mentions two types of indications for hair loss and benign prostatic hyperplasia.
  • Step 305 Screen the feature data set according to the preset pharmacological knowledge, and generate a screened feature data set.
  • the execution subject may sort and filter the feature dimensions of the feature data set obtained in step 304 according to the preset pharmacological knowledge and in combination with the drug instructions, to generate a filtered feature data set. For example, when generating the rule that the diagnosis does not match the medication, only the generic name feature of the drug and the diagnostic feature of the patient need to be used; when generating the rule that the medication does not match the gender, only the generic name feature of the drug and the gender feature of the patient need to be used; When generating the rules of the drug, only the characteristics of the generic name of the drug and the route of administration of the drug need to be used; when generating rules such as too high a single dose and too high dosing frequency, only the age, population, diagnostic characteristics of the patient and the generic characteristics of the drug need to be used. Name, single dose, dosing frequency characteristics. Accurate and targeted feature data is obtained through screening, which improves the efficiency and accuracy of the system.
  • Step 306 according to the current review rule, judge the feature data set, and obtain a review result corresponding to the prescription review information.
  • the execution subject may judge the feature data set according to the current review rule, and obtain the review result corresponding to the prescription review information.
  • the audit rules are updated based on the training results of the classification decision model obtained by training. An example of an audit rule is shown in Figure 4b.
  • the method further includes: training the classification decision model according to preset conditions; the classification decision model is obtained by training based on the following steps: acquiring a training sample set, wherein the training samples in the training sample set are The samples include: various feature datasets of different hospitals obtained from the extraction and review results corresponding to various feature datasets of different hospitals; using machine learning methods, the various feature datasets of different hospitals included in the training samples are collected in the training sample set.
  • the review results corresponding to various feature data sets of different hospitals are used as the expected output of the detection network, and the classification decision model is obtained by training.
  • the review rule is updated based on the training result of the classification decision model obtained by training, including: judging whether the training of the classification decision model is completed; in response to the completion of the training of the classification decision model, according to the classification decision model
  • the input parameters and output parameters of the audit rules are updated.
  • the audit rules are updated after each model training, so that the audit rules are consistent with the classification decision model, which simplifies the update process of the audit rules and ensures the optimal solution of the audit rules.
  • the method further includes: optimizing the structure of the product and/or the product strategy based on the correlation between the review result and the product. According to the review results, the pharmacist will review the classification decision model and review rules to ensure the correctness of the review results of the review system.
  • steps 301, 303, and 306 are basically the same as the operations of steps 101, 102, and 103 in the embodiment shown in FIG. 1, and are not repeated here.
  • the schematic diagram 300 of the information judgment method in this embodiment adopts cleaning prescription information and review information to generate cleaned prescription review information, and analyzes the feature data.
  • the characteristic data set is screened, and the filtered characteristic data set is generated, which solves the sparse problem of diagnostic names and reduces the dimension of features. Accurate and targeted feature data is obtained through screening, which improves the efficiency and accuracy of the system.
  • the present application provides an embodiment of an information judgment apparatus, the apparatus embodiment corresponds to the method embodiment shown in FIG. 1 , and the apparatus may specifically be Used in various electronic devices.
  • the information judging device 500 of this embodiment includes: an obtaining unit 501 , a feature extraction unit 502 and a judging unit 503 , wherein the obtaining unit is configured to obtain prescription review information, wherein the prescription review information includes: The prescription information and the comment information given by the pharmacist according to the prescription information; the feature extraction unit is configured to perform feature extraction on the prescription information and the comment information, and generate a feature data set corresponding to the prescription comment information; the judgment unit is configured to Review rules, judge the feature data set, and obtain the review results corresponding to the prescription review information, wherein the review rules are used to represent the correspondence between the feature data sets and review results, and the review rules are based on the training results of the classification decision model obtained by training. renew.
  • the specific processing of the acquiring unit 501 , the feature extracting unit 502 and the determining unit 503 of the information determining apparatus 500 and the technical effects brought about by them may refer to steps 101 to 103 in the embodiment corresponding to FIG. 1 , respectively. The related descriptions are not repeated here.
  • the filtering unit includes: a clustering module, which is configured to, after clustering all product pairs in the first product pair set, calculate to obtain each product pair in the first product pair set product repurchase cycle; the first judgment module is configured to determine whether each product in the first product pair exceeds the repurchase time limit according to the repurchase cycle of each product and the purchase time of the corresponding product; the first storage module is configured If the product in the first product pair collection exceeds the repurchase time limit, the product pair of this product is stored in the filtered first product pair collection, and if the product in the first product pair collection does not exceed the repurchase time limit, the product The product pairs are stored in the unfiltered first product pair collection.
  • a clustering module which is configured to, after clustering all product pairs in the first product pair set, calculate to obtain each product pair in the first product pair set product repurchase cycle
  • the first judgment module is configured to determine whether each product in the first product pair exceeds the repurchase time limit according to the repurchase cycle of each product and the purchase time of the
  • the filtering unit further includes: a second judgment module, configured to judge the first product pair in the set according to the time of occurrence of the user behavior and/or the number of occurrences of the user behavior Whether each product satisfies the exemption conditions, wherein the exemption conditions are used to characterize the threshold judgment on the occurrence time of user behavior and/or the number of occurrences of user behavior, and the judgment is completed based on the standard product unit SPU of the product; the second storage module, configured If the products in the first product pair set meet the exemption conditions, the product pair corresponding to the product is stored in the filtered first product pair set, and if the products in the first product pair set do not meet the exemption conditions, the product corresponds to The product pairs are stored in the unfiltered first product pair collection.
  • a second judgment module configured to judge the first product pair in the set according to the time of occurrence of the user behavior and/or the number of occurrences of the user behavior Whether each product satisfies the exemption conditions, wherein the exemption conditions are used to characterize the threshold
  • the filtering unit further includes: an obtaining module, configured to obtain a third set of product pairs from the first set of product pairs, wherein the third set of product pairs is used to represent unidentified Obtain a set of product pairs in the repurchase cycle; the filtering module is configured to filter the third product pair set based on the product repurchase selection model, and generate a repurchase cycle with each product in the third product pair set, wherein, The product repurchase selection model is used to characterize the selection of the third product pair set based on the product repurchase cycle.
  • the product filtering strategy in the first selection unit performs combined screening on the set of unfiltered first product pairs based on multiple dimensions;
  • the first selection unit includes: a judgment module, which is It is configured to determine whether each product in the unfiltered first product pair set is a purchased commodity of the user according to the product filtering model, wherein the product filtering model is used to characterize the product word based on the product, the standard product unit SPU of the product and the value of the product.
  • the unfiltered first product is combined and filtered for each product in the collection; the deletion module is configured so that if the unfiltered first product is the product in the collection of the user's purchased product , the product pair corresponding to the product is deleted from the unfiltered first product pair set to obtain the second product pair set.
  • the apparatus further includes: a second selection unit, configured to select a set of second product pairs based on the category of the product to obtain a selected set of second product pairs.
  • the apparatus further includes: a generating unit configured to sort the user's second commodity set according to the commodity display strategy, and generate a user's commodity list.
  • the present application further provides an electronic device and a readable storage medium.
  • FIG. 6 it is a block diagram of an electronic device according to an information judgment method of an embodiment of the present application.
  • Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers.
  • Electronic devices may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smart phones, wearable devices, and other similar computing devices.
  • the components shown herein, their connections and relationships, and their functions are by way of example only, and are not intended to limit implementations of the application described and/or claimed herein.
  • the electronic device includes: one or more processors 601, a memory 602, and interfaces for connecting various components, including a high-speed interface and a low-speed interface.
  • the various components are interconnected using different buses and may be mounted on a common motherboard or otherwise as desired.
  • the processor may process instructions executed within the electronic device, including storing in or on the memory to display a GUI (Graphical User Interface) on an external input/output device such as a display device coupled to the interface ) instructions for graphics information.
  • GUI Graphic User Interface
  • multiple processors and/or multiple buses may be used with multiple memories and multiple memories, if desired.
  • multiple electronic devices may be connected, each providing some of the necessary operations (eg, as a server array, a group of blade servers, or a multiprocessor system).
  • a processor 601 is taken as an example in FIG. 6 .
  • the memory 602 is the non-transitory computer-readable storage medium provided by the present application.
  • the memory stores instructions executable by at least one processor, so that the at least one processor executes the information judgment method provided by the present application.
  • the non-transitory computer-readable storage medium of the present application stores computer instructions, and the computer instructions are used to cause the computer to execute the information judgment method provided by the present application.
  • the memory 602 can be used to store non-transitory software programs, non-transitory computer-executable programs, and modules, such as program instructions/modules corresponding to the information judgment method in the embodiments of the present application (for example, appendix).
  • the processor 601 executes various functional applications and data processing of the server by running the non-transitory software programs, instructions and modules stored in the memory 602, ie, implements the information judgment method in the above method embodiments.
  • the memory 602 may include a storage program area and a storage data area, wherein the storage program area may store an operating system and an application program required by at least one function; the storage data area may store data created by judging the use of the electronic device according to the information. Additionally, memory 602 may include high-speed random access memory, and may also include non-transitory memory, such as at least one magnetic disk storage device, flash memory device, or other non-transitory solid state storage device. In some embodiments, the memory 602 may optionally include memory disposed remotely relative to the processor 601, and these remote memories may be connected to the information determination electronic device through a network. Examples of such networks include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.
  • the electronic device of the information judgment method may further include: an input device 603 and an output device 604 .
  • the processor 601 , the memory 602 , the input device 603 and the output device 604 may be connected by a bus or in other ways, and the connection by a bus is taken as an example in FIG. 6 .
  • the input device 603 can receive input numerical or character information, and generate key signal input related to user settings and function control of the information judging electronic device, such as a touch screen, a keypad, a mouse, a trackpad, a touchpad, a pointing stick, an or Multiple input devices such as mouse buttons, trackballs, joysticks, etc.
  • Output devices 604 may include display devices, auxiliary lighting devices (eg, LEDs), haptic feedback devices (eg, vibration motors), and the like.
  • the display device may include, but is not limited to, a liquid crystal display (LCD), a light emitting diode (LED) display, and a plasma display. In some implementations, the display device may be a touch screen.
  • Various implementations of the systems and techniques described herein can be implemented in digital electronic circuitry, integrated circuit systems, application specific ASICs (application specific integrated circuits), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include being implemented in one or more computer programs executable and/or interpretable on a programmable system including at least one programmable processor that The processor, which may be a special purpose or general-purpose programmable processor, may receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device an output device.
  • the processor which may be a special purpose or general-purpose programmable processor, may receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device an output device.
  • machine-readable medium and “computer-readable medium” refer to any computer program product, apparatus, and/or apparatus for providing machine instructions and/or data to a programmable processor ( For example, magnetic disks, optical disks, memories, programmable logic devices (PLDs), including machine-readable media that receive machine instructions as machine-readable signals.
  • machine-readable signal refers to any signal used to provide machine instructions and/or data to a programmable processor.
  • the systems and techniques described herein may be implemented on a computer having a display device (eg, a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user ); and a keyboard and pointing device (eg, a mouse or trackball) through which a user can provide input to the computer.
  • a display device eg, a CRT (cathode ray tube) or LCD (liquid crystal display) monitor
  • a keyboard and pointing device eg, a mouse or trackball
  • Other kinds of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (eg, visual feedback, auditory feedback, or tactile feedback); and can be in any form (including acoustic input, voice input, or tactile input) to receive input from the user.
  • the systems and techniques described herein may be implemented on a computing system that includes back-end components (eg, as a data server), or a computing system that includes middleware components (eg, an application server), or a computing system that includes front-end components (eg, as a data server).
  • back-end components eg, as a data server
  • middleware components eg, an application server
  • front-end components eg, as a data server
  • a user computer having a graphical user interface or web browser through which a user may interact with implementations of the systems and techniques described herein
  • the components of the system may be interconnected by any form or medium of digital data communication (eg, a communication network). Examples of communication networks include: Local Area Networks (LANs), Wide Area Networks (WANs), and the Internet.
  • a computer system can include clients and servers.
  • Clients and servers are generally remote from each other and usually interact through a communication network.
  • the relationship of client and server arises by computer programs running on the respective computers and having a client-server relationship to each other.
  • the prescription review information is obtained, feature extraction is performed on the prescription information and the review information, a feature data set corresponding to the prescription review information is generated, and a prescription is obtained by judging the feature data set according to the current review rules.
  • the review results corresponding to the review information in which the review rules are updated based on the training results of the classification decision model obtained by training, which solves the problem of configuration errors caused by manual formulation of rules, and liberates pharmacists from complicated rule configuration work.

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • Public Health (AREA)
  • Data Mining & Analysis (AREA)
  • Primary Health Care (AREA)
  • Epidemiology (AREA)
  • General Health & Medical Sciences (AREA)
  • Business, Economics & Management (AREA)
  • Biomedical Technology (AREA)
  • Human Resources & Organizations (AREA)
  • Pathology (AREA)
  • Chemical & Material Sciences (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Databases & Information Systems (AREA)
  • Medicinal Chemistry (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Strategic Management (AREA)
  • General Physics & Mathematics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Economics (AREA)
  • Software Systems (AREA)
  • Educational Administration (AREA)
  • Operations Research (AREA)
  • Mathematical Physics (AREA)
  • General Engineering & Computer Science (AREA)
  • Computing Systems (AREA)
  • Evolutionary Computation (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Artificial Intelligence (AREA)
  • General Business, Economics & Management (AREA)
  • Tourism & Hospitality (AREA)
  • Quality & Reliability (AREA)
  • Marketing (AREA)
  • Game Theory and Decision Science (AREA)
  • Development Economics (AREA)
  • Pharmacology & Pharmacy (AREA)
  • Toxicology (AREA)
  • Medical Treatment And Welfare Office Work (AREA)

Abstract

一种信息判断方法和装置，具体实现方案为：获取处方点评信息（101），其中处方点评信息包括：医生开具的处方信息和药师根据处方信息给出的点评信息；对处方信息和点评信息进行特征提取，生成处方点评信息对应的特征数据集（102）；根据当前的审核规则，对特征数据集进行判断，得到处方点评信息对应的点评结果（103），其中审核规则用于表征特征数据集与点评结果之间的对应关系，审核规则基于训练得到的分类决策模型的训练结果而更新。该方案实现一种利用学习得到的审核规则对处方点评信息进行判断的方法，提高了处方审核系统的审核效率和准确率。

Description

信息判断方法和装置
交叉引用
本专利申请要求于2020年07月28日提交的、申请号为202010735811.1、发明名称为“信息判断方法和装置”的中国专利申请的优先权，该申请的全文以引用的方式并入本申请中。
技术领域
本申请涉及计算机技术领域，具体涉及智慧医疗、信息处理技术领域，尤其涉及信息判断方法和装置。
背景技术
目前因医疗资源分配不均，导致部分医院药师的专业领域知识缺乏，对医院开具的处方审核不够完善以及准确，需要一种处方审核系统来协助完成。处方审核系统的核心在于审核规则，目前通用的获取审核规则的方式有两种：一种是药师根据经验配置，一种是系统从药学知识图谱中抽取。而人工配制规则是一项繁重的工作，且很容易由于疏忽配置错误；从药学知识图谱中抽取规则作为一种新方法能够降低药师的工作量，但获取的规则是静态规则，无法动态的学习处方的审核规则，并且因每家医院对相同药品的审核规则不同，从通用药学知识图谱提取的规则无法满足多家医院多样化的审方需求。因此如何通过每家医院的大量处方点评信息学习出审核规则就显得尤为重要。
发明内容
本申请提供了一种信息判断方法、装置、设备以及存储介质。
根据本申请的第一方面，提供了一种信息判断方法，该方法包括：获取处方点评信息，其中处方点评信息包括：医生开具的处方信息和药师根据处方信息给出的点评信息；对处方信息和点评信息进行特征提取，生成 处方点评信息对应的特征数据集；根据当前的审核规则，对特征数据集进行判断，得到处方点评信息对应的点评结果，其中审核规则用于表征特征数据集与点评结果之间的对应关系，审核规则基于训练得到的分类决策模型的训练结果而更新。
在一些实施例中，方法还包括：根据预设条件，对分类决策模型进行训练；分类决策模型基于以下步骤训练得到：获取训练样本集，其中，训练样本集中的训练样本包括：提取得到的不同医院的各类特征数据集和与不同医院的各类特征数据集对应的点评结果；利用机器学习方法，将训练样本集中训练样本包括的不同医院的各类特征数据集作为检测网络的输入，将与不同医院的各类特征数据集对应的点评结果作为检测网络的期望输出，训练得到分类决策模型。
在一些实施例中，审核规则基于训练得到的分类决策模型的训练结果而更新，包括：判断分类决策模型是否训练完成；响应于分类决策模型训练完成，根据分类决策模型的输入参数和输出参数，对审核规则进行更新。
在一些实施例中，在对处方信息和点评信息进行特征提取，生成处方点评信息对应的特征数据集之前以及在获取训练样本集之前，还包括：对处方信息和点评信息进行清洗，生成清洗后的处方点评信息，其中清洗基于对处方信息和点评信息进行数据结构化处理。
在一些实施例中，对处方信息和点评信息进行清洗，生成清洗后的处方点评信息，包括：对处方信息进行信息提取，得到处方信息对应的个人信息、处方信息对应的诊断信息和处方信息对应的药品信息；根据个人信息，确定处方信息对应的人群标签；对诊断信息进行标准化，得到处方信息对应的诊断名称；对药品信息进行标准化和单位归一化，得到处方信息对应的药品标准信息；根据文本分类技术，对点评信息进行分类，确定点评信息的所属类别；根据人群标签、诊断名称、药品标准信息和点评信息的所属类别，生成清洗后的处方点评信息。
在一些实施例中，在对处方信息和点评信息进行特征提取，生成处方点评信息对应的特征数据集之后以及在获取训练样本集之后，还包括：对特征数据集进行编码化处理，得到处理后的特征数据集。
在一些实施例中，在对处方信息和点评信息进行特征提取，生成处方点评信息对应的特征数据集之后以及在获取训练样本集之后，还包括：根据预设的药理知识，对特征数据集进行筛选，生成筛选后的特征数据集。
在一些实施例中，方法还包括：基于点评结果与产品的相关性，优化产品的结构和/或产品策略。
在一些实施例中，特征数据集包括：特征相对独立的离散特征和特征之间具有关联关系的关联特征。
根据本申请的第二方面，提供了一种信息判断装置，装置包括：获取单元，被配置成获取处方点评信息，其中处方点评信息包括：医生开具的处方信息和药师根据处方信息给出的点评信息；特征提取单元，被配置成对处方信息和点评信息进行特征提取，生成处方点评信息对应的特征数据集；判断单元，被配置成根据当前的审核规则，对特征数据集进行判断，得到处方点评信息对应的点评结果，其中审核规则用于表征特征数据集与点评结果之间的对应关系，审核规则基于训练得到的分类决策模型的训练结果而更新。
在一些实施例中，装置还包括：训练单元，被配置成根据预设条件，对分类决策模型进行训练；训练单元中的分类决策模型基于以下步骤训练得到：获取训练样本集，其中，训练样本集中的训练样本包括：提取得到的不同医院的各类特征数据集和与不同医院的各类特征数据集对应的点评结果；利用机器学习方法，将训练样本集中训练样本包括的不同医院的各类特征数据集作为检测网络的输入，将与不同医院的各类特征数据集对应的点评结果作为检测网络的期望输出，训练得到分类决策模型。
在一些实施例中，判断单元，包括：判断模块，被配置成判断分类决策模型是否训练完成；更新模块，被配置成响应于分类决策模型训练完成，根据分类决策模型的输入参数和输出参数，对审核规则进行更新。
在一些实施例中，装置还包括：清洗单元，被配置成对处方信息和点评信息进行清洗，生成清洗后的处方点评信息，其中清洗基于对处方信息和点评信息进行数据结构化处理。
在一些实施例中，清洗单元，包括：提取模块，被配置成对处方信息 进行信息提取，得到处方信息对应的个人信息、处方信息对应的诊断信息和处方信息对应的药品信息；确定模块，被配置成根据个人信息，确定处方信息对应的人群标签；对诊断信息进行标准化，得到处方信息对应的诊断名称；对药品信息进行标准化和单位归一化，得到处方信息对应的药品标准信息；分类模块，被配置成根据文本分类技术，对点评信息进行分类，确定点评信息的所属类别；生成模块，被配置成根据人群标签、诊断名称、药品标准信息和点评信息的所属类别，生成清洗后的处方点评信息。
在一些实施例中，装置还包括：处理单元，被配置成对特征数据集进行编码化处理，得到处理后的特征数据集。
在一些实施例中，装置还包括：筛选单元，被配置成根据预设的药理知识，对特征数据集进行筛选，生成筛选后的特征数据集。
在一些实施例中，装置还包括：优化单元，被配置成基于点评结果与产品的相关性，优化产品的结构和/或产品策略。
根据本申请的第三方面，提供了一种电子设备，包括：至少一个处理器；以及与至少一个处理器通信连接的存储器；其中，存储器存储有可被至少一个处理器执行的指令，指令被至少一个处理器执行，以使至少一个处理器能够执行如第一方面中任一实现方式描述的方法。
根据本申请的第四方面，本申请提供了一种存储有计算机指令的非瞬时计算机可读存储介质，其特征在于，计算机指令用于使计算机执行如第一方面中任一实现方式描述的方法。
根据本申请的技术采用获取处方点评信息，对处方信息和点评信息进行特征提取，生成处方点评信息对应的特征数据集，根据当前的审核规则，对特征数据集进行判断，得到处方点评信息对应的点评结果，其中，审核规则基于训练得到的分类决策模型的训练结果而更新，解决了因人工配制规则而导致的配置错误的问题，将药师从繁杂的规则配置工作中解放出来，同时还解决了现有技术中因获取的规则是静态规则，无法动态的学习处方的审核规则的问题，实现一种利用学习得到的审核规则对处方点评信息进行判断的方法，提高了处方审核系统的审核效率和准确率。
应当理解，本部分所描述的内容并非旨在标识本申请的实施例的关键 或重要特征，也不用于限制本申请的范围。本申请的其它特征将通过以下的说明书而变得容易理解。
附图说明
附图用于更好地理解本方案，不构成对本申请的限定。
图1是根据本申请的信息判断方法的第一实施例的示意图；
图2是可以实现本申请实施例的信息判断方法的场景图；
图3是根据本申请的信息判断方法的第二实施例的示意图；
图4a是清洗后的处方点评信息的示例图；
图4b是审核规则的一个示例图；
图5是根据本申请的信息判断装置的一个实施例的结构示意图；
图6是用来实现本申请实施例的信息判断方法的电子设备的框图。
具体实施方式
以下结合附图对本申请的示范性实施例做出说明，其中包括本申请实施例的各种细节以助于理解，应当将它们认为仅仅是示范性的。因此，本领域普通技术人员应当认识到，可以对这里描述的实施例做出各种改变和修改，而不会背离本申请的范围和精神。同样，为了清楚和简明，以下的描述中省略了对公知功能和结构的描述。
需要说明的是，在不冲突的情况下，本申请中的实施例及实施例中的特征可以相互组合。下面将参考附图并结合实施例来详细说明本申请。
图1示出了根据本申请的信息判断方法的第一实施例的示意图100。该信息判断方法，包括以下步骤：
步骤101，获取处方点评信息。
在本实施例中，执行主体可以通过有线连接方式或者无线连接方式从其他电子设备或者本地获取处方点评信息，也可以通过对处方审核请求进行解析得到处方点评信息。其中处方点评信息包括：处方信息和点评信息。处方信息表示由医生开具，包含患者的个人信息(例如年龄、性别等)、患者的诊断信息(包括所患疾病、过敏史、体检指标等)以及药品信息(包括药品的通用名、规格、单次使用剂量、给药途径、给药频次等)。点评 信息表示由药师根据处方信息给出，包括药师对处方是否合理的判断信息，点评信息可以包括：有用药与诊断不符、用药与性别不符、单次剂量过高等。需要指出的是，上述无线连接方式可以包括但不限于3G、4G、5G连接、WiFi连接、蓝牙连接、WiMAX连接、Zigbee连接、UWB(ultra wideband)连接、以及其他现在已知或将来开发的无线连接方式。
步骤102，对处方信息和点评信息进行特征提取，生成处方点评信息对应的特征数据集。
在本实施例中，执行主体可以基于特征提取方法，对处方信息和点评信息进行特征提取，生成处方点评信息对应的特征数据集。其中，特征数据集包括：特征相对独立的离散特征和特征之间具有关联关系的关联特征。例如，特征数据集中的特征可以包括：患者的性别、年龄、人群特征，患者的诊断特征，药品的通用名、给药频次、单次剂量、给药途径特征。其中性别、人群、诊断、通用名、给药途径等属于离散特征，年龄、给药频次、单次剂量等属于关联特征。
步骤103，根据当前的审核规则，对特征数据集进行判断，得到处方点评信息对应的点评结果。
在本实施例中，执行主体可以根据当前的审核规则，对特征数据集进行判断，得到处方点评信息对应的点评结果。其中审核规则用于表征特征数据集与点评结果之间的对应关系，审核规则基于训练得到的分类决策模型的训练结果而更新，使得审核规则经动态学习得到最优的审核规则。
需要说明的是，上述更新方法是目前广泛研究和应用的公知技术，在此不再赘述。
继续参见图2，本实施例的信息判断方法200运行于电子设备201中。当电子设备201接收到审核请求后，电子设备201首先获取处方点评信息202，然后电子设备201对处方点评信息中的处方信息和处方点评信息中的点评信息进行特征提取，生成处方点评信息对应的特征数据集203，最后电子设备201根据当前的审核规则，对特征数据集进行判断，得到处方点评信息对应的点评结果204。
本申请的上述实施例提供的信息判断方法采用获取处方点评信息，对处方信息和点评信息进行特征提取，生成处方点评信息对应的特征数据集， 根据当前的审核规则，对特征数据集进行判断，得到处方点评信息对应的点评结果，其中，审核规则基于训练得到的分类决策模型的训练结果而更新，解决了因人工配制规则而导致的配置错误的问题，将药师从繁杂的规则配置工作中解放出来，同时还解决了现有技术中因获取的规则是静态规则，无法动态的学习处方的审核规则的问题，实现一种利用学习得到的审核规则对处方点评信息进行判断的方法，提高了处方审核系统的审核效率和准确率。
进一步参考图3，其示出了信息判断方法的第二实施例的示意图300。该方法的流程包括以下步骤：
步骤301，获取处方点评信息。
步骤302，对处方信息和点评信息进行清洗，生成清洗后的处方点评信息。
在本实施例中，执行主体可以分别对处方信息和点评信息进行清洗，生成清洗后的处方信息和点评信息，将清洗后的处方信息和点评信息汇总，得到清洗后的处方点评信息，其中清洗基于对处方信息和点评信息进行数据结构化处理。
在本实施例的一些可选的实现方式中，对处方信息和点评信息进行清洗，生成清洗后的处方点评信息，包括：对处方信息进行信息提取，得到处方信息对应的个人信息、处方信息对应的诊断信息和处方信息对应的药品信息；根据个人信息，确定处方信息对应的人群标签；对诊断信息进行标准化，得到处方信息对应的诊断名称；对药品信息进行标准化和单位归一化，得到处方信息对应的药品标准信息；根据文本分类技术，对点评信息进行分类，确定点评信息的所属类别；根据人群标签、诊断名称、药品标准信息和点评信息的所属类别，生成清洗后的处方点评信息，具体示例如图4a所示。
进一步说明，人群标签可以包括：新生儿、儿童、成人、老人等。常见的诊断名称标准化方法分为两种：第一种利用国际通用的ICD(International Classification of Diseases,国际疾病分类)10编码表进行标准化，训练词向量并利用余弦相似度将诊断名称映射为ICD10编码表中 的标准疾病名称；第二种是采用句法分析算法提取诊断名称中短语的上下位关系，并将基础诊断名称作为诊断的标准化疾病名称。点评信息的所属类别可以包括：合理处方、诊断与用药不符、用药与性别不符、重复用药、单次剂量过高、给药频次过高和给药途径不对等。对药品信息进行标准化和单位归一化，具体可以包括：从处方中提取药品的用法用量信息，对用法用量中给药途径和给药品率描述进行标准化，对用法用量中单次剂量的单位进行归一化。通过统一处方点评信息中各个特征的描述，降低了处方审核规则的复杂度。
步骤303，对处方信息和点评信息进行特征提取，生成处方点评信息对应的特征数据集。
步骤304，对特征数据集进行编码化处理，得到处理后的特征数据集。
在本实施例中，执行主体可以对步骤303生成的特征数据集中的诊断描述的相关特征进行分类和编码，并结合药品说明书对编码进行降维处理，得到处理后的特征数据集。由于诊断名称描述的复杂性和多样性的特点，若直接使用独热编码ONE-HOT会导致特征的维度过高，不利于分类决策模型的训练，通过对特征数据集进行分类编码，降低特征的维度，解决诊断名称的稀疏问题。
举例说明，以非那雄胺片的药品说明书为例，该药品说明书提到了脱发和前列腺增生两类适应症。我们将涉及到非那雄胺片的处方中的诊断信息编码为2位编码，并利用文本相似度将处方中的诊断名称映射成上述2位编码，以此来解决诊断名称稀疏的问题。比如我们将秃发映射成脱发，将前列腺炎增生映射为前列腺增生。
步骤305，根据预设的药理知识，对特征数据集进行筛选，生成筛选后的特征数据集。
在本实施例中，执行主体可以根据预设的药理知识并结合药品说明书，对步骤304得到的特征数据集的特征维度进行重要性排序和筛选，生成筛选后的特征数据集。例如，生成诊断与用药不符的规则时，只需要使用药品通用名特征、患者的诊断特征；生成用药与性别不符的规则时，只需要使用药品的通用名特征、患者的性别特征；生成重复用药的规则时，只需 要使用药品的通用名特征、药品的给药途径特征；生成单次剂量过高、给药频次过高等规则时，只需要使用患者的年龄、人群、诊断特征以及药品的通用名、单次剂量、给药频次特征。通过筛选得到准确且具有针对性的特征数据，提升了系统的效率和准确性。
步骤306，根据当前的审核规则，对特征数据集进行判断，得到处方点评信息对应的点评结果。
在本实施例中，执行主体可以根据当前的审核规则，对特征数据集进行判断，得到处方点评信息对应的点评结果。其中审核规则基于训练得到的分类决策模型的训练结果而更新。审核规则的一个示例参见图4b所示。
在本实施例的一些可选的实现方式中，方法还包括：根据预设条件，对分类决策模型进行训练；分类决策模型基于以下步骤训练得到：获取训练样本集，其中，训练样本集中的训练样本包括：提取得到的不同医院的各类特征数据集和与不同医院的各类特征数据集对应的点评结果；利用机器学习方法，将训练样本集中训练样本包括的不同医院的各类特征数据集作为检测网络的输入，将与不同医院的各类特征数据集对应的点评结果作为检测网络的期望输出，训练得到分类决策模型。基于不同医院的各类特征数据集进行训练，解决了因每家医院对相同药品的审核规则不同，无法满足多家医院多样化的审方需求的问题，提高了系统效率和应用范围。利用传统的机器学习模型具备的非常强的可解释性，提升了系统效能。
在本实施例的一些可选的实现方式中，审核规则基于训练得到的分类决策模型的训练结果而更新，包括：判断分类决策模型是否训练完成；响应于分类决策模型训练完成，根据分类决策模型的输入参数和输出参数，对审核规则进行更新。在每次模型训练完对审核规则进行更新，使审核规则与分类决策模型保持一致，简化审核规则更新过程的同时保证了审核规则的最优解。
在本实施例的一些可选的实现方式中，方法还包括：基于点评结果与产品的相关性，优化产品的结构和/或产品策略。根据点评结果，药师会对分类决策模型和审核规则进行复核，保证审核系统审核结果的正确性。
在本实施例中，步骤301、303和306的具体操作与图1所示的实施 例中的步骤101、102和103的操作基本相同，在此不再赘述。
从图3中可以看出，与图1对应的实施例相比，本实施例中的信息判断方法的示意图300采用对处方信息和点评信息进行清洗，生成清洗后的处方点评信息，对特征数据集进行编码化处理，得到处理后的特征数据集，根据预设的药理知识，对特征数据集进行筛选，生成筛选后的特征数据集，解决了诊断名称的稀疏问题，降低了特征的维度，通过筛选得到准确且具有针对性的特征数据，提升了系统的效率和准确性。
进一步参考图5，作为对上述各图所示方法的实现，本申请提供了一种信息判断装置的一个实施例，该装置实施例与图1所示的方法实施例相对应，该装置具体可以应用于各种电子设备中。
如图5所示，本实施例的信息判断装置500包括：获取单元501、特征提取单元502和判断单元503，其中，获取单元，被配置成获取处方点评信息，其中处方点评信息包括：医生开具的处方信息和药师根据处方信息给出的点评信息；特征提取单元，被配置成对处方信息和点评信息进行特征提取，生成处方点评信息对应的特征数据集；判断单元，被配置成根据当前的审核规则，对特征数据集进行判断，得到处方点评信息对应的点评结果，其中审核规则用于表征特征数据集与点评结果之间的对应关系，审核规则基于训练得到的分类决策模型的训练结果而更新。
在本实施例中，信息判断装置500的获取单元501、特征提取单元502和判断单元503的具体处理及其所带来的技术效果可分别参考图1对应的实施例中的步骤101到步骤103的相关说明，在此不再赘述。
在本实施例的一些可选的实现方式中，过滤单元，包括：聚类模块，被配置成对第一产品对集合中的所有产品对进行聚类后，计算得到第一产品对集合中各个产品的复购周期；第一判断模块，被配置成根据各个产品的复购周期和相应产品的购买时刻，判断第一产品对集合中各个产品是否超过复购时限；第一存储模块，被配置成若第一产品对集合中的产品超过复购时限，将该产品的产品对存入已过滤的第一产品对集合，若第一产品对集合中的产品未超过复购时限，将该产品的产品对存入未过滤的第一产品对集合。
在本实施例的一些可选的实现方式中，过滤单元，还包括：第二判断模块，被配置成根据用户行为的发生时刻和/或用户行为的发生次数，判断第一产品对集合中的各个产品是否满足豁免条件，其中，豁免条件用于表征对用户行为的发生时刻和/或用户行为的发生次数进行阈值判定，判断基于产品的标准产品单位SPU来完成；第二存储模块，被配置成若第一产品对集合中的产品满足豁免条件，将该产品对应的产品对存入已过滤的第一产品对集合，若第一产品对集合中的产品不满足豁免条件，将该产品对应的产品对存入未过滤的第一产品对集合。
在本实施例的一些可选的实现方式中，过滤单元，还包括：获取模块，被配置成从第一产品对集合中获取第三产品对集合，其中，第三产品对集合用于表征未获得复购周期的各个产品对的集合；过滤模块，被配置成基于产品复购选取模型，对第三产品对集合进行过滤，生成与第三产品对集合中各个产品的复购周期，其中，产品复购选取模型用于表征基于产品的复购周期对第三产品对集合进行选取。
在本实施例的一些可选的实现方式中，第一选取单元中的产品过滤策略基于多个维度对未过滤的第一产品对集合进行组合筛选；第一选取单元，包括：判断模块，被配置成根据产品过滤模型，判断未过滤的第一产品对集合中的各个产品是否为用户的购买商品，其中，产品过滤模型用于表征基于产品的产品词、产品的标准产品单位SPU和产品的库存保有单位SKU中的至少两项，对未过滤的第一产品对集合中的各个产品进行组合过滤；删除模块，被配置成若未过滤的第一产品对集合中的产品是用户的购买商品，则将该产品对应的产品对从未过滤的第一产品对集合中删除，得到第二产品对集合。
在本实施例的一些可选的实现方式中，装置还包括：第二选取单元，被配置成基于产品的类目，对第二产品对集合进行选取，得到选取后的第二产品对集合。
在本实施例的一些可选的实现方式中，装置还包括：生成单元，被配置成根据商品展示策略，对用户的第二商品集进行排序，生成用户的商品列表。
根据本申请的实施例，本申请还提供了一种电子设备和一种可读存储介质。
如图6所示，是根据本申请实施例的信息判断方法的电子设备的框图。电子设备旨在表示各种形式的数字计算机，诸如，膝上型计算机、台式计算机、工作台、个人数字助理、服务器、刀片式服务器、大型计算机、和其它适合的计算机。电子设备还可以表示各种形式的移动装置，诸如，个人数字处理、蜂窝电话、智能电话、可穿戴设备和其它类似的计算装置。本文所示的部件、它们的连接和关系、以及它们的功能仅仅作为示例，并且不意在限制本文中描述的和/或者要求的本申请的实现。
如图6所示，该电子设备包括：一个或多个处理器601、存储器602，以及用于连接各部件的接口，包括高速接口和低速接口。各个部件利用不同的总线互相连接，并且可以被安装在公共主板上或者根据需要以其它方式安装。处理器可以对在电子设备内执行的指令进行处理，包括存储在存储器中或者存储器上以在外部输入/输出装置(诸如，耦合至接口的显示设备)上显示GUI(Graphical User Interface，图形用户界面)的图形信息的指令。在其它实施方式中，若需要，可以将多个处理器和/或多条总线与多个存储器和多个存储器一起使用。同样，可以连接多个电子设备，各个设备提供部分必要的操作(例如，作为服务器阵列、一组刀片式服务器、或者多处理器系统)。图6中以一个处理器601为例。
存储器602即为本申请所提供的非瞬时计算机可读存储介质。其中，存储器存储有可由至少一个处理器执行的指令，以使至少一个处理器执行本申请所提供的信息判断方法。本申请的非瞬时计算机可读存储介质存储计算机指令，该计算机指令用于使计算机执行本申请所提供的信息判断方法。
存储器602作为一种非瞬时计算机可读存储介质，可用于存储非瞬时软件程序、非瞬时计算机可执行程序以及模块，如本申请实施例中的信息判断方法对应的程序指令/模块(例如，附图5所示的获取单元501、特征提取单元502和判断单元503)。处理器601通过运行存储在存储器602中的非瞬时软件程序、指令以及模块，从而执行服务器的各种功能应用以及数据处理，即实现上述方法实施例中的信息判断方法。
存储器602可以包括存储程序区和存储数据区，其中，存储程序区可存储操作系统、至少一个功能所需要的应用程序；存储数据区可存储根据信息判断电子设备的使用所创建的数据等。此外，存储器602可以包括高速随机存取存储器，还可以包括非瞬时存储器，例如至少一个磁盘存储器件、闪存器件、或其他非瞬时固态存储器件。在一些实施例中，存储器602可选包括相对于处理器601远程设置的存储器，这些远程存储器可以通过网络连接至信息判断电子设备。上述网络的实例包括但不限于互联网、企业内部网、局域网、移动通信网及其组合。
信息判断方法的电子设备还可以包括：输入装置603和输出装置604。处理器601、存储器602、输入装置603和输出装置604可以通过总线或者其他方式连接，图6中以通过总线连接为例。
输入装置603可接收输入的数字或字符信息，以及产生与信息判断电子设备的用户设置以及功能控制有关的键信号输入，例如触摸屏、小键盘、鼠标、轨迹板、触摸板、指示杆、一个或者多个鼠标按钮、轨迹球、操纵杆等输入装置。输出装置604可以包括显示设备、辅助照明装置(例如，LED)和触觉反馈装置(例如，振动电机)等。该显示设备可以包括但不限于，液晶显示器(LCD)、发光二极管(LED)显示器和等离子体显示器。在一些实施方式中，显示设备可以是触摸屏。
此处描述的系统和技术的各种实施方式可以在数字电子电路系统、集成电路系统、专用ASIC(专用集成电路)、计算机硬件、固件、软件、和/或它们的组合中实现。这些各种实施方式可以包括：实施在一个或者多个计算机程序中，该一个或者多个计算机程序可在包括至少一个可编程处理器的可编程系统上执行和/或解释，该可编程处理器可以是专用或者通用可编程处理器，可以从存储系统、至少一个输入装置、和至少一个输出装置接收数据和指令，并且将数据和指令传输至该存储系统、该至少一个输入装置、和该至少一个输出装置。
这些计算程序(也称作程序、软件、软件应用、或者代码)包括可编程处理器的机器指令，并且可以利用高级过程和/或面向对象的编程语言、和/或汇编/机器语言来实施这些计算程序。如本文使用的，术语“机器可读介质”和“计算机可读介质”指的是用于将机器指令和/或数据提供给可 编程处理器的任何计算机程序产品、设备、和/或装置(例如，磁盘、光盘、存储器、可编程逻辑装置(PLD))，包括，接收作为机器可读信号的机器指令的机器可读介质。术语“机器可读信号”指的是用于将机器指令和/或数据提供给可编程处理器的任何信号。
为了提供与用户的交互，可以在计算机上实施此处描述的系统和技术，该计算机具有：用于向用户显示信息的显示装置(例如，CRT(阴极射线管)或者LCD(液晶显示器)监视器)；以及键盘和指向装置(例如，鼠标或者轨迹球)，用户可以通过该键盘和该指向装置来将输入提供给计算机。其它种类的装置还可以用于提供与用户的交互；例如，提供给用户的反馈可以是任何形式的传感反馈(例如，视觉反馈、听觉反馈、或者触觉反馈)；并且可以用任何形式(包括声输入、语音输入或者、触觉输入)来接收来自用户的输入。
可以将此处描述的系统和技术实施在包括后台部件的计算系统(例如，作为数据服务器)、或者包括中间件部件的计算系统(例如，应用服务器)、或者包括前侧部件的计算系统(例如，具有图形用户界面或者网络浏览器的用户计算机，用户可以通过该图形用户界面或者该网络浏览器来与此处描述的系统和技术的实施方式交互)、或者包括这种后台部件、中间件部件、或者前侧部件的任何组合的计算系统中。可以通过任何形式或者介质的数字数据通信(例如，通信网络)来将系统的部件相互连接。通信网络的示例包括：局域网(LAN)、广域网(WAN)和互联网。
计算机系统可以包括客户端和服务器。客户端和服务器一般远离彼此并且通常通过通信网络进行交互。通过在相应的计算机上运行并且彼此具有客户端-服务器关系的计算机程序来产生客户端和服务器的关系。
根据本申请实施例的技术方案，采用获取处方点评信息，对处方信息和点评信息进行特征提取，生成处方点评信息对应的特征数据集，根据当前的审核规则，对特征数据集进行判断，得到处方点评信息对应的点评结果，其中，审核规则基于训练得到的分类决策模型的训练结果而更新，解决了因人工配制规则而导致的配置错误的问题，将药师从繁杂的规则配置工作中解放出来，同时还解决了现有技术中因获取的规则是静态规则，无法动态的学习处方的审核规则的问题，实现一种利用学习得到的审核规则 对处方点评信息进行判断的方法，提高了处方审核系统的审核效率和准确率。
应该理解，可以使用上面所示的各种形式的流程，重新排序、增加或删除步骤。例如，本申请中记载的各步骤可以并行地执行也可以顺序地执行也可以不同的次序执行，只要能够实现本申请公开的技术方案所期望的结果，本文在此不进行限制。
上述具体实施方式，并不构成对本申请保护范围的限制。本领域技术人员应该明白的是，根据设计要求和其他因素，可以进行各种修改、组合、子组合和替代。任何在本申请的精神和原则之内所作的修改、等同替换和改进等，均应包含在本申请保护范围之内。

Claims (18)

  1. 一种信息判断方法，所述方法包括：
    获取处方点评信息，其中所述处方点评信息包括：医生开具的处方信息和药师根据所述处方信息给出的点评信息；
    对所述处方信息和所述点评信息进行特征提取，生成所述处方点评信息对应的特征数据集；以及
    根据当前的审核规则，对所述特征数据集进行判断，得到所述处方点评信息对应的点评结果，其中所述审核规则用于表征特征数据集与点评结果之间的对应关系，所述审核规则基于训练得到的分类决策模型的训练结果而更新。
  2. 根据权利要求1所述方法，还包括：
    根据预设条件，对所述分类决策模型进行训练；
    所述分类决策模型基于以下步骤训练得到：
    获取训练样本集，其中，所述训练样本集中的训练样本包括：提取得到的不同医院的各类特征数据集和与不同医院的各类特征数据集对应的点评结果；以及
    利用机器学习方法，将所述训练样本集中训练样本包括的不同医院的各类特征数据集作为检测网络的输入，将与不同医院的各类特征数据集对应的点评结果作为所述检测网络的期望输出，训练得到分类决策模型。
  3. 根据权利要求1-2任一项所述方法，其中，所述审核规则基于训练得到的分类决策模型的训练结果而更新，包括：
    判断所述分类决策模型是否训练完成；以及
    响应于所述分类决策模型训练完成，根据所述分类决策模型的输入参数和输出参数，对所述审核规则进行更新。
  4. 根据权利要求1-3任一项所述方法，其中，在所述对所述处 方信息和所述点评信息进行特征提取，生成所述处方点评信息对应的特征数据集之前以及在所述获取训练样本集之前，还包括：
    对所述处方信息和所述点评信息进行清洗，生成清洗后的处方点评信息，其中所述清洗基于对所述处方信息和所述点评信息进行数据结构化处理。
  5. 根据权利要求4所述方法，其中，所述对所述处方信息和所述点评信息进行清洗，生成清洗后的处方点评信息，包括：
    对所述处方信息进行信息提取，得到所述处方信息对应的个人信息、所述处方信息对应的诊断信息和所述处方信息对应的药品信息；
    根据所述个人信息，确定所述处方信息对应的人群标签；对所述诊断信息进行标准化，得到所述处方信息对应的诊断名称；对所述药品信息进行标准化和单位归一化，得到所述处方信息对应的药品标准信息；
    根据文本分类技术，对所述点评信息进行分类，确定所述点评信息的所属类别；以及
    根据所述人群标签、所述诊断名称、所述药品标准信息和所述点评信息的所属类别，生成清洗后的处方点评信息。
  6. 根据权利要求1-5任一项所述方法，其中，在所述对所述处方信息和所述点评信息进行特征提取，生成所述处方点评信息对应的特征数据集之后以及在所述获取训练样本集之后，还包括：
    对特征数据集进行编码化处理，得到处理后的特征数据集。
  7. 根据权利要求1-6任一项所述方法，其中，在所述对所述处方信息和所述点评信息进行特征提取，生成所述处方点评信息对应的特征数据集之后以及在所述获取训练样本集之后，还包括：
    根据预设的药理知识，对特征数据集进行筛选，生成筛选后的特征数据集。
  8. 根据权利要求1-7任一项所述方法，还包括：
    基于所述点评结果与产品的相关性，优化产品的结构和/或产品策略。
  9. 一种信息判断装置，所述装置包括：
    获取单元，被配置成获取处方点评信息，其中所述处方点评信息包括：医生开具的处方信息和药师根据所述处方信息给出的点评信息；
    特征提取单元，被配置成对所述处方信息和所述点评信息进行特征提取，生成所述处方点评信息对应的特征数据集；以及
    判断单元，被配置成根据当前的审核规则，对所述特征数据集进行判断，得到所述处方点评信息对应的点评结果，其中所述审核规则用于表征特征数据集与点评结果之间的对应关系，所述审核规则基于训练得到的分类决策模型的训练结果而更新。
  10. 根据权利要求9所述装置，还包括：
    训练单元，被配置成根据预设条件，对所述分类决策模型进行训练；以及
    所述训练单元中的所述分类决策模型基于以下步骤训练得到：获取训练样本集，其中，所述训练样本集中的训练样本包括：提取得到的不同医院的各类特征数据集和与不同医院的各类特征数据集对应的点评结果；利用机器学习方法，将所述训练样本集中训练样本包括的不同医院的各类特征数据集作为检测网络的输入，将与不同医院的各类特征数据集对应的点评结果作为所述检测网络的期望输出，训练得到分类决策模型。
  11. 根据权利要求9-10任一项所述装置，其中，所述判断单元，包括：
    判断模块，被配置成判断所述分类决策模型是否训练完成；以 及
    更新模块，被配置成响应于所述分类决策模型训练完成，根据所述分类决策模型的输入参数和输出参数，对所述审核规则进行更新。
  12. 根据权利要求9-11任一项所述装置，还包括：
    清洗单元，被配置成对所述处方信息和所述点评信息进行清洗，生成清洗后的处方点评信息，其中所述清洗基于对所述处方信息和所述点评信息进行数据结构化处理。
  13. 根据权利要求12所述装置，其中，所述清洗单元，包括：
    提取模块，被配置成对所述处方信息进行信息提取，得到所述处方信息对应的个人信息、所述处方信息对应的诊断信息和所述处方信息对应的药品信息；
    确定模块，被配置成根据所述个人信息，确定所述处方信息对应的人群标签；对所述诊断信息进行标准化，得到所述处方信息对应的诊断名称；对所述药品信息进行标准化和单位归一化，得到所述处方信息对应的药品标准信息；
    分类模块，被配置成根据文本分类技术，对所述点评信息进行分类，确定所述点评信息的所属类别；以及
    生成模块，被配置成根据所述人群标签、所述诊断名称、所述药品标准信息和所述点评信息的所属类别，生成清洗后的处方点评信息。
  14. 根据权利要求9-13任一项所述装置，还包括：
    处理单元，被配置成对特征数据集进行编码化处理，得到处理后的特征数据集。
  15. 根据权利要求9-14任一项所述装置，还包括：
    筛选单元，被配置成根据预设的药理知识，对特征数据集进行 筛选，生成筛选后的特征数据集。
  16. 根据权利要求9-15任一项所述装置，还包括：
    优化单元，被配置成基于所述点评结果与产品的相关性，优化产品的结构和/或产品策略。
  17. 一种电子设备，其特征在于，包括：
    至少一个处理器；以及
    与所述至少一个处理器通信连接的存储器；其中，
    所述存储器存储有可被所述至少一个处理器执行的指令，所述指令被所述至少一个处理器执行，以使所述至少一个处理器能够执行权利要求1-8中任一项所述的方法。
  18. 一种存储有计算机指令的非瞬时计算机可读存储介质，其特征在于，所述计算机指令用于使所述计算机执行权利要求1-8中任一项所述的方法。
PCT/CN2021/092019 2020-07-28 2021-05-07 信息判断方法和装置 Ceased WO2022021990A1 (zh)

Priority Applications (3)

Application Number Priority Date Filing Date Title
EP21851572.4A EP4191599A4 (en) 2020-07-28 2021-05-07 METHOD AND DEVICE FOR DETERMINING INFORMATION
US18/014,073 US20240296926A1 (en) 2020-07-28 2021-05-07 Method and apparatus for judging information
JP2023512260A JP7518971B2 (ja) 2020-07-28 2021-05-07 情報判断の方法、装置、電子機器、記憶媒体およびコンピュータプログラム

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010735811.1 2020-07-28
CN202010735811.1A CN112309531A (zh) 2020-07-28 2020-07-28 信息判断方法和装置

Publications (1)

Publication Number Publication Date
WO2022021990A1 true WO2022021990A1 (zh) 2022-02-03

Family

ID=74483598

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/092019 Ceased WO2022021990A1 (zh) 2020-07-28 2021-05-07 信息判断方法和装置

Country Status (5)

Country Link
US (1) US20240296926A1 (zh)
EP (1) EP4191599A4 (zh)
JP (1) JP7518971B2 (zh)
CN (1) CN112309531A (zh)
WO (1) WO2022021990A1 (zh)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112309531A (zh) * 2020-07-28 2021-02-02 北京沃东天骏信息技术有限公司 信息判断方法和装置
CN118629579B (zh) * 2024-08-12 2025-02-14 长春中医药大学 一种基于互联网的临床药学智慧管理方法

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20190259482A1 (en) * 2018-02-20 2019-08-22 Mediedu Oy System and method of determining a prescription for a patient
CN110223751A (zh) * 2019-05-16 2019-09-10 平安科技（深圳）有限公司 基于医疗知识图谱的处方评价方法、系统及计算机设备
CN111445976A (zh) * 2020-03-24 2020-07-24 屹嘉智创(厦门)科技有限公司 一种智能合理用药系统及方法
CN112309531A (zh) * 2020-07-28 2021-02-02 北京沃东天骏信息技术有限公司 信息判断方法和装置

Family Cites Families (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH10124578A (ja) * 1996-10-18 1998-05-15 Sanyo Electric Co Ltd チェックルールの更新装置
US8265959B2 (en) * 2008-01-31 2012-09-11 Humana Inc. Prescription drug prior authorization system and method
WO2012122347A1 (en) * 2011-03-10 2012-09-13 Teva Pharmaceutical Industries Ltd. A method, system and program for improved health care
CN110504035B (zh) * 2013-01-16 2023-05-30 梅达器材 医疗资料库及系统
JP2016527060A (ja) * 2013-08-09 2016-09-08 パーセプティメッド インコーポレイテッドＰｅｒｃｅｐｔｉｍｅｄ， Ｉｎｃ． 遠隔医薬検証
JP6432281B2 (ja) * 2014-08-19 2018-12-05 株式会社湯山製作所 病棟薬剤業務支援システム
US20190171767A1 (en) * 2017-12-04 2019-06-06 Paypal, Inc. Machine Learning and Automated Persistent Internet Domain Monitoring
US20200058381A1 (en) * 2018-08-14 2020-02-20 Miteshkumar Ishwarbhai Patel System and Method for Auditing, Monitoring, Recording, and Executing Healthcare Transactions, Communications, and Decisions
CN109670788A (zh) * 2018-12-13 2019-04-23 平安医疗健康管理股份有限公司 基于数据分析的医保审核方法、装置、设备和存储介质
CN109920508B (zh) * 2018-12-28 2022-11-01 安徽省立医院 处方审核方法及系统

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20190259482A1 (en) * 2018-02-20 2019-08-22 Mediedu Oy System and method of determining a prescription for a patient
CN110223751A (zh) * 2019-05-16 2019-09-10 平安科技（深圳）有限公司 基于医疗知识图谱的处方评价方法、系统及计算机设备
CN111445976A (zh) * 2020-03-24 2020-07-24 屹嘉智创(厦门)科技有限公司 一种智能合理用药系统及方法
CN112309531A (zh) * 2020-07-28 2021-02-02 北京沃东天骏信息技术有限公司 信息判断方法和装置

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP4191599A4 *

Also Published As

Publication number Publication date
CN112309531A (zh) 2021-02-02
US20240296926A1 (en) 2024-09-05
JP7518971B2 (ja) 2024-07-18
EP4191599A4 (en) 2024-08-21
JP2023525410A (ja) 2023-06-15
EP4191599A1 (en) 2023-06-07

Similar Documents

Publication Publication Date Title
CN107799160B (zh) 用药辅助决策方法及装置、存储介质、电子设备
US20210210213A1 (en) Method, apparatus, medium and electronic device for recommending medicine
CN109670054B (zh) 知识图谱构建方法、装置、存储介质及电子设备
CN113345577B (zh) 诊疗辅助信息的生成方法、模型训练方法、装置、设备以及存储介质
US8700589B2 (en) System for linking medical terms for a medical knowledge base
US20190371475A1 (en) Generating and applying subject event timelines
Santos et al. A data mining system for providing analytical information on brain tumors to public health decision makers
Kunjir et al. Data mining and visualization for prediction of multiple diseases in healthcare
JP2020518050A (ja) エンティティ間のコンテキスト的類似度の学習及び適用
CN109448859B (zh) 数据处理方法及装置、电子设备、存储介质
CN111382275A (zh) 医疗知识图谱的构建方法、装置、介质及电子设备
CN104715013A (zh) 一种基于Hadoop的用户健康数据分析方法和系统
CN113360490B (zh) 数据处理方法、装置、设备、介质及程序产品
Pang et al. A multi-modal data platform for diagnosis and prediction of Alzheimer’s disease using machine learning methods
CN111274397A (zh) 建立实体关系检测模型的方法以及装置
CN112562807A (zh) 医疗数据分析方法、装置、设备、存储介质和程序产品
CN111832298A (zh) 病历的质检方法、装置、设备以及存储介质
CN112562808B (zh) 患者画像的生成方法、装置、电子设备及存储介质
CN108335749A (zh) 抑郁症数据分析方法及装置
CN112786131A (zh) 识别就诊信息的方法及装置、电子设备和存储介质
WO2022021990A1 (zh) 信息判断方法和装置
CN114373539A (zh) 中西医临床数据的处理方法、系统、存储介质及终端
Bannach et al. Visual analytics for radiomics: Combining medical imaging with patient data for clinical research
CN110176283A (zh) 时空大数据关联的慢病管理方法、存储介质和终端
Pradeepa et al. DEODORANT: a novel approach for early detection and prevention of polycystic ovary syndrome using association rule in hypergraph with the dominating set property

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21851572

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2023512260

Country of ref document: JP

Kind code of ref document: A

WWE Wipo information: entry into national phase

Ref document number: 2021851572

Country of ref document: EP

ENP Entry into the national phase

Ref document number: 2021851572

Country of ref document: EP

Effective date: 20230228

NENP Non-entry into the national phase

Ref country code: DE