Question and answer management method and system for traditional Chinese medicine inquiry system

Abstract

The invention belongs to the field of online inquiry, and provides a question and answer management method and system for a traditional Chinese medicine medical inquiry system, and the method comprises the steps: obtaining the online inquiry content of a doctor; extracting target symptom information for diagnosis; determining symptom elements which need to be obtained through inspection diagnosis, generating image acquisition prompt information, and prompting a patient to upload corresponding image information; deducing the auscultation information and the cut diagnosis information which need to be further acquired to form a diagnosis element set which needs to be supplemented; carrying out obtainability analysis on the diagnosis element set needing to be supplemented, and determining contents which can be obtained through an inquiry problem and / or an image acquisition mode; and generating a complementary inquiry problem and / or complementary image acquisition prompt information based on the obtainability analysis result, and guiding the patient to provide complementary information related to auscultation and clinics.

Description

Technical Field

[0001] The present invention belongs to the field of online medical consultation, and in particular relates to a question and answer management method and system for a traditional Chinese medical consultation system. Background Art

[0002] Traditional Chinese medicine diagnosis and treatment relies heavily on the syndrome differentiation method of "four examinations combined", that is, to comprehensively judge the patient's physical condition and symptom characteristics through four diagnostic methods: inspection, auscultation, questioning and palpation. Among them, auscultation usually includes the identification of the patient's breath odor, breath, body odor, cough sound, etc., and palpation mainly includes the grasp of the pulse, such as pulse frequency, rhythm, pulse shape (smooth, rapid, slow, astringent, etc.), both of which require the doctor's on-site perception and contact.

[0003] In recent years, with the rise of telemedicine technology, online TCM consultation systems have been widely used in clinical practice, realizing remote interaction between doctors and patients through network terminals. Although consultation and observation can be partially realized through text communication and image uploading in an online environment, the implementation of olfactory and palpation is severely limited. Since doctors cannot perceive the patient's odor characteristics face to face, nor can they touch the patient's pulse for pulse diagnosis, olfactory and palpation information is difficult to collect or completely missing, making the information obtained from online TCM consultations significantly incomplete.

[0004] In addition, in actual applications, especially when doctors deal with a large number of online consultations, they tend to neglect the collection and judgment of auscultation and palpation elements. Especially for doctors with relatively insufficient clinical experience, there is a lack of a systematic guidance mechanism to assist them in identifying gaps in diagnostic information. They often find it difficult to realize that the current information cannot support accurate dialectics, and thus fail to guide patients to provide relevant supplementary information in a timely manner, affecting the quality of diagnosis and the accuracy of treatment recommendations. Summary of the invention

[0005] In order to solve the problems in the prior art, the present invention provides a question-answering management method for a traditional Chinese medicine medical consultation system, comprising the following steps: Obtain the content of the doctor's online consultation, which includes the doctor's questions asked to the patient and the patient's response information; Perform semantic analysis on the content of the doctor's online consultation to extract target symptom information for diagnosis; Based on the target symptom information, determine the symptom elements that need to be obtained through visual examination, generate image acquisition prompt information, and prompt the patient to upload corresponding image information; Combining the target symptom information with the image information uploaded by the patient, inferring the auscultation information and palpation information that need to be further acquired, and forming a set of diagnostic elements that need to be supplemented; Conducting an accessibility analysis on the set of diagnostic elements that need to be supplemented, and determining the content that can be obtained through asking questions and / or collecting images; Based on the results of the availability analysis, supplementary diagnostic questions and / or supplementary image acquisition prompts are generated, and the patient is guided to provide supplementary information related to auscultation and palpation.

[0006] Furthermore, the step of semantically analyzing the content of the doctor's online consultation includes: using natural language processing technology to perform word segmentation, keyword extraction, semantic classification and standardization on the consultation questions and answer information, and combining with the traditional Chinese medicine symptom terminology database to extract the patient's main symptoms, accompanying symptoms and their corresponding location characteristics, time characteristics and trait characteristics.

[0007] Furthermore, the step of determining the symptom elements that need to be obtained through visual diagnosis includes: the system calls the association rule library between symptoms and visual diagnosis elements, matches the target symptom information, generates image acquisition prompt information including image acquisition part, image acquisition requirements and image upload method, and prompts the image information of the corresponding part in the user interface.

[0008] Furthermore, in the step of forming a set of diagnostic elements that need to be supplemented, the system calls the symptom-syndrome-four-diagnosis information association rule library, and determines the syndrome candidate set and the missing auscultation or palpation elements through forward reasoning. The missing elements are recorded as a set of diagnostic elements that need to be supplemented.

[0009] Furthermore, in the step of performing an accessibility analysis on the set of diagnostic elements that need to be supplemented, the system calls a preset diagnostic element-acquisition method mapping table to determine whether each diagnostic element can be acquired through medical questions, image acquisition, or a combination of the two, and automatically generates corresponding supplementary medical questions or image acquisition prompt information based on the analysis results.

[0010] Another aspect of the present invention further provides a question-answering management system for a traditional Chinese medical consultation system, comprising the following modules: An acquisition module is used to acquire the content of the doctor's online consultation, which includes the doctor's questions asked to the patient and the patient's answer information; A semantic analysis module, used to perform semantic analysis on the content of the doctor's online consultation and extract target symptom information for diagnosis; A visual diagnosis prompt module is used to determine the symptom elements that need to be obtained through visual diagnosis based on the target symptom information, generate image acquisition prompt information, and prompt the patient to upload the corresponding image information; An inference module is used to combine the target symptom information with the image information uploaded by the patient to infer the auscultation information and palpation information that need to be further obtained, so as to form a set of diagnostic elements that need to be supplemented; An availability analysis module, used to perform an availability analysis on the set of diagnostic elements to be supplemented, and determine the content that can be obtained by asking questions and / or collecting images; The supplementary prompt generation module is used to generate supplementary diagnostic questions and / or supplementary image acquisition prompt information based on the availability analysis results, and guide the patient to provide supplementary information related to auscultation and palpation.

[0011] Furthermore, the semantic analysis module includes: a natural language processing unit, which is used to perform word segmentation, keyword extraction, semantic classification and standardization on medical questions and answer information; a symptom extraction unit, which is used to extract the patient's main symptoms, accompanying symptoms and their corresponding location characteristics, time characteristics and trait characteristics in combination with the traditional Chinese medicine symptom terminology database.

[0012] Furthermore, the visual diagnosis prompt module includes: a rule matching unit, which is used to call the association rule library between symptoms and visual diagnosis elements to match the target symptom information; an image prompt generation unit, which is used to generate image acquisition prompt information including image acquisition part, image acquisition requirements and image upload method, and prompt the patient to upload image information of the corresponding part in the user interface.

[0013] Furthermore, the inference module includes: a syndrome reasoning unit, which is used to call the symptom-syndrome-four-diagnosis information association rule library and determine the syndrome candidate set through forward reasoning; an element missing judgment unit, which is used to identify the missing auscultation or palpation elements in the current diagnosis and treatment information and record them as a set of diagnostic elements that need to be supplemented.

[0014] Furthermore, the availability analysis module includes: a mapping table calling unit, used to call a preset diagnostic element-acquisition method mapping table; an acquisition method determination unit, used to determine whether each diagnostic element can be acquired through medical questioning, image acquisition, or a combination of the two; a supplementary prompt generation module, used to automatically generate corresponding supplementary medical question or image acquisition prompt information based on the analysis results.

[0015] The present invention provides a question-answering management method and system for a traditional Chinese medicine medical consultation system, which can effectively solve the technical problems in the existing online traditional Chinese medicine consultation system such as the inability to conveniently obtain auscultation and palpation information, incomplete diagnosis and treatment information, and difficulty for doctors to guide patients to provide auxiliary information, and has the following beneficial effects: The present invention combines the patient's medical interview content and image information to automatically infer the auscultation and palpation information that needs to be further collected, and converts this type of information into content that can be obtained through additional medical interviews or image collection, effectively making up for the limitation that doctors cannot directly perform auscultation and palpation in a remote environment, and realizing the intelligent completion of missing information in the "four examinations in one" of traditional Chinese medicine.

[0016] Based on rule reasoning and information integrity judgment, the system automatically identifies whether the current diagnostic information is complete, and automatically generates supplementary questions and prompt information to assist doctors in efficiently guiding patients to provide key auxiliary information. It is especially helpful for inexperienced doctors to avoid missing key diagnostic elements and improve the scientific nature of syndrome differentiation and conditioning recommendations.

[0017] Through interactive methods such as graphic prompts and question guidance, the system can clearly guide patients to complete image acquisition and symptom description, enabling patients to provide high-quality diagnosis and treatment information without professional guidance, thereby improving information collection efficiency and user experience.

[0018] When the system establishes the mapping table of diagnostic elements and collection methods, it generates mapping results through formula reasoning and dynamic parameters, eliminating the tedious maintenance of static tables. At the same time, it comprehensively considers multiple dimensions such as symptom importance, method reliability, user operability, historical feedback, etc. to improve the scientific nature of the recommended collection method.

[0019] In summary, the method of the present invention not only optimizes the online consultation process of traditional Chinese medicine, improves the completeness of diagnosis and treatment information and the accuracy of syndrome differentiation, but also has the advantages of being highly intelligent, practical, and user-friendly, and is suitable for a variety of traditional Chinese medicine telemedicine service scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0021] Figure 1 is a flow chart of the method of the present invention; Figure 2 It is a system block diagram of the present invention. DETAILED DESCRIPTION

[0022] Below, the invention is preferably described in conjunction with the accompanying drawings and specific implementation methods.

[0023] This embodiment solves the above problem through the following steps: In one embodiment, reference Figure 1 The present invention provides a question-and-answer management method for a traditional Chinese medicine medical consultation system. The method is based on the diagnosis and treatment concept of "four examinations combined" in traditional Chinese medicine. It intelligently analyzes the doctor's online consultation content, determines the visual diagnosis image information that needs to be collected, and intelligently infers the auscultation and palpation content that needs to be further obtained by combining the consultation and visual diagnosis results. Supplementary consultation questions and image collection prompts are generated accordingly, thereby assisting in the comprehensive acquisition of diagnosis and treatment information and accurate syndrome differentiation in a remote environment.

[0024] The content of the doctor's online consultation is obtained, which includes the doctor's questions asked to the patient and the patient's response information.

[0025] The consultation content information generated when the doctor and the patient remotely interact through a network terminal in the traditional Chinese medicine medical consultation system is obtained, and the consultation content information at least includes: one or more consultation questions with diagnostic purposes asked by the doctor to the patient in accordance with the principle of traditional Chinese medicine syndrome differentiation and treatment based on the main symptoms and basic health information described by the patient, and the patient's reply to the consultation questions, and the reply content is the symptom description information used for diagnosis. The consultation questions and reply content can be input in text form and / or voice form, and automatically collected, data recorded and structured stored by the system for subsequent generation of visual diagnosis prompt information and inference of diagnosis and treatment content.

[0026] For example, the doctor obtains online the patient's chief complaint of "cough and sore throat", then the doctor asks the patient questions through the system according to the diagnosis and treatment needs: "Do you have fever symptoms?", "What color is your sputum?", "Do you have a bad taste in your mouth?", the patient replies through system text input or voice input: "I have a fever, the sputum is yellow and thick, and my mouth tastes bitter." The system collects the reply content and records it as structured data, which serves as the basis for subsequent judgment on whether it is necessary to collect tongue coating images and infer olfactory diagnosis information.

[0027] The contents of the doctor's online consultation are semantically parsed to extract target symptom information for diagnosis.

[0028] The content of the doctor's online consultation is semantically parsed, and the semantic parsing includes using natural language processing technology to identify language features, extract keywords, restore sentence meanings, and semantically classify the doctor's questions and the patient's responses. In combination with the TCM symptom terminology database and symptom ontology model, the core diagnostic elements in the content are standardized to extract target symptom information for subsequent diagnostic analysis. The target symptom information includes the patient's chief symptoms, accompanying symptoms, time characteristics, location characteristics, trait characteristics, etc., which is used to support the system in determining the content of inspection prompts and inferring the needs of auscultation and palpation.

[0029] For example, a doctor asks a question through the online consultation system: "Do you have phlegm when you cough? What is the color and properties of the phlegm?" The patient replies: "When I cough, I have yellow phlegm, and it's quite sticky." The system performs semantic analysis on the reply information, identifies "cough" as the main symptom, and "yellow phlegm" and "sticky" as the properties of phlegm. Through standardized matching of the keywords "yellow" and "sticky", the system extracts the target symptom information as: "cough with yellow and thick phlegm", and classifies it into the "lung heat cough" related symptom set of the TCM symptom ontology, providing a basis for subsequently prompting patients to upload tongue coating images and inferring "bad breath" information during auscultation.

[0030] Based on the target symptom information, the symptom elements that need to be obtained through visual examination are determined, image acquisition prompt information is generated, and the patient is prompted to upload the corresponding image information.

[0031] Based on the target symptom information, the system preset symptom and visual diagnosis element association rule library is called to perform feature matching on the target symptom information, identify the symptom elements corresponding to the symptom information that need to be obtained through visual diagnosis, and generate corresponding image acquisition prompt information by the system. The image acquisition prompt information includes the target part of image acquisition, acquisition requirements and upload method. The system prompts the patient to acquire and upload corresponding image information through the user terminal according to the image acquisition prompt information. The image information is input as visual diagnosis data for auxiliary diagnosis to support subsequent diagnosis and treatment analysis.

[0032] Specifically: The system has a built-in symptom-inspection element association rule base, which records the mapping relationship between various target symptom information and corresponding inspection requirements. For example, the symptoms of "cough and yellow sputum" correspond to the inspection elements of "tongue coating" and "complexion", and the symptoms of "abdominal distension and poor appetite" correspond to "tongue body" and "thickness of tongue coating".

[0033] Based on the target symptom information, the system searches the association rule library to obtain a set of symptom elements that need to be confirmed through inspection, such as tongue coating color, tongue fatness, complexion redness, eyelid color, etc.

[0034] The system generates image acquisition prompts for each symptom element, including: Image acquisition site (e.g., tongue, face, fingers); Image acquisition requirements (such as natural light shooting, no obstructions, clear focus); Image upload interface link and operation guide.

[0035] The system displays image acquisition prompts through the user interface. Patients collect images according to the prompts and upload them to the system. The system performs preliminary quality verification on the images to confirm that they can be used for visual diagnosis and analysis.

[0036] For example: The target symptoms extracted from the patients’ response information were: “cough with yellow sputum, dry mouth, and sore throat”.

[0037] The system matches association rules and determines that the elements that need to be obtained through inspection are "tongue coating color", "tongue shape", and "throat color"; The system generates the following image acquisition prompts: "Please upload a clear photo of your tongue coating to determine its color and quality"; "Please upload a photo of your throat to help determine whether your throat is red and swollen"; “It is recommended to shoot in natural light and keep the image as focused and clear as possible”; The system pushes prompts to the patient's terminal, the patient collects and uploads images, and the system completes image quality checks and stores them for subsequent symptom inference.

[0038] Through this step, the system can automatically guide patients to collect and upload key visual images based on their subjective symptom information, effectively making up for the limitation that doctors cannot see patients face to face in remote environments, and improving the integrity of diagnosis and treatment information and the accuracy of syndrome differentiation. At the same time, the intelligent generation of image acquisition prompts optimizes the user experience, reduces the error rate of image acquisition, and enhances the practicality and interaction efficiency of the system.

[0039] By combining the target symptom information with the image information uploaded by the patient, the auscultation information and palpation information that need to be further obtained are inferred to form a set of diagnostic elements that need to be supplemented.

[0040] Combining the target symptom information obtained through semantic analysis with the image information uploaded by the patient, the system conducts a comprehensive analysis of the currently acquired diagnostic information based on the preset TCM syndrome differentiation reasoning rule library, and infers the auscultation information and palpation information that need to be further acquired, thereby forming a set of diagnostic elements to be supplemented. The set of diagnostic elements includes but is not limited to the patient's breath characteristics, body odor, pulse characteristics, pulse frequency, etc. The inference process is based on the TCM four diagnostic theories and syndrome association rules to ensure automatic recognition and information completion of auscultation and palpation needs in a remote environment.

[0041] Specifically, the target symptom information (text data) and the image analysis results (image features) are input into the diagnostic information integration module to form the acquired diagnostic data set: Text feature set: (such as cough, yellow sputum, sore throat); Image feature set: (such as red tongue with yellow coating, red and swollen throat); Calling the Symptom-Syndrome-Four Diagnostic Information Association Rule Base Will and Map to a set of possible syndrome types and determine whether the current information is sufficient to support the establishment of the syndrome type.

[0042] If the diagnosis information is insufficient, the system infers the set of auscultation and palpation information that needs further confirmation based on the inference results. , such as "Do you have bad breath?", "Do you have a rapid pulse?", "Do you have a slippery pulse?"

[0043] right Each item in the test is structurally marked as supplementary information to be collected, and the supplementary type (smell or palpation), content, collection method and other information are recorded to provide a basis for subsequent information acquisition.

[0044] Furthermore, when obtaining the text feature set, a natural language processing (NLP) toolkit can be used, such as Python's spaCy, Jieba, or Transformers library; a TCM symptom dictionary can be established, including standard symptom words, synonyms, and feature attributes (such as location, nature, and time), and the corresponding text features can be extracted through the TCM symptom dictionary.

[0045] Furthermore, when obtaining the image feature set, doctors can manually identify and annotate it. You can also use image recognition models (such as lightweight CNNs such as MobileNet and ResNet); for images such as tongue images and faces, train or call ready-made models to identify features such as color, thickness, and moisture.

[0046] Furthermore, in constructing the symptom-syndrome-four diagnostic information association rules, we collected the symptom-syndrome records in the classic Chinese medicine literature (such as "Huangdi Neijing", "Shanghan Lun", "Jingui Yaolue", etc.); summarized the modern Chinese medicine clinical guidelines and medical record database to obtain standardized symptom-syndrome-four diagnostic information relationship data; and structured it into the correspondence between symptom items, inspection, auscultation, questioning and palpation characteristic items and syndrome type items.

[0047] Establish a TCM symptom ontology library, define standard terms, synonyms, symptom classifications and attributes (such as location, nature, time characteristics, etc.); establish a syndrome standard library, standardize syndrome names according to "Diagnosis in Traditional Chinese Medicine", and define the key symptoms and four diagnostic features required for each syndrome.

[0048] Based on the structured data, a rule base is defined. Each rule is as follows: Rule form: If [symptom set] + [four diagnostic characteristics set] meet a specific combination, then it is inferred as [syndrome type]; Example rule: If symptoms include "cough, yellow sputum, and dry mouth" and a red tongue with yellow fur, the syndrome may be "wind-heat invading the lungs."

[0049] Furthermore, the rule base supports: Necessary condition rules (if missing, it is invalid); Sufficient condition rule (existence means high probability); Assign weight coefficients and perform confidence calculations (for probabilistic reasoning models).

[0050] Furthermore, based on the rule base, a reasoning engine is built to support forward reasoning (inferring syndrome type from symptoms) and reverse reasoning (checking symptom completeness from syndrome type); a logical reasoning framework (such as the Drools rule engine) or a syndrome type reasoning algorithm based on Bayesian network / graph neural network is selected; comprehensive reasoning of multiple rules is supported, and multiple alternative syndrome types and their confidence rankings are output.

[0051] In this step, the smell and palpation information is automatically identified and inferred in an online environment to make up for the lack of information in the absence of face-to-face and palpation conditions, thereby improving the accuracy of syndrome differentiation. Automatically prompting for additional information needs through rule reasoning can assist doctors in efficient syndrome differentiation and reduce the risk of repeated inquiries and omissions. The resulting set of diagnostic elements is highly structured, which is convenient for subsequent decision-making engines, conditioning suggestion systems and other modules to call, and realize automated diagnosis and treatment. By prompting supplementary content, user participation and trust in the system can be enhanced, and user stickiness and satisfaction of online consultations can be improved.

[0052] An availability analysis is performed on the set of diagnostic elements that need to be supplemented to determine the content that can be obtained through asking questions and / or collecting images.

[0053] Based on the system preset diagnostic factor collection method mapping table, the collection method is determined for each diagnostic factor to be supplemented, and it is determined whether the diagnostic factor can be obtained through additional medical questions or through image collection, or whether a combination of the two is required for auxiliary acquisition; the analysis process forms a diagnostic factor acquisition plan including collection method annotations, and generates medical questions and / or image upload prompt information for supplementary collection, thereby realizing alternative or auxiliary collection of auscultation and palpation information.

[0054] Specifically:

[0055] A mapping table of diagnostic elements and collection methods is established. Each diagnostic element defines its availability and acquisition method. The table format is shown in Table 1:

[0056] Table 1 Diagnostic element items

[0057] From the above steps, the system obtains the set of diagnostic elements that need to be supplemented .

[0058] For each ,The system retrieves the mapping table, determines its recommended collection method and alternative collection methods, and determines whether information acquisition can be achieved through medical questioning or image acquisition.

[0059] The system generates a collection plan for each diagnostic element: If it is possible to ask → generate additional questions; If image is needed → generate image acquisition prompt; If it can be obtained in combination → Generate questions and image prompts at the same time.

[0060] For example: A collection of diagnostic elements: bad breath, slippery pulse, rapid pulse, and body odor.

[0061] System analysis results include: Bad breath: Smell diagnosis is recommended, which can be replaced by asking → question: "Do you feel a bad smell in your mouth?"; Smooth pulse and rapid pulse: Palpation is recommended, which can be partially confirmed by imaging + questioning → question: "Do you feel a fast pulse?"; Image prompt: "Please upload a photo of the end of your finger"; Body odor: Smell diagnosis is recommended, which can be done by asking → Question: "Have you noticed a noticeable body odor recently?"

[0062] This step uses accessibility analysis to reasonably convert the information that originally requires the doctor to perceive into patient-operable questions and answers or image uploads, thereby improving the information acquisition rate.

[0063] Based on the results of the availability analysis, supplementary diagnostic questions and / or supplementary image acquisition prompts are generated, and the patient is guided to provide supplementary information related to auscultation and palpation.

[0064] Based on the results of the accessibility analysis of the set of diagnostic elements that need to be supplemented, the system automatically generates targeted question content and image acquisition prompt information according to the correspondence between the diagnostic elements and the acquisition methods. The supplementary medical questions are used to guide the patient to make a self-description or selection of the information related to auscultation and / or palpation that needs to be collected, and the image acquisition prompt information is used to prompt the patient to collect and upload image information related to auscultation and / or palpation; the system displays the supplementary medical questions and / or image acquisition prompt information to the patient through the terminal interface, and guides the patient to complete the provision of the supplementary information according to the prompts, so as to achieve effective acquisition of the diagnostic elements that need to be supplemented, and assist the "four examinations in one" of traditional Chinese medicine in a remote environment.

[0065] Based on the determined collection method, the system converts the olfactory information that needs to be supplemented into actionable inquiry questions, such as "Do you feel a bad smell in your mouth?", "Have you felt that your body odor has worsened recently?".

[0066] For the physical examination information that needs to be acquired through image collection, image collection prompt information is generated, including the collection site (such as the end of the finger), shooting requirements (such as natural light, no obstruction) and upload method.

[0067] The system presents supplementary prompts on the user side, guiding patients to complete information supplementation through a combination of graphics and text or voice assistance.

[0068] For example, as shown in Table 2: Table 2 Prompt items for diagnostic factors

[0069] In a further improved embodiment, the diagnostic factor-collection method mapping table is established by: In the TCM consultation process, the system combines the doctor's completed consultation content with the image information uploaded by the patient, and through semantic analysis and image recognition processing, determines the set of diagnostic elements that still need to be further obtained in the current diagnosis and treatment process: ; in, Indicates a missing diagnostic element, such as "bad breath", "slippery pulse", "body odor", "rapid pulse", etc.

[0070] For each diagnostic factor, the system sets the available collection methods. , where the subscript Indicates collection test; Preferably, the collection method includes: It means Question (Q) - providing information through the patient's subjective answer; Indicates image acquisition (I) - such as uploading tongue, finger, facial images, etc.; It means the combination of question and image (Q+I) - joint acquisition to improve reliability; Furthermore, the collection method may also include voice collection, video shooting, physiological sensor interface, etc.

[0071] For each With each , the system extracts parameters from the following four dimensions: 1) Contribution to diagnosis : Indicates diagnostic elements The degree of importance in current dialectical reasoning. The methods of obtaining it include: The system has built-in weights in the TCM syndrome-symptom association database; Or the forward reasoning model can automatically assign values ​​to missing information through sensitivity analysis; Value range: [0,1].

[0072] 2) Collection reliability : Indicates the way Collect elements The accuracy estimate is: Labeled by clinical experts or evaluated through training of historical medical consultation big data; For example, "Ask a doctor about bad breath" Can be set to 0.6.

[0073] 3) User feasibility score : Indicates the current user's Operational ability and coordination: Dynamic adjustment based on terminal type (PC / mobile phone), whether it supports taking photos, user age, etc. Value range: [0,1]. For example, when elderly people use image acquisition, set it to a lower value.

[0074] 4) Historical success rate : Indicates the probability of this combination successfully collecting valid data in past consultation records: ; in, is the smoothing factor, is the number of successful attempts, is the total number of attempts.

[0075] In order to improve the personalization and accuracy of the strategy, the system also introduces: 1) Context-sensitive factors : Reflecting the "sensitivity" of the diagnostic factor in the current complaint, symptom combination, and image manifestation, it is defined as: ; For example, in the context of "red tongue with yellow fur + cough + yellow sputum", the sensitivity of "bad breath" is higher.

[0076] 2) Multimodal Validation Factor : Indicates whether a certain collection method can be jointly verified by other methods (such as image + voice) to improve credibility. The specific value can be preset in the system.

[0077] 3) Method Penalty Factor : It indicates the complexity of the collection method, the degree of user resistance, etc. It is mainly used to suppress complex and tedious operation processes. The specific value can be preset in the system.

[0078] The system calculates each diagnostic factor In all collection methods The overall rating is: ; in: λ 1 , λ 2 , λ 3 ∈ [ 0 , 1 ] It is an adjustment coefficient, which can be set by the system or can be adjusted.

[0079] Final recommendation: ; All scoring results and recommended collection methods are output as structured mapping table entries.

[0080] Based on the availability analysis, the system automatically pushes the necessary supplementary medical consultation content and image prompts, effectively supplementing the missing auscultation and palpation information in the remote environment, and improving the comprehensiveness of diagnosis and treatment. At the same time, through the standard template display of questions and image collection prompts, patients can efficiently complete the collection of required information without professional guidance, and the operation is simple.

[0081] See also Figure 2 In another embodiment, the present invention further provides a question-answering management system for a traditional Chinese medical consultation system, comprising: An acquisition module is used to acquire the content of the doctor's online consultation, which includes the doctor's questions asked to the patient and the patient's answer information; A semantic analysis module, used to perform semantic analysis on the content of the doctor's online consultation and extract target symptom information for diagnosis; A visual diagnosis prompt module is used to determine the symptom elements that need to be obtained through visual diagnosis based on the target symptom information, generate image acquisition prompt information, and prompt the patient to upload the corresponding image information; An inference module is used to combine the target symptom information with the image information uploaded by the patient to infer the auscultation information and palpation information that need to be further obtained, so as to form a set of diagnostic elements that need to be supplemented; An availability analysis module, used to perform an availability analysis on the set of diagnostic elements to be supplemented, and determine the content that can be obtained by asking questions and / or collecting images; The supplementary prompt generation module is used to generate supplementary diagnostic questions and / or supplementary image acquisition prompt information based on the availability analysis results, and guide the patient to provide supplementary information related to auscultation and palpation.

[0082] It should be noted that the explanation of the embodiment of the question and answer management method of the aforementioned traditional Chinese medicine medical consultation system is also applicable to the device of the embodiment of the present application and will not be repeated here.

[0083] Those of ordinary skill in the art will appreciate that the various units and algorithm steps described in the embodiments disclosed herein can be implemented in a combination of electronic hardware, computer software, and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0084] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0085] In several embodiments provided in the present application, if any function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory; hereinafter referred to as: ROM), random access memory (Random Access Memory; hereinafter referred to as: RAM), disk or optical disk, and other media that can store program codes.

[0086] The above is only a specific implementation of the present application. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in this application, which should be included in the protection scope of this application. For some module structures that are not particularly clear in the present invention, the content recorded in the prior art shall prevail. The prior art mentioned in the above background technology section and the specific embodiment section of the present invention can be used as a part of the present invention to understand the meaning of some technical features or parameters.

Claims

1. A question-answer management method for a traditional Chinese medicine medical consultation system, characterized in that: The steps include: Obtain the content of the doctor's online consultation, which includes the doctor's questions asked to the patient and the patient's response information; Perform semantic analysis on the content of the doctor's online consultation to extract target symptom information for diagnosis; Based on the target symptom information, determine the symptom elements that need to be obtained through visual examination, generate image acquisition prompt information, and prompt the patient to upload corresponding image information; Combining the target symptom information with the image information uploaded by the patient, inferring the auscultation information and palpation information that need to be further acquired, and forming a set of diagnostic elements that need to be supplemented; Conducting an accessibility analysis on the set of diagnostic elements that need to be supplemented, and determining the content that can be obtained through asking questions and / or collecting images; Based on the results of the availability analysis, additional diagnostic questions and / or additional image acquisition prompts are generated, and the patient is guided to provide additional information related to auscultation and palpation.

2. The question-answer management method of the TCM medical consultation system according to claim 1, characterized in that: The step of semantically analyzing the content of the doctor's online consultation includes: using natural language processing technology to perform word segmentation, keyword extraction, semantic classification and standardization on the consultation questions and answer information, and combining with the traditional Chinese medicine symptom terminology database to extract the patient's main symptoms, accompanying symptoms and their corresponding location characteristics, time characteristics and trait characteristics.

3. The question-answer management method of the TCM medical consultation system according to claim 1, characterized in that: The steps of determining the symptom elements that need to be obtained through visual diagnosis include: the system calls the association rule library between symptoms and visual diagnosis elements, matches the target symptom information, generates image acquisition prompt information including image acquisition part, image acquisition requirements and image upload method, and prompts the image information of the corresponding part in the user interface.

4. The question-answer management method of the TCM medical consultation system according to claim 1, characterized in that: In the step of forming a set of diagnostic elements that need to be supplemented, the system calls the symptom-syndrome-four-diagnosis information association rule library, and determines the syndrome candidate set and the missing auscultation or palpation elements through forward reasoning. The missing elements are recorded as the set of diagnostic elements that need to be supplemented.

5. The question-answer management method of the TCM medical consultation system according to claim 1, characterized in that: In the step of performing an accessibility analysis on the set of diagnostic elements that need to be supplemented, the system calls a preset diagnostic element-acquisition method mapping table to determine whether each diagnostic element can be obtained through medical questions, image acquisition, or a combination of the two, and automatically generates corresponding supplementary medical questions or image acquisition prompt information based on the analysis results.

6. A question-answering management system for a traditional Chinese medical consultation system, characterized in that: The system includes the following modules: An acquisition module is used to acquire the content of the doctor's online consultation, which includes the doctor's questions asked to the patient and the patient's answer information; A semantic analysis module, used to perform semantic analysis on the content of the doctor's online consultation and extract target symptom information for diagnosis; A visual diagnosis prompt module, used to determine the symptom elements to be obtained through visual diagnosis based on the target symptom information, generate image acquisition prompt information, and prompt the patient to upload the corresponding image information; An inference module is used to combine the target symptom information with the image information uploaded by the patient to infer the auscultation information and palpation information that need to be further obtained, so as to form a set of diagnostic elements that need to be supplemented; An availability analysis module, used to perform an availability analysis on the set of diagnostic elements to be supplemented, and determine the content that can be obtained by asking questions and / or collecting images; The supplementary prompt generation module is used to generate supplementary diagnostic questions and / or supplementary image acquisition prompt information based on the availability analysis results, and guide the patient to provide supplementary information related to auscultation and palpation.

7. The question-answer management system of the traditional Chinese medical consultation system according to claim 6, characterized in that: The semantic analysis module includes: a natural language processing unit, which is used to perform word segmentation, keyword extraction, semantic classification and standardization on medical questions and answer information; a symptom extraction unit, which is used to extract the patient's main symptoms, accompanying symptoms and their corresponding location characteristics, time characteristics and trait characteristics in combination with the traditional Chinese medicine symptom terminology database.

8. The question-answer management system of the traditional Chinese medical consultation system according to claim 6, characterized in that: The visual diagnosis prompt module includes: a rule matching unit, which is used to call the association rule library between symptoms and visual diagnosis elements to match the target symptom information; an image prompt generation unit, which is used to generate image acquisition prompt information including image acquisition part, image acquisition requirements and image upload method, and prompt the patient to upload the image information of the corresponding part in the user interface.

9. The question-answer management system of the traditional Chinese medical consultation system according to claim 6, characterized in that: The inference module includes: a syndrome reasoning unit, which is used to call the symptom-syndrome-four-diagnosis information association rule library and determine the syndrome candidate set through forward reasoning; an element missing judgment unit, which is used to identify the missing auscultation or palpation elements in the current diagnosis and treatment information and record them as a set of diagnostic elements that need to be supplemented.

10. The question-answer management system of the traditional Chinese medical consultation system according to claim 6, characterized in that: The availability analysis module includes: a mapping table calling unit, which is used to call a preset diagnostic element-acquisition method mapping table; an acquisition method determination unit, which is used to determine whether each diagnostic element can be acquired through medical questioning, image acquisition, or a combination of the two; and a supplementary prompt generation module, which is used to automatically generate corresponding supplementary medical questioning or image acquisition prompt information based on the analysis results.

Images