Auxiliary medical consultation method, device, storage medium and electronic device

Abstract

The present application relates to the field of digital medical technology, and discloses an auxiliary medical consultation method, device, storage medium, and electronic device. The method includes: parsing the current symptom description text of the user asking the question to generate a number of first symptom keywords; based on each first symptom keyword and each historical symptom keyword corresponding to the user asking the question, using a preset disease diagnosis model for diagnosis to obtain the target disease and target TCM syndrome type of the current user asking the question; based on the target disease and target TCM syndrome type, performing a diagnostic path query to obtain a first symptom set corresponding to the target disease and a second symptom set corresponding to the target TCM syndrome type; based on each first symptom keyword and each historical symptom keyword, using a preset TCM knowledge graph to filter each symptom in the first symptom set and the second symptom set to obtain a symptom set to be asked for auxiliary medical consultation corresponding to the user asking the question. The auxiliary medical consultation method in the present application can improve the efficiency of medical consultation.

Description

Technical Field

[0001] The present invention relates to the field of traditional Chinese medicine consultation technology, applicable to the field of digital medical technology, and in particular to an auxiliary consultation method, device, storage medium and electronic equipment. Background Art

[0002] In today's rapidly advancing technological landscape, artificial intelligence (AI) has gradually become a technological leader in China's industrial evolution and a key support for enhancing its international competitiveness. Combining AI with Traditional Chinese Medicine (TCM) to help modern people improve their health is highly anticipated. As more and more AI technologies are successfully implemented in various fields, related intelligent dialogue-based medical consultation systems are also emerging in the field of smart TCM. TCM intelligent dialogue-based medical consultation systems aim to alleviate the shortage of medical personnel by helping doctors collect user information and symptoms in advance and provide preliminary diagnostic results for their reference. Based on TCM knowledge, the consultation system collects symptoms through multiple rounds of question-and-answer sessions with users. This system can support disease-assisted diagnosis, health management, remote consultation, and other functions. Traditional assisted medical consultation methods are mostly based on a symptom set for the target disease and syndrome type. However, during the consultation process, the same symptom attributes are asked multiple times, resulting in low symptom collection efficiency. Summary of the Invention

[0003] In view of this, the present invention provides an auxiliary medical consultation method, device, storage medium and electronic device, the main purpose of which is to solve the current problem of low efficiency in collecting medical consultation symptoms.

[0004] To solve the above problems, the present application provides an auxiliary diagnosis method, including:

[0005] Parse the user's current symptom description text to generate several first symptom keywords;

[0006] Based on each of the first symptom keywords and each of the historical symptom keywords corresponding to the user asking the question, a preset disease diagnosis model is used to perform a diagnosis to obtain the target disease and target TCM syndrome type of the current user asking the question;

[0007] Perform a diagnostic path query based on the target disease and the target TCM syndrome type to obtain a first symptom set corresponding to the target disease and a second symptom set corresponding to the target TCM syndrome type;

[0008] Based on each of the first symptom keywords and each of the historical symptom keywords, a preset traditional Chinese medicine knowledge graph is used to filter each symptom in the first symptom set and the second symptom set to obtain a set of symptoms to be asked corresponding to the consulting user for auxiliary consultation.

[0009] Optionally, the current symptom description text of the consulting user is parsed to generate a number of first symptom keywords, specifically including:

[0010] Get the current symptom description text of the consulting user;

[0011] Processing the symptom description text using a natural language processing method to obtain status description information of the consulting user;

[0012] Using a preset intent classification model to perform intent classification on the state description information to obtain a classification result;

[0013] When the classification result indicates that the status description information is an inquiry intention, performing word segmentation processing on the status description information of the inquiry user to obtain a word segmentation result;

[0014] Symptom keyword matching is performed based on the word segmentation result to obtain each of the first symptom keywords.

[0015] Optionally, the method further includes:

[0016] When the classification result shows that the status description information is not intended for a medical consultation, obtaining a preset question text;

[0017] The user who inquires about the patient is asked a question based on the question text, so as to update the current symptom description text based on the answer information received from the user who inquires about the question text.

[0018] Optionally, before filtering the first symptom set and each symptom in the second symptom set based on a preset TCM knowledge graph, the method further includes: constructing the preset TCM knowledge graph, specifically including:

[0019] Obtain several TCM symptoms;

[0020] According to each symptom relationship category and the filtering rules corresponding to each symptom relationship category, for the target TCM symptom, a filtering relationship between each TCM symptom and the target TCM symptom is constructed to obtain a plurality of target filtering relationships corresponding to the target symptom, so as to obtain filtering relationships corresponding to each TCM symptom respectively;

[0021] Based on the filtering relationships, a TCM knowledge graph is constructed that reflects the relationships between the TCM symptoms.

[0022] Optionally, the method of using a preset disease diagnosis model to perform diagnosis based on each of the first symptom keywords and each of the historical symptom keywords corresponding to the consulting user to obtain the target disease and target TCM syndrome type of the current consulting user specifically includes:

[0023] Recombining the first symptom keywords and the historical symptom keywords corresponding to the consulting user to obtain second symptom keywords;

[0024] Based on each of the second symptom keywords, using the disease diagnosis model to perform disease diagnosis, calculating the score values ​​of various diseases corresponding to each of the second symptom keywords, and determining the disease with the largest score value as the target disease;

[0025] Based on each of the second symptom keywords, the disease diagnosis model is used to perform TCM syndrome diagnosis, the score values ​​of various TCM syndromes corresponding to each of the second symptom keywords are calculated, and the TCM syndrome with the largest score value is determined as the target TCM syndrome.

[0026] Optionally, the first symptoms in the first symptom set and the second symptoms in the second symptom set are filtered using a preset TCM knowledge graph based on the first symptom keywords and the historical symptom keywords to obtain a set of symptoms to be asked for auxiliary consultation corresponding to the consulting user, specifically including:

[0027] Based on each of the second symptom keywords, a preset TCM knowledge graph is searched to obtain a filtering relationship corresponding to each of the second symptom keywords, each of the filtering relationships carrying a filtering rule;

[0028] The first symptoms in the first symptom set and the second symptoms in the second symptom set are filtered based on the filtering rules to obtain a set of symptoms to be asked corresponding to the consulting user and used for auxiliary consultation.

[0029] Optionally, after obtaining the set of symptoms to be asked corresponding to the consulting user and used for auxiliary consultation, the method further includes:

[0030] Constructing questioning words based on the preset template and the set of symptoms to be asked, and generating questions to be asked;

[0031] Ask the user who is asking questions based on the question to be asked.

[0032] To solve the above problems, the present application provides an auxiliary medical consultation device, comprising:

[0033] Parsing module: used to parse the current symptom description text of the consulting user and generate several first symptom keywords;

[0034] Diagnosis module: used for performing diagnosis using a preset disease diagnosis model based on each of the first symptom keywords and each of the historical symptom keywords corresponding to the consulting user, to obtain the target disease and target TCM syndrome type of the current consulting user;

[0035] Query module: used to perform diagnostic path query based on the target disease and the target TCM syndrome type, and obtain a first symptom set corresponding to the target disease and a second symptom set corresponding to the target TCM syndrome type;

[0036] Filtering module: used to filter the symptoms in the first symptom set and the second symptom set based on the first symptom keywords and the historical symptom keywords using a preset traditional Chinese medicine knowledge graph to obtain a set of symptoms to be asked corresponding to the consulting user for auxiliary consultation.

[0037] To solve the above problem, the present application provides a storage medium, which stores a computer program. When the computer program is executed by a processor, the steps of the above-mentioned auxiliary medical consultation method are implemented.

[0038] To solve the above problems, the present application provides an electronic device, characterized in that it includes at least a memory and a processor, wherein a computer program is stored on the memory, and the processor implements the steps of the auxiliary diagnosis method when executing the computer program on the memory.

[0039] This application generates several first symptom keywords by parsing the current symptom description text of the user asking the question; based on each of the first symptom keywords and each of the historical symptom keywords corresponding to the user asking the question, a preset disease diagnosis model is used for diagnosis to obtain the target disease and target TCM syndrome type of the current user asking the question; based on the target disease and the target TCM syndrome type, a diagnostic path query is performed to obtain a first symptom set corresponding to the target disease and a second symptom set corresponding to the target TCM syndrome type; based on each of the first symptom keywords and each of the historical symptom keywords, a preset TCM knowledge graph is used to filter each symptom in the first symptom set and the second symptom set to obtain a symptom set to be asked for auxiliary consultation corresponding to the user asking the question, so as to ask questions to the user asking the question based on each symptom in the symptom set to be asked. The auxiliary consultation method in this application can filter the symptoms to be asked, reduce questions about symptoms with the same attributes, and improve the efficiency of system symptom collection.

[0040] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are specifically listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Various other advantages and benefits will become apparent to those skilled in the art upon reading the detailed description of the preferred embodiment below. The accompanying drawings are for illustration purposes only and are not to be considered as limiting the present invention. The same reference symbols are used throughout the drawings to represent the same components. In the drawings:

[0042] Figure 1 A schematic diagram of a flow chart of an auxiliary medical consultation method provided in an embodiment of the present application is shown;

[0043] Figure 2 A flowchart of an auxiliary medical consultation method provided by another embodiment of the present application is shown;

[0044] Figure 3 A structural block diagram of an auxiliary medical consultation device provided in another embodiment of the present application is shown. DETAILED DESCRIPTION

[0045] Various aspects and features of the present application are described herein with reference to the accompanying drawings.

[0046] It should be understood that various modifications may be made to the embodiments of the present application. Therefore, the above description should not be considered as limiting, but merely as an example of an embodiment. Other modifications within the scope and spirit of the present application will occur to those skilled in the art.

[0047] The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate embodiments of the present application and, together with the general description of the present application given above and the detailed description of the embodiments given below, serve to explain the principles of the present application.

[0048] These and other characteristics of the present application will become apparent from the following description of a preferred form of embodiment given as a non-limiting example with reference to the accompanying drawings.

[0049] It should also be understood that although the present application has been described with reference to certain specific examples, those skilled in the art will readily be able to implement many other equivalent forms of the present application.

[0050] The above and other aspects, features and advantages of the present application will become more apparent from the following detailed description when taken in conjunction with the accompanying drawings.

[0051] Specific embodiments of the present application will be described hereinafter with reference to the accompanying drawings; however, it should be understood that the embodiments described are merely examples of the present application and may be implemented in a variety of ways. Familiar and / or repetitive functions and structures are not described in detail to avoid obscuring the present application with unnecessary or redundant details. Therefore, the specific structural and functional details described herein are not intended to be limiting, but rather serve merely as a basis and representative basis for the claims to teach those skilled in the art to variously utilize the present application with substantially any suitable detailed structure.

[0052] This specification may use the phrases "in one embodiment," "in another embodiment," "in yet another embodiment," or "in other embodiments," which may all refer to one or more of the same or different embodiments according to the present application.

[0053] The present application embodiment provides an auxiliary diagnosis method, such as Figure 1 Shown, including:

[0054] Step S101: Parse the current symptom description text of the consulting user to generate a number of first symptom keywords;

[0055] During the specific implementation of this step, the current symptom description text of the consulting user is obtained; the symptom description text is processed using a natural language processing method to obtain the status description information of the consulting user; the status description information is classified using a preset consultation intention classification model to obtain a classification result; when the classification result is that the status description information is a consultation intention, the status description information of the consulting user is segmented to obtain a segmentation result; symptom keyword matching is performed based on the segmentation result to obtain each of the first symptom keywords. Specifically, the preset entity recognition model and the preset entity alignment model can be used to process and obtain each of the first symptom keywords. The first symptom keywords include: migraine, mild headache, dull headache and other symptom keywords.

[0056] Step S102: Based on each of the first symptom keywords and each of the historical symptom keywords corresponding to the user asking the question, a preset disease diagnosis model is used to perform diagnosis to obtain the target disease and target TCM syndrome type of the current user asking the question;

[0057] During the specific implementation of this step, each of the first symptom keywords and each of the historical symptom keywords corresponding to the consulting user are reorganized to obtain each of the second symptom keywords; based on each of the second symptom keywords, the disease diagnosis model is used to diagnose the disease, the score values ​​of various diseases corresponding to each of the second symptom keywords are calculated, and the disease with the largest score value is determined as the target disease; based on each of the second symptom keywords, the disease diagnosis model is used to diagnose the TCM syndrome type, the score values ​​of various TCM syndrome types corresponding to each of the second symptom keywords are calculated, and the TCM syndrome type with the largest score value is determined as the target TCM syndrome type.

[0058] Step S103: performing a diagnostic path query based on the target disease and the target TCM syndrome type to obtain a first symptom set corresponding to the target disease and a second symptom set corresponding to the target TCM syndrome type;

[0059] During the specific implementation of this step, the disease diagnostic pathway is queried based on the target disease to obtain the first symptom set corresponding to the target disease; the TCM syndrome type diagnostic pathway is queried based on the target TCM syndrome type to obtain the second symptom set corresponding to the target TCM syndrome type. Specifically, each disease and each TCM syndrome type corresponds to a symptom set, and each symptom set contains dimensional data such as main symptoms and secondary related symptoms. After the target disease and target TCM syndrome type results are output based on the disease diagnosis model, the first symptom set corresponding to the target disease and the second symptom set corresponding to the target TCM syndrome type are recommended based on the output target disease and target TCM syndrome type in combination with the diagnostic pathway.

[0060] Step S104: Based on each of the first symptom keywords and each of the historical symptom keywords, the preset TCM knowledge graph is used to filter each symptom in the first symptom set and the second symptom set to obtain questions corresponding to the user asking the question.

[0061] During the specific implementation of this step, the preset TCM knowledge graph is queried based on each second symptom keyword to obtain a filtering relationship corresponding to each second symptom keyword, and each filtering relationship carries a filtering rule; based on each filtering rule, each first symptom in the first symptom set and each second symptom in the second symptom set are filtered to obtain a set of symptoms to be asked for auxiliary consultation corresponding to the consulting user. Then, based on the preset template and the set of symptoms to be asked, questioning words are generated to obtain questions to be asked; questions are asked to the consulting user based on the questions to be asked. The above-mentioned intelligent question-answering method can be applied to intelligent diagnosis and treatment and remote consultation.

[0062] This application generates several first symptom keywords by parsing the current symptom description text of the user asking the question; based on each of the first symptom keywords and each of the historical symptom keywords corresponding to the user asking the question, a preset disease diagnosis model is used for diagnosis to obtain the target disease and target TCM syndrome type of the current user asking the question; based on the target disease and the target TCM syndrome type, a diagnostic path query is performed to obtain a first symptom set corresponding to the target disease and a second symptom set corresponding to the target TCM syndrome type; based on each of the first symptom keywords and each of the historical symptom keywords, a preset TCM knowledge graph is used to filter each symptom in the first symptom set and the second symptom set to obtain a symptom set to be asked for auxiliary consultation corresponding to the user asking the question, so as to ask questions to the user asking the question based on each symptom in the symptom set to be asked. The auxiliary consultation method in this application can filter the symptoms to be asked, reduce questions about symptoms with the same attributes, and improve the efficiency of system symptom collection.

[0063] Another embodiment of the present application provides another auxiliary diagnosis method, such as Figure 2 Shown, including:

[0064] Step S201: Parse the current symptom description text of the consulting user to generate a number of first symptom keywords;

[0065] During the specific implementation of this step, the current symptom description text of the consulting user is obtained; the patient's current symptom description text is the words used by the consulting user during the conversation, which is generally more colloquial, and the words of the consulting user need to be processed. The symptom description text is processed by a natural language processing method to obtain the status description information of the consulting user; the status description information is classified using a preset consultation intention classification model to obtain a classification result; the training method of the consultation intention classification model is: obtain a number of consultation intention sentences, mark each of the consultation intention sentences, obtain a number of label sentences, and train the initial consultation intention model based on each of the consultation intention sentences as training samples. When the model accuracy meets the design requirements, the consultation intention classification model is obtained. When the classification result is that the status description information is a consultation intention, the status description information of the consulting user is segmented to obtain a segmentation result; symptom keyword matching is performed based on the segmentation result to obtain each of the first symptom keywords. Specifically, when the classification result is that the state description information is a medical consultation intention, a preset entity recognition method is used to identify the entities in the state description information, extract each entity in the state description information, and then use a preset entity alignment method to input them one by one into a preset sliding window algorithm for symptom keyword matching to obtain each first symptom keyword. When the classification result is that the state description information is not a medical consultation intention, a preset question text is obtained; based on the question text, questions are asked to the medical consultation user, and the current symptom description text is updated based on the received reply information of the medical consultation user to the question text. A number of first symptom keywords are generated based on the updated current symptom description text. The first symptom keywords are: migraine, headache extending to the back of the neck, posterior headache, pain in the forehead and brow bone, headache, splitting headache, headache, mild headache, repeated headaches, headache caused by wind and sun, headache caused by emotional fluctuations and insomnia, and other physical conditions causing headaches.

[0066] Step S202: Recombining the first symptom keywords and the historical symptom keywords corresponding to the consulting user to obtain second symptom keywords;

[0067] During the specific implementation of this step, the first symptom keywords obtained in the current communication round and several historical symptom keywords obtained in previous communication rounds are recombined to obtain second symptom keywords that are the sum of the first symptom keywords and the historical symptom keywords. For example, if the first symptom keyword of the current round of the current user is "migraine" and the historical symptom keywords obtained during the consultation and diagnosis process in previous rounds are "mild headache" and "headache", then the second symptom keywords obtained by summing the first symptom keywords and the historical symptom keywords are "migraine", "mild headache", and "headache".

[0068] Step S203: Based on each of the second symptom keywords, the disease diagnosis model is used to perform disease diagnosis, and the scores of various diseases corresponding to each of the second symptom keywords are calculated, and the disease with the largest score is determined as the target disease;

[0069] During the specific implementation of this step, disease diagnosis is performed based on each of the second symptom keywords and the disease diagnosis model. Scores for the various diseases corresponding to each of the second symptom keywords are calculated, and the disease with the highest score is determined as the target disease. For example, if the second symptom keywords (migraine, mild headache, and head pain) are input into the disease diagnosis model for disease diagnosis, and the disease with the highest score is headache, headache is determined as the target disease.

[0070] Step S204: Based on each of the second symptom keywords, the disease diagnosis model is used to perform TCM syndrome diagnosis, and the scores of various TCM syndromes corresponding to each of the second symptom keywords are calculated, and the TCM syndrome with the largest score is determined as the target TCM syndrome;

[0071] During the specific implementation of this step, a TCM syndrome type diagnosis is performed based on each of the second symptom keywords and the disease diagnosis model. Scores for the various TCM syndrome types corresponding to each of the second symptom keywords are calculated, and the TCM syndrome type with the highest score is determined as the target TCM syndrome type. For example, if the second symptom keywords (migraine, mild headache, and head pain) are input into the disease diagnosis model for TCM syndrome type diagnosis, and the TCM syndrome type with the highest score is Liver Yang Headache, Liver Yang Headache is determined as the target TCM syndrome type.

[0072] Step S205: constructing a TCM knowledge graph;

[0073] During the specific implementation of this step, several TCM symptoms are obtained;

[0074] According to each symptom relationship category and the filtering rules corresponding to each symptom relationship category, for the target TCM symptom, a filtering relationship between each TCM symptom and the target TCM symptom is constructed, and a plurality of target filtering relationships corresponding to the target symptom are obtained, thereby obtaining filtering relationships corresponding to each TCM symptom; wherein the symptom relationship categories include: location symptom relationship category, property symptom relationship category, degree symptom relationship category, existence state symptom relationship category, and cause symptom relationship category. Based on each of the filtering relationships, a TCM knowledge graph reflecting the relationship between each of the TCM symptoms is constructed. Figure 3The schematic diagram of the TCM knowledge graph constructed for headache symptoms is shown. Specifically, the TCM knowledge graph is constructed using a graphical data structure. The obtained TCM symptoms are numbered, and the numbers are identified by nodes. The relationships between the TCM symptoms are encoded and identified by relationship codes Relationships. Nodes and Relationships contain attributes in the form of key / value. Nodes are connected through the relationships defined by Relationships to form a relational network structure. The data between nodes is stored in the form of triples. For example, the triple storage method of node 1 and node 2 is: [Node1, Relationship1, Node2] and so on. The data of the TCM symptom knowledge graph is stored using triple data, as follows:

[0075] [Headache, location (multiple choices), migraine],

[0076] [Headache, location (multiple choices), headache extending to the neck and back],

[0077] [Headache, location (multiple choices), occipital headache],

[0078] [Headache, location (multiple choices), pain in the forehead and brow bone],

[0079] [Headache, location (multiple choices), headache on both sides],

[0080] [Headache, location (multiple choices), pain at the top of the head],

[0081] [Headache, nature (single choice), head pain],

[0082] [Headache, nature (single choice), head swelling and splitting],

[0083] [Headache, nature (single choice), headache],

[0084] [Headache, nature (single choice), head tightness],

[0085] [Headache, nature (single choice), headache likes to wrap],

[0086] [Headache, nature (single choice), dull headache],

[0087] [Headache, nature (single choice), headache like a wrap],

[0088] [Headache, nature (single choice), empty pain in the head],

[0089] [Headache, nature (single choice), headache, heaviness, dizziness],

[0090] [Headache, nature (single choice), headache throbbing pain],

[0091] [Headache, nature (single choice), cold head pain],

[0092] [Headache, degree (single choice), mild headache],

[0093] [Headache, degree (single choice), mild headache],

[0094] [Headache, degree (single choice), moderate headache],

[0095] …

[0096] Among them, location (multiple choices), property (single choice), etc. represent filtering relationships, and multiple choices and single choices correspond to filtering rules.

[0097] Step S206: performing a diagnostic path query based on the target disease and the target TCM syndrome type to obtain a first symptom set corresponding to the target disease and a second symptom set corresponding to the target TCM syndrome type;

[0098] During the specific implementation of this step, the disease diagnosis path is queried based on the target disease to obtain the first symptom set corresponding to the target disease; the TCM syndrome type diagnostic path is queried based on the target TCM syndrome type to obtain the second symptom set corresponding to the target TCM syndrome type. Specifically, each disease and each TCM syndrome type corresponds to a symptom set, and each symptom set contains dimensional data such as main symptoms and secondary related symptoms. After the target disease and target TCM syndrome type results are output based on the disease diagnosis model, the first symptom set corresponding to the target disease and the second symptom set corresponding to the target TCM syndrome type are recommended based on the output target disease and target TCM syndrome type and the diagnostic path. For example: after searching the disease diagnosis path for the target disease of headache, the first symptom set corresponding to the headache obtained is: migraine, occipital headache, headache, splitting headache and other first symptoms; after searching the disease diagnosis path for the target TCM syndrome type of liver yang headache, the second symptom set corresponding to the liver yang headache obtained is: irritability, irritability, dizziness and other second symptoms.

[0099] Step S207: querying a preset TCM knowledge graph based on each of the second symptom keywords to obtain a filtering relationship corresponding to each of the second symptom keywords, each of the filtering relationships carrying a filtering rule;

[0100] During the specific implementation of this step, a preset TCM knowledge graph is queried based on each second symptom keyword to obtain a filtering relationship entity corresponding to each second symptom keyword. For example, when the second symptom keyword is: "migraine", "mild headache", or "headache", the symptoms (headache) and filtering relationship entity (part (multiple choices)) related to "migraine" are queried in the constructed graph. Because the filtering rule of the filtering relationship entity is a multiple-choice type, it is not necessary to filter the symptoms whose relationship entity with the "headache" symptom in the symptom set to be asked is "part (multiple choices)"; the symptoms (headache) and filtering relationship entity (degree (single choice)) related to "mild headache" are queried in the constructed graph. Because the filtering rule of the relationship entity is a single-choice type, it is necessary to filter the symptoms whose relationship entity with the "headache" symptom in the symptom set to be asked is "degree (single choice)"; the symptoms (headache) and relationship entity (property (single choice)) related to "headache" are queried in the constructed graph. Because the relationship entity is a single-choice type, it is necessary to filter the symptoms whose relationship entity with the "headache" symptom in the symptom set to be asked is "property (single choice)".

[0101] Step S208: filtering each first symptom in the first symptom set and each second symptom in the second symptom set based on each filtering rule to obtain a symptom set to be asked corresponding to the consulting user and used for auxiliary consultation;

[0102] During the specific implementation of this step, each first symptom in the first symptom set and each second symptom in the second symptom set are filtered based on each filtering rule to obtain a set of symptoms to be asked corresponding to the consulting user and used for auxiliary consultation. For example: when the second symptom keyword is: "migraine", "mild headache", "headache", the symptoms (headache) and filtering relationship entities (part (multiple choices)) related to "migraine" are searched in the constructed graph. Because the filtering rule of filtering relationship entities is of multiple choice type, it is not necessary to filter the symptoms in the set of symptoms to be asked whose relationship entity with the symptom of "headache" is "part (multiple choices)"; the symptoms (headache) and filtering relationship entities (degree (single choice)) related to "mild headache" are searched in the constructed graph. Because the filtering rule of relationship entities is of single choice type, it is necessary to filter the symptoms in the set of symptoms to be asked whose relationship entity with the symptom of "headache" is "part (multiple choices)". The relationship entity of the "headache" symptom is "degree (single choice)", and the first symptom and / or second symptom such as "moderate headache" are filtered out. In the constructed graph, the symptoms (headache) and relationship entity (property (single choice)) related to "headache" are searched. Because the relationship entity is a single choice type, it is necessary to filter out the symptoms in the set of symptoms to be asked that have the relationship entity of "property (single choice)" with the "headache" symptom, and filter out the first symptoms and / or second symptoms such as "splitting head" and "dull head pain". After filtering, the set of symptoms to be asked is: bilateral headache, recurrent headache, irritability, irritability, and dizziness.

[0103] Step S209: constructing questioning words based on the preset template and the set of symptoms to be asked, and generating questions to be asked;

[0104] During the specific implementation of this step, the preset template may be: Do you have symptoms such as xxx, xxx, etc. This application does not restrict the format of the preset template. For example, using the preset template and the set of symptoms to be asked to construct a question, the generated question to be asked may be: Do you have symptoms such as bilateral headaches, recurrent headaches, irritability, dizziness, etc.

[0105] Step S210: Asking questions to the consulting user based on the question to be asked.

[0106] During the specific implementation of this step, the questions to be asked are displayed on the artificial intelligence consultation page based on the questions to be asked, and the symptom description text of the user asking the question for the displayed questions to be asked is obtained, so as to parse the symptom description text and collect the next round of symptoms to be asked, so as to help the doctor collect user information and user symptoms in advance and provide preliminary diagnosis results for the doctor's reference.

[0107] The present application generates a number of first symptom keywords by parsing the current symptom description text of the consulting user; recombines each of the first symptom keywords and each of the historical symptom keywords corresponding to the consulting user to obtain each of the second symptom keywords; based on each of the second symptom keywords, uses the disease diagnosis model to diagnose the disease, calculates the score values ​​of various diseases corresponding to each of the second symptom keywords, and determines the disease with the largest score value as the target disease; based on each of the second symptom keywords, uses the disease diagnosis model to diagnose TCM syndrome types, calculates the score values ​​of various TCM syndrome types corresponding to each of the second symptom keywords, and determines the TCM syndrome type with the largest score value as the target TCM syndrome type; constructs a TCM knowledge graph; based on the target The diagnostic path query is performed on the target disease and the target TCM syndrome type to obtain the first symptom set corresponding to the target disease and the second symptom set corresponding to the target TCM syndrome type; based on each of the second symptom keywords, the preset TCM knowledge graph is queried to obtain the filtering relationship corresponding to each of the second symptom keywords, and each of the filtering relationships carries a filtering rule; based on each of the filtering rules, each of the first symptoms in the first symptom set and each of the second symptoms in the second symptom set is filtered to obtain the symptom set to be asked for auxiliary consultation corresponding to the consulting user; based on the preset template and the symptom set to be asked, questioning words are constructed to generate questions to be asked; based on the questions to be asked, questions are asked to the consulting user. The method in this application can filter the first symptom set corresponding to the target disease and the second symptom set corresponding to the target TCM syndrome type, thereby reducing questions about symptoms with the same attributes and improving consultation efficiency.

[0108] Another embodiment of the present application provides an auxiliary medical consultation device, such as Figure 3 Shown, including:

[0109] Parsing module 1: used to parse the current symptom description text of the user asking the question and generate a number of first symptom keywords;

[0110] Diagnosis module 2: configured to diagnose the target disease and target TCM syndrome type of the current user by using a preset disease diagnosis model based on the first symptom keywords and the historical symptom keywords corresponding to the user asking the question;

[0111] Query module 3: for performing a diagnostic path query based on the target disease and the target TCM syndrome type, and obtaining a first symptom set corresponding to the target disease and a second symptom set corresponding to the target TCM syndrome type;

[0112] Filtering module 4: used to filter the symptoms in the first symptom set and the second symptom set based on the first symptom keywords and the historical symptom keywords using a preset traditional Chinese medicine knowledge graph to obtain a set of symptoms to be asked corresponding to the consulting user for auxiliary consultation.

[0113] During the specific implementation process, the parsing module 1 is specifically used to: obtain the current symptom description text of the consulting user; use a natural language processing method to process the symptom description text to obtain the status description information of the consulting user; use a preset intention classification model to perform intent classification on the status description information to obtain a classification result; when the classification result is that the status description information is a consultation intention, perform word segmentation processing on the status description information of the consulting user to obtain a word segmentation result; perform symptom keyword matching based on the word segmentation result to obtain each of the first symptom keywords.

[0114] During the specific implementation process, the analysis module 1 is also used to: obtain a preset question text when the classification result is that the status description information is not intended to be a medical consultation; ask questions to the medical consultation user based on the question text, and update the current symptom description text based on the received reply information of the medical consultation user to the question text.

[0115] During the specific implementation process, the auxiliary medical consultation device also includes: a preset TCM knowledge graph construction module, and the preset TCM knowledge graph construction module is specifically used to: obtain a number of TCM symptoms; according to each symptom relationship category and the filtering rules corresponding to each symptom relationship category, for the target TCM symptom, construct a filtering relationship between each TCM symptom and the target TCM symptom, and obtain a number of target filtering relationships corresponding to the target symptom, so as to obtain each filtering relationship corresponding to each TCM symptom; based on each filtering relationship, a TCM knowledge graph reflecting the relationship between each TCM symptom is constructed.

[0116] During the specific implementation process, the diagnosis module 2 is specifically used to: reorganize each first symptom keyword and each historical symptom keyword corresponding to the consulting user to obtain each second symptom keyword; based on each second symptom keyword, use the disease diagnosis model to diagnose the disease, calculate the score values ​​of various diseases corresponding to each second symptom keyword, and determine the disease with the largest score value as the target disease; based on each second symptom keyword, use the disease diagnosis model to diagnose TCM syndrome types, calculate the score values ​​of various TCM syndrome types corresponding to each second symptom keyword, and determine the TCM syndrome type with the largest score value as the target TCM syndrome type.

[0117] During the specific implementation process, the filtering module 4 specifically includes: querying the preset traditional Chinese medicine knowledge graph based on each second symptom keyword, obtaining the filtering relationship corresponding to each second symptom keyword, and each filtering relationship carries a filtering rule; filtering each first symptom in the first symptom set and each second symptom in the second symptom set based on each filtering rule, and obtaining a set of symptoms to be asked corresponding to the consulting user for auxiliary consultation.

[0118] In a specific implementation, the auxiliary consultation device further includes: a question generation module, which is specifically used to construct questioning words based on a preset template and the set of symptoms to be asked, and generate questions to be asked. The auxiliary consultation device also includes: a questioning module, which is specifically used to ask questions to the consultation user based on the questions to be asked.

[0119] This application generates several first symptom keywords by parsing the current symptom description text of the user asking the question; based on each of the first symptom keywords and each of the historical symptom keywords corresponding to the user asking the question, a preset disease diagnosis model is used for diagnosis to obtain the target disease and target TCM syndrome type of the current user asking the question; based on the target disease and the target TCM syndrome type, a diagnostic path query is performed to obtain a first symptom set corresponding to the target disease and a second symptom set corresponding to the target TCM syndrome type; based on each of the first symptom keywords and each of the historical symptom keywords, a preset TCM knowledge graph is used to filter each symptom in the first symptom set and the second symptom set to obtain a symptom set to be asked for auxiliary consultation corresponding to the user asking the question, so as to ask questions to the user asking the question based on each symptom in the symptom set to be asked. The auxiliary consultation method in this application can filter the symptoms to be asked, reduce questions about symptoms with the same attributes, and improve the efficiency of system symptom collection.

[0120] Another embodiment of the present application provides a storage medium storing a computer program. When the computer program is executed by a processor, the following method steps are implemented:

[0121] Step 1: Parse the user's current symptom description text to generate several first symptom keywords;

[0122] Step 2: Based on the first symptom keywords and the historical symptom keywords corresponding to the user asking the question, a preset disease diagnosis model is used to perform a diagnosis to obtain the target disease and target TCM syndrome type of the current user asking the question;

[0123] Step 3: performing a diagnostic path query based on the target disease and the target TCM syndrome type to obtain a first symptom set corresponding to the target disease and a second symptom set corresponding to the target TCM syndrome type;

[0124] Step 4: Based on each of the first symptom keywords and each of the historical symptom keywords, a preset TCM knowledge graph is used to filter each symptom in the first symptom set and the second symptom set to obtain a set of symptoms to be asked corresponding to the consulting user for auxiliary consultation.

[0125] Those skilled in the art will understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct RAMbus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM), etc.

[0126] Those skilled in the art will clearly understand that for the sake of convenience and brevity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0127] The specific implementation process of the above method steps can be found in the embodiments of any of the above auxiliary medical consultation methods, and this embodiment will not be repeated here.

[0128] This application generates several first symptom keywords by parsing the current symptom description text of the user asking the question; based on each of the first symptom keywords and each of the historical symptom keywords corresponding to the user asking the question, a preset disease diagnosis model is used for diagnosis to obtain the target disease and target TCM syndrome type of the current user asking the question; based on the target disease and the target TCM syndrome type, a diagnostic path query is performed to obtain a first symptom set corresponding to the target disease and a second symptom set corresponding to the target TCM syndrome type; based on each of the first symptom keywords and each of the historical symptom keywords, a preset TCM knowledge graph is used to filter each symptom in the first symptom set and the second symptom set to obtain a symptom set to be asked for auxiliary consultation corresponding to the user asking the question, so as to ask questions to the user asking the question based on each symptom in the symptom set to be asked. The auxiliary consultation method in this application can filter the symptoms to be asked, reduce questions about symptoms with the same attributes, and improve the efficiency of system symptom collection.

[0129] Another embodiment of the present application provides an electronic device, which may be a server, and the electronic device includes a processor, a memory, a network interface, and a database connected via a system bus. The processor of the electronic device is used to provide computing and control capabilities. The memory of the electronic device includes a non-volatile and / or volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the electronic device is used to communicate with an external client via a network connection. When the electronic device program is executed by the processor, it implements a function or step on the server side of an auxiliary diagnosis method.

[0130] In one embodiment, an electronic device is provided, which may be a client. The electronic device includes a processor, a memory, a network interface, a display screen, and an input device connected via a system bus. The processor of the electronic device is used to provide computing and control capabilities. The memory of the electronic device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the electronic device is used to communicate with an external server via a network connection. When the electronic device program is executed by the processor, it implements the functions or steps of the client side of an auxiliary consultation method.

[0131] Another embodiment of the present application provides an electronic device, comprising at least a memory and a processor, wherein the memory stores a computer program, and the processor implements the following method steps when executing the computer program in the memory:

[0132] Step 1: Parse the user's current symptom description text to generate several first symptom keywords;

[0133] Step 2: Based on the first symptom keywords and the historical symptom keywords corresponding to the user asking the question, a preset disease diagnosis model is used to perform a diagnosis to obtain the target disease and target TCM syndrome type of the current user asking the question;

[0134] Step 3: performing a diagnostic path query based on the target disease and the target TCM syndrome type to obtain a first symptom set corresponding to the target disease and a second symptom set corresponding to the target TCM syndrome type;

[0135] Step 4: Based on each of the first symptom keywords and each of the historical symptom keywords, a preset TCM knowledge graph is used to filter each symptom in the first symptom set and the second symptom set to obtain a set of symptoms to be asked corresponding to the consulting user for auxiliary consultation.

[0136] The specific implementation process of the above method steps can be found in the embodiments of any of the above auxiliary medical consultation methods, and this embodiment will not be repeated here.

[0137] This application generates several first symptom keywords by parsing the current symptom description text of the user asking the question; based on each of the first symptom keywords and each of the historical symptom keywords corresponding to the user asking the question, a preset disease diagnosis model is used for diagnosis to obtain the target disease and target TCM syndrome type of the current user asking the question; based on the target disease and the target TCM syndrome type, a diagnostic path query is performed to obtain a first symptom set corresponding to the target disease and a second symptom set corresponding to the target TCM syndrome type; based on each of the first symptom keywords and each of the historical symptom keywords, a preset TCM knowledge graph is used to filter each symptom in the first symptom set and the second symptom set to obtain a symptom set to be asked for auxiliary consultation corresponding to the user asking the question, so as to ask questions to the user asking the question based on each symptom in the symptom set to be asked. The auxiliary consultation method in this application can filter the symptoms to be asked, reduce questions about symptoms with the same attributes, and improve the efficiency of system symptom collection.

[0138] The above embodiments are merely exemplary embodiments of the present application and are not intended to limit the scope of the present application. The scope of protection of the present application is defined by the claims. Those skilled in the art may make various modifications or equivalent substitutions to the present application within the essence and scope of protection of the present application, and such modifications or equivalent substitutions shall also be deemed to fall within the scope of protection of the present application.

Claims

1. An auxiliary medical inquiry method, characterized in that: include: Parse the user's current symptom description text to generate several first symptom keywords; Based on each of the first symptom keywords and each of the historical symptom keywords corresponding to the user asking the question, a preset disease diagnosis model is used to perform a diagnosis to obtain the target disease and target TCM syndrome type of the user asking the question; Perform a diagnostic path query based on the target disease and the target TCM syndrome type to obtain a first symptom set corresponding to the target disease and a second symptom set corresponding to the target TCM syndrome type; Based on each of the first symptom keywords and each of the historical symptom keywords, a preset TCM knowledge graph is used to filter each symptom in the first symptom set and the second symptom set to obtain a symptom set to be asked for auxiliary consultation corresponding to the consulting user; Before using the preset TCM knowledge graph to filter the symptoms in the first symptom set and the second symptom set, the method further includes: constructing the preset TCM knowledge graph, specifically including: Obtain several TCM symptoms; According to each symptom relationship category and the filtering rules corresponding to each symptom relationship category, for the target TCM symptom, a filtering relationship between each TCM symptom and the target TCM symptom is constructed to obtain a plurality of target filtering relationships corresponding to the target TCM symptom, so as to obtain filtering relationships corresponding to each TCM symptom respectively; A TCM knowledge graph reflecting the relationship between the TCM symptoms is constructed based on the filtering relationships; The method further comprises filtering each first symptom in the first symptom set and each second symptom in the second symptom set based on each first symptom keyword and each historical symptom keyword using a preset TCM knowledge graph to obtain a symptom set to be asked for auxiliary consultation corresponding to the consultation user, specifically including: Recombining the first symptom keywords and the historical symptom keywords corresponding to the consulting user to obtain second symptom keywords; Based on each of the second symptom keywords, a preset TCM knowledge graph is searched to obtain a filtering relationship corresponding to each of the second symptom keywords, each of the filtering relationships carrying a filtering rule; Filtering each first symptom in the first symptom set and each second symptom in the second symptom set based on each filtering rule to obtain a symptom set to be asked for auxiliary consultation corresponding to the consultation user; specifically including: When the filtering rule is of a multiple-selection type, it is not necessary to filter the first symptoms and the second symptoms that have the same keyword filtering relationship as the second symptom; When the filtering rule is a single-select type, it is necessary to filter each of the first symptoms and each of the second symptoms that have the same filtering relationship as the second symptom keyword; The current symptom description text is the conversation data of the current communication round; The historical symptom keywords are keywords obtained by parsing the conversation data of each historical communication round; The symptom relationship categories include location symptom relationship categories, nature symptom relationship categories, degree symptom relationship categories, duration symptom relationship categories, and cause symptom relationship categories; The filtering rules include multiple-selection filtering rules and single-selection filtering rules.

2. The method according to claim 1, wherein The current symptom description text of the consulting user is parsed to generate a number of first symptom keywords, specifically including: Get the current symptom description text of the consulting user; Processing the symptom description text using a natural language processing method to obtain status description information of the consulting user; Using a preset intent classification model to perform intent classification on the state description information to obtain a classification result; When the classification result indicates that the status description information is an inquiry intention, performing word segmentation processing on the status description information of the inquiry user to obtain a word segmentation result; Symptom keyword matching is performed based on the word segmentation result to obtain each of the first symptom keywords.

3. The method according to claim 2, wherein The method further comprises: When the classification result shows that the status description information is not intended for a medical consultation, obtaining a preset question text; The user who inquires about the patient is asked a question based on the question text, so as to update the current symptom description text based on the answer information received from the user who inquires about the question text.

4. The method according to claim 1, wherein The method of performing diagnosis based on each of the first symptom keywords and each of the historical symptom keywords corresponding to the user asking the question using a preset disease diagnosis model to obtain the target disease and target TCM syndrome type of the user asking the question specifically includes: Recombining the first symptom keywords and the historical symptom keywords corresponding to the consulting user to obtain second symptom keywords; Based on each of the second symptom keywords, using the disease diagnosis model to perform disease diagnosis, calculating the score values ​​of various diseases corresponding to each of the second symptom keywords, and determining the disease with the largest score value as the target disease; Based on each of the second symptom keywords, the disease diagnosis model is used to perform TCM syndrome diagnosis, the score values ​​of various TCM syndromes corresponding to each of the second symptom keywords are calculated, and the TCM syndrome with the largest score value is determined as the target TCM syndrome.

5. The method according to claim 4, wherein After obtaining the set of symptoms to be asked corresponding to the user asking the question and used for auxiliary consultation, the method further includes: Constructing questioning words based on the preset template and the set of symptoms to be asked, and generating questions to be asked; Ask the user who is asking questions based on the question to be asked.

6. A medical consultation auxiliary device, characterized in that: include: Parsing module: used to parse the current symptom description text of the consulting user and generate several first symptom keywords; The current symptom description text is the conversation data of the current communication round; Diagnosis module: used to diagnose the user's target disease and target TCM syndrome type using a preset disease diagnosis model based on each of the first symptom keywords and each of the historical symptom keywords corresponding to the user asking the question; the historical symptom keywords are keywords obtained by parsing the conversation data of each historical communication round; A TCM knowledge graph construction module is preset to obtain a number of TCM symptoms; according to each symptom relationship category and the filtering rules corresponding to each symptom relationship category, for the target TCM symptom, a filtering relationship between each TCM symptom and the target TCM symptom is constructed to obtain a number of target filtering relationships corresponding to the target TCM symptom, so as to obtain filtering relationships corresponding to each TCM symptom; based on each filtering relationship, a TCM knowledge graph reflecting the relationship between each TCM symptom is constructed; the symptom relationship category includes a part symptom relationship category, a property symptom relationship category, a degree symptom relationship category, a duration symptom relationship category and a cause symptom relationship category; the filtering rules include a multiple-selection filtering rule and a single-selection filtering rule; Query module: used to perform diagnostic path query based on the target disease and the target TCM syndrome type, and obtain a first symptom set corresponding to the target disease and a second symptom set corresponding to the target TCM syndrome type; Filtering module: used to filter each symptom in the first symptom set and the second symptom set based on each first symptom keyword and each historical symptom keyword using a preset Chinese medicine knowledge graph to obtain a symptom set to be asked for auxiliary consultation corresponding to the inquiring user, specifically used to reorganize each first symptom keyword and each historical symptom keyword corresponding to the inquiring user to obtain each second symptom keyword; query the preset Chinese medicine knowledge graph based on each second symptom keyword to obtain a filtering relationship corresponding to each second symptom keyword, each filtering relationship carries a filtering rule; filter each first symptom in the first symptom set and each second symptom in the second symptom set based on each filtering rule to obtain a symptom set to be asked for auxiliary consultation corresponding to the inquiring user; when the filtering rule is a multiple-choice type, it is not necessary to filter each first symptom and each second symptom that have the same filtering relationship as the second symptom keyword; when the filtering rule is a single-choice type, it is necessary to filter each first symptom and each second symptom that have the same filtering relationship as the second symptom keyword.

7. A storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the auxiliary medical consultation method according to any one of claims 1 to 5 are implemented.

8. An electronic device, characterized in that: The system comprises at least a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the auxiliary medical consultation method according to any one of claims 1 to 5 when executing the computer program on the memory.

Images