A traditional Chinese medicine personalized diagnosis and treatment data summary processing method, system and program product

Abstract

The application belongs to the technical field of data management, and specifically discloses a traditional Chinese medicine individualized diagnosis and treatment data summary processing method, system and program product, which can grasp the deep and nonlinear relationship between traditional Chinese medicine diagnosis and treatment symptoms and syndromes, realize deep semantic understanding, significantly improve data matching accuracy, classify scattered cases according to syndromes, dynamically absorb the individualized experience of different traditional Chinese medicine doctors, continuously evolve the experience database, ensure the safety and privacy of traditional Chinese medicine diagnosis and treatment data, provide a technical basis for diagnosis and treatment knowledge tracing and right confirmation, facilitate decryption of the individualized diagnosis and treatment knowledge of a specific doctor group, and help digital inheritance of traditional Chinese medicine diagnosis and treatment experience. Through innovative feature fusion and similarity calculation, the application can realize deep simulation of traditional Chinese medicine diagnosis and treatment logic, and through a dynamic and safe encryption archiving mechanism, an experience database capable of continuous growth and effectively organizing traditional Chinese medicine doctor individualized diagnosis and treatment data is formed.

Description

Technical Field

[0001] This invention belongs to the field of data management technology, specifically relating to a method, system, and program product for summarizing and processing personalized TCM diagnosis and treatment data. Background Technology

[0002] With the development of the "Internet + Traditional Chinese Medicine (TCM) diagnosis and treatment" system, a large amount of TCM diagnosis and treatment data has been digitally recorded. This data contains each TCM practitioner's unique diagnostic and treatment principles and medication experience, representing a valuable intellectual asset. However, current medical diagnosis and treatment data management systems still face the following problems in managing this type of personalized TCM diagnosis and treatment knowledge:

[0003] 1. Data silos and the challenge of summarization: TCM diagnosis and treatment data is highly unstructured, and the diagnosis and treatment experience of different TCM doctors varies, making it difficult to effectively aggregate and summarize knowledge.

[0004] 2. Low efficiency of knowledge reuse: Existing systems mostly use simple keyword matching and database query methods, which cannot deeply analyze the complex internal relationship between symptoms and diagnosis, resulting in low accuracy when matching similar historical cases for new cases.

[0005] 3. Rigid knowledge storage methods: Traditional experience bases usually store diagnosis and treatment data independently in the form of isolated cases, lacking effective knowledge extraction and structured summarization.

[0006] 4. Weak protection of knowledge and privacy: The personalized treatment experience of TCM doctors is their core knowledge, but the existing system lacks a mechanism to strongly link doctors' identity information with encrypted core knowledge assets, which is not conducive to the traceability of doctors' knowledge and the protection of their rights.

[0007] Therefore, there is an urgent need for a personalized TCM diagnosis and treatment data management method that can deeply analyze TCM diagnostic and treatment information, intelligently match similar cases, and safely and dynamically build and enrich a TCM experience knowledge base. Summary of the Invention

[0008] The purpose of this invention is to provide a method, system, and program product for summarizing and processing personalized TCM diagnosis and treatment data, in order to solve the above-mentioned problems existing in the prior art.

[0009] To achieve the above objectives, the present invention adopts the following technical solution:

[0010] Firstly, a method for summarizing and processing personalized TCM diagnosis and treatment data is provided, including:

[0011] Obtain personalized diagnosis and treatment datasets and identity information of TCM physicians. The personalized diagnosis and treatment datasets include patient symptom information, diagnosis result information, treatment plan information, TCM prescription information, and TCM annotation information.

[0012] The patient's symptom information is extracted and vectorized to obtain several symptom keyword vectors, and the co-occurrence strength between the symptom keyword vectors is determined. The diagnostic result information is extracted and vectorized to obtain diagnostic keyword vectors.

[0013] A symptom relationship graph is constructed based on the vectors of each symptom keyword and the strength of their co-occurrence relationships. The symptom relationship graph and the diagnostic keyword vectors are then used to construct a symptom-diagnosis heterogeneity graph.

[0014] The symptom diagnosis heterogeneity graph is input into a pre-trained graph neural network model for feature association analysis to obtain a fused feature vector;

[0015] The experience summary items in the experience base are matched using the fused feature vector and the symptom relationship graph. The matched experience summary items are used as the target experience summary items. The experience base contains several experience summary items. Each experience summary item is associated with a corresponding reference fused feature vector and a reference symptom relationship graph. Each experience summary item also contains a corresponding encrypted knowledge tree.

[0016] The treatment plan information, TCM prescription information, and TCM annotation information are encrypted using identity information to obtain encrypted experience entry data, and the identity verification code is calculated using the identity information.

[0017] The identity verification code is combined with the encrypted experience entry data and added as a new branch to the encrypted knowledge tree of the target experience summary project.

[0018] In one possible design, the patient symptom information is subjected to keyword extraction and vectorization to obtain several symptom keyword vectors, and the co-occurrence strength between the symptom keyword vectors is determined. Similarly, the diagnostic result information is subjected to keyword extraction and vectorization to obtain diagnostic keyword vectors, including:

[0019] Keyword extraction was performed on patient symptom information to obtain several symptom keywords, and the co-occurrence strength among each symptom keyword was determined. Keyword extraction was also performed on diagnostic result information to obtain diagnostic keywords.

[0020] Each symptom keyword is vectorized to obtain the corresponding symptom keyword vector. The co-occurrence strength between each symptom keyword is used as the co-occurrence strength between each symptom keyword vector. The diagnostic keywords are also vectorized to obtain the diagnostic keyword vector.

[0021] In one possible design, determining the co-occurrence strength among the symptom keywords includes:

[0022] Each symptom keyword is input into a pre-set TCM symptom co-occurrence relationship prior knowledge graph for prior knowledge analysis to determine the strength of the co-occurrence relationship between each symptom keyword. The TCM symptom co-occurrence relationship prior knowledge graph contains several symptom keywords and the strength of the co-occurrence relationship between each symptom keyword.

[0023] In one possible design, the symptom relationship graph includes several first nodes, which are connected by undirected edges. Each first node is associated with a corresponding symptom keyword vector, and each undirected edge is associated with the co-occurrence strength between the corresponding symptom keyword vectors. The construction of a symptom diagnosis heterogeneous graph using the symptom relationship graph and diagnostic keyword vectors includes:

[0024] The diagnostic keyword vectors are mapped to the corresponding second nodes, the second nodes are added to the symptom relationship graph, and directed edges are constructed from the second nodes to each first node to obtain the symptom diagnosis heterogeneous graph.

[0025] In one possible design, the step of matching corresponding experience summary items in the experience base using fused feature vectors and symptom relationship graphs, and using the matched experience summary items as target experience summary items, includes:

[0026] Retrieve the reference fusion feature vectors and reference symptom relationship diagrams corresponding to each experience summary item in the experience base;

[0027] Calculate the cosine similarity between the fused feature vector and the reference fused feature vector of the corresponding experience summary item, and use the graph kernel function to calculate the graph structure similarity between the symptom relationship graph and the reference symptom relationship graph of the corresponding experience summary item;

[0028] The comprehensive matching degree parameter of the corresponding experience summary project is obtained by weighted summing the cosine similarity between the fused feature vector and the reference fused feature vector of the corresponding experience summary project and the graph structure similarity between the symptom relationship graph and the reference symptom relationship graph of the corresponding experience summary project.

[0029] The experience summary items with the highest comprehensive matching degree parameters that exceed the set parameter threshold are selected as the target experience summary items.

[0030] In one possible design, the encryption of treatment plan information, traditional Chinese medicine prescription information, and traditional Chinese medicine annotation information using identity information yields encrypted experience entry data, including:

[0031] A three-element information group is formed by utilizing treatment plan information, traditional Chinese medicine prescription information, and traditional Chinese medicine annotation information;

[0032] Using the identity information as a key, the three-element information group is encrypted to obtain encrypted experience entry data.

[0033] In one possible design, the calculation of the identity verification code using identity information includes:

[0034] The identity information is hashed to obtain the corresponding hash value, and the hash value is used as the identity verification code.

[0035] Secondly, a personalized TCM diagnosis and treatment data summary and processing system is provided, comprising a data acquisition unit, a feature extraction unit, a graph construction unit, a feature analysis unit, an item matching unit, a data encryption unit, and a knowledge construction unit, wherein:

[0036] The data acquisition unit is used to acquire personalized diagnosis and treatment datasets and identity information of TCM physicians. The personalized diagnosis and treatment datasets include patient symptom information, diagnosis result information, treatment plan information, TCM prescription information, and TCM annotation information.

[0037] The feature extraction unit is used to extract and vectorize patient symptom information to obtain several symptom keyword vectors, and to determine the co-occurrence strength between symptom keyword vectors. It also extracts and vectorizes diagnostic result information to obtain diagnostic keyword vectors.

[0038] The graph construction unit is used to construct a symptom relationship graph based on each symptom keyword vector and the strength of their co-occurrence relationships, and to construct a symptom diagnosis heterogeneity graph using the symptom relationship graph and the diagnostic keyword vector;

[0039] The feature analysis unit is used to input the symptom diagnosis heterogeneity map into a pre-trained graph neural network model for feature association analysis to obtain a fused feature vector;

[0040] The project matching unit is used to match the corresponding experience summary project in the experience base using the fused feature vector and the symptom relationship graph, and to take the matched experience summary project as the target experience summary project. The experience base contains several experience summary projects, each of which is associated with a corresponding reference fused feature vector and a reference symptom relationship graph, and each experience summary project contains a corresponding encrypted knowledge tree.

[0041] The data encryption unit is used to encrypt treatment plan information, traditional Chinese medicine prescription information and traditional Chinese medicine annotation information using identity identification information to obtain encrypted experience entry data, and to calculate identity verification code using identity identification information.

[0042] The knowledge building unit is used to combine the identity verification code with the encrypted experience entry data and add it as a new branch to the encrypted knowledge tree of the target experience summary project.

[0043] Thirdly, a system for summarizing and processing personalized TCM diagnosis and treatment data is provided, including:

[0044] Memory, used to store instructions;

[0045] The processor is configured to read instructions stored in the memory and execute any one of the TCM personalized diagnosis and treatment data summary and processing methods described in the first aspect above, according to the instructions.

[0046] Fourthly, a computer-readable storage medium is provided, on which instructions are stored, which, when executed on a computer, cause the computer to perform any one of the methods for summarizing and processing personalized TCM diagnosis and treatment data as described in the first aspect. Simultaneously, a computer program product is also provided, which, when executed on a computer, performs any one of the methods for summarizing and processing personalized TCM diagnosis and treatment data as described in the first aspect.

[0047] Beneficial Effects: This invention, through the fusion feature extraction of graph neural networks, can grasp the deep, non-linear relationship between symptoms and syndromes in Traditional Chinese Medicine (TCM) diagnosis and treatment, achieving deep semantic understanding that surpasses simple keyword matching and significantly improves data matching accuracy. Centered on experience summary projects, it categorizes scattered cases according to syndromes and dynamically incorporates personalized experiences from different TCM physicians through an encrypted knowledge tree, enabling the experience base to continuously evolve. Employing differentiated encryption based on identity information ensures the security and privacy of TCM diagnosis and treatment data storage, while providing a technical foundation for tracing the source of diagnosis and treatment knowledge and confirming rights. It facilitates the decryption of personalized diagnosis and treatment knowledge of specific physician groups with physician authorization, contributing to the digital inheritance of TCM diagnosis and treatment experience. Through innovative feature fusion and similarity calculation, this invention can achieve deep simulation of TCM diagnosis and treatment logic and, through a dynamic and secure encrypted archiving mechanism, form a continuously growing experience base that effectively organizes personalized diagnosis and treatment data from TCM physicians. Attached Figure Description

[0048] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0049] Figure 1 This is a flowchart illustrating the method in Embodiment 1 of the present invention;

[0050] Figure 2 This is a schematic diagram of the system configuration in Embodiment 2 of the present invention;

[0051] Figure 3 This is a schematic diagram of the system configuration in Embodiment 3 of the present invention. Detailed Implementation

[0052] It should be noted that the descriptions of these embodiments are intended to aid in understanding the invention and do not constitute a limitation thereof. The specific structural and functional details disclosed herein are merely for describing exemplary embodiments of the invention. However, the invention may be embodied in many alternative forms and should not be construed as being limited to the embodiments described herein.

[0053] It should be understood that, unless otherwise explicitly specified and limited, the corresponding terms should be interpreted broadly. For example, "connection" can be a fixed connection, a detachable connection, or an integral connection; it can be a direct connection or an indirect connection through an intermediate medium; it can be a connection within two components. Those skilled in the art can understand the specific meaning of the above terms in the embodiments according to the specific circumstances.

[0054] Specific details are provided in the following description to provide a complete understanding of the exemplary embodiments. However, those skilled in the art will understand that the exemplary embodiments can be implemented without these specific details. For example, apparatus may be shown in block diagrams to avoid obscuring the examples with unnecessary details. In other embodiments, well-known processes, structures, and techniques may be omitted with non-essential details to avoid obscuring the embodiments.

[0055] Example 1:

[0056] This embodiment provides a method for summarizing and processing personalized TCM diagnosis and treatment data, which can be applied to relevant data servers, such as... Figure 1 As shown, the method includes the following steps:

[0057] S1. Obtain personalized diagnosis and treatment datasets and identity information of TCM physicians. The personalized diagnosis and treatment datasets include patient symptom information, diagnosis result information, treatment plan information, TCM prescription information, and TCM annotation information.

[0058] In practice, the server can first obtain the personalized diagnosis and treatment dataset and identity information of TCM physicians through the electronic medical record system. The personalized diagnosis and treatment dataset includes patient symptom information, diagnosis result information, treatment plan information, TCM prescription information, and TCM annotation information. For example, patient symptom information S = "distending pain in the hypochondrium, loss of appetite, and emotional depression"; diagnosis result information D = "liver stagnation and spleen deficiency syndrome"; treatment plan information T = "soothing the liver and strengthening the spleen"; TCM prescription information P = "Bupleurum, white peony root, atractylodes macrocephala, poria cocos..."; and TCM annotation information N = "This case focuses on harmonizing the liver and spleen; Bupleurum soothes the liver, and atractylodes macrocephala and poria cocos strengthen the spleen...". The identity information may include the TCM physician's ID or certificate number, etc.

[0059] S2. Extract and vectorize the patient's symptom information to obtain several symptom keyword vectors, determine the co-occurrence strength between the symptom keyword vectors, extract and vectorize the diagnostic results information to obtain diagnostic keyword vectors.

[0060] In specific implementation, after acquiring the personalized diagnosis and treatment dataset, the server can extract keywords from the patient's symptom information to obtain several symptom keywords and determine the co-occurrence strength between each symptom keyword. Specifically, this includes inputting each symptom keyword into a pre-constructed TCM symptom co-occurrence relationship prior knowledge graph for prior knowledge analysis to determine the co-occurrence strength between each symptom keyword. The TCM symptom co-occurrence relationship prior knowledge graph contains several symptom keywords and the co-occurrence strength between each symptom keyword. Then, keywords are extracted from the diagnostic result information to obtain diagnostic keywords. Next, each symptom keyword is vectorized to obtain the corresponding symptom keyword vector, and the co-occurrence strength between each symptom keyword is used as the co-occurrence strength between each symptom keyword vector. Simultaneously, the diagnostic keywords are vectorized to obtain diagnostic keyword vectors.

[0061] S3. Construct a symptom relationship graph based on the symptom keyword vectors and the strength of their co-occurrence relationships, and use the symptom relationship graph and diagnostic keyword vectors to construct a symptom diagnosis heterogeneity graph.

[0062] In practice, the server can construct a symptom relationship graph using each symptom keyword vector and the strength of their co-occurrence relationships. This graph contains several first nodes, connected by undirected edges, each associated with a corresponding symptom keyword vector, and each undirected edge associated with the strength of the co-occurrence relationship between the corresponding symptom keyword vectors. Then, the diagnostic keyword vectors are mapped to corresponding second nodes, which are added to the symptom relationship graph. Directed edges are then constructed from the second nodes to each first node, resulting in a symptom diagnosis heterogeneous graph.

[0063] S4. Input the symptom diagnosis heterogeneous graph into the pre-trained graph neural network model for feature association analysis to obtain the fused feature vector.

[0064] In practice, after constructing the symptom relationship graph and the symptom diagnosis heterogeneity graph, the server inputs the symptom diagnosis heterogeneity graph into a pre-trained graph neural network model (GNN, which is an algorithm framework based on deep learning to process graph structure data, and completes tasks such as classification, prediction and generation by extracting the features of nodes, edges and the graph as a whole) for feature association analysis to obtain a fused feature vector.

[0065] S5. Use the fusion feature vector and symptom relationship graph to match the corresponding experience summary items in the experience base, and take the matched experience summary items as the target experience summary items. The experience base contains several experience summary items, each of which is associated with a corresponding reference fusion feature vector and reference symptom relationship graph, and each experience summary item contains a corresponding encrypted knowledge tree.

[0066] In practice, the server can retrieve the reference fusion feature vector and reference symptom relationship graph corresponding to each experience summary item in the experience database; then, it calculates the cosine similarity between the fusion feature vector and the reference fusion feature vector of the corresponding experience summary item, and uses a graph kernel function to calculate the graph structure similarity between the symptom relationship graph and the reference symptom relationship graph of the corresponding experience summary item; then, it weights and sums the cosine similarity between the fusion feature vector and the reference fusion feature vector of the corresponding experience summary item and the graph structure similarity between the symptom relationship graph and the reference symptom relationship graph of the corresponding experience summary item to obtain the comprehensive matching degree parameter of the corresponding experience summary item; finally, the experience summary item with the highest comprehensive matching degree parameter that exceeds the set parameter threshold (e.g., 0.85) is taken as the target experience summary item.

[0067] S6. Use the identity information to encrypt the treatment plan information, TCM prescription information, and TCM annotation information to obtain encrypted experience entry data, and use the identity information to calculate the identity verification code.

[0068] In practice, the server first uses the treatment plan information T, the traditional Chinese medicine prescription information P, and the traditional Chinese medicine annotation information N to form a three-element information group.<T,P,N> Then, the identity information is used as a key to pair the three-element information group.<T,P,N> Encryption is performed to obtain encrypted experience entry data. Simultaneously, the identity information can be hashed to obtain the corresponding hash value, which is then used as the identity verification code.

[0069] S7. Combine the identity verification code with the encrypted experience entry data and add it as a new branch to the encrypted knowledge tree of the target experience summary project.

[0070] In practice, the server ultimately combines the identity verification code with the encrypted experience entry data and adds it as a new branch to the encrypted knowledge tree of the target experience summary project. In subsequent data access, when it is necessary to retrieve the encrypted experience entry data of each TCM physician under the corresponding experience summary project in the experience database, and after obtaining the authorization and identity information of the corresponding TCM physician, the server can first perform a hash calculation using the identity information. The resulting hash value is then verified using the identity verification code under each branch in the encrypted knowledge tree of the specified experience summary project. The encrypted experience entry data corresponding to the verified branch is retrieved, and decrypted using the identity information to obtain the corresponding three-element information group, thus achieving secure and effective data utilization.

[0071] This method, through fusion feature extraction using graph neural networks, can grasp the deep, non-linear relationship between symptoms and syndromes in Traditional Chinese Medicine (TCM) diagnosis and treatment, achieving deep semantic understanding that surpasses simple keyword matching and significantly improves data matching accuracy. Centered on experience summary projects, it categorizes scattered cases by syndrome and dynamically incorporates personalized experiences from different TCM physicians through an encrypted knowledge tree, enabling the experience database to continuously evolve. Differential encryption based on identity information ensures the security and privacy of TCM diagnosis and treatment data, while providing a technical foundation for knowledge tracing and rights confirmation. With physician authorization, it facilitates the decryption of personalized diagnosis and treatment knowledge for specific physician groups, contributing to the digital inheritance of TCM diagnosis and treatment experience.

[0072] Example 2:

[0073] This embodiment provides a system for summarizing and processing personalized TCM diagnosis and treatment data, such as... Figure 2 As shown, it includes a data acquisition unit, a feature extraction unit, a graph construction unit, a feature analysis unit, an item matching unit, a data encryption unit, and a knowledge construction unit, wherein:

[0074] The data acquisition unit is used to acquire personalized diagnosis and treatment datasets and identity information of TCM physicians. The personalized diagnosis and treatment datasets include patient symptom information, diagnosis result information, treatment plan information, TCM prescription information, and TCM annotation information.

[0075] The feature extraction unit is used to extract and vectorize patient symptom information to obtain several symptom keyword vectors, and to determine the co-occurrence strength between symptom keyword vectors. It also extracts and vectorizes diagnostic result information to obtain diagnostic keyword vectors.

[0076] The graph construction unit is used to construct a symptom relationship graph based on each symptom keyword vector and the strength of their co-occurrence relationships, and to construct a symptom diagnosis heterogeneity graph using the symptom relationship graph and the diagnostic keyword vector;

[0077] The feature analysis unit is used to input the symptom diagnosis heterogeneity map into a pre-trained graph neural network model for feature association analysis to obtain a fused feature vector;

[0078] The project matching unit is used to match the corresponding experience summary project in the experience base using the fused feature vector and the symptom relationship graph, and to take the matched experience summary project as the target experience summary project. The experience base contains several experience summary projects, each of which is associated with a corresponding reference fused feature vector and a reference symptom relationship graph, and each experience summary project contains a corresponding encrypted knowledge tree.

[0079] The data encryption unit is used to encrypt treatment plan information, traditional Chinese medicine prescription information and traditional Chinese medicine annotation information using identity identification information to obtain encrypted experience entry data, and to calculate identity verification code using identity identification information.

[0080] The knowledge building unit is used to combine the identity verification code with the encrypted experience entry data and add it as a new branch to the encrypted knowledge tree of the target experience summary project.

[0081] Example 3:

[0082] This embodiment provides a system for summarizing and processing personalized TCM diagnosis and treatment data, such as... Figure 3 As shown, at the hardware level, it includes:

[0083] The data interface is used to establish data communication between the processor and external data terminals;

[0084] Memory, used to store instructions;

[0085] The processor is used to read the instructions stored in the memory and execute the TCM personalized diagnosis and treatment data summary and processing method in Embodiment 1 according to the instructions.

[0086] Optionally, the system also includes an internal bus, through which the processor, memory, and data interface can be interconnected. This internal bus can be a PCIe (Peripheral Component Interconnect Eexpress) bus, which can be divided into an address bus, a data bus, a control bus, etc. The memory can include, but is not limited to, Random Access Memory (RAM), Read Only Memory (ROM), Flash Memory, First Input First Output (FIFO), and / or First In Last Out (FILO). The processor can be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc.; it can also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0087] Example 4:

[0088] This embodiment provides a computer-readable storage medium storing instructions. When these instructions are executed on a computer, the computer performs the TCM personalized diagnosis and treatment data summary and processing method described in Embodiment 1. The computer-readable storage medium refers to a data storage medium, which may include, but is not limited to, floppy disks, optical disks, hard disks, flash memory, USB flash drives, and / or Memory Sticks. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices.

[0089] This embodiment also provides a computer program product that, when run on a computer, executes the TCM personalized diagnosis and treatment data summary and processing method described in Embodiment 1. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device.

[0090] Finally, it should be noted that the above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A traditional Chinese medicine personalized diagnosis and treatment data summary processing method, characterized in that, The application comprises the following steps: Obtaining the personalized diagnosis and treatment data set and the identity information of a traditional Chinese medicine doctor, wherein the personalized diagnosis and treatment data set comprises patient symptom information, diagnosis result information, treatment scheme information, traditional Chinese medicine prescription information and traditional Chinese medicine annotation information; Extracting keywords from the patient symptom information and performing vectorization processing to obtain a plurality of symptom keyword vectors, and determining the co-occurrence relationship strength between the symptom keyword vectors; extracting keywords from the diagnosis result information and performing vectorization processing to obtain diagnosis keyword vectors; Constructing a symptom relationship graph based on the symptom keyword vectors and the co-occurrence relationship strength therebetween, and using the symptom relationship graph and the diagnosis keyword vectors to form a symptom diagnosis heterogeneous graph; Inputting the symptom diagnosis heterogeneous graph into a pre-trained graph neural network model for feature correlation analysis to obtain a fusion feature vector; Matching the fusion feature vector and the symptom relationship graph with corresponding experience summary items in an experience library, and taking the matched experience summary items as target experience summary items, comprising: calling the reference fusion feature vector and the reference symptom relationship graph corresponding to each experience summary item in the experience library, wherein the experience library comprises a plurality of experience summary items, each experience summary item is associated with a corresponding reference fusion feature vector and a reference symptom relationship graph, and each experience summary item comprises a corresponding encrypted knowledge tree; calculating the cosine similarity between the fusion feature vector and the reference fusion feature vector of the corresponding experience summary item, and using a graph kernel function to calculate the graph structure similarity between the symptom relationship graph and the reference symptom relationship graph of the corresponding experience summary item; weighting and summing the cosine similarity between the fusion feature vector and the reference fusion feature vector of the corresponding experience summary item and the graph structure similarity between the symptom relationship graph and the reference symptom relationship graph of the corresponding experience summary item to obtain a comprehensive matching degree parameter of the corresponding experience summary item; taking the experience summary item with the highest comprehensive matching degree parameter and exceeding the set parameter threshold as the target experience summary item; Encrypting the treatment scheme information, the traditional Chinese medicine prescription information and the traditional Chinese medicine annotation information using the identity information to obtain encrypted experience item data, and calculating an identity verification code using the identity information; Combining the identity verification code and the encrypted experience item data to add a new branch to the encrypted knowledge tree of the target experience summary item.

2. The method of claim 1, wherein, The method comprises the following steps: Extracting keywords from the patient symptom information to obtain a plurality of symptom keywords, and determining the co-occurrence relationship strength between the symptom keywords; extracting keywords from the diagnosis result information to obtain diagnosis keywords; Performing vectorization processing on each symptom keyword to obtain a corresponding symptom keyword vector, and taking the co-occurrence relationship strength between the symptom keywords as the co-occurrence relationship strength between the symptom keyword vectors; performing vectorization processing on the diagnosis keywords to obtain diagnosis keyword vectors.

3. The method of claim 2, wherein the method further comprises: The method comprises the following steps: The symptom keywords are input into a preset TCM symptom co-occurrence relationship priori knowledge graph for priori knowledge analysis to determine the co-occurrence relationship strength between the symptom keywords.

4. The method of claim 1, wherein the method further comprises: The symptom relationship graph comprises a plurality of first nodes connected by undirected edges, and each first node is associated with a corresponding symptom keyword vector, and each undirected edge is associated with the co-occurrence relationship strength between corresponding symptom keyword vectors. The symptom relationship graph and the diagnosis keyword vector are used to build a symptom diagnosis heterogeneous graph, including:

5. The method of claim 1, wherein the method further comprises: The diagnosis keyword vector is mapped to a corresponding second node, the second node is added to the symptom relationship graph, and a directed edge from the second node to each first node is constructed to obtain the symptom diagnosis heterogeneous graph. The treatment scheme information, the TCM prescription information and the TCM annotation information are encrypted using the identity information to obtain encrypted experience entry data, including: The treatment scheme information, the TCM prescription information and the TCM annotation information are used to form a ternary information group; 6. The method of claim 1, wherein the method further comprises: The identity information is used as a key to encrypt the ternary information group to obtain the encrypted experience entry data. The identity verification code is calculated using the identity information, including:

7. A traditional Chinese medicine personalized diagnosis and treatment data summary processing system, characterized in that, The identity information is hashed to obtain a corresponding hash value, and the calculated hash value is used as the identity verification code. The data acquisition unit, the feature extraction unit, the graph construction unit, the feature analysis unit, the project matching unit, the data encryption unit and the knowledge construction unit are included, wherein: The data acquisition unit is configured to acquire the individualized diagnosis and treatment data set of the TCM physician and the identity information, wherein the individualized diagnosis and treatment data set comprises patient symptom information, diagnosis result information, treatment scheme information, TCM prescription information and TCM annotation information; The feature extraction unit is configured to extract and vectorize the patient symptom information to obtain a plurality of symptom keyword vectors and determine the co-occurrence relationship strength between the symptom keyword vectors, and extract and vectorize the diagnosis result information to obtain a diagnosis keyword vector; The graph construction unit is configured to construct a symptom relationship graph based on the symptom keyword vectors and the co-occurrence relationship strength therebetween, and use the symptom relationship graph and the diagnosis keyword vector to build a symptom diagnosis heterogeneous graph; The feature analysis unit is configured to input the symptom diagnosis heterogeneous graph into a pre-trained graph neural network model for feature correlation analysis to obtain a fusion feature vector; The project matching unit is configured to match the corresponding experience summary project in the experience library by using the fusion feature vector and the symptom relationship graph, and take the matched experience summary project as the target experience summary project, and includes: calling the reference fusion feature vector and the reference symptom relationship graph corresponding to each experience summary project in the experience library, wherein the experience library contains a plurality of experience summary projects, each experience summary project is associated with a corresponding reference fusion feature vector and a reference symptom relationship graph, and each experience summary project contains a corresponding encrypted knowledge tree; calculating the cosine similarity between the fusion feature vector and the reference fusion feature vector of the corresponding experience summary project, and calculating the graph structure similarity between the symptom relationship graph and the reference symptom relationship graph of the corresponding experience summary project by using a graph kernel function; weighting and summing the cosine similarity between the fusion feature vector and the reference fusion feature vector of the corresponding experience summary project and the graph structure similarity between the symptom relationship graph and the reference symptom relationship graph of the corresponding experience summary project to obtain a comprehensive matching degree parameter of the corresponding experience summary project; taking the experience summary project with the highest comprehensive matching degree parameter and exceeding the set parameter threshold as the target experience summary project; The data encryption unit is configured to encrypt the treatment scheme information, the Chinese medicine prescription information and the Chinese medicine annotation information by using the identity information to obtain encrypted experience item data, and calculate an identity check code by using the identity information. The knowledge construction unit is configured to combine the identity check code and the encrypted experience item data and add them as a new branch to the encrypted knowledge tree of the target experience summary project.

8. A traditional Chinese medicine personalized diagnosis and treatment data summary processing system, characterized in that, The computer program product includes: A memory configured to store instructions; A processor configured to read the instructions stored in the memory and execute the TCM individualized diagnosis and treatment data summary processing method according to any one of claims 1-6.

9. A computer program product, characterised in that, When the computer program product runs on the computer, the TCM individualized diagnosis and treatment data summary processing method according to any one of claims 1-6 is executed.

Images