CN-121983281-A - Lymphoma traditional Chinese medicine syndrome recognition model construction method and device and electronic equipment

CN121983281ACN 121983281 ACN121983281 ACN 121983281ACN-121983281-A

Abstract

The invention discloses a method and a device for constructing a traditional Chinese medicine syndrome identification model of lymphoma, electronic equipment, a storage medium and a program product. The method comprises the steps of obtaining a traditional Chinese medicine waiting symptom list of a plurality of lymphoma patients, wherein the traditional Chinese medicine waiting symptom list comprises a plurality of symptom columns and symptoms corresponding to each symptom column, the symptoms comprise single symptoms and compound symptoms, determining numerical codes corresponding to the single symptoms and numerical code strings corresponding to the compound symptoms in the traditional Chinese medicine waiting symptom list according to a predefined symptom code mapping dictionary, generating a waiting symptom code list, and training a machine learning model according to the waiting symptom code list to obtain a lymphoma traditional Chinese medicine syndrome identification model. The technical scheme of the embodiment of the invention can improve the efficiency and accuracy of identifying the traditional Chinese medicine symptoms of the lymphoma patient.

Inventors

YE BAODONG
ZHANG YU
LI HANGCHAO
DONG NANXI
LIU WENBIN
LIU JINGJING

Assignees

浙江省中医院、浙江中医药大学附属第一医院（浙江省东方医院）

Dates

Publication Date: 20260505
Application Date: 20260119

Claims (10)

1. The method for constructing the traditional Chinese medicine syndrome recognition model of the lymphoma is characterized by comprising the following steps of: Obtaining a traditional Chinese medicine waiting symptom list of a plurality of lymphoma patients, wherein the traditional Chinese medicine waiting symptom list comprises a plurality of symptom columns and symptoms corresponding to each symptom column, and the symptoms comprise single symptoms and compound symptoms; according to a predefined symptom code mapping dictionary, determining a numerical code corresponding to a single symptom and a numerical code character string corresponding to a compound symptom in the traditional Chinese medicine waiting symptom table, and generating a waiting symptom code table; Training a machine learning model according to the waiting symptom coding table to obtain a lymphoma traditional Chinese medicine symptom identification model.
2. The method according to claim 1, wherein determining the numerical code corresponding to the single symptom and the numerical code string corresponding to the compound symptom in the traditional Chinese medicine waiting symptom table according to the predefined symptom code mapping dictionary comprises: Matching the single symptom with a preset symptom in the symptom code mapping dictionary aiming at the single symptom, and taking a numerical code corresponding to the preset symptom successfully matched with the single symptom as the numerical code of the single symptom; aiming at the compound symptoms, splitting the compound symptoms into a plurality of single symptoms according to a first preset separator in the compound symptoms, determining a numerical code corresponding to each single symptom, and splicing the numerical codes corresponding to the plurality of single symptoms into a numerical code character string through a second preset separator.
3. The method of claim 1, wherein training the machine learning model according to the waiting symptom encoding table to obtain a lymphoma traditional Chinese medicine syndrome identification model comprises: determining a first symptom characteristic of a single symptom and a second symptom characteristic of a compound symptom in the waiting symptom coding table through single-heat coding, and generating a waiting symptom characteristic table corresponding to the waiting symptom coding table; performing recursive clustering on the waiting symptom characteristic table, and determining traditional Chinese medicine symptoms corresponding to each type of waiting symptom characteristic table as clustering labels; Labeling the waiting symptom characteristic table according to the clustering label, and training the random forest model according to the labeled waiting symptom characteristic table to obtain the lymphoma traditional Chinese medicine symptom recognition model.
4. A method according to claim 3, wherein said determining by one-hot encoding a first symptom characteristic of a single symptom and a second symptom characteristic of a compound symptom in the list of candidate symptom codes comprises: Determining, for each symptom column in the waiting symptom encoding table, a symptom type included in the symptom column; determining a binary dummy variable of a single symptom as a first symptom feature according to the symptom type and a numerical code corresponding to the single symptom in the symptom column through single-heat coding; and determining a symptom existence binary matrix of the compound symptoms as a second symptom characteristic according to the symptom type and a numerical code character string corresponding to the compound symptoms in the symptom column through single-heat coding.
5. The method of claim 3, wherein the labeling the waiting symptom feature table according to the cluster label, and training the random forest model according to the labeled waiting symptom feature table to obtain the lymphoma traditional Chinese medicine symptom recognition model, comprises: inputting the marked waiting symptom characteristic table into a random forest model, taking marked cluster labels as classification targets, and determining optimal parameter combinations in a plurality of parameter combinations through cross-verified grid search; and determining a random forest model when the optimal parameter combination is used as a model parameter, and using the random forest model as a traditional Chinese medicine syndrome identification model of lymphoma.
6. The method of claim 3, wherein the labeling the waiting symptom code table according to the cluster tag, and training the random forest model according to the labeled waiting symptom code table to obtain a lymphoma traditional Chinese medicine symptom recognition model, further comprises: verifying the performance of the traditional Chinese medicine syndrome recognition model of the lymphoma through a preset test set, and calculating the contribution value of each symptom feature in a waiting symptom feature table in the test set when the model is predicted; and generating a global feature importance ranking according to the contribution values, and generating a force guide graph and a summary bee colony graph corresponding to each waiting symptom feature table.
7. The method of claim 1, further comprising, after training the machine learning model according to the waiting symptom encoding table to obtain a lymphoma traditional Chinese medicine syndrome identification model: Obtaining a target traditional Chinese medicine waiting symptom table of a target lymphoma patient, and generating a target waiting symptom characteristic table according to the target traditional Chinese medicine waiting symptom table; Inputting the target waiting symptom characteristic table into the lymphoma traditional Chinese medicine syndrome identification model to obtain lymphoma traditional Chinese medicine syndrome identification results corresponding to the target traditional Chinese medicine waiting symptom table, and generating a force guide graph corresponding to each target waiting symptom characteristic table and a summary bee colony graph.
8. An electronic device, the electronic device comprising: at least one processor, and A memory communicatively coupled to the at least one processor, wherein, The memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the method of creating a pattern recognition model in lymphoma according to any one of claims 1 to 7.
9. A computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions for causing a processor to implement the method for creating a pattern recognition model in lymphoma according to any one of claims 1 to 7 when executed.
10. A computer program product comprising a computer program, characterized in that the computer program, when executed by a processor, implements the method for constructing a model for identifying a symptom of lymphoma according to any of claims 1-7.

Description

Lymphoma traditional Chinese medicine syndrome recognition model construction method and device and electronic equipment Technical Field The invention relates to the technical field of traditional Chinese medicine syndrome recognition, in particular to a method and a device for constructing a lymphoma traditional Chinese medicine syndrome recognition model, electronic equipment, a storage medium and a program product. Background Lymphoma is the highest-incidence malignant tumor of the blood system, and the current common small-molecule drugs, monoclonal antibodies, antibody coupling drugs and cell therapies still have unmet needs although the prognosis is greatly improved, and the treatment comprises the steps of 1, lack of effective intervention means for patients with node therapy to prevent relapse, 2, observation of patients with lymphoma in waiting period, no special intervention means, and 3, lack of effective intervention means for complications of Western medicine treatment. The traditional Chinese medicine has the advantages of innate immunity regulation, and has definite curative effects on the immune system abnormality of lymphoma by using the intervention treatment of the traditional Chinese medicine in the observation waiting period, the maintenance treatment of the node therapy and the patients with complications. However, the traditional Chinese medicine lacks authoritative consensus or guidelines for treating the disease, and the knowledge of each person is not uniform, so that the dialectical system is complex, the traditional Chinese medicine symptoms of the disease cannot be accurately identified, and the application of clinicians is not facilitated. Disclosure of Invention The invention provides a method, a device, electronic equipment, a storage medium and a program product for constructing a traditional Chinese medicine syndrome identification model for lymphoma, which can improve the efficiency and accuracy of traditional Chinese medicine syndrome identification of lymphoma patients. According to an aspect of the invention, there is provided a method for constructing a model for identifying traditional Chinese medicine symptoms of lymphoma, the method comprising: Obtaining a traditional Chinese medicine waiting symptom list of a plurality of lymphoma patients, wherein the traditional Chinese medicine waiting symptom list comprises a plurality of symptom columns and symptoms corresponding to each symptom column, and the symptoms comprise single symptoms and compound symptoms; according to a predefined symptom code mapping dictionary, determining a numerical code corresponding to a single symptom and a numerical code character string corresponding to a compound symptom in the traditional Chinese medicine waiting symptom table, and generating a waiting symptom code table; Training a machine learning model according to the waiting symptom coding table to obtain a lymphoma traditional Chinese medicine symptom identification model. According to another aspect of the present invention, there is provided a device for constructing a model for identifying a syndrome in lymphoma, the device comprising: the system comprises a data acquisition module, a data processing module and a data processing module, wherein the data acquisition module is used for acquiring a traditional Chinese medicine waiting symptom list of a plurality of lymphoma patients, the traditional Chinese medicine waiting symptom list comprises a plurality of symptom columns and symptoms corresponding to each symptom column, and the symptoms comprise single symptoms and compound symptoms; The data coding module is used for determining a numerical code corresponding to a single symptom and a numerical code character string corresponding to a compound symptom in the traditional Chinese medicine waiting symptom table according to a predefined symptom code mapping dictionary, and generating a waiting symptom code table; And the model training module is used for training the machine learning model according to the waiting symptom coding table to obtain a lymphoma traditional Chinese medicine symptom identification model. According to another aspect of the present invention, there is provided an electronic apparatus including: at least one processor, and A memory communicatively coupled to the at least one processor, wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor, so that the at least one processor can execute the method for constructing the pattern recognition model in the lymphoma according to any embodiment of the present invention. According to another aspect of the present invention, there is provided a computer readable storage medium storing computer instructions for causing a processor to implement the method for constructing a pattern recognition model in lymphoma according to any one of the embodiments of the present in