Traditional Chinese medicine digital public service platform

Abstract

The invention provides a traditional Chinese medicine digital public service platform, and relates to the technical field of digital public service platforms. Wherein the multi-party consultation module comprises a consultation request unit for receiving a consultation request of a user; the data verification unit verifies request data corresponding to the consultation request, and constructs a consultation task when the verification is passed; the execution unit executes the consultation task and constructs a consultation virtual space; after the construction of the consultation virtual space is completed and the execution of the consultation task is completed, the access unit permits the user to enter the consultation virtual space; when the user enters the consultation virtual space, the consultation process control unit controls information access and output of each terminal entering the consultation virtual space based on a consultation process synchronously generated when the consultation virtual space is constructed; and the consultation receipt unit records the consultation process, generates a process record and sends the process record to the user. According to the invention, the multi-party consultation module is used for providing multi-party consultation suitable for the user demand for the user.

Description

Technical Field

[0001] This invention relates to the field of digital public service platform technology, and in particular to a digital public service platform for traditional Chinese medicine. Background Technology

[0002] The digital public service platform for traditional Chinese medicine (TCM) is a comprehensive service system built upon technologies such as big data and artificial intelligence, aiming to improve the efficiency and accessibility of TCM services. Its core functions include intelligent diagnosis and treatment, and knowledge services.

[0003] Among them, the knowledge service mainly integrates resources such as ancient Chinese medicine books and famous doctors' experience to provide knowledge base query services; intelligent diagnosis and treatment conducts health assessments based on data uploaded by users to obtain assessment data to assist doctors in their diagnosis and treatment activities; in some cases, users may require multi-party consultations, but the existing intelligent diagnosis and treatment cannot meet this need. Summary of the Invention

[0004] One of the objectives of this invention is to provide a digital public service platform for traditional Chinese medicine, which provides users with multi-party consultations tailored to their needs through a multi-party consultation module.

[0005] This invention provides a digital public service platform for traditional Chinese medicine, including: a multi-party consultation module; The multi-party consultation module includes: The consultation request unit receives the user's consultation request; the data verification unit verifies the request data corresponding to the consultation request, and when the verification is successful, a consultation task is constructed; the execution unit executes the consultation task and constructs a consultation virtual space; when the consultation virtual space is constructed and the consultation task is completed, the access unit allows the user to enter the consultation virtual space; when the user enters the consultation virtual space, the consultation process control unit controls the information access and output of each terminal entering the consultation virtual space based on the consultation process synchronously generated when the consultation virtual space is constructed; the consultation receipt unit records the consultation process and generates a process record to send to the user.

[0006] Preferably, the data verification unit performs the following operations: The requested data is parsed and filtered to determine the case data and examination data; The case data and examination data are validated based on the pre-configured first validation rule; And / or, The case data and examination data were analyzed to obtain the first analysis results; The request category corresponding to the consultation request is verified based on the first analysis result.

[0007] Preferably, the execution unit performs the following operations: Analyze the consultation task to determine the consultation items and required parameters; Based on the consultation project and demand parameters, determine the terminals to participate in the consultation from the current consultation selection list; these terminals will be the target terminals. Send a consultation participation inquiry to the target terminal, and when a positive response is received, establish a communication connection with the target terminal and construct a virtual consultation space.

[0008] Preferably, the execution unit determines the terminals participating in the consultation from the current consultation selection list based on the consultation project and requirement parameters; the target terminals include: When the parameter item corresponding to the specified person in the demand parameters is empty, the nearest terminal is determined from the consultation selection list based on the user's location as the first target terminal; When the parameter item corresponding to the specified person in the requirement parameters is not empty, the terminal pointed to by the parameter item is used as the first target terminal; Based on the parameters in the demand parameters corresponding to the number of consultation experts required, and the correlation between each terminal in the consultation selection list and the selected target terminal, the remaining target terminals are determined.

[0009] Preferably, the steps for determining the correlation between the terminals in the consultation selection list are as follows: The location between terminals is quantized to obtain the first correlation value; Analyze the correlation between the historical consultation records of the two terminals to obtain the second correlation value; Analyze the correlation between the history of the two terminals to obtain a third correlation value; The correlation degree is obtained by weighting and summing the first, second, and third correlation values.

[0010] Preferably, the digital public service platform for traditional Chinese medicine also includes: an intelligent diagnosis and treatment unit, which is used as one of the terminals participating in the consultation when constructing the virtual consultation space, and is also used to perform preliminary analysis based on the requested data to obtain preliminary analysis results.

[0011] Preferably, the digital public service platform for traditional Chinese medicine also includes: a docking module for docking with various external systems, providing standardized data access interfaces, and uniformly managing application access control, load balancing, caching, and security functions.

[0012] Preferably, the digital public service platform for traditional Chinese medicine also includes: a data management module for business database segmentation and read / write separation; the business database is divided into a medical record database and a medical knowledge base, and a traditional Chinese medicine knowledge graph is constructed on this basis.

[0013] Preferably, the digital public service platform for traditional Chinese medicine also includes: The multimodal data preprocessing and feature extraction module is responsible for standardizing and extracting features from the four diagnostic methods of observation, auscultation, inquiry, and palpation, providing high-quality input for subsequent intelligent analysis. The multimodal data preprocessing and feature extraction module includes: The visual diagnosis data processing unit uses a fully convolutional network algorithm for intelligent segmentation and recognition of tongue and facial image data; The auscultation data processing unit uses a pre-trained audio model to extract acoustic features; The consultation data processing unit uses a domain-adaptive pre-trained language model to process text information of patients' chief complaints and symptom descriptions. The palpation data processing unit uses a pre-trained time series model to extract pulse waveform features.

[0014] Preferably, the digital public service platform for traditional Chinese medicine also includes: The multimodal information fusion and diagnosis module is used to fuse and diagnose the multimodal features extracted by the multimodal data preprocessing and feature extraction module. It can also be connected to the intelligent diagnosis and treatment unit during multi-party consultations to participate in the analysis of preliminary results. The Personalized Intervention and Health Management module is used to output personalized intervention recommendations based on the preliminary analysis results.

[0015] The present invention has the following beneficial effects: The multi-party consultation module provides users with multi-party consultations tailored to their needs; when responding to users' multi-party consultation requests, it verifies the case data and examination data provided by users to ensure that the requirements of multi-party consultation are met; when selecting terminals to participate in multi-party consultations, it introduces correlation to ensure the diversity of opinions during multi-party consultations and to ensure the effective conduct of multi-party consultations.

[0016] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description and the accompanying drawings.

[0017] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0018] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings: Figure 1 This is a schematic diagram of a digital public service platform for traditional Chinese medicine in an embodiment of the present invention; Figure 2This is a schematic diagram illustrating the steps for determining the correlation degree in an embodiment of the present invention.

[0019] In the picture: 1. Consultation request unit; 2. Data verification unit; 3. Execution unit; 4. Access unit; 5. Consultation process control unit; 6. Consultation receipt unit. Detailed Implementation

[0020] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0021] Example 1: This embodiment of the invention provides a digital public service platform for traditional Chinese medicine, including: a multi-party consultation module; wherein, as... Figure 1 As shown, the multi-party consultation module includes: Consultation request unit 1 receives the user's consultation request; data verification unit 2 verifies the request data corresponding to the consultation request, and when the verification is successful, a consultation task is constructed; execution unit 3 executes the consultation task and constructs a consultation virtual space; when the consultation virtual space is constructed and the consultation task is completed, access unit 4 allows the user to enter the consultation virtual space; when the user enters the consultation virtual space, consultation process control unit 5 controls the information access and output of each terminal entering the consultation virtual space based on the consultation process synchronously generated when the consultation virtual space is constructed; consultation receipt unit 6 records the consultation process and generates a process record to be sent to the user.

[0022] This embodiment constructs a consultation virtual space based on the user's consultation request through consultation request unit 1, data verification unit 2, and execution unit 3. In the consultation virtual space, multiple experts can use terminals to conduct multi-party consultations with the user. During the consultation, the consultation process control unit 5 controls the consultation process within the consultation virtual space. The consultation virtual space can be specifically implemented as either a multi-party video conference or a multi-party audio conference. The consultation process includes: sorting and numbering each terminal, confirming whether the personnel corresponding to each terminal are online, sending the case data and examination data parsed from the request data to each terminal, after sending the case data and examination data and waiting for a preset time (any of 2 to 20 minutes), obtaining the input information of each terminal in numerical order and synchronizing it to the user, and receiving the information input by the user through the terminal after each terminal's information is transmitted to the user's terminal and establishing a discussion and communication channel with the terminal corresponding to the information transmitted to the user's previous terminal.

[0023] The data verification unit performs the following operations: The requested data is parsed and filtered to determine the case data and examination data; The case data and examination data are validated based on the pre-configured first validation rule; And / or, The case data and examination data were analyzed to obtain the first analysis results; The request category corresponding to the consultation request is verified based on the first analysis result.

[0024] When a user requires multi-party consultation, case data and examination data are input through the terminal. After input, a consultation request (multi-party consultation) is issued through the terminal based on the input data, and the data is embedded in the request data corresponding to the consultation request. The request data can be directly parsed and filtered to obtain case data and examination data. The first verification rule is matched one-to-one with the request category corresponding to the consultation request, including: an examination item table for examination data verification and the time permission range corresponding to each item in the examination item table, and a case verification item for case data verification. The case verification item includes: whether the etiology, examination and conclusion corresponding to each historical case data is complete, and whether the total number of historical case data is greater than or equal to a preset value. The verification of case data and examination data mainly verifies completeness and timeliness, that is, whether the case data and examination data are complete and whether the examination is recent, thereby ensuring the accuracy and effectiveness of the consultation. The first analysis result is mainly a preliminary judgment of the user's symptoms. The preliminary judgment determines whether it is consistent with the symptoms corresponding to the request category. If they are inconsistent, the multi-party consultation request is rejected. The execution unit performs the following operations: The consultation task is analyzed to determine the consultation items and required parameters. The consultation items correspond to the user's symptoms. The required parameters include: parameters indicating the designated personnel, parameters indicating the minimum number of terminals to participate in the multi-party consultation, parameters indicating the maximum number of terminals to participate in the multi-party consultation, and parameters indicating whether virtual terminals are required. Based on the consultation project and demand parameters, determine the terminals to participate in the consultation from the current consultation selection list; these terminals will be the target terminals. Send a consultation participation inquiry to the target terminal, and when a positive response is received, establish a communication connection with the target terminal and construct a virtual consultation space.

[0025] The execution unit determines the participating terminals from the current consultation selection list based on the consultation project and requirement parameters; the target terminals include: When the parameter item corresponding to the specified personnel in the demand parameters is empty, the nearest terminal is determined from the consultation selection list based on the user's location as the first target terminal; the location of the terminal and the user's location can be obtained by analyzing their respective location data, and then the distance between the two can be determined; when not specified, the nearest principle is used to facilitate direct connection with the personnel of the nearest terminal when the user needs on-site guidance after multi-party consultation. When the parameter item corresponding to the specified person in the requirement parameters is not empty, the terminal pointed to by the parameter item is used as the first target terminal; Based on the parameters in the demand parameters corresponding to the number of consultation experts required, and the correlation between each terminal in the consultation selection list and the selected target terminals, the remaining target terminals are determined. After determining the first target terminal, the terminal with the lowest correlation with the first target terminal is selected as the second, and then other terminals are determined sequentially based on the principle of having the lowest sum of correlation with the determined target terminals.

[0026] Among them, such as Figure 2 As shown, the steps for determining the correlation between the various terminals in the consultation selection list are as follows: Step 1: Quantize the positions between terminals to obtain the first correlation value; The first correlation value is obtained by standardizing the position between terminals, that is, the first correlation value is proportional to the distance; Step 2: Analyze the correlation between the historical consultation records of the two terminals to obtain the second correlation value; The second correlation value is obtained by quantitative analysis of the correlation between the historical consultation records of the two terminals. It mainly analyzes the ratio of the historical consultations jointly participated in by the two terminals to the total number of historical consultations participated in by the two terminals, and then determines the second correlation value. Step 3: Analyze the correlation between the history of the two terminals to obtain the third correlation value; By analyzing whether there are common projects in the resumes of the personnel corresponding to the two terminals, the common projects are quantified and weighted to obtain the third association value; Step 4: Calculate the weighted sum of the first, second, and third correlation values ​​to obtain the correlation degree. By comprehensively analyzing location, historical consultation records, and user history (of the terminal), the correlation degree is obtained. This degree guides multi-party consultations, ensuring their accuracy and effectiveness.

[0027] Example 2: This embodiment of the invention provides a digital public service platform for traditional Chinese medicine, including: a multi-party consultation module; The multi-party consultation module includes: The consultation request unit receives the user's consultation request; the data verification unit verifies the request data corresponding to the consultation request, and when the verification is successful, a consultation task is constructed; the execution unit executes the consultation task and constructs a consultation virtual space; when the consultation virtual space is constructed and the consultation task is completed, the access unit allows the user to enter the consultation virtual space; when the user enters the consultation virtual space, the consultation process control unit controls the information access and output of each terminal entering the consultation virtual space based on the consultation process synchronously generated when the consultation virtual space is constructed; the consultation receipt unit records the consultation process and generates a process record to send to the user.

[0028] In addition, the digital public service platform for traditional Chinese medicine also includes: an intelligent diagnosis and treatment unit, which is used as one of the terminals participating in the consultation when constructing the virtual consultation space, and is also used to perform preliminary analysis based on the requested data to obtain preliminary analysis results.

[0029] This embodiment provides an intelligent diagnosis and treatment unit during multi-party consultations, outputting preliminary analysis results of intelligent diagnosis and treatment to participate in multi-party consultations, thereby improving the accuracy and effectiveness of consultations.

[0030] To address situations where participating terminals fail to reach a consensus for an extended period during multi-party consultations, the first step is to assess this situation. A threshold can be configured, ranging from half an hour to two hours; exceeding this threshold indicates a prolonged failure to reach an agreement. Once this is identified, an arbitration request is submitted through the arbitration request unit of the Traditional Chinese Medicine Digital Public Service Platform. A pre-configured list of terminals with arbitration qualifications is used to select a predetermined number of terminals to form an arbitration group, which is then integrated into the multi-party consultation. The terminals in the arbitration qualification list are configured through the following analysis: analysis of the personnel qualifications associated with each terminal (which can be done through recommendations from other terminals), and / or analysis of the terminal's multi-party consultation records (ensuring that the number of participations, the number of opinions expressed, and the number of points of agreement in the final opinion all meet certain criteria).

[0031] In addition, the multi-party consultation module also includes: a mid-intervention unit, which is used to receive mid-intervention requests from licensed terminals in the list of licensed terminals configured for each terminal participating in the multi-party consultation, and then authorize the corresponding terminal to join the consultation virtual space as an intervention terminal, and / or, receive mid-intervention requests from each terminal participating in the multi-party consultation and designate the terminal as an intervention terminal, send an inquiry to the intervention terminal, and authorize the intervention terminal to join the meeting virtual space when a positive feedback is received; The interventional terminal is associated with the corresponding terminal participating in the multi-party consultation; after the interventional terminal is added to the conference virtual space, it acts as an observer terminal of the associated terminal (it can view information, but cannot send information). After receiving a request to exchange identities with the listening terminal and / or associated terminal, the system sends an inquiry to another terminal. Upon receiving a positive response, the system exchanges identities with the listening terminal and associated terminal.

[0032] By intervening midway and exchanging roles, the personnel in the multi-party consultation are replaced by those led by the participating terminal. This is to ensure that the multi-party consultation can continue even if the participating terminal personnel need to deal with an emergency and are unable to continue the consultation.

[0033] Example 3: This embodiment of the invention provides a digital public service platform for traditional Chinese medicine, including: a multi-party consultation module; The multi-party consultation module includes: The consultation request unit receives the user's consultation request; the data verification unit verifies the request data corresponding to the consultation request, and when the verification is successful, a consultation task is constructed; the execution unit executes the consultation task and constructs a consultation virtual space; when the consultation virtual space is constructed and the consultation task is completed, the access unit allows the user to enter the consultation virtual space; when the user enters the consultation virtual space, the consultation process control unit controls the information access and output of each terminal entering the consultation virtual space based on the consultation process synchronously generated when the consultation virtual space is constructed; the consultation receipt unit records the consultation process and generates a process record to send to the user.

[0034] To enable data integration with other systems, the TCM digital public service platform also includes an integration module, which connects with various external systems, provides standardized data access interfaces, and uniformly manages application access control, load balancing, caching, and security functions.

[0035] To manage data within the platform, the TCM Digital Public Service Platform also includes: a data management module for business database partitioning and read / write separation; dividing the business database into a medical record database and a medical knowledge base, and building a TCM knowledge graph on top of these. This employs a master-slave high-availability architecture for business database partitioning and read / write separation. The data layer ensures accuracy, integrity, and multimodal availability of data through a rigorous data quality control process, including data acquisition, cleaning, annotation, and management. Furthermore, it includes a capability management module, providing core AI capabilities such as knowledge graphs, natural language processing, and diagnostic inference algorithms to the application side. This layer integrates various algorithms, including the Multi-Feature Fusion Health Detection Pre-trained Model (MFFHPT) and Bayesian network probabilistic inference models, supporting the platform's intelligent diagnosis and decision-making functions. A business application management module provides application services including a remote consultation platform, intelligent assisted diagnosis, intelligent prescription review, and health management. This layer is based on a microservice architecture, ensuring that each functional module can be deployed independently and scaled elastically.

[0036] To handle requests for multi-party consultations and daily user-uploaded data, the Traditional Chinese Medicine Digital Public Service Platform also includes: The multimodal data preprocessing and feature extraction module is responsible for standardizing and extracting features from the four diagnostic methods of observation, auscultation, inquiry, and palpation, providing high-quality input for subsequent intelligent analysis. The multimodal data preprocessing and feature extraction module includes: The visual diagnosis data processing unit employs a fully convolutional network algorithm for intelligent segmentation and recognition of tongue and facial image data. It further extracts features such as color, texture, and shape, with key features like tongue color, coating color, tongue texture, and coating texture quantified according to the "Classification and Determination of Traditional Chinese Medicine Constitution" standard. For infrared thermal imaging information, a specific pre-trained model (such as ThermalGAN) is used to process thermal imaging features and extract temperature distribution and abnormal heat zone information. The auscultation data processing unit uses pre-trained audio models to extract acoustic features; audio models include VGGish or Wav2Vec 2.0; acoustic features include parameters such as pitch and rhythm; used to analyze abnormalities in the patient's voice (such as coughing and wheezing). Signal processing techniques are used to reduce noise and enhance features in the sound signals to improve recognition accuracy. The consultation data processing unit uses a domain-adaptive pre-trained language model to process text information such as patient complaints and symptom descriptions; for example, it uses a Traditional Chinese Medicine BERT model to process text information such as patient complaints and symptom descriptions. Through intelligent consultation functions, it obtains intelligent follow-up questions and recommendations, and combines ASR speech recognition technology to achieve real-time transcription of doctor-patient dialogue content, extracting key symptom and examination indicator information. The palpation data processing unit uses a pre-trained time-series model to extract pulse waveform features. This involves using a pre-trained time-series model (such as LSTM or Transformer) to extract pulse waveform features, including parameters such as frequency, amplitude, and rhythm. Modern signal processing techniques are then used to automatically analyze and identify the pulse, providing the pulse diagnosis result and pulse name.

[0037] To address the small sample size of TCM data, the multimodal data preprocessing and feature extraction module employs a pre-trained model fine-tuning strategy. By utilizing a model pre-trained on a large-scale general dataset and fine-tuning it with TCM data, the module effectively solves the problem of insufficient learning of modal representations due to the small number of training samples.

[0038] To enable preliminary diagnosis, the digital public service platform for traditional Chinese medicine also includes: The multimodal information fusion and diagnosis module is used to fuse and diagnose the multimodal features extracted by the multimodal data preprocessing and feature extraction module. It can also be connected to the intelligent diagnosis and treatment unit during multi-party consultations to participate in the analysis of preliminary results. The multimodal information fusion and diagnosis module employs attention-based multimodal fusion: This method calculates the attention weights of each modality based on the obtained feature representations of each individual modality. Specifically, for each modality's feature vector, an attention score is calculated using an attention network (such as a multilayer perceptron), normalized using a softmax function, and then the features of each modality are weighted and summed to obtain the fused representation. This method not only improves the fusion effect but also clearly demonstrates the contribution of each modality to diagnosis, enhancing the model's interpretability. A multi-task learning framework is implemented: By introducing related modal sub-tasks into the learning process of the main disease diagnosis task, the model's generalization ability is improved. Specific task designs include: Inspection sub-task: tongue image classification or facial color recognition; Auscultation sub-task: sound type classification (e.g., cough, wheezing); Inquiry sub-task: symptom entity recognition or sentiment analysis; Palpation sub-task: pulse waveform classification (e.g., floating pulse, deep pulse). A hard parameter sharing mechanism is adopted, with a shared encoder at the bottom layer of the model extracting common features; and a task-specific head at the top layer, including a main task head and sub-task heads. All tasks are simultaneously optimized using weighted and multi-task loss functions to balance their importance. Probabilistic reasoning based on Bayesian networks: A probabilistic reasoning model is built based on Bayesian networks to achieve intelligent computer identification of TCM syndromes. Network nodes represent different symptoms, signs, and syndromes. Conditional probability tables describe the relationships between nodes. Combining prior probabilities and observational evidence, the posterior probabilities of various syndromes are calculated, providing quantitative references for doctors' diagnoses. Diagnostic result generation: Feature inputs are fused into a classifier (e.g., multilayer perceptron + softmax) for disease or syndrome classification. The output includes constitution type, syndrome name, and corresponding probability distribution, while also providing contribution analysis of different modalities to enhance the interpretability of the results.

[0039] The personalized intervention and health management module outputs personalized intervention recommendations based on preliminary analysis results. Based on diagnostic results, it provides patients with personalized treatment and health management plans: Model compression based on knowledge distillation: Using knowledge distillation technology, a large teacher model (such as a multi-diagnosis, multi-parameter model) is trained into a small student model. By minimizing the KL divergence loss of soft labels, the student model retains the performance of the teacher model but has fewer parameters. The distilled model is easier to interpret and deploy on portable devices, adapting to home and community use scenarios. Intelligent evolutionary algorithm generates intervention recommendations: A genetic algorithm is used as a heuristic stochastic optimization strategy to generate optimal clinical intervention recommendations. Intervention recommendations (such as traditional Chinese medicine prescriptions and acupuncture programs) are encoded into chromosomes, with each gene representing an intervention parameter (such as herbal dosage and treatment frequency). The fitness function is designed based on the patient outcomes predicted by the model, comprehensively considering the probability of recovery, intervention cost, and side effect risk. The population is iteratively optimized through selection, crossover, and mutation operations, ultimately outputting the intervention plan with the highest fitness. Personalized health management: Personalized health management plans are recommended based on the user's medical history, and health and wellness suggestions are provided in conjunction with their physical constitution. Based on test data, it generates a combined constitution chart and a five-organ early warning chart, intuitively displaying the user's health status. Based on seasonal solar terms and sub-health big data, it predicts the user's health status and intelligently recommends a systematic and personalized conditioning plan including diet, exercise, emotional well-being, moxibustion, massage, and topical treatments. Full-process health tracking: It establishes a health record for the user, saves all test information, and allows real-time viewing across multiple devices, enabling health early warnings, health data analysis, and comprehensive health management. It can connect to blood pressure, heart rate, blood sugar, weight, and other monitoring devices, forming a complete personal health data ecosystem.

[0040] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A digital public service platform for traditional Chinese medicine, characterized in that, include: Multi-party consultation module; The multi-party consultation module includes: The consultation request unit receives the user's consultation request; the data verification unit verifies the request data corresponding to the consultation request, and when the verification is successful, a consultation task is constructed; the execution unit executes the consultation task and constructs a consultation virtual space; when the consultation virtual space is constructed and the consultation task is completed, the access unit allows the user to enter the consultation virtual space; when the user enters the consultation virtual space, the consultation process control unit controls the information access and output of each terminal entering the consultation virtual space based on the consultation process synchronously generated when the consultation virtual space is constructed; the consultation receipt unit records the consultation process and generates a process record to send to the user.

2. The digital public service platform for traditional Chinese medicine as described in claim 1, characterized in that, The data validation unit performs the following operations: The requested data is parsed and filtered to determine the case data and examination data; The case data and examination data are validated based on the pre-configured first validation rule; And / or, The case data and examination data were analyzed to obtain the first analysis results; The request category corresponding to the consultation request is verified based on the first analysis result.

3. The digital public service platform for traditional Chinese medicine as described in claim 1, characterized in that, The execution unit performs the following operations: Analyze the consultation task to determine the consultation items and required parameters; Based on the consultation project and demand parameters, determine the terminals to participate in the consultation from the current consultation selection list; these terminals will be the target terminals. Send a consultation participation inquiry to the target terminal. When a positive response is received, establish a communication connection with the target terminal and construct a virtual consultation space.

4. The digital public service platform for traditional Chinese medicine as described in claim 3, characterized in that, The execution unit determines the terminals to participate in the consultation from the current consultation selection list based on the consultation project and requirement parameters; As the target terminal, it includes: When the parameter item corresponding to the specified person in the demand parameters is empty, the nearest terminal is determined from the consultation selection list based on the user's location as the first target terminal; When the parameter item corresponding to the specified person in the requirement parameters is not empty, the terminal pointed to by the parameter item is used as the first target terminal; Based on the parameters in the demand parameters corresponding to the number of consultation experts required, and the correlation between each terminal in the consultation selection list and the selected target terminal, the remaining target terminals are determined.

5. The digital public service platform for traditional Chinese medicine as described in claim 4, characterized in that, The steps for determining the correlation between the various terminals in the consultation selection list are as follows: The location between terminals is quantized to obtain the first correlation value; Analyze the correlation between the historical consultation records of the two terminals to obtain the second correlation value; Analyze the correlation between the history of the two terminals to obtain a third correlation value; The correlation degree is obtained by weighting and summing the first, second, and third correlation values.

6. The digital public service platform for traditional Chinese medicine as described in claim 1, characterized in that, Also includes: The intelligent diagnosis and treatment unit is used as one of the terminals participating in the consultation when constructing the virtual consultation space, and is also used to perform preliminary analysis based on the requested data to obtain preliminary analysis results.

7. The digital public service platform for traditional Chinese medicine as described in claim 1, characterized in that, Also includes: The interface module is used to interface with various external systems, providing standardized data access interfaces and unified management of application access control, load balancing, caching and security functions.

8. The digital public service platform for traditional Chinese medicine as described in claim 1, characterized in that, Also includes: The data management module is used for business database sharding and read / write separation; The business database is divided into a medical record database and a medical knowledge base, and a traditional Chinese medicine knowledge graph is built on this basis.

9. The digital public service platform for traditional Chinese medicine as described in claim 1, characterized in that, Also includes: The multimodal data preprocessing and feature extraction module is responsible for standardizing and extracting features from the information obtained from the four diagnostic methods of observation, auscultation, inquiry, and palpation, providing high-quality input for subsequent intelligent analysis. The multimodal data preprocessing and feature extraction module includes: The visual diagnosis data processing unit uses a fully convolutional network algorithm for intelligent segmentation and recognition of tongue and facial image data; The auscultation data processing unit uses a pre-trained audio model to extract acoustic features; The consultation data processing unit uses a domain-adaptive pre-trained language model to process text information of patients' chief complaints and symptom descriptions. The palpation data processing unit uses a pre-trained time series model to extract pulse waveform features.

10. The digital public service platform for traditional Chinese medicine as described in claim 9, characterized in that, Also includes: The multimodal information fusion and diagnosis module is used to fuse and diagnose the multimodal features extracted by the multimodal data preprocessing and feature extraction module. It can also be connected to the intelligent diagnosis and treatment unit during multi-party consultations to participate in the analysis of preliminary results. The Personalized Intervention and Health Management module is used to output personalized intervention recommendations based on the preliminary analysis results.

Images