WO2023178970A1 - 医疗数据处理方法、装置、设备及存储介质 - Google Patents

医疗数据处理方法、装置、设备及存储介质 Download PDF

Info

Publication number
WO2023178970A1
WO2023178970A1 PCT/CN2022/121724 CN2022121724W WO2023178970A1 WO 2023178970 A1 WO2023178970 A1 WO 2023178970A1 CN 2022121724 W CN2022121724 W CN 2022121724W WO 2023178970 A1 WO2023178970 A1 WO 2023178970A1
Authority
WO
WIPO (PCT)
Prior art keywords
user
medical data
disease
doctor
end node
Prior art date
Application number
PCT/CN2022/121724
Other languages
English (en)
French (fr)
Inventor
赵璐偲
Original Assignee
康键信息技术（深圳）有限公司
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 康键信息技术（深圳）有限公司 filed Critical 康键信息技术（深圳）有限公司
Publication of WO2023178970A1 publication Critical patent/WO2023178970A1/zh

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
    • G16H40/00—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices
    • G16H40/20—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices for the management or administration of healthcare resources or facilities, e.g. managing hospital staff or surgery rooms
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR 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
    • 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
    • G16H10/00—ICT specially adapted for the handling or processing of patient-related medical or healthcare data
    • G16H10/60—ICT specially adapted for the handling or processing of patient-related medical or healthcare data for patient-specific data, e.g. for electronic patient records
    • 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
    • G16H80/00—ICT specially adapted for facilitating communication between medical practitioners or patients, e.g. for collaborative diagnosis, therapy or health monitoring

Definitions

  • This application relates to the field of artificial intelligence, and in particular, to a medical data processing method, device, equipment and storage medium.
  • This application provides a medical data processing method, device, equipment and storage medium to improve the accuracy of doctor-patient matching.
  • the first aspect of this application provides a medical data processing method, which includes: obtaining medical data corresponding to the user to be processed, and judging whether the user meets the preset user type based on the medical data, wherein , the medical data includes user gender, user age and symptom data; if the user meets the preset user type, a disease template corresponding to the user is matched according to the medical data, wherein the disease template includes user The disease diagnosis path to be filled in; fill in the disease template with information based on the disease diagnosis path, obtain the end node and user answer path corresponding to the disease template, and generate a user based on the end node and the user answer path demand; if the user demand is a doctor's consultation, input the medical data and the user answer path into the preset medical data processing model for disease data processing to obtain a prediction node; compare the prediction node and the end The nodes perform comparison analysis to obtain the comparison result; if the comparison result is that the prediction node and the end node are the same, then perform doctor
  • a second aspect of the present application provides a medical data processing device, including: a memory and at least one processor, with instructions stored in the memory; and the at least one processor calls the instructions in the memory, so that the The medical data processing equipment performs the following steps: obtains the medical data corresponding to the user to be processed, and determines whether the user meets the preset user type based on the medical data, wherein the medical data includes user gender, user age and Symptom data; if the user meets the preset user type, match the disease template corresponding to the user according to the medical data, wherein the disease template includes a disease diagnosis path to be filled in by the user; based on the disease diagnosis The path fills the disease template with information, obtains the end node and user answer path corresponding to the disease template, and generates user demand based on the end node and the user answer path; if the user demand is a doctor's consultation, Then the medical data and the user answer path are input into the preset medical data processing model for disease data processing to obtain a prediction node; a comparison analysis
  • a third aspect of the present application provides a computer-readable storage medium. Instructions are stored on the computer-readable storage medium. When the instructions are executed by a processor, the following steps are implemented: Obtain the medical data corresponding to the user to be processed. , and determine whether the user meets the preset user type based on the medical data, where the medical data includes user gender, user age and symptom data; if the user meets the preset user type, then based on the The medical data matches a disease template corresponding to the user, where the disease template includes a disease diagnosis path to be filled in by the user; information is filled in the disease template based on the disease diagnosis path to obtain the end corresponding to the disease template nodes and user answer paths, and generate user needs based on the end node and the user answer path; if the user needs are doctor consultations, input the medical data and the user answer path into preset medical data
  • the processing model performs disease data processing to obtain prediction nodes; performs comparison analysis on the prediction node and the end node to obtain a comparison result; if the comparison
  • a fourth aspect of the present application provides a medical data processing device, wherein the medical data processing device includes: an acquisition module for acquiring medical data corresponding to a user to be processed, and determining whether the user is a user based on the medical data. Comply with the preset user type, wherein the medical data includes user gender, user age and symptom data; a matching module configured to match the user with the user according to the medical data if the user complies with the preset user type.
  • the corresponding disease template wherein the disease template includes a disease diagnosis path to be filled in by the user; a filling module, used to fill the disease template with information based on the disease diagnosis path to obtain the end node corresponding to the disease template and The user answers the path, and generates user requirements according to the end node and the user answer path; a processing module, used to input and preset the medical data and the user answer path if the user requirement is a doctor's consultation.
  • the medical data processing model performs disease data processing to obtain prediction nodes; an analysis module is used to compare and analyze the prediction node and the end node to obtain a comparison result; a generation module is used to compare the comparison result If the prediction node and the end node are the same, then perform doctor matching on the user based on the prediction node and the end node to obtain the target doctor.
  • the medical data includes user gender, user age and symptom data; if the user meets the preset user type, the medical data is matched with The disease template corresponding to the user, where the disease template includes the disease diagnosis path to be filled in by the user; the disease template is filled with information based on the disease diagnosis path to obtain the end node and user answer path corresponding to the disease template, and based on the end node and user answer path Generate user needs; if the user needs are doctor consultation, input the medical data and user answer path into the preset medical data processing model for disease data processing, and obtain the prediction node; perform a comparison and analysis on the prediction node and the end node to obtain the comparison If the comparison result is that the prediction node and the end node are the same, the user will be matched with doctors based on the prediction node and the end node to obtain the target doctor.
  • This application obtains preliminary diagnosis results and other medical information by performing self-diagnosis on the user. This more comprehensive information can be used to more accurately match doctors and patients, avoiding some invalid communication or wrong matching. User self-diagnosis and medical need identification can be carried out based on the disease diagnosis path template, which improves the efficiency of the online consultation process.
  • Figure 1 is a schematic diagram of an embodiment of the medical data processing method in the embodiment of the present application.
  • Figure 2 is a schematic diagram of another embodiment of the medical data processing method in the embodiment of the present application.
  • Figure 3 is a schematic diagram of an embodiment of the medical data processing device in the embodiment of the present application.
  • Figure 4 is a schematic diagram of another embodiment of the medical data processing device in the embodiment of the present application.
  • Figure 5 is a schematic diagram of an embodiment of the medical data processing equipment in the embodiment of the present application.
  • This application provides a medical data processing method, device, equipment and storage medium to improve the accuracy of doctor-patient matching.
  • One embodiment of the medical data processing method in the embodiment of the present application includes:
  • the medical data includes user gender, user age and symptom data;
  • the medical data input by the user to be processed includes personal gender, age and symptom data.
  • User types include those who do not know the disease they suffer from, have unclear consultation needs, or actively choose smart disease diagnosis.
  • the server identifies the user type based on the medical data input by the user, and determines whether the user to be processed meets the above user types.
  • the symptom data can be obtained from medical websites, medical institution databases, etc.
  • Symptom data is text data that includes disease words and symptom words. Symptom data can be obtained from medical websites or medical institution data through preset crawlers.
  • the medical website or medical institution data records the user's disease words and symptom words to generate symptom data. , or the user can directly input disease words and symptom words to obtain symptom data.
  • the execution subject of this application can be a medical data processing device, or a terminal or a server, which is not specifically limited here.
  • the embodiments of this application are explained by taking the server as the execution subject as an example.
  • the embodiments of this application can obtain and process relevant data based on artificial intelligence technology.
  • Artificial Intelligence is a theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.
  • Basic artificial intelligence technologies generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technology, operation/interaction systems, mechatronics and other technologies.
  • Artificial intelligence software technology mainly includes computer vision technology, robotics technology, biometric technology, speech processing technology, natural language processing technology, and machine learning/deep learning.
  • the server can be an independent server, or it can provide cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, and content delivery network (Content Delivery Network, CDN), as well as cloud servers for basic cloud computing services such as big data and artificial intelligence platforms.
  • the user meets the preset user type, match the disease template corresponding to the user based on the medical data, where the disease template includes the disease diagnosis path to be filled in by the user;
  • the server enters the self-diagnosis module.
  • the server first matches the disease template based on the previous information.
  • the disease template contains the disease diagnosis path, similar to a tree diagram, including multiple question options and different answers. May lead to different branches. By the user answering the questions in the template, the system will fill in the template nodes and throw new questions until it reaches the diagnosis node or the end node. It should be noted that the disease template is obtained through computer-assisted medical literature mining and doctor editing and review.
  • the server when the user fills in the questions in the template, the server will fill in the template nodes and throw new questions until all diagnostic nodes or end nodes are filled. Based on the final node and the intermediate user answer path, the server predicts a user's needs through a deep learning network model, including doctor consultation, finding a hospital, buying medicine, physical examination, registration, chronic disease management, comprehensive consultation, etc. The server will also predict a user's needs based on the user's needs. Recommend relevant functions and guide users to the next service module.
  • the server will input the medical data and user answer path into a preset medical data processing model for disease data processing.
  • the preset medical data processing model includes an input layer and a multi-layer convolution. Neural network and normalization layer, among which, the multi-layer convolutional neural network is used to perform convolution operation on the input vector, and the normalization layer is used to perform logistic regression operation on the feature vector, and finally outputs the prediction node.
  • the server performs comparison analysis on the prediction node and the end node to obtain the comparison result.
  • the server calculates the similarity between the prediction node and the end node.
  • the server determines the prediction node and the end node.
  • the end nodes are the same.
  • the server determines that the predicted node is different from the end node.
  • the comparison result is that the prediction node and the end node are the same, the user will be matched with doctors based on the prediction node and the end node to obtain the target doctor.
  • the server calculates the credibility of the prediction node and the end node to obtain the target credibility.
  • the server sorts the target credibility and selects the target credibility.
  • the degree corresponds to the maximum value of the candidate doctor, and the maximum value of the candidate doctor is obtained.
  • the server uses the candidate doctor corresponding to the maximum value as the target doctor.
  • the server stores the user requirements in the blockchain database, and the details are not limited here.
  • the medical data includes the user's gender, user age and symptom data; if the user meets the preset user type, the corresponding user is matched based on the medical data.
  • disease template where the disease template includes a disease diagnosis path to be filled in by the user; fill in the disease template with information based on the disease diagnosis path, obtain the end node and user answer path corresponding to the disease template, and generate a user based on the end node and user answer path Demand; if the user's demand is a doctor's consultation, input the medical data and user answer path into the preset medical data processing model for disease data processing, and obtain the prediction node; perform a comparison and analysis on the prediction node and the end node to obtain the comparison result. ; If the comparison result is that the prediction node and the end node are the same, the user is matched with doctors based on the prediction node and the end node to obtain the target doctor.
  • This application obtains preliminary diagnosis results and other medical information by self-diagnosing the user, so that it can be used This part of more comprehensive information allows for more accurate doctor-patient matching, avoiding some ineffective communication or mismatching.
  • User self-diagnosis and medical need identification are performed based on the disease diagnosis path template, which improves the efficiency of the online consultation process.
  • Another embodiment of the medical data processing method in the embodiment of this application includes:
  • the server receives the medical data input by the user to be processed, where the medical data includes user gender, user age and symptom data; the server generates a user portrait corresponding to the user based on the user's gender, user age and symptom data; the server calculates whether the user portrait Match the similarity corresponding to the preset user type, and compare the similarity with the preset target value; if the similarity is greater than or equal to the preset target value, the server determines that the user matches the preset user type. Specifically, if the similarity is less than the preset target value, the server determines that the user does not meet the preset user type.
  • the server receives the medical data input by the user to be processed. The user can input medical data through the preset terminal.
  • the medical data includes The user's personal information and symptom data; the server generates a user portrait corresponding to the user based on the user's gender, user age, and symptom data, and the server generates a triplet corresponding to the user, which includes gender, age, and symptoms; the server calculates whether the user portrait Comply with the similarity corresponding to the preset user type, and compare the similarity with the preset target value; if the similarity is greater than or equal to the preset target value, the server determines that the user matches the preset user type. If the similarity is below If the preset target value is set, the server determines that the user does not meet the above three categories of users.
  • the server will extract keywords from the medical data to obtain the target keywords; the server will match the target keywords with multiple preset disease templates to obtain the corresponding information for each disease template. keyword hit rate; the server sorts the keyword hit rate corresponding to each disease template, and uses the disease template with the highest keyword hit rate as the disease template corresponding to the user, where the disease template includes the disease diagnosis path to be filled in by the user. .
  • the server performs keyword extraction on the medical data to obtain the target keywords. The server performs keyword extraction by first converting the medical data into text data, and then using the preset OCR text recognition The model extracts text data to obtain standard text data.
  • the target keywords are obtained by removing duplicate content from the standard text data.
  • the server matches the target keywords with multiple preset disease templates to obtain each disease template.
  • the keyword hit rate is the number of keyword hits; the server sorts the keyword hit rate corresponding to each disease template, and uses the disease template with the highest keyword hit rate as the disease corresponding to the user Template, wherein the disease template includes a disease diagnosis path to be filled in by the user.
  • the server obtains the run-through rate and recommended click-through rate based on the disease diagnosis path; the server fills the disease template with information based on the run-through rate and recommended click-through rate to obtain the filled disease template; the server extracts data from the filled disease template , obtain the end node and user answer path corresponding to the disease template; the server generates user requirements based on the end node and user answer path, where user requirements include doctor consultation and offline consultation.
  • the server obtains the pass rate and recommended click rate based on the disease diagnosis path.
  • the server optimizes the template through indicators such as the pass rate and recommended click rate obtained from the statistical analysis of online business data.
  • the pass rate refers to whether the user can go
  • the server fills the disease template with information based on the run-through rate and recommended click rate to obtain the filled disease template; the server extracts data from the filled disease template and obtains the end node and user answer corresponding to the disease template. Path; the server generates user requirements based on the end node and user answer path, where user requirements include doctor consultation and offline consultation.
  • a template can be used as a node of other templates, that is, a sub-template.
  • a general ophthalmology template may include several eye disease templates as sub-templates, and the weights of template nodes will be adjusted based on the user's historical medical records such as consultation information.
  • the server will input the medical data and user answer path into the preset medical data processing model; the server will use the convolutional neural network in the medical data processing model to convolve the medical data and user answer path.
  • Product operation is performed to obtain the feature vector corresponding to the medical data; the server performs feature normalization processing on the feature vector to obtain the prediction node.
  • the preset medical data processing model includes an input layer, a convolutional neural network and an output layer.
  • the input layer one-hot vector coding layer (one-hot vector); the hidden layer: the convolution operation function, which is linear unit; output layer: the dimensions are the same as those of the input layer, and logistic regression is used.
  • the server performs one-hot vector encoding on the target information through the input layer to obtain a low-dimensional vector.
  • the low-dimensional vector is [0,0,0,1,0,1,0,0].
  • the server performs one-hot vector encoding through the convolutional neural network layer.
  • the low-dimensional vector performs feature abstraction operations to obtain abstract feature values, which are also feature vectors;
  • the server performs logistic regression operations on the abstract feature values through the output layer to obtain prediction nodes, where the logistic regression operation is a softmax regression operation, where,
  • the prediction node can be the result of the first online diagnosis, such as: online first diagnosis of blepharitis, conjunctivitis, respiratory infection, etc.
  • the server performs comparison analysis on the prediction node and the end node to obtain the comparison result.
  • the server calculates the similarity between the prediction node and the end node.
  • the server determines the prediction node and the end node.
  • the end nodes are the same.
  • the server determines that the predicted node is different from the end node.
  • the comparison result is that the prediction node and the end node are the same, then the user is matched with a doctor based on the prediction node and the end node to obtain the target doctor;
  • the server obtains multiple candidate doctors in the preset candidate doctor set; the server calculates the credibility of the multiple candidate doctors based on the prediction node and the end node, and obtains each The target credibility corresponding to each candidate doctor; the server ranks multiple candidate doctors in the candidate doctor set according to the target credibility corresponding to each candidate doctor, and uses the highest-ranked candidate doctor as the doctor matching the user, and obtains Target Doctor.
  • the server sorts the preset candidate doctor list based on the target credibility to obtain the ranking.
  • the server is based on the user and considers the user preference and candidate doctor workload.
  • the server defines the user's preference degree to measure the user's preference for candidate doctors.
  • the server obtains the preference index corresponding to the user through the weighted average method of doctor-patient matching index. After the server calculates the matching degree of all candidate doctors to the user, it recommends the target doctor based on the highest target credibility and preference index.
  • the server matches users with highly relevant target doctors to improve the user's consultation experience. In addition, some target doctors with specific expertise can also be matched with relevant users, thus avoiding the waste of medical resources.
  • the server calculates the credibility of the prediction node and the end node to obtain the target credibility.
  • the server sorts the target credibility and selects the target credibility.
  • the degree corresponds to the maximum value of the candidate doctor, and the maximum value of the candidate doctor is obtained.
  • the server uses the candidate doctor corresponding to the maximum value as the target doctor.
  • the matching doctor model will remove the diagnosis result information and use the department, chief complaint and some node information to match the doctor.
  • the template questions answered by the user will also be transmitted to the doctor as auxiliary information.
  • the server matches the user with doctors based on the auxiliary information and the end node, and obtains the target doctor corresponding to the user.
  • the server stores the user requirements in the blockchain database, and the details are not limited here.
  • the medical data includes the user's gender, user age and symptom data; if the user meets the preset user type, the corresponding user is matched based on the medical data.
  • disease template where the disease template includes a disease diagnosis path to be filled in by the user; fill in the disease template with information based on the disease diagnosis path, obtain the end node and user answer path corresponding to the disease template, and generate a user based on the end node and user answer path Demand; if the user's demand is a doctor's consultation, input the medical data and user answer path into the preset medical data processing model for disease data processing, and obtain the prediction node; perform a comparison and analysis on the prediction node and the end node to obtain the comparison result. ; If the comparison result is that the prediction node and the end node are the same, the user is matched with doctors based on the prediction node and the end node to obtain the target doctor.
  • This application obtains preliminary diagnosis results and other medical information by self-diagnosing the user, so that it can be used This part of more comprehensive information allows for more accurate doctor-patient matching, avoiding some ineffective communication or mismatching.
  • User self-diagnosis and medical need identification are performed based on the disease diagnosis path template, which improves the efficiency of the online consultation process.
  • An example of the medical data processing device in the embodiment of the present application includes:
  • the acquisition module 301 is used to obtain the medical data corresponding to the user to be processed, and determine whether the user meets the preset user type based on the medical data, where the medical data includes user gender, user age and symptom data;
  • the matching module 302 is configured to match a disease template corresponding to the user according to the medical data if the user meets the preset user type, where the disease template includes a disease diagnosis path to be filled in by the user;
  • the filling module 303 is used to fill the disease template with information based on the disease diagnosis path, obtain the end node and user answer path corresponding to the disease template, and generate user requirements based on the end node and the user answer path. ;
  • the processing module 304 is used to input the medical data and the user answer path into a preset medical data processing model to perform disease data processing if the user demand is a doctor's consultation to obtain a prediction node;
  • the analysis module 305 is used to compare and analyze the prediction node and the end node to obtain a comparison result
  • the generation module 306 is configured to perform doctor matching on the user based on the prediction node and the end node to obtain a target doctor if the comparison result is that the prediction node and the end node are the same.
  • the server stores the user requirements in the blockchain database, and the details are not limited here.
  • the medical data includes the user's gender, user age and symptom data; if the user meets the preset user type, the corresponding user is matched based on the medical data.
  • disease template where the disease template includes a disease diagnosis path to be filled in by the user; fill in the disease template with information based on the disease diagnosis path, obtain the end node and user answer path corresponding to the disease template, and generate a user based on the end node and user answer path Demand; if the user's demand is a doctor's consultation, input the medical data and user answer path into the preset medical data processing model for disease data processing, and obtain the prediction node; perform a comparison and analysis on the prediction node and the end node to obtain the comparison result. ; If the comparison result is that the prediction node and the end node are the same, the user is matched with doctors based on the prediction node and the end node to obtain the target doctor.
  • This application obtains preliminary diagnosis results and other medical information by self-diagnosing the user, so that it can be used This part of more comprehensive information allows for more accurate doctor-patient matching, avoiding some ineffective communication or mismatching.
  • User self-diagnosis and medical need identification are performed based on the disease diagnosis path template, which improves the efficiency of the online consultation process.
  • Another embodiment of the medical data processing device in the embodiment of the present application includes:
  • the acquisition module 301 is used to obtain the medical data corresponding to the user to be processed, and determine whether the user meets the preset user type based on the medical data, where the medical data includes user gender, user age and symptom data;
  • the matching module 302 is configured to match a disease template corresponding to the user according to the medical data if the user meets the preset user type, where the disease template includes a disease diagnosis path to be filled in by the user;
  • the filling module 303 is used to fill the disease template with information based on the disease diagnosis path, obtain the end node and user answer path corresponding to the disease template, and generate user requirements based on the end node and the user answer path. ;
  • the processing module 304 is used to input the medical data and the user answer path into a preset medical data processing model to perform disease data processing if the user demand is a doctor's consultation to obtain prediction nodes;
  • the analysis module 305 is used to compare and analyze the prediction node and the end node to obtain a comparison result
  • the generation module 306 is configured to perform doctor matching on the user based on the prediction node and the end node to obtain a target doctor if the comparison result is that the prediction node and the end node are the same.
  • the acquisition module 301 is specifically used for:
  • Receive medical data input by the user to be processed wherein the medical data includes user gender, user age, and symptom data; generate a user portrait corresponding to the user based on the user gender, the user age, and the symptom data; Calculate whether the user portrait conforms to the similarity corresponding to the preset user type, and compare the similarity with the preset target value; if the similarity is greater than or equal to the preset target value, determine the similarity The above user matches the preset user type.
  • the matching module 302 is specifically used for:
  • keyword extraction is performed on the medical data to obtain target keywords; the target keywords are matched with multiple preset disease templates to obtain each disease template. Corresponding keyword hit rate; sort the keyword hit rate corresponding to each disease template, and use the disease template with the highest keyword hit rate as the disease template corresponding to the user, where the disease template includes the user Disease diagnosis path to be filled in.
  • the filling module 303 is specifically used for:
  • the template performs data extraction to obtain the end node and user answer path corresponding to the disease template; user requirements are generated based on the end node and the user answer path, where the user requirements include doctor consultation and offline consultation.
  • processing module 304 is specifically used to:
  • the medical data and the user answer path are input into the preset medical data processing model; the medical data and the medical data are processed through the convolutional neural network in the medical data processing model.
  • the user answer path is subjected to a convolution operation to obtain the feature vector corresponding to the medical data; the feature vector is subjected to feature normalization processing to obtain the prediction node.
  • the generation module 306 is specifically used for:
  • the comparison result is that the prediction node and the end node are the same, obtain multiple candidate doctors in the preset candidate doctor set; perform a search on the multiple candidate doctors based on the prediction node and the end node. Calculate the credibility to obtain the target credibility corresponding to each candidate doctor; rank the multiple candidate doctors in the candidate doctor set according to the target credibility corresponding to each candidate doctor, and assign the highest-ranked The candidate doctor is used as the doctor matching the user, and the target doctor is obtained.
  • the medical data processing device also includes:
  • Deletion module 307 configured to delete the prediction node if the comparison result is that the prediction node and the end node are not the same, and generate auxiliary information corresponding to the user based on the disease template; based on The auxiliary information and the end node perform doctor matching on the user to obtain the target doctor corresponding to the user.
  • the server stores the user requirements in the blockchain database, and the details are not limited here.
  • the medical data includes the user's gender, user age and symptom data; if the user meets the preset user type, the corresponding user is matched based on the medical data.
  • disease template where the disease template includes a disease diagnosis path to be filled in by the user; fill in the disease template with information based on the disease diagnosis path, obtain the end node and user answer path corresponding to the disease template, and generate a user based on the end node and user answer path Demand; if the user's demand is a doctor's consultation, input the medical data and user answer path into the preset medical data processing model for disease data processing, and obtain the prediction node; perform a comparison and analysis on the prediction node and the end node to obtain the comparison result. ; If the comparison result is that the prediction node and the end node are the same, the user is matched with doctors based on the prediction node and the end node to obtain the target doctor.
  • This application obtains preliminary diagnosis results and other medical information by self-diagnosing the user, so that it can be used This part of more comprehensive information allows for more accurate doctor-patient matching, avoiding some ineffective communication or mismatching.
  • User self-diagnosis and medical need identification are performed based on the disease diagnosis path template, which improves the efficiency of the online consultation process.
  • FIG. 5 is a schematic structural diagram of a medical data processing device provided by an embodiment of the present application.
  • the medical data processing device 500 may vary greatly due to different configurations or performance, and may include one or more processors (central processing units). , CPU) 510 (eg, one or more processors), one or more storage media 530 (eg, one or more mass storage devices) storing application programs 533 or data 532.
  • the storage medium 530 may be short-term storage or persistent storage.
  • the program stored in the storage medium 530 may include one or more modules (not shown in the figure), and each module may include a series of instruction operations in the medical data processing device 500 .
  • the processor 510 may be configured to communicate with the storage medium 530 and execute a series of instruction operations in the storage medium 530 on the medical data processing device 500 .
  • the medical data processing device 500 may also include one or more power supplies 540, one or more wired or wireless network interfaces 550, one or more input and output interfaces 520, and/or, one or more operating systems 531, such as Windows Serve , Mac OS X, Unix, Linux, FreeBSD and more.
  • operating systems 531 such as Windows Serve , Mac OS X, Unix, Linux, FreeBSD and more.
  • the medical data processing device includes a processor.
  • the processor performs the steps of the medical data processing method in the above embodiments.
  • the computer-readable storage medium can be a non-volatile computer-readable storage medium.
  • the computer-readable storage medium can also be a volatile computer-readable storage medium. Instructions are stored in the computer-readable storage medium. When the instructions are run on the computer, they cause the computer to perform the following steps: obtain the medical data corresponding to the user to be processed, and determine whether the user meets the preset based on the medical data.
  • the medical data includes user gender, user age and symptom data; if the user meets the preset user type, the disease template corresponding to the user is matched according to the medical data, wherein the The disease template includes a disease diagnosis path to be filled in by the user; the disease template is filled with information based on the disease diagnosis path to obtain the end node corresponding to the disease template and the user answer path, and based on the end node and the The user answer path generates user needs; if the user need is a doctor's consultation, the medical data and the user answer path are input into the preset medical data processing model for disease data processing to obtain prediction nodes; the prediction is The node and the end node are compared and analyzed to obtain a comparison result; if the comparison result is that the predicted node and the end node are the same, then the user is evaluated based on the predicted node and the end node. Doctor matching, get the target doctor.
  • the computer-readable storage medium may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, at least one application program required for a function, etc.; the storage data area may store information based on the blockchain node. Use the created data, etc.
  • Blockchain is a new application model of computer technology such as distributed data storage, point-to-point transmission, consensus mechanism, and encryption algorithm.
  • Blockchain is essentially a decentralized database. It is a series of data blocks generated using cryptographic methods. Each data block contains a batch of network transaction information and is used to verify its Validity of information (anti-counterfeiting) and generation of the next block.
  • Blockchain can include the underlying platform of the blockchain, the platform product service layer, and the application service layer.
  • the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it may be stored in a computer-readable storage medium.
  • the technical solution of the present application is essentially or contributes to the existing technology, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium , including several instructions to cause a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of this application.
  • the aforementioned storage media include: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), magnetic disk or optical disk and other media that can store program code. .

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • Public Health (AREA)
  • Biomedical Technology (AREA)
  • General Health & Medical Sciences (AREA)
  • Primary Health Care (AREA)
  • Epidemiology (AREA)
  • Pathology (AREA)
  • General Business, Economics & Management (AREA)
  • Data Mining & Analysis (AREA)
  • Theoretical Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Physics & Mathematics (AREA)
  • Computational Linguistics (AREA)
  • Artificial Intelligence (AREA)
  • Biophysics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Evolutionary Computation (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Databases & Information Systems (AREA)
  • Medical Treatment And Welfare Office Work (AREA)

Abstract

本申请涉及人工智能领域，公开了一种医疗数据处理方法、装置、设备及存储介质，用于提高医患匹配的准确率。所述医疗数据处理方法包括：根据医疗数据判断用户是否符合用户类型；若用户符合用户类型，则根据医疗数据匹配与用户对应的疾病模板；基于疾病诊断路径对疾病模板进行信息填充，得到结束节点和用户回答路径，并根据结束节点和用户回答路径生成用户需求；若用户需求为医生问诊，则将医疗数据和用户回答路径输入医疗数据处理模型进行疾病数据处理，得到预测节点；对预测节点和结束节点进行比对分析，若比对结果为相同，则对用户进行医生匹配，得到目标医生。此外，本申请还涉及区块链技术，用户需求可存储于区块链节点中。

Description

医疗数据处理方法、装置、设备及存储介质
本申请要求于2022年3月24日提交中国专利局、申请号为202210296654.8、发明名称为“医疗数据处理方法、装置、设备及存储介质”的中国专利申请的优先权，其全部内容通过引用结合在申请中。
技术领域
本申请涉及人工智能领域，尤其涉及一种医疗数据处理方法、装置、设备及存储介质。
背景技术
随着医疗互联网和信息化的发展，医疗资源不平衡的问题有所缓解，网上预约挂号、远程医生咨询、处方推荐和开设及购药等服务，给患者提供了便利性，也帮助消化了一部分线下医院的繁重业务。同时，互联网平台或信息系统中都积累了大量的医生信息、患者信息、对话数据，形成了宝贵的互联网医疗大数据，特别是每个患者的历史活动记录(例如问诊、挂号、购药等)，对于该患者未来的疾病诊断和医疗需求识别有极大的作用。
发明人意识到，在线问诊的普遍做法会首先通过患者主诉和基本信息为其匹配科室和医生，通常利用一些分类模型对患者进行分类，目前患者和医生匹配的满意度还有待提升，主要是因为用户信息少，去匹配量大、来源多、数据结构不同的医院医生数据，很难做到精确，即现有方案准确率低。
发明内容
本申请提供了一种医疗数据处理方法、装置、设备及存储介质，用于提高医患匹配的准确率。
为实现上述目的，本申请第一方面提供了一种医疗数据处理方法，包括：获取待处理的用户对应的医疗数据，并根据所述医疗数据判断所述用户是否符合预置的用户类型，其中，所述医疗数据包括用户性别、用户年龄和症状数据；若所述用户符合预置的用户类型，则根据所述医疗数据匹配与所述用户对应的疾病模板，其中，所述疾病模板包括用户待填写的疾病诊断路径；基于所述疾病诊断路径对所述疾病模板进行信息填充，得到所述疾病模板对应的结束节点和用户回答路径，并根据所述结束节点和所述用户回答路径生成用户需求；若所述用户需求为医生问诊，则将所述医疗数据和所述用户回答路径输入预置的医疗数据处理模型进行疾病数据处理，得到预测节点；对所述预测节点和所述结束节点进行比对分析，得到比对结果；若所述比对结果为所述预测节点和所述结束节点相同，则根据所述预测节点和所述结束节点对所述用户进行医生匹配，得到目标医生。
本申请第二方面提供了一种医疗数据处理设备，包括：存储器和至少一个处理器，所述存储器中存储有指令；所述至少一个处理器调用所述存储器中的所述指令，以使得所述医疗数据处理设备执行如下步骤：获取待处理的用户对应的医疗数据，并根据所述医疗数据判断所述用户是否符合预置的用户类型，其中，所述医疗数据包括用户性别、用户年龄和症状数据；若所述用户符合预置的用户类型，则根据所述医疗数据匹配与所述用户对应的疾病模板，其中，所述疾病模板包括用户待填写的疾病诊断路径；基于所述疾病诊断路径对所述疾病模板进行信息填充，得到所述疾病模板对应的结束节点和用户回答路径，并根据所述结束节点和所述用户回答路径生成用户需求；若所述用户需求为医生问诊，则将所述医疗数据和所述用户回答路径输入预置的医疗数据处理模型进行疾病数据处理，得到预测节点；对所述预测节点和所述结束节点进行比对分析，得到比对结果；若所述比对结果为所述预测节点和所述结束节点相同，则根据所述预测节点和所述结束节点对所述用户进行医生匹配，得到目标医生。
本申请第三方面提供了一种计算机可读存储介质，所述计算机可读存储介质上存储有指令，其中，所述指令被处理器执行时实现如下步骤：获取待处理的用户对应的医疗数据，并根据所述医疗数据判断所述用户是否符合预置的用户类型，其中，所述医疗数据包括用 户性别、用户年龄和症状数据；若所述用户符合预置的用户类型，则根据所述医疗数据匹配与所述用户对应的疾病模板，其中，所述疾病模板包括用户待填写的疾病诊断路径；基于所述疾病诊断路径对所述疾病模板进行信息填充，得到所述疾病模板对应的结束节点和用户回答路径，并根据所述结束节点和所述用户回答路径生成用户需求；若所述用户需求为医生问诊，则将所述医疗数据和所述用户回答路径输入预置的医疗数据处理模型进行疾病数据处理，得到预测节点；对所述预测节点和所述结束节点进行比对分析，得到比对结果；若所述比对结果为所述预测节点和所述结束节点相同，则根据所述预测节点和所述结束节点对所述用户进行医生匹配，得到目标医生。
本申请第四方面提供了一种医疗数据处理装置，其中，所述医疗数据处理装置包括：获取模块，用于获取待处理的用户对应的医疗数据，并根据所述医疗数据判断所述用户是否符合预置的用户类型，其中，所述医疗数据包括用户性别、用户年龄和症状数据；匹配模块，用于若所述用户符合预置的用户类型，则根据所述医疗数据匹配与所述用户对应的疾病模板，其中，所述疾病模板包括用户待填写的疾病诊断路径；填充模块，用于基于所述疾病诊断路径对所述疾病模板进行信息填充，得到所述疾病模板对应的结束节点和用户回答路径，并根据所述结束节点和所述用户回答路径生成用户需求；处理模块，用于若所述用户需求为医生问诊，则将所述医疗数据和所述用户回答路径输入预置的医疗数据处理模型进行疾病数据处理，得到预测节点；分析模块，用于对所述预测节点和所述结束节点进行比对分析，得到比对结果；生成模块，用于若所述比对结果为所述预测节点和所述结束节点相同，则根据所述预测节点和所述结束节点对所述用户进行医生匹配，得到目标医生。
本申请提供的技术方案中，根据医疗数据判断用户是否符合预置的用户类型，其中，医疗数据包括用户性别、用户年龄和症状数据；若用户符合预置的用户类型，则根据医疗数据匹配与用户对应的疾病模板，其中，疾病模板包括用户待填写的疾病诊断路径；基于疾病诊断路径对疾病模板进行信息填充，得到疾病模板对应的结束节点和用户回答路径，并根据结束节点和用户回答路径生成用户需求；若用户需求为医生问诊，则将医疗数据和用户回答路径输入预置的医疗数据处理模型进行疾病数据处理，得到预测节点；对预测节点和结束节点进行比对分析，得到比对结果；若比对结果为预测节点和结束节点相同，则根据预测节点和结束节点对用户进行医生匹配，得到目标医生，本申请通过对用户进行自诊得到初步诊断结果和其他医疗信息，从而能够利用这部分更全面的信息进行更精确地医患匹配，避免了一些无效沟通或错误匹配，根据疾病诊断路径模版的方式进行用户自诊和医疗需求识别，提高了在线问诊流程的效率。
附图说明
图1为本申请实施例中医疗数据处理方法的一个实施例示意图；
图2为本申请实施例中医疗数据处理方法的另一个实施例示意图；
图3为本申请实施例中医疗数据处理装置的一个实施例示意图；
图4为本申请实施例中医疗数据处理装置的另一个实施例示意图；
图5为本申请实施例中医疗数据处理设备的一个实施例示意图。
具体实施方式
本申请提供了一种医疗数据处理方法、装置、设备及存储介质，用于提高医患匹配的准确率。
为了使本技术领域的人员更好地理解本申请方案，下面将结合本申请实施例中的附图，对本申请实施例进行描述。
本申请的说明书和权利要求书及上述附图中的术语“第一”、“第二”、“第三”、“第四”等(如果存在)是用于区别类似的对象，而不必用于描述特定的顺序或先后次序。应该理 解这样使用的数据在适当情况下可以互换，以便这里描述的实施例能够以除了在这里图示或描述的内容以外的顺序实施。此外，术语“包括”或“具有”及其任何变形，意图在于覆盖不排他的包含，例如，包含了一系列步骤或单元的过程、方法、系统、产品或设备不必限于清楚地列出的那些步骤或单元，而是可包括没有清楚地列出的或对于这些过程、方法、产品或设备固有的其它步骤或单元。
为便于理解，下面对本申请实施例的具体流程进行描述，请参阅图1，本申请实施例中医疗数据处理方法的一个实施例包括：
101、获取待处理的用户对应的医疗数据，并根据医疗数据判断用户是否符合预置的用户类型，其中，医疗数据包括用户性别、用户年龄和症状数据；
需要说明的是，待处理用户输入的医疗数据包括个人性别、年龄和症状数据。用户类型包括不清楚自己所患疾病、问诊需求不明确、或主动选择了智能疾病诊断，服务器根据用户输入的医疗数据对用户类型进行识别，并判断待处理用户是否符合上述用户类型。需要说明的是，为了保证症状数据的真实性，症状数据的获取可以为医疗网站、医疗机构数据库等。症状数据是包括疾病词汇和症状词汇的文本数据，症状数据可以通过预置的爬虫从医疗网站或者医疗机构数据中获取，其中，医疗网站或者医疗机构数据记录用户的疾病词汇和症状词汇生成症状数据，也可以由用户直接输入疾病词汇和症状词汇从而得到症状数据。
可以理解的是，本申请的执行主体可以为医疗数据处理装置，还可以是终端或者服务器，具体此处不做限定。本申请实施例以服务器为执行主体为例进行说明。本申请实施例可以基于人工智能技术对相关的数据进行获取和处理。其中，人工智能(Artificial Intelligence，AI)是利用数字计算机或者数字计算机控制的机器模拟、延伸和扩展人的智能，感知环境、获取知识并使用知识获得最佳结果的理论、方法、技术及应用系统。人工智能基础技术一般包括如传感器、专用人工智能芯片、云计算、分布式存储、大数据处理技术、操作/交互系统、机电一体化等技术。人工智能软件技术主要包括计算机视觉技术、机器人技术、生物识别技术、语音处理技术、自然语言处理技术以及机器学习/深度学习等几大方向。服务器可以是独立的服务器，也可以是提供云服务、云数据库、云计算、云函数、云存储、网络服务、云通信、中间件服务、域名服务、安全服务、内容分发网络(Content Delivery Network，CDN)、以及大数据和人工智能平台等基础云计算服务的云服务器。
102、若用户符合预置的用户类型，则根据医疗数据匹配与用户对应的疾病模板，其中，疾病模板包括用户待填写的疾病诊断路径；
具体的，如果用户符合预置的用户类型则服务器进入自诊模块，服务器首先根据前面的信息匹配疾病模版，疾病模版包含了疾病诊断路径，类似树图，包含了多个问题选项，不同的回答可能导向不同的分支。通过用户回答模版中的问题，系统将填充模版节点并抛出新的问题，直到走到了诊断节点或结束节点。需要说明的是，疾病模版通过计算机辅助进行医疗文献挖掘加上医生编辑和审核得到。
103、基于疾病诊断路径对疾病模板进行信息填充，得到疾病模板对应的结束节点和用户回答路径，并根据结束节点和用户回答路径生成用户需求；
需要说明的是，用户填充模版中的问题，服务器将填充模版节点并抛出新的问题，直到填充完所有的诊断节点或结束节点。服务器根据最终节点和中间的用户回答路径，通过深度学习网络模型预测一个用户需求，包括医生问诊、找医院、买药、体检、挂号、慢病管理、综合咨询等，服务器还会根据用户需求推荐相关功能，引导用户进入下一服务模块。
104、若用户需求为医生问诊，则将医疗数据和用户回答路径输入预置的医疗数据处理模型进行疾病数据处理，得到预测节点；
具体的，若用户需求为医生问诊，则服务器将医疗数据和用户回答路径输入预置的医 疗数据处理模型进行疾病数据处理，其中，预置的医疗数据处理模型包括输入层、多层卷积神经网络和归一化层，其中，多层卷积神经网络用于对输入向量进行卷积运算，归一化层用于对特征向量进行逻辑回归运算，最终输出预测节点。
105、对预测节点和结束节点进行比对分析，得到比对结果；
具体的，服务器对预测节点和结束节点进行比对分析，得到比对结果，其中，服务器通过计算预测节点和结束节点的相似度，服务器当该相似度超过预置阈值时，确定该预测节点和结束节点相同，服务器当该相似度不超过该预置阈值时，确定该预测节点与该结束节点不相同。
106、若比对结果为预测节点和结束节点相同，则根据预测节点和结束节点对用户进行医生匹配，得到目标医生。
具体的，若比对结果为预测节点和结束节点相同，则服务器计算预测节点和结束节点的可信度，得到目标可信度，服务器对目标可信度的大小进行排序，并选取目标可信度对应候选医生的最大值，得到候选医生的最大值，服务器将最大值对应的候选医生作为目标医生。
进一步地，服务器将用户需求存储于区块链数据库中，具体此处不做限定。
本申请实施例中，根据医疗数据判断用户是否符合预置的用户类型，其中，医疗数据包括用户性别、用户年龄和症状数据；若用户符合预置的用户类型，则根据医疗数据匹配与用户对应的疾病模板，其中，疾病模板包括用户待填写的疾病诊断路径；基于疾病诊断路径对疾病模板进行信息填充，得到疾病模板对应的结束节点和用户回答路径，并根据结束节点和用户回答路径生成用户需求；若用户需求为医生问诊，则将医疗数据和用户回答路径输入预置的医疗数据处理模型进行疾病数据处理，得到预测节点；对预测节点和结束节点进行比对分析，得到比对结果；若比对结果为预测节点和结束节点相同，则根据预测节点和结束节点对用户进行医生匹配，得到目标医生，本申请通过对用户进行自诊得到初步诊断结果和其他医疗信息，从而能够利用这部分更全面的信息进行更精确地医患匹配，避免了一些无效沟通或错误匹配，根据疾病诊断路径模版的方式进行用户自诊和医疗需求识别，提高了在线问诊流程的效率。
请参阅图2，本申请实施例中医疗数据处理方法的另一个实施例包括：
201、获取待处理的用户对应的医疗数据，并根据医疗数据判断用户是否符合预置的用户类型，其中，医疗数据包括用户性别、用户年龄和症状数据；
具体的，服务器接收待处理的用户输入的医疗数据，其中，医疗数据包括用户性别、用户年龄和症状数据；服务器基于用户性别、用户年龄和症状数据生成用户对应的用户画像；服务器计算用户画像是否符合预置的用户类型对应的相似度，并对相似度和预设目标值进行比对；服务器若相似度大于或等于预设目标值，则确定用户符合预置的用户类型。具体的，服务器若相似度小于预设目标值，则确定用户不符合预置的用户类型，服务器接收待处理的用户输入的医疗数据，用户可以通过预置的终端输入医疗数据，医疗数据中包含用户的个人信息和症状数据；服务器基于用户性别、用户年龄和症状数据生成用户对应的用户画像，服务器生成用户对应的三元组，三元组中包括性别、年龄和症状；服务器计算用户画像是否符合预置的用户类型对应的相似度，并对相似度和预设目标值进行比对；服务器若相似度大于或等于预设目标值，则确定用户符合预置的用户类型，若相似度下雨预设目标值，则服务器确定用户不符合上述三类用户。
202、若用户符合预置的用户类型，则根据医疗数据匹配与用户对应的疾病模板，其中，疾病模板包括用户待填写的疾病诊断路径；
具体的，服务器若用户符合预置的用户类型，则对医疗数据进行关键词提取，得到目 标关键词；服务器分别将目标关键词和预置的多个疾病模板进行匹配，得到每个疾病模板对应的关键词命中率；服务器对每个疾病模板对应的关键词命中率进行排序，并将关键词命中率最高的疾病模板作为用户对应的疾病模板，其中，疾病模板包括用户待填写的疾病诊断路径。具体的，若用户符合预置的用户类型，则服务器对医疗数据进行关键词提取，得到目标关键词，服务器进行关键词提取首先是将医疗数据转换为文本数据，再通过预置的OCR文字识别模型对文本数据进行提取，得到标准文本数据，最后通过对标准文本数据进行重复内容去除，得到目标关键词；服务器分别将目标关键词和预置的多个疾病模板进行匹配，得到每个疾病模板对应的关键词命中率，关键词命中率也就是关键词命中的个数；服务器对每个疾病模板对应的关键词命中率进行排序，并将关键词命中率最高的疾病模板作为用户对应的疾病模板，其中，疾病模板包括用户待填写的疾病诊断路径。
203、基于疾病诊断路径对疾病模板进行信息填充，得到疾病模板对应的结束节点和用户回答路径，并根据结束节点和用户回答路径生成用户需求；
具体的，服务器基于疾病诊断路径获取跑通率和推荐点击率；服务器基于跑通率和推荐点击率对疾病模板进行信息填充，得到填充后的疾病模板；服务器对填充后的疾病模板进行数据提取，得到疾病模板对应的结束节点和用户回答路径；服务器基于结束节点和用户回答路径生成用户需求，其中，用户需求包括医生问诊和线下问诊。具体的，服务器基于疾病诊断路径获取跑通率和推荐点击率，其中，服务器通过线上业务数据统计分析得到的跑通率和推荐点击率等指标去优化模版，跑通率指用户是否能走到模版的最终节点，服务器基于跑通率和推荐点击率对疾病模板进行信息填充，得到填充后的疾病模板；服务器对填充后的疾病模板进行数据提取，得到疾病模板对应的结束节点和用户回答路径；服务器基于结束节点和用户回答路径生成用户需求，其中，用户需求包括医生问诊和线下问诊。其中，一个模版可作为别的模版的节点，即子模版。模版之间可能会出现包含关系，例如眼科通用模版可能包含若干眼科疾病模版作为子模版，模版节点的权重会根据用户历史医疗记录如问诊信息进行调整。
204、若用户需求为医生问诊，则将医疗数据和用户回答路径输入预置的医疗数据处理模型进行疾病数据处理，得到预测节点；
具体的，服务器若用户需求为医生问诊，则将医疗数据和用户回答路径输入预置的医疗数据处理模型；服务器通过医疗数据处理模型中的卷积神经网络对医疗数据和用户回答路径进行卷积运算，得到医疗数据对应的特征向量；服务器对特征向量进行特征归一化处理，得到预测节点。需要说明的是，预置的医疗数据处理模型包括输入层、卷积神经网络和输出层，输入层：独热向量编码层(one-hot vector)；隐藏层：卷积运算函数，也就是线性的单元；输出层：维度跟输入层的维度一样，用的是逻辑回归。服务器通过输入层对目标填报信息进行独热向量编码，得到低维度向量，低维度向量例如：[0,0,0,1,0,1,0,0]，服务器通过卷积神经网络层对低维度向量进行特征抽象运算，得到抽象特征值，抽象特征值也就是特征向量；服务器通过输出层对抽象特征值进行逻辑回归运算，得到预测节点，其中，逻辑回归运算为softmax回归运算，其中，预测节点可以为线上初诊结果，例如：线上初诊眼睑炎、结膜炎、呼吸道感染等。
205、对预测节点和结束节点进行比对分析，得到比对结果；
具体的，服务器对预测节点和结束节点进行比对分析，得到比对结果，其中，服务器通过计算预测节点和结束节点的相似度，服务器当该相似度超过预置阈值时，确定该预测节点和结束节点相同，服务器当该相似度不超过该预置阈值时，确定该预测节点与该结束节点不相同。
206、若比对结果为预测节点和结束节点相同，则根据预测节点和结束节点对用户进行医生匹配，得到目标医生；
具体的，服务器若比对结果为预测节点和结束节点相同，则获取预置候选医生集合中的多个候选医生；服务器基于预测节点和结束节点对多个候选医生进行可信度计算，得到每个候选医生对应的目标可信度；服务器根据每个候选医生对应的目标可信度对候选医生集合中的多个候选医生进行排名，并将排名最高的候选医生作为与用户匹配的医生，得到目标医生。具体的，服务器基于目标可信度对预置的候选医生列表进行排序，得到排名，服务器基于用户并考虑用户偏好和候选医生工作量，服务器定义了用户的偏好程度来衡量用户对候选医生的偏好指数，服务器通过医患匹配指数加权平均的方法，得到用户对应的偏好指数，服务器计算完所有候选医生中的医生对用户的匹配程度后，根据最高目标可信度和偏好指数推荐目标医生。服务器为用户匹配到相关度较高的目标医生，提高用户的问诊体验，此外某些有特定专长的目标医生也可以匹配到相关的用户，进而可以避免医疗资源的浪费。
207、若比对结果为预测节点和结束节点不相同，则对预测节点进行删除，并基于疾病模板生成用户对应的辅助信息；
具体的，若比对结果为预测节点和结束节点相同，则服务器计算预测节点和结束节点的可信度，得到目标可信度，服务器对目标可信度的大小进行排序，并选取目标可信度对应候选医生的最大值，得到候选医生的最大值，服务器将最大值对应的候选医生作为目标医生。
208、基于辅助信息和结束节点对用户进行医生匹配，得到用户对应的目标医生。
具体的，服务器如果结果不符合，匹配医生模型会去掉诊断结果信息，利用科室和主诉及部分节点的信息去匹配医生。用户回答过的模版问题也会被传输给医生作为辅助信息。服务器基于辅助信息和结束节点对用户进行医生匹配，得到用户对应的目标医生。
进一步地，服务器将用户需求存储于区块链数据库中，具体此处不做限定。
本申请实施例中，根据医疗数据判断用户是否符合预置的用户类型，其中，医疗数据包括用户性别、用户年龄和症状数据；若用户符合预置的用户类型，则根据医疗数据匹配与用户对应的疾病模板，其中，疾病模板包括用户待填写的疾病诊断路径；基于疾病诊断路径对疾病模板进行信息填充，得到疾病模板对应的结束节点和用户回答路径，并根据结束节点和用户回答路径生成用户需求；若用户需求为医生问诊，则将医疗数据和用户回答路径输入预置的医疗数据处理模型进行疾病数据处理，得到预测节点；对预测节点和结束节点进行比对分析，得到比对结果；若比对结果为预测节点和结束节点相同，则根据预测节点和结束节点对用户进行医生匹配，得到目标医生，本申请通过对用户进行自诊得到初步诊断结果和其他医疗信息，从而能够利用这部分更全面的信息进行更精确地医患匹配，避免了一些无效沟通或错误匹配，根据疾病诊断路径模版的方式进行用户自诊和医疗需求识别，提高了在线问诊流程的效率。
上面对本申请实施例中医疗数据处理方法进行了描述，下面对本申请实施例中医疗数据处理装置进行描述，请参阅图3，本申请实施例中医疗数据处理装置一个实施例包括：
获取模块301，用于获取待处理的用户对应的医疗数据，并根据所述医疗数据判断所述用户是否符合预置的用户类型，其中，所述医疗数据包括用户性别、用户年龄和症状数据；
匹配模块302，用于若所述用户符合预置的用户类型，则根据所述医疗数据匹配与所述用户对应的疾病模板，其中，所述疾病模板包括用户待填写的疾病诊断路径；
填充模块303，用于基于所述疾病诊断路径对所述疾病模板进行信息填充，得到所述疾病模板对应的结束节点和用户回答路径，并根据所述结束节点和所述用户回答路径生成用户需求；
处理模块304，用于若所述用户需求为医生问诊，则将所述医疗数据和所述用户回答 路径输入预置的医疗数据处理模型进行疾病数据处理，得到预测节点；
分析模块305，用于对所述预测节点和所述结束节点进行比对分析，得到比对结果；
生成模块306，用于若所述比对结果为所述预测节点和所述结束节点相同，则根据所述预测节点和所述结束节点对所述用户进行医生匹配，得到目标医生。
进一步地，服务器将用户需求存储于区块链数据库中，具体此处不做限定。
本申请实施例中，根据医疗数据判断用户是否符合预置的用户类型，其中，医疗数据包括用户性别、用户年龄和症状数据；若用户符合预置的用户类型，则根据医疗数据匹配与用户对应的疾病模板，其中，疾病模板包括用户待填写的疾病诊断路径；基于疾病诊断路径对疾病模板进行信息填充，得到疾病模板对应的结束节点和用户回答路径，并根据结束节点和用户回答路径生成用户需求；若用户需求为医生问诊，则将医疗数据和用户回答路径输入预置的医疗数据处理模型进行疾病数据处理，得到预测节点；对预测节点和结束节点进行比对分析，得到比对结果；若比对结果为预测节点和结束节点相同，则根据预测节点和结束节点对用户进行医生匹配，得到目标医生，本申请通过对用户进行自诊得到初步诊断结果和其他医疗信息，从而能够利用这部分更全面的信息进行更精确地医患匹配，避免了一些无效沟通或错误匹配，根据疾病诊断路径模版的方式进行用户自诊和医疗需求识别，提高了在线问诊流程的效率。
请参阅图4，本申请实施例中医疗数据处理装置另一个实施例包括：
获取模块301，用于获取待处理的用户对应的医疗数据，并根据所述医疗数据判断所述用户是否符合预置的用户类型，其中，所述医疗数据包括用户性别、用户年龄和症状数据；
匹配模块302，用于若所述用户符合预置的用户类型，则根据所述医疗数据匹配与所述用户对应的疾病模板，其中，所述疾病模板包括用户待填写的疾病诊断路径；
填充模块303，用于基于所述疾病诊断路径对所述疾病模板进行信息填充，得到所述疾病模板对应的结束节点和用户回答路径，并根据所述结束节点和所述用户回答路径生成用户需求；
处理模块304，用于若所述用户需求为医生问诊，则将所述医疗数据和所述用户回答路径输入预置的医疗数据处理模型进行疾病数据处理，得到预测节点；
分析模块305，用于对所述预测节点和所述结束节点进行比对分析，得到比对结果；
生成模块306，用于若所述比对结果为所述预测节点和所述结束节点相同，则根据所述预测节点和所述结束节点对所述用户进行医生匹配，得到目标医生。
可选的，获取模块301具体用于：
接收待处理的用户输入的医疗数据，其中，所述医疗数据包括用户性别、用户年龄和症状数据；基于所述用户性别、所述用户年龄和所述症状数据生成所述用户对应的用户画像；计算所述用户画像是否符合预置的用户类型对应的相似度，并对所述相似度和预设目标值进行比对；若所述相似度大于或等于所述预设目标值，则确定所述用户符合预置的用户类型。
可选的，匹配模块302具体用于：
若所述用户符合预置的用户类型，则对所述医疗数据进行关键词提取，得到目标关键词；分别将所述目标关键词和预置的多个疾病模板进行匹配，得到每个疾病模板对应的关键词命中率；对每个疾病模板对应的关键词命中率进行排序，并将所述关键词命中率最高的疾病模板作为所述用户对应的疾病模板，其中，所述疾病模板包括用户待填写的疾病诊断路径。
可选的，填充模块303具体用于：
基于所述疾病诊断路径获取跑通率和推荐点击率；基于所述跑通率和所述推荐点击率 对所述疾病模板进行信息填充，得到填充后的疾病模板；对所述填充后的疾病模板进行数据提取，得到所述疾病模板对应的结束节点和用户回答路径；基于所述结束节点和所述用户回答路径生成用户需求，其中，所述用户需求包括医生问诊和线下问诊。
可选的，处理模块304具体用于：
若所述用户需求为医生问诊，则将所述医疗数据和所述用户回答路径输入预置的医疗数据处理模型；通过所述医疗数据处理模型中的卷积神经网络对所述医疗数据和所述用户回答路径进行卷积运算，得到所述医疗数据对应的特征向量；对所述特征向量进行特征归一化处理，得到预测节点。
可选的，生成模块306具体用于：
若所述比对结果为所述预测节点和所述结束节点相同，则获取预置候选医生集合中的多个候选医生；基于所述预测节点和所述结束节点对所述多个候选医生进行可信度计算，得到每个候选医生对应的目标可信度；根据每个候选医生对应的目标可信度对所述候选医生集合中的所述多个候选医生进行排名，并将排名最高的候选医生作为与所述用户匹配的医生，得到目标医生。
可选的，医疗数据处理装置还包括：
删除模块307，用于若所述比对结果为所述预测节点和所述结束节点不相同，则对所述预测节点进行删除，并基于所述疾病模板生成所述用户对应的辅助信息；基于所述辅助信息和所述结束节点对所述用户进行医生匹配，得到所述用户对应的目标医生。
进一步地，服务器将用户需求存储于区块链数据库中，具体此处不做限定。
本申请实施例中，根据医疗数据判断用户是否符合预置的用户类型，其中，医疗数据包括用户性别、用户年龄和症状数据；若用户符合预置的用户类型，则根据医疗数据匹配与用户对应的疾病模板，其中，疾病模板包括用户待填写的疾病诊断路径；基于疾病诊断路径对疾病模板进行信息填充，得到疾病模板对应的结束节点和用户回答路径，并根据结束节点和用户回答路径生成用户需求；若用户需求为医生问诊，则将医疗数据和用户回答路径输入预置的医疗数据处理模型进行疾病数据处理，得到预测节点；对预测节点和结束节点进行比对分析，得到比对结果；若比对结果为预测节点和结束节点相同，则根据预测节点和结束节点对用户进行医生匹配，得到目标医生，本申请通过对用户进行自诊得到初步诊断结果和其他医疗信息，从而能够利用这部分更全面的信息进行更精确地医患匹配，避免了一些无效沟通或错误匹配，根据疾病诊断路径模版的方式进行用户自诊和医疗需求识别，提高了在线问诊流程的效率。
上面图3和图4从模块化功能实体的角度对本申请实施例中的医疗数据处理装置进行详细描述，下面从硬件处理的角度对本申请实施例中医疗数据处理设备进行详细描述。
图5是本申请实施例提供的一种医疗数据处理设备的结构示意图，该医疗数据处理设备500可因配置或性能不同而产生比较大的差异，可以包括一个或一个以上处理器(central processing units，CPU)510(例如，一个或一个以上处理器)，一个或一个以上存储应用程序533或数据532的存储介质530(例如一个或一个以上海量存储设备)。其中，存储介质530可以是短暂存储或持久存储。存储在存储介质530的程序可以包括一个或一个以上模块(图示没标出)，每个模块可以包括对医疗数据处理设备500中的一系列指令操作。更进一步地，处理器510可以设置为与存储介质530通信，在医疗数据处理设备500上执行存储介质530中的一系列指令操作。
医疗数据处理设备500还可以包括一个或一个以上电源540，一个或一个以上有线或无线网络接口550，一个或一个以上输入输出接口520，和/或，一个或一个以上操作系统531，例如Windows Serve，Mac OS X，Unix，Linux，FreeBSD等等。本领域技术人员可以理解，图5示出的医疗数据处理设备结构并不构成对医疗数据处理设备的限定，可以包括 比图示更多或更少的部件，或者组合某些部件，或者不同的部件布置。
本申请还提供一种医疗数据处理设备，所述医疗数据处理设备包括处理器，计算机可读指令被处理器执行时，使得处理器执行上述各实施例中的所述医疗数据处理方法的步骤。
本申请还提供一种计算机可读存储介质，该计算机可读存储介质可以为非易失性计算机可读存储介质，该计算机可读存储介质也可以为易失性计算机可读存储介质，所述计算机可读存储介质中存储有指令，当所述指令在计算机上运行时，使得计算机执行如下步骤：获取待处理的用户对应的医疗数据，并根据所述医疗数据判断所述用户是否符合预置的用户类型，其中，所述医疗数据包括用户性别、用户年龄和症状数据；若所述用户符合预置的用户类型，则根据所述医疗数据匹配与所述用户对应的疾病模板，其中，所述疾病模板包括用户待填写的疾病诊断路径；基于所述疾病诊断路径对所述疾病模板进行信息填充，得到所述疾病模板对应的结束节点和用户回答路径，并根据所述结束节点和所述用户回答路径生成用户需求；若所述用户需求为医生问诊，则将所述医疗数据和所述用户回答路径输入预置的医疗数据处理模型进行疾病数据处理，得到预测节点；对所述预测节点和所述结束节点进行比对分析，得到比对结果；若所述比对结果为所述预测节点和所述结束节点相同，则根据所述预测节点和所述结束节点对所述用户进行医生匹配，得到目标医生。
进一步地，计算机可读存储介质可主要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、至少一个功能所需的应用程序等；存储数据区可存储根据区块链节点的使用所创建的数据等。
本申请所指区块链是分布式数据存储、点对点传输、共识机制、加密算法等计算机技术的新型应用模式。区块链(Blockchain)，本质上是一个去中心化的数据库，是一串使用密码学方法相关联产生的数据块，每一个数据块中包含了一批次网络交易的信息，用于验证其信息的有效性(防伪)和生成下一个区块。区块链可以包括区块链底层平台、平台产品服务层以及应用服务层等。
所属领域的技术人员可以清楚地了解到，为描述的方便和简洁，上述描述的系统，装置和单元的具体工作过程，可以参考前述方法实施例中的对应过程，在此不再赘述。
所述集成的单元如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储介质中。基于这样的理解，本申请的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的全部或部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)执行本申请各个实施例所述方法的全部或部分步骤。而前述的存储介质包括：U盘、移动硬盘、只读存储器(read-only memory，ROM)、随机存取存储器(random access memory，RAM)、磁碟或者光盘等各种可以存储程序代码的介质。
以上所述，以上实施例仅用以说明本申请的技术方案，而非对其限制；尽管参照前述实施例对本申请进行了详细的说明，本领域的普通技术人员应当理解：其依然可以对前述各实施例所记载的技术方案进行修改，或者对其中部分技术特征进行等同替换；而这些修改或者替换，并不使相应技术方案的本质脱离本申请各实施例技术方案的精神和范围。

Claims (20)

  1. 一种医疗数据处理方法，其中，所述医疗数据处理方法包括：
    获取待处理的用户对应的医疗数据，并根据所述医疗数据判断所述用户是否符合预置的用户类型，其中，所述医疗数据包括用户性别、用户年龄和症状数据；
    若所述用户符合预置的用户类型，则根据所述医疗数据匹配与所述用户对应的疾病模板，其中，所述疾病模板包括用户待填写的疾病诊断路径；
    基于所述疾病诊断路径对所述疾病模板进行信息填充，得到所述疾病模板对应的结束节点和用户回答路径，并根据所述结束节点和所述用户回答路径生成用户需求；
    若所述用户需求为医生问诊，则将所述医疗数据和所述用户回答路径输入预置的医疗数据处理模型进行疾病数据处理，得到预测节点；
    对所述预测节点和所述结束节点进行比对分析，得到比对结果；
    若所述比对结果为所述预测节点和所述结束节点相同，则根据所述预测节点和所述结束节点对所述用户进行医生匹配，得到目标医生。
  2. 根据权利要求1所述的医疗数据处理方法，其中，所述获取待处理的用户对应的医疗数据，并根据所述医疗数据判断所述用户是否符合预置的用户类型，包括：
    接收待处理的用户输入的医疗数据，其中，所述医疗数据包括用户性别、用户年龄和症状数据；
    基于所述用户性别、所述用户年龄和所述症状数据生成所述用户对应的用户画像；
    计算所述用户画像是否符合预置的用户类型对应的相似度，并对所述相似度和预设目标值进行比对；
    若所述相似度大于或等于所述预设目标值，则确定所述用户符合预置的用户类型。
  3. 根据权利要求1所述的医疗数据处理方法，其中，所述若所述用户符合预置的用户类型，则根据所述医疗数据匹配与所述用户对应的疾病模板，包括：
    若所述用户符合预置的用户类型，则对所述医疗数据进行关键词提取，得到目标关键词；
    分别将所述目标关键词和预置的多个疾病模板进行匹配，得到每个疾病模板对应的关键词命中率；
    对每个疾病模板对应的关键词命中率进行排序，并将所述关键词命中率最高的疾病模板作为所述用户对应的疾病模板，其中，所述疾病模板包括用户待填写的疾病诊断路径。
  4. 根据权利要求1所述的医疗数据处理方法，其中，所述基于所述疾病诊断路径对所述疾病模板进行信息填充，得到所述疾病模板对应的结束节点和用户回答路径，并根据所述结束节点和所述用户回答路径生成用户需求，包括：
    基于所述疾病诊断路径获取跑通率和推荐点击率；
    基于所述跑通率和所述推荐点击率对所述疾病模板进行信息填充，得到填充后的疾病模板；
    对所述填充后的疾病模板进行数据提取，得到所述疾病模板对应的结束节点和用户回答路径；
    基于所述结束节点和所述用户回答路径生成用户需求，其中，所述用户需求包括医生问诊和线下问诊。
  5. 根据权利要求1所述的医疗数据处理方法，其中，所述若所述用户需求为医生问诊，则将所述医疗数据和所述用户回答路径输入预置的医疗数据处理模型进行疾病数据处理，得到预测节点，包括：
    若所述用户需求为医生问诊，则将所述医疗数据和所述用户回答路径输入预置的医疗数据处理模型；
    通过所述医疗数据处理模型中的卷积神经网络对所述医疗数据和所述用户回答路径进行卷积运算，得到所述医疗数据对应的特征向量；
    对所述特征向量进行特征归一化处理，得到预测节点。
  6. 根据权利要求1所述的医疗数据处理方法，其中，所述若所述比对结果为所述预测节点和所述结束节点相同，则根据所述预测节点和所述结束节点对所述用户进行医生匹配，得到目标医生，包括：
    若所述比对结果为所述预测节点和所述结束节点相同，则获取预置候选医生集合中的多个候选医生；
    基于所述预测节点和所述结束节点对所述多个候选医生进行可信度计算，得到每个候选医生对应的目标可信度；
    根据每个候选医生对应的目标可信度对所述候选医生集合中的所述多个候选医生进行排名，并将排名最高的候选医生作为与所述用户匹配的医生，得到目标医生。
  7. 根据权利要求1-6中任一项所述的医疗数据处理方法，其中，在所述若所述比对结果为所述预测节点和所述结束节点相同，则根据所述预测节点和所述结束节点对所述用户进行医生匹配，得到目标医生之后，所述医疗数据处理方法还包括：
    若所述比对结果为所述预测节点和所述结束节点不相同，则对所述预测节点进行删除，并基于所述疾病模板生成所述用户对应的辅助信息；
    基于所述辅助信息和所述结束节点对所述用户进行医生匹配，得到所述用户对应的目标医生。
  8. 一种医疗数据处理设备，其中，所述医疗数据处理设备包括：存储器和至少一个处理器，所述存储器中存储有指令；所述至少一个处理器调用所述存储器中的所述指令，以使得所述医疗数据处理设备执行如下步骤：
    获取待处理的用户对应的医疗数据，并根据所述医疗数据判断所述用户是否符合预置的用户类型，其中，所述医疗数据包括用户性别、用户年龄和症状数据；
    若所述用户符合预置的用户类型，则根据所述医疗数据匹配与所述用户对应的疾病模板，其中，所述疾病模板包括用户待填写的疾病诊断路径；
    基于所述疾病诊断路径对所述疾病模板进行信息填充，得到所述疾病模板对应的结束节点和用户回答路径，并根据所述结束节点和所述用户回答路径生成用户需求；
    若所述用户需求为医生问诊，则将所述医疗数据和所述用户回答路径输入预置的医疗数据处理模型进行疾病数据处理，得到预测节点；
    对所述预测节点和所述结束节点进行比对分析，得到比对结果；
    若所述比对结果为所述预测节点和所述结束节点相同，则根据所述预测节点和所述结束节点对所述用户进行医生匹配，得到目标医生。
  9. 根据权利要求8所述的医疗数据处理设备，其中，所述获取待处理的用户对应的医疗数据，并根据所述医疗数据判断所述用户是否符合预置的用户类型，包括：
    接收待处理的用户输入的医疗数据，其中，所述医疗数据包括用户性别、用户年龄和症状数据；
    基于所述用户性别、所述用户年龄和所述症状数据生成所述用户对应的用户画像；
    计算所述用户画像是否符合预置的用户类型对应的相似度，并对所述相似度和预设目标值进行比对；
    若所述相似度大于或等于所述预设目标值，则确定所述用户符合预置的用户类型。
  10. 根据权利要求8所述的医疗数据处理设备，其中，所述若所述用户符合预置的用户类型，则根据所述医疗数据匹配与所述用户对应的疾病模板，包括：
    若所述用户符合预置的用户类型，则对所述医疗数据进行关键词提取，得到目标关键 词；
    分别将所述目标关键词和预置的多个疾病模板进行匹配，得到每个疾病模板对应的关键词命中率；
    对每个疾病模板对应的关键词命中率进行排序，并将所述关键词命中率最高的疾病模板作为所述用户对应的疾病模板，其中，所述疾病模板包括用户待填写的疾病诊断路径。
  11. 根据权利要求8所述的医疗数据处理设备，其中，所述基于所述疾病诊断路径对所述疾病模板进行信息填充，得到所述疾病模板对应的结束节点和用户回答路径，并根据所述结束节点和所述用户回答路径生成用户需求，包括：
    基于所述疾病诊断路径获取跑通率和推荐点击率；
    基于所述跑通率和所述推荐点击率对所述疾病模板进行信息填充，得到填充后的疾病模板；
    对所述填充后的疾病模板进行数据提取，得到所述疾病模板对应的结束节点和用户回答路径；
    基于所述结束节点和所述用户回答路径生成用户需求，其中，所述用户需求包括医生问诊和线下问诊。
  12. 根据权利要求8所述的医疗数据处理设备，其中，所述若所述用户需求为医生问诊，则将所述医疗数据和所述用户回答路径输入预置的医疗数据处理模型进行疾病数据处理，得到预测节点，包括：
    若所述用户需求为医生问诊，则将所述医疗数据和所述用户回答路径输入预置的医疗数据处理模型；
    通过所述医疗数据处理模型中的卷积神经网络对所述医疗数据和所述用户回答路径进行卷积运算，得到所述医疗数据对应的特征向量；
    对所述特征向量进行特征归一化处理，得到预测节点。
  13. 根据权利要求8所述的医疗数据处理设备，其中，所述若所述比对结果为所述预测节点和所述结束节点相同，则根据所述预测节点和所述结束节点对所述用户进行医生匹配，得到目标医生，包括：
    若所述比对结果为所述预测节点和所述结束节点相同，则获取预置候选医生集合中的多个候选医生；
    基于所述预测节点和所述结束节点对所述多个候选医生进行可信度计算，得到每个候选医生对应的目标可信度；
    根据每个候选医生对应的目标可信度对所述候选医生集合中的所述多个候选医生进行排名，并将排名最高的候选医生作为与所述用户匹配的医生，得到目标医生。
  14. 根据权利要求8-13中任一项所述的医疗数据处理设备，其中，在所述若所述比对结果为所述预测节点和所述结束节点相同，则根据所述预测节点和所述结束节点对所述用户进行医生匹配，得到目标医生之后，所述医疗数据处理设备还包括：
    若所述比对结果为所述预测节点和所述结束节点不相同，则对所述预测节点进行删除，并基于所述疾病模板生成所述用户对应的辅助信息；
    基于所述辅助信息和所述结束节点对所述用户进行医生匹配，得到所述用户对应的目标医生。
  15. 一种计算机可读存储介质，所述计算机可读存储介质上存储有指令，其中，所述指令被处理器执行时实现如下步骤：
    获取待处理的用户对应的医疗数据，并根据所述医疗数据判断所述用户是否符合预置的用户类型，其中，所述医疗数据包括用户性别、用户年龄和症状数据；
    若所述用户符合预置的用户类型，则根据所述医疗数据匹配与所述用户对应的疾病模 板，其中，所述疾病模板包括用户待填写的疾病诊断路径；
    基于所述疾病诊断路径对所述疾病模板进行信息填充，得到所述疾病模板对应的结束节点和用户回答路径，并根据所述结束节点和所述用户回答路径生成用户需求；
    若所述用户需求为医生问诊，则将所述医疗数据和所述用户回答路径输入预置的医疗数据处理模型进行疾病数据处理，得到预测节点；
    对所述预测节点和所述结束节点进行比对分析，得到比对结果；
    若所述比对结果为所述预测节点和所述结束节点相同，则根据所述预测节点和所述结束节点对所述用户进行医生匹配，得到目标医生。
  16. 根据权利要求15所述的计算机可读存储介质，其中，所述获取待处理的用户对应的医疗数据，并根据所述医疗数据判断所述用户是否符合预置的用户类型，包括：
    接收待处理的用户输入的医疗数据，其中，所述医疗数据包括用户性别、用户年龄和症状数据；
    基于所述用户性别、所述用户年龄和所述症状数据生成所述用户对应的用户画像；
    计算所述用户画像是否符合预置的用户类型对应的相似度，并对所述相似度和预设目标值进行比对；
    若所述相似度大于或等于所述预设目标值，则确定所述用户符合预置的用户类型。
  17. 根据权利要求15所述的计算机可读存储介质，其中，所述若所述用户符合预置的用户类型，则根据所述医疗数据匹配与所述用户对应的疾病模板，包括：
    若所述用户符合预置的用户类型，则对所述医疗数据进行关键词提取，得到目标关键词；
    分别将所述目标关键词和预置的多个疾病模板进行匹配，得到每个疾病模板对应的关键词命中率；
    对每个疾病模板对应的关键词命中率进行排序，并将所述关键词命中率最高的疾病模板作为所述用户对应的疾病模板，其中，所述疾病模板包括用户待填写的疾病诊断路径。
  18. 根据权利要求15所述的计算机可读存储介质，其中，所述基于所述疾病诊断路径对所述疾病模板进行信息填充，得到所述疾病模板对应的结束节点和用户回答路径，并根据所述结束节点和所述用户回答路径生成用户需求，包括：
    基于所述疾病诊断路径获取跑通率和推荐点击率；
    基于所述跑通率和所述推荐点击率对所述疾病模板进行信息填充，得到填充后的疾病模板；
    对所述填充后的疾病模板进行数据提取，得到所述疾病模板对应的结束节点和用户回答路径；
    基于所述结束节点和所述用户回答路径生成用户需求，其中，所述用户需求包括医生问诊和线下问诊。
  19. 根据权利要求15所述的计算机可读存储介质，其中，所述若所述用户需求为医生问诊，则将所述医疗数据和所述用户回答路径输入预置的医疗数据处理模型进行疾病数据处理，得到预测节点，包括：
    若所述用户需求为医生问诊，则将所述医疗数据和所述用户回答路径输入预置的医疗数据处理模型；
    通过所述医疗数据处理模型中的卷积神经网络对所述医疗数据和所述用户回答路径进行卷积运算，得到所述医疗数据对应的特征向量；
    对所述特征向量进行特征归一化处理，得到预测节点。
  20. 一种医疗数据处理装置，其中，所述医疗数据处理装置包括：
    获取模块，用于获取待处理的用户对应的医疗数据，并根据所述医疗数据判断所述用 户是否符合预置的用户类型，其中，所述医疗数据包括用户性别、用户年龄和症状数据；
    匹配模块，用于若所述用户符合预置的用户类型，则根据所述医疗数据匹配与所述用户对应的疾病模板，其中，所述疾病模板包括用户待填写的疾病诊断路径；
    填充模块，用于基于所述疾病诊断路径对所述疾病模板进行信息填充，得到所述疾病模板对应的结束节点和用户回答路径，并根据所述结束节点和所述用户回答路径生成用户需求；
    处理模块，用于若所述用户需求为医生问诊，则将所述医疗数据和所述用户回答路径输入预置的医疗数据处理模型进行疾病数据处理，得到预测节点；
    分析模块，用于对所述预测节点和所述结束节点进行比对分析，得到比对结果；
    生成模块，用于若所述比对结果为所述预测节点和所述结束节点相同，则根据所述预测节点和所述结束节点对所述用户进行医生匹配，得到目标医生。
PCT/CN2022/121724 2022-03-24 2022-09-27 医疗数据处理方法、装置、设备及存储介质 WO2023178970A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202210296654.8A CN114743647A (zh) 2022-03-24 2022-03-24 医疗数据处理方法、装置、设备及存储介质
CN202210296654.8 2022-03-24

Publications (1)

Publication Number Publication Date
WO2023178970A1 true WO2023178970A1 (zh) 2023-09-28

Family

ID=82277325

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/121724 WO2023178970A1 (zh) 2022-03-24 2022-09-27 医疗数据处理方法、装置、设备及存储介质

Country Status (2)

Country Link
CN (1) CN114743647A (zh)
WO (1) WO2023178970A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN118248297A (zh) * 2024-05-28 2024-06-25 山东佰泰丰信息科技有限公司 一种多元化信息管理系统

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114743647A (zh) * 2022-03-24 2022-07-12 康键信息技术（深圳）有限公司 医疗数据处理方法、装置、设备及存储介质

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080091631A1 (en) * 2006-10-11 2008-04-17 Henry Joseph Legere Method and Apparatus for an Algorithmic Approach to Patient-Driven Computer-Assisted Diagnosis
CN109273101A (zh) * 2018-09-05 2019-01-25 百度在线网络技术（北京）有限公司 基于决策树的人机交互方法、装置、计算机设备及介质
CN110111884A (zh) * 2019-04-30 2019-08-09 杭州电子科技大学 一种基于cmkmc的人机协同智慧医疗辅助决策系统
CN110489566A (zh) * 2019-08-22 2019-11-22 上海软中信息系统咨询有限公司 一种智能导诊服务机器人的导诊方法
CN114743647A (zh) * 2022-03-24 2022-07-12 康键信息技术（深圳）有限公司 医疗数据处理方法、装置、设备及存储介质

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080091631A1 (en) * 2006-10-11 2008-04-17 Henry Joseph Legere Method and Apparatus for an Algorithmic Approach to Patient-Driven Computer-Assisted Diagnosis
CN109273101A (zh) * 2018-09-05 2019-01-25 百度在线网络技术（北京）有限公司 基于决策树的人机交互方法、装置、计算机设备及介质
CN110111884A (zh) * 2019-04-30 2019-08-09 杭州电子科技大学 一种基于cmkmc的人机协同智慧医疗辅助决策系统
CN110489566A (zh) * 2019-08-22 2019-11-22 上海软中信息系统咨询有限公司 一种智能导诊服务机器人的导诊方法
CN114743647A (zh) * 2022-03-24 2022-07-12 康键信息技术（深圳）有限公司 医疗数据处理方法、装置、设备及存储介质

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN118248297A (zh) * 2024-05-28 2024-06-25 山东佰泰丰信息科技有限公司 一种多元化信息管理系统

Also Published As

Publication number Publication date
CN114743647A (zh) 2022-07-12

Similar Documents

Publication Publication Date Title
CN111986770B (zh) 药方用药审核方法、装置、设备及存储介质
CN111191020B (zh) 基于机器学习和知识图谱的处方推荐方法和系统
WO2023178970A1 (zh) 医疗数据处理方法、装置、设备及存储介质
US20200050949A1 (en) Digital assistant platform
WO2021151327A1 (zh) 分诊数据处理方法、装置、设备及介质
WO2022062353A1 (zh) 医疗数据处理方法、装置、计算机设备和存储介质
CN112037880A (zh) 用药推荐方法、装置、设备及存储介质
CN112035741B (zh) 基于用户体检数据的预约方法、装置、设备及存储介质
CN112069413B (zh) 医生推荐方法、装置、服务器和存储介质
CN107705842A (zh) 智能分诊系统及其工作方法
Li et al. Reliable medical diagnosis from crowdsourcing: Discover trustworthy answers from non-experts
WO2023178971A1 (zh) 就医的互联网挂号方法、装置、设备及存储介质
CN113223735B (zh) 基于对话表征的分诊方法、装置、设备及存储介质
CN111933291A (zh) 医疗信息推荐装置、方法、系统、设备及可读存储介质
CN111667890A (zh) 基于人工智能的用药推荐方法、装置、设备及存储介质
CN112201359A (zh) 基于人工智能的重症问诊数据识别方法及装置
CN111370102A (zh) 科室导诊方法、装置以及设备
CN113658712A (zh) 医患匹配方法、装置、设备及存储介质
EP4131279A1 (en) Experience engine-method and apparatus of learning from similar patients
US20150106111A1 (en) System, method and computer program product for diagnostic and treatment enhancement
CN110473636B (zh) 基于深度学习的智能医嘱推荐方法及系统
CN111415760A (zh) 医生推荐方法、系统、计算机设备及存储介质
CN117594206A (zh) 基于医疗互联平台的患者一体化分诊系统及方法
CN113779954A (zh) 相似文本的推荐方法、装置和电子设备
CN113241193A (zh) 药物推荐模型训练方法、推荐方法、装置、设备及介质

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: 22933022

Country of ref document: EP

Kind code of ref document: A1