An intelligent classification method and system for electronic medical records of traditional Chinese medicine based on knowledge graph

Abstract

The present invention relates to the field of data management technology, and in particular to a method and system for intelligent classification of electronic medical records of traditional Chinese medicine based on a knowledge graph. The method comprises: obtaining a knowledge graph of traditional Chinese medicine, which is composed of a plurality of first knowledge entities of traditional Chinese medicine; extracting a second knowledge entity of the electronic medical record of traditional Chinese medicine, and matching the first knowledge entity of traditional Chinese medicine in the knowledge graph of traditional Chinese medicine based on the second knowledge entity of traditional Chinese medicine; obtaining a storage address of the electronic medical record of traditional Chinese medicine, and linking the storage address and the first knowledge entity of traditional Chinese medicine matching the corresponding second knowledge entity of traditional Chinese medicine; setting a target knowledge entity of traditional Chinese medicine, and matching the first knowledge entity of traditional Chinese medicine in the knowledge graph of traditional Chinese medicine based on the target knowledge entity; obtaining the electronic medical record of traditional Chinese medicine matching the target knowledge entity based on the storage address linked to the first knowledge entity of traditional Chinese medicine; the present invention can not only improve the intelligent management level of electronic medical records, but also provide a convenient tool for clinical diagnosis and research in the field of traditional Chinese medicine.

Description

Technical Field

[0001] The present invention relates to the technical field of data management, and in particular to a method and system for intelligent classification of electronic medical records of traditional Chinese medicine based on knowledge graph. Background Art

[0002] With the rapid development of medical technology and Internet technology, electronic medical records, as an important form of digital medical records, have been promoted and applied in more and more hospitals. Electronic medical records are different from traditional handwritten paper medical records. They use electronic devices (such as computers, health cards, etc.) as carriers to realize the storage, management, transmission and reproduction of medical record data, greatly improving the efficiency and quality of medical services.

[0003] TCM electronic medical records are developed on the basis of traditional electronic medical records and combined with the characteristics of TCM diagnosis and treatment. TCM diagnosis and treatment emphasizes syndrome differentiation and treatment, involving the interaction and association of complex knowledge entities such as symptoms, syndromes and diseases, thus placing higher requirements on the management of TCM medical records. Traditional electronic medical record systems are mostly centered on Western medicine diagnosis and treatment, and it is difficult to fully support the unique needs of TCM. In addition, existing technologies for the classification of TCM medical records mainly rely on manual experience, rule matching or simple text analysis methods. These methods are usually unable to effectively handle the associations between complex knowledge entities, and it is difficult to achieve fast and accurate classification in massive medical record data. Summary of the invention

[0004] The present invention provides a method and system for intelligent classification of electronic medical records of traditional Chinese medicine based on knowledge graph, aiming to achieve accurate classification and intelligent processing of electronic medical records of traditional Chinese medicine based on existing knowledge graph technology.

[0005] In a first aspect, the present invention provides a method for intelligent classification of electronic medical records of traditional Chinese medicine based on knowledge graph, comprising the following steps:

[0006] Acquire a TCM knowledge graph, wherein the TCM knowledge graph is composed of a plurality of first TCM knowledge entities and connecting edges between the first TCM knowledge entities;

[0007] Extracting a second TCM knowledge entity from the TCM electronic medical record, and matching the first TCM knowledge entity in the TCM knowledge graph based on the second TCM knowledge entity;

[0008] Acquire the storage address of the TCM electronic medical record, and link the storage address with the first TCM knowledge entity that matches the corresponding second TCM knowledge entity;

[0009] Setting a target TCM knowledge entity, and matching a first TCM knowledge entity in the TCM knowledge graph based on the target TCM knowledge entity;

[0010] Based on the storage address linked to the first TCM knowledge entity, a TCM electronic medical record matching the target TCM knowledge entity is obtained.

[0011] Optionally, the first TCM knowledge entity includes a first TCM symptom knowledge entity, a first TCM syndrome type knowledge entity and a first TCM disease knowledge entity.

[0012] Optionally, in the TCM knowledge graph, any first TCM disease knowledge entity has a connection edge with at least one first TCM syndrome type knowledge entity, and any first TCM syndrome type knowledge entity has a connection edge with at least one first TCM symptom knowledge entity.

[0013] Optionally, the step of extracting the second TCM knowledge entity from the TCM electronic medical record comprises the following steps:

[0014] All second TCM knowledge entities extracted from TCM electronic medical records;

[0015] From all the second TCM knowledge entities, the second TCM symptom knowledge entities, the second TCM syndrome type knowledge entities and the second TCM disease knowledge entities are screened out.

[0016] Optionally, matching the first TCM knowledge entity in the TCM knowledge graph based on the second TCM knowledge entity comprises the following steps:

[0017] Based on the second TCM symptom knowledge entity, matching the first TCM symptom knowledge entity in the TCM knowledge graph;

[0018] Based on the second TCM syndrome type knowledge entity, matching the first TCM syndrome type knowledge entity in the TCM knowledge graph;

[0019] Based on the second TCM disease knowledge entity, the first TCM disease knowledge entity in the TCM knowledge graph is matched.

[0020] Optionally, any second TCM knowledge entity is matched with the first TCM knowledge entity through the following matching model: , where i and j are numbers, i has a value not exceeding I, I is the total number of first TCM symptom knowledge entities in the TCM knowledge graph, j has a value not exceeding J, J is the total number of second TCM knowledge entities, represents the first TCM knowledge entity that matches the second TCM knowledge entity, Indicates that among all the first TCM knowledge entities, The first TCM knowledge entity with the maximum value, The first TCM knowledge entity The Second TCM Knowledge Entity The semantic similarity of .

[0021] Optionally, matching the first TCM knowledge entity in the TCM knowledge graph based on the target TCM knowledge entity comprises the following steps:

[0022] Based on the target TCM symptom knowledge entity, matching the first TCM symptom knowledge entity in the TCM knowledge graph;

[0023] Based on the target TCM syndrome type knowledge entity, matching the first TCM syndrome type knowledge entity in the TCM knowledge graph;

[0024] Based on the target TCM disease knowledge entity, the first TCM disease knowledge entity in the TCM knowledge graph is matched.

[0025] Optionally, any target TCM knowledge entity is matched with the first TCM knowledge entity through the following matching model:

[0026] , where i and k are numbers, i has a value not exceeding I, I is the total number of first TCM symptom knowledge entities in the TCM knowledge graph, and k has a value not exceeding K, K is the total number of target TCM knowledge entities, represents the first TCM knowledge entity that matches the target TCM knowledge entity, Indicates that among all the first TCM knowledge entities, The first TCM knowledge entity with the maximum value, The first TCM knowledge entity Target TCM knowledge entity The semantic similarity of .

[0027] Optionally, when the target TCM knowledge entity includes multiple TCM knowledge entities that are connected, the first TCM knowledge entity that matches it also includes TCM knowledge entities that are equal in number to the TCM knowledge entities it contains, and obtaining the TCM electronic medical record that matches the target TCM knowledge entity based on the storage address linked to the first TCM knowledge entity includes the following steps:

[0028] The storage addresses linked to each first TCM knowledge entity are merged, and based on the merged storage addresses, a TCM electronic medical record matching the target TCM knowledge entity is obtained.

[0029] In a second aspect, the intelligent classification system for electronic medical records of traditional Chinese medicine based on knowledge graph provided by the present invention comprises an input device, a processor, a memory and an output device, wherein the input device, the processor, the memory and the output device are interconnected, the memory is used to store a computer program, the computer program comprises program instructions, and the processor is configured to call the program instructions to execute the intelligent classification method for electronic medical records of traditional Chinese medicine based on knowledge graph provided by the first aspect of the present invention.

[0030] The benefits of the intelligent classification method and system for TCM electronic medical records based on knowledge graph provided by the present invention are:

[0031] The present invention uses the knowledge graph to match the TCM knowledge entities in the electronic medical record with the known standardized knowledge such as symptoms, syndromes and diseases, thus realizing the accurate classification and retrieval of medical records; and by linking the medical records with the entities in the knowledge graph, users can quickly locate the relevant medical records based on the target knowledge entities such as symptoms, syndromes or diseases, thus greatly improving the efficiency of medical record classification and reducing the need for manual intervention; at the same time, the semantic similarity matching model is adopted so that similar concepts in different medical records can be effectively associated, thus ensuring the accuracy and consistency of classification. The present invention not only improves the intelligent management level of electronic medical records, but also provides a convenient tool for clinical diagnosis and research in the field of TCM, and has broad application prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 A flow chart of the intelligent classification method for electronic medical records of traditional Chinese medicine based on knowledge graph provided in an embodiment of the present invention;

[0033] Figure 2 A schematic diagram of knowledge entity connections in a TCM knowledge graph provided for the implementation of the present invention;

[0034] Figure 3 Schematic diagram of the intelligent classification system for electronic medical records of traditional Chinese medicine based on knowledge graph provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0035] In the following description, for the purpose of explanation rather than limitation, specific details such as specific system structures and technologies are proposed to facilitate a thorough understanding of the embodiments of the present application.

[0036] In one embodiment, see Figure 1 , Figure 1 Flowchart of the intelligent classification method for electronic medical records of traditional Chinese medicine based on knowledge graph provided in an embodiment of the present invention.

[0037] like Figure 1 As shown, the intelligent classification method of TCM electronic medical records based on knowledge graph provided by the present invention comprises the following steps:

[0038] S01. Obtain a TCM knowledge graph, wherein the TCM knowledge graph is composed of a plurality of first TCM knowledge entities and connecting edges between the first TCM knowledge entities.

[0039] It can be understood that the traditional Chinese medicine knowledge graph described in the present invention is a structured network graph used to represent knowledge in the field of traditional Chinese medicine, which can be constructed using existing knowledge graph technology.

[0040] Furthermore, the first TCM knowledge entity of the present invention is a basic node in the TCM knowledge graph, and each first TCM knowledge entity represents a specific knowledge unit in the field of TCM, and is used to describe key concepts or objects in TCM theory, TCM clinical diagnosis or TCM treatment process.

[0041] In this embodiment, the TCM knowledge graph constructed based on the existing knowledge graph technology includes at least the first TCM symptom knowledge entity, the first TCM syndrome knowledge entity and the first TCM disease knowledge entity. In one or some other embodiments, the constructed or acquired TCM knowledge graph may also include other types of first TCM knowledge entities.

[0042] Furthermore, the first TCM symptom knowledge entity refers to the specific symptoms described in the TCM diagnosis and treatment process, such as fear of cold, clear runny nose, cough, excessive sputum, itchy throat, headache, fatigue, shortness of breath, spontaneous sweating, itchy skin, erythema, papules, blisters, etc.

[0043] Furthermore, the first TCM syndrome type knowledge entity refers to the classification of pathological conditions obtained through analysis of symptoms, signs, etc. in TCM syndrome differentiation and treatment, such as wind-cold, wind-heat, qi deficiency, yin deficiency, yang deficiency, blood stasis, qi stagnation, dampness and toxicity, kidney yin deficiency, kidney yang deficiency, etc.

[0044] Furthermore, the first TCM disease knowledge entity refers to a specific disease or condition diagnosed by TCM, such as cold, pneumonia, eczema, insomnia, hair loss, etc.

[0045] In this embodiment, in the TCM knowledge graph provided by the present invention, any first TCM disease knowledge entity has a connection edge with at least one first TCM syndrome type knowledge entity, and any first TCM syndrome type knowledge entity has a connection edge with at least one first TCM symptom knowledge entity. In one or some other embodiments, the connection relationship between TCM knowledge entities in the TCM knowledge graph, that is, the setting of the connection edge, may be the same as or different from that in this embodiment.

[0046] Specifically, see Figure 2 , Figure 2 A schematic diagram of knowledge entity connections in the TCM knowledge graph provided for the implementation of the present invention. Figure 2 As shown, there is a connection edge between any first TCM disease knowledge entity A and at least one first TCM syndrome type knowledge entity B, and there is a connection edge between any first TCM syndrome type knowledge entity B and at least one first TCM symptom knowledge entity C.

[0047] S02. Extract a second TCM knowledge entity from the TCM electronic medical record, and match a first TCM knowledge entity in the TCM knowledge graph based on the second TCM knowledge entity.

[0048] It is understandable that the TCM electronic medical records described in step S02 are historical TCM electronic medical records to be accurately classified and efficiently managed. It is known that TCM electronic files of different departments are usually stored in different storage addresses.

[0049] In this embodiment, the extraction of the second TCM knowledge entity of the TCM electronic medical record in step S02 can be achieved through NLP (Natural Language Processing, NLP) technology.

[0050] It is known that NLP technology refers to a set of technologies used to enable computers to understand, analyze, generate and process human language. It can implement word segmentation, part-of-speech tagging, named entity recognition, syntactic analysis, semantic analysis, sentiment analysis and other processing for speech or text language.

[0051] Furthermore, based on the NLP technology, the step S02 of extracting the second TCM knowledge entity of the TCM electronic medical record includes the following steps:

[0052] S021. All second TCM knowledge entities extracted from TCM electronic medical records.

[0053] It should be noted that since the recorders of TCM electronic medical records are inconsistent, the types of TCM knowledge entities extracted from any two TCM electronic medical records may be the same or different.

[0054] In this embodiment, the recorded TCM electronic medical record is provided with required content, such as the second TCM symptom knowledge entity, the second TCM syndrome type knowledge entity and the second TCM disease knowledge entity.

[0055] In one or some other embodiments, other mandatory contents are provided in the recorded TCM electronic medical record, such as a second TCM drug knowledge entity, a second TCM prescription knowledge entity, a second TCM constitution type knowledge entity, and the like.

[0056] S022. Filter out the second TCM symptom knowledge entity, the second TCM syndrome type knowledge entity and the second TCM disease knowledge entity from all the second TCM knowledge entities.

[0057] Step S022 can also be implemented through NLP technology, which will not be described here.

[0058] Furthermore, the matching of the first TCM knowledge entity in the TCM knowledge graph based on the second TCM knowledge entity in step S02 includes the following steps:

[0059] S0221. Based on the second TCM symptom knowledge entity, match the first TCM symptom knowledge entity in the TCM knowledge graph.

[0060] S0222. Based on the second TCM syndrome type knowledge entity, match the first TCM syndrome type knowledge entity in the TCM knowledge graph.

[0061] S0223. Based on the second TCM disease knowledge entity, match the first TCM disease knowledge entity in the TCM knowledge graph.

[0062] It can be understood that since different types of first TCM knowledge entities in the TCM knowledge graph have been classified, the matching operation with the second TCM symptom knowledge entity in step S0221, the matching operation with the second TCM syndrome type knowledge entity in step S0222, and the matching operation with the second TCM disease knowledge entity in step S0223 can be performed synchronously or asynchronously.

[0063] Furthermore, in step S0221, step S0222 and step S0223, any second TCM knowledge entity is matched with the first TCM knowledge entity through the following matching model: .

[0064] Furthermore, i and j are both numbers, i has a value not exceeding I, I is the total number of first TCM symptom knowledge entities in the TCM knowledge graph, j has a value not exceeding J, J is the total number of second TCM knowledge entities, represents the first TCM knowledge entity that matches the second TCM knowledge entity, Indicates that among all the first TCM knowledge entities, The first TCM knowledge entity with the maximum value, The first TCM knowledge entity The Second TCM Knowledge Entity The semantic similarity of .

[0065] Furthermore, the semantic similarity between any two TCM knowledge entities can be calculated using existing semantic similarity calculation models, such as a cosine similarity calculation model based on a word vector model, a Euclidean distance calculation model based on a word vector model, a cosine similarity calculation model based on TF-IDF, and the like.

[0066] It should be noted that, in the execution of steps S0221, S0222 and S0223 of the above matching model, the second TCM knowledge entity respectively represents different types of second TCM knowledge entities: in step S0221, the second TCM knowledge entity refers to the second TCM symptom knowledge entity; in step S0221, the second TCM knowledge entity refers to the second TCM syndrome type knowledge entity; in step S0221, the second TCM knowledge entity refers to the second TCM disease knowledge entity.

[0067] S03. Obtain the storage address of the TCM electronic medical record, and link the storage address with the first TCM knowledge entity that matches the corresponding second TCM knowledge entity.

[0068] The storage address described in the present invention refers to the specific path or location used to locate and access the traditional Chinese medicine electronic medical record data in the storage system, and its form depends on how the system stores and manages the electronic medical record data, such as database address, file system path, URL / network address, data index address, etc.

[0069] Furthermore, in step S03, linking the storage address with the first TCM knowledge entity that matches the second TCM knowledge entity in the TCM knowledge graph includes the following steps:

[0070] S031. Link the storage address of the second TCM symptom knowledge entity with the first TCM symptom knowledge entity that matches the corresponding second TCM symptom knowledge entity.

[0071] S032. Link the storage address of the second TCM syndrome type knowledge entity with the first TCM syndrome type knowledge entity that matches the corresponding second TCM syndrome type knowledge entity.

[0072] S033. Link the storage address of the second TCM disease knowledge entity with the first TCM disease knowledge entity that matches the corresponding second TCM disease knowledge entity.

[0073] Likewise, steps S031 to S033 may be performed synchronously or asynchronously.

[0074] It should be noted that in the TCM knowledge graph, any storage address linked to the first TCM knowledge entity will form a connection edge with the corresponding first TCM knowledge entity in the TCM knowledge graph.

[0075] Typically, the connection edges formed by the storage address connection edges are only used to implement logical link relationships for the storage and management of TCM electronic medical records without performing visual operations.

[0076] S04. Set a target TCM knowledge entity, and match the first TCM knowledge entity in the TCM knowledge graph based on the target TCM knowledge entity.

[0077] It can be understood that the target TCM symptom entity described in the present invention is a specific TCM knowledge entity used for query, analysis or matching; usually, the user can manually input the target TCM knowledge entity through the system interface as a retrieval condition or query condition.

[0078] Likewise, the target TCM knowledge entity includes at least a target TCM symptom knowledge entity, a target TCM syndrome type knowledge entity, and a target TCM disease knowledge entity.

[0079] Furthermore, the step S04 of matching the first TCM knowledge entity in the TCM knowledge graph based on the target TCM knowledge entity comprises the following steps:

[0080] S041. Based on the target TCM symptom knowledge entity, matching the first TCM symptom knowledge entity in the TCM knowledge graph;

[0081] S042. Based on the target TCM syndrome type knowledge entity, matching the first TCM syndrome type knowledge entity in the TCM knowledge graph;

[0082] S043. Based on the target TCM disease knowledge entity, match the first TCM disease knowledge entity in the TCM knowledge graph.

[0083] Similarly, the matching operation with the target TCM symptom knowledge entity in step S041, the matching operation with the target TCM syndrome type knowledge entity in step S042, and the matching operation with the target TCM disease knowledge entity in step S043 can be performed synchronously or asynchronously.

[0084] Furthermore, in step S041, step S042 and step S043, any second TCM knowledge entity is matched with the first TCM knowledge entity through the following matching model: .

[0085] Among them, i and k are numbers, the value of i does not exceed I, I is the total number of the first TCM symptom knowledge entities in the TCM knowledge graph, and the value of k does not exceed K, K is the total number of target TCM knowledge entities. represents the first TCM knowledge entity that matches the target TCM knowledge entity, Indicates that among all the first TCM knowledge entities, The first TCM knowledge entity with the maximum value, The first TCM knowledge entity Target TCM knowledge entity The semantic similarity of .

[0086] Similarly, the semantic similarity between any two TCM knowledge entities can be calculated using existing semantic similarity calculation models, such as a cosine similarity calculation model based on a word vector model, a Euclidean distance calculation model based on a word vector model, a cosine similarity calculation model based on TF-IDF, and the like.

[0087] It should be noted that in the execution of steps S041, S042 and S043 of the above matching model, the target TCM knowledge entities respectively represent different types of target TCM knowledge entities: in step S041, the target TCM knowledge entity refers to the target TCM symptom knowledge entity; in step S042, the target TCM knowledge entity refers to the target TCM syndrome type knowledge entity; in step S043, the target TCM knowledge entity refers to the target TCM disease knowledge entity.

[0088] S05. Based on the storage address linked to the first TCM knowledge entity, obtain a TCM electronic medical record that matches the target TCM knowledge entity.

[0089] It is easy to understand that when the target TCM knowledge entity set in step S04 is a single or multiple unrelated TCM knowledge entities, step S05 will obtain one or more corresponding TCM electronic medical records based on each target TCM knowledge entity.

[0090] In this embodiment, when the target TCM knowledge entity set in step S04 includes multiple TCM knowledge entities having a connection relationship, the first TCM knowledge entity matching it also includes TCM knowledge entities equal in number to the TCM knowledge entities it contains.

[0091] Furthermore, the step S05 of acquiring the TCM electronic medical record matching the target TCM knowledge entity based on the storage address linked to the first TCM knowledge entity includes the following steps:

[0092] The storage addresses linked to each first TCM knowledge entity are merged, and based on the merged storage addresses, a TCM electronic medical record matching the target TCM knowledge entity is obtained.

[0093] It can be understood that, in this embodiment, by merging the storage addresses, all medical record data related to all target TCM knowledge entities can be acquired at one time.

[0094] In one embodiment, see Figure 3 , Figure 3 Schematic diagram of the intelligent classification system for electronic medical records of traditional Chinese medicine based on knowledge graph provided in an embodiment of the present invention.

[0095] like Figure 3 As shown, the intelligent classification system for electronic medical records of traditional Chinese medicine based on knowledge graph provided in an embodiment of the present invention includes an input device, a processor, a memory and an output device, wherein the input device, the processor, the memory and the output device are interconnected.

[0096] Furthermore, the input device includes a keyboard, a touch screen or a voice input device, allowing the user to input knowledge entities or electronic medical record data related to traditional Chinese medicine.

[0097] Furthermore, the processor, as the core of the system, calls computer programs and instructions in the memory, and extracts knowledge entities in the TCM electronic medical records through natural language processing technology, including TCM symptoms, syndrome types and diseases.

[0098] It can be understood that the processor performs intelligent classification and matching operations based on the extracted entity information and by comparing the first TCM knowledge entity in the knowledge graph through a matching model.

[0099] Furthermore, the memory is used to store the core data and instructions of the system, including the constructed traditional Chinese medicine knowledge graph, extraction and matching algorithm models, electronic medical record data storage addresses, etc.

[0100] During the operation, the processor will continuously call the data and instructions in the memory to extract TCM knowledge entities, calculate semantic similarity, and perform matching and linking operations. For example, when extracting the "Qi Deficiency" syndrome entity in a medical record, the processor will call the relevant algorithm stored in the memory to match the entity, and then through the existing knowledge graph, realize the linking of other symptoms, syndromes, and disease entities associated with "Qi Deficiency" to achieve accurate classification.

[0101] Furthermore, the output device is responsible for displaying the processing results, which may include display screens, printers and other devices. Users can view the electronic medical record classification results that match the target TCM knowledge entity through the output device. For example, when a user enters a target entity of "kidney yang deficiency", the system will automatically match the corresponding first TCM knowledge entity, extract the associated medical record data and display it on the output device.

[0102] To sum up, the operation process of the entire system forms an efficient closed loop through data reception by input devices, intelligent analysis and matching by processors, data management by storage devices, and result display by output devices, which significantly improves the intelligent classification and management level of TCM electronic medical records. This not only realizes the accurate retrieval and classification of medical record data, but also plays an important role in TCM diagnosis, research, and data mining.

Claims

1. A method for intelligent classification of TCM electronic medical records based on knowledge graph, characterized in that: The steps include: S01. Obtain a TCM knowledge graph, wherein the TCM knowledge graph is composed of a plurality of first TCM knowledge entities and connecting edges between the first TCM knowledge entities; the TCM knowledge graph includes a first TCM symptom knowledge entity, a first TCM syndrome knowledge entity, and a first TCM disease knowledge entity, wherein the first TCM symptom knowledge entity refers to a specific symptom described in the TCM diagnosis and treatment process; S02, extracting a second TCM knowledge entity from the TCM electronic medical record, and matching a first TCM knowledge entity in the TCM knowledge graph based on the second TCM knowledge entity; S03, obtaining the storage address of the TCM electronic medical record, and linking the storage address with the first TCM knowledge entity that matches the corresponding second TCM knowledge entity; S04, setting a target TCM knowledge entity, and matching a first TCM knowledge entity in the TCM knowledge graph based on the target TCM knowledge entity; S05. Based on the storage address linked to the first TCM knowledge entity, obtain a TCM electronic medical record matching the target TCM knowledge entity; The method of extracting the second TCM knowledge entity from the TCM electronic medical record comprises the following steps: S021. All second TCM knowledge entities extracted from TCM electronic medical records; S022. Filtering out second TCM symptom knowledge entities, second TCM syndrome type knowledge entities and second TCM disease knowledge entities from all the second TCM knowledge entities; The matching of the first TCM knowledge entity in the TCM knowledge graph based on the second TCM knowledge entity comprises the following steps: S0221. Based on the second TCM symptom knowledge entity, match the first TCM symptom knowledge entity in the TCM knowledge graph; S0222. Based on the second TCM syndrome type knowledge entity, match the first TCM syndrome type knowledge entity in the TCM knowledge graph; S0223. Based on the second TCM disease knowledge entity, match the first TCM disease knowledge entity in the TCM knowledge graph; Any second TCM knowledge entity is matched with the first TCM knowledge entity through the following matching model: , where i and j are numbers, i has a value not exceeding I, I is the total number of first TCM symptom knowledge entities in the TCM knowledge graph, j has a value not exceeding J, J is the total number of second TCM knowledge entities, represents the first TCM knowledge entity that matches the second TCM knowledge entity, Indicates that among all the first TCM knowledge entities, The first TCM knowledge entity with the maximum value, The first TCM knowledge entity The Second TCM Knowledge Entity The semantic similarity of Step S03 links the storage address with the first TCM knowledge entity that matches the second TCM knowledge entity in the TCM knowledge graph, including the following steps: S031, linking the storage address of the second TCM symptom knowledge entity with the first TCM symptom knowledge entity that matches the corresponding second TCM symptom knowledge entity; S032, linking the storage address of the second TCM syndrome type knowledge entity with the first TCM syndrome type knowledge entity that matches the corresponding second TCM syndrome type knowledge entity; S033, linking the storage address of the second TCM disease knowledge entity with the first TCM disease knowledge entity that matches the corresponding second TCM disease knowledge entity; Step S04, matching the first TCM knowledge entity in the TCM knowledge graph based on the target TCM knowledge entity, comprises the following steps: S041. Based on the target TCM symptom knowledge entity, matching the first TCM symptom knowledge entity in the TCM knowledge graph; S042. Based on the target TCM syndrome type knowledge entity, matching the first TCM syndrome type knowledge entity in the TCM knowledge graph; S043. Based on the target TCM disease knowledge entity, match the first TCM disease knowledge entity in the TCM knowledge graph.

2. According to claim 1, the intelligent classification method for electronic medical records of traditional Chinese medicine based on knowledge graph is characterized in that: The first TCM knowledge entity includes a first TCM symptom knowledge entity, a first TCM syndrome type knowledge entity and a first TCM disease knowledge entity.

3. The intelligent classification method for electronic medical records of traditional Chinese medicine based on knowledge graph according to claim 2 is characterized in that: In the TCM knowledge graph, any first TCM disease knowledge entity has a connection edge with at least one first TCM syndrome type knowledge entity, and any first TCM syndrome type knowledge entity has a connection edge with at least one first TCM symptom knowledge entity.

4. The intelligent classification method for electronic medical records of traditional Chinese medicine based on knowledge graph according to claim 3 is characterized in that: Any target TCM knowledge entity is matched with the first TCM knowledge entity through the following matching model: Among them, i and k are numbers, the value of i does not exceed I, I is the total number of the first TCM symptom knowledge entities in the TCM knowledge graph, and the value of k does not exceed K, K is the total number of target TCM knowledge entities. represents the first TCM knowledge entity that matches the target TCM knowledge entity, Indicates that among all the first TCM knowledge entities, The first TCM knowledge entity with the maximum value, The first TCM knowledge entity Target TCM knowledge entity The semantic similarity of .

5. The intelligent classification method for electronic medical records of traditional Chinese medicine based on knowledge graph according to claim 1 is characterized in that: When the target TCM knowledge entity includes multiple TCM knowledge entities that are connected, the first TCM knowledge entity that matches it also includes TCM knowledge entities that are equal in number to the TCM knowledge entities it contains, and the method of obtaining the TCM electronic medical record that matches the target TCM knowledge entity based on the storage address linked to the first TCM knowledge entity includes the following steps: The storage addresses linked to each first TCM knowledge entity are merged, and based on the merged storage addresses, a TCM electronic medical record matching the target TCM knowledge entity is obtained.

6. An intelligent classification system for TCM electronic medical records based on knowledge graph, characterized in that: The intelligent classification system for electronic medical records of traditional Chinese medicine includes an input device, a processor, a memory and an output device, wherein the input device, the processor, the memory and the output device are interconnected, the memory is used to store a computer program, the computer program includes program instructions, and the processor is configured to call the program instructions to execute the intelligent classification method for electronic medical records of traditional Chinese medicine based on a knowledge graph as described in any one of claims 1 to 5.

Images