Traditional Chinese medicine physique intelligent identification method based on conventional test data

Abstract

The invention belongs to the technical field of traditional Chinese medicine physique identification, and discloses a traditional Chinese medicine physique intelligent identification method based on conventional inspection data, comprising the following steps: acquiring personal basic information and conventional inspection data of each participant; performing data annotation on the acquired participant information data; training a traditional Chinese medicine syndrome type intelligent prediction model; the feasibility of the traditional Chinese medicine syndrome type identification method based on the conventional test data is evaluated; according to the mapping relation between the traditional Chinese medicine syndrome type and the physique, traditional Chinese medicine physique prediction is achieved. According to the traditional Chinese medicine physique intelligent identification method based on the conventional test data, the subjective influence in a traditional scale method and a doctor four-diagnosis method is effectively weakened, automatic processing and analysis of mass data are realized based on a machine learning algorithm, the time period of physique identification is shortened, the accuracy of a result is ensured, and the efficiency of physique identification is improved. The data quality is improved, seamless integration and comparative analysis of data between different researches and applications are promoted, and deep development of traditional Chinese medicine physique research is promoted.

Description

Technical Field

[0001] The present invention relates to the technical field of TCM constitution identification, and in particular to a TCM constitution intelligent identification method based on conventional test data. Background Art

[0002] The concept of traditional Chinese medicine profoundly reveals the subtle connection between the human body as a complex organic whole and the natural world. Constitution, as a unique attribute naturally formed in this process, profoundly affects the body's susceptibility to disease, tolerance and recovery process after illness. The wisdom of traditional Chinese medicine aims to achieve the far-reaching goal of preventing disease invasion and strengthening the foundation of constitution by subtly harmonizing the balance of yin and yang in the body and promoting the harmonious operation of the functions of the internal organs. The traditional Chinese medicine constitution classification system is based on the uniqueness of individual morphological structure, differences in physiological functions, diversity of psychological characteristics and adaptability to environmental changes. Each constitution carries its unique physiological manifestations and potential pathological tendencies. Therefore, in-depth understanding and accurate identification of individual constitution characteristics are of immeasurable value for early prevention of diseases, accurate diagnosis and formulation of personalized treatment plans.

[0003] In clinical practice, TCM constitution identification demonstrates its irreplaceable superiority. It not only provides doctors with a window to understand the patient's overall constitution, helping them to develop more appropriate and effective treatment strategies; it also opens the door to self-cognition and adjustment for patients, enhancing their enthusiasm and satisfaction in participating in the treatment process. In key areas such as comprehensive management of chronic diseases, adjustment of sub-health status, and health maintenance of the elderly, TCM constitution identification plays a core role, effectively improving people's health level and quality of life.

[0004] On the technical level, the methods and techniques of TCM constitution identification keep pace with the times and are constantly enriched and improved. From the traditional scale method (such as the questionnaire survey of the nine constitution classification methods), the constitution characteristics are captured through carefully designed standardized questions; to the exquisite use of the four diagnostic methods by doctors, relying on profound medical experience and keen intuition, through observation, smell, questioning and palpation to fully understand the patient's condition; to multimodal data analysis enabled by modern technology, using the power of artificial intelligence and big data technology to deeply mine multi-dimensional information such as tongue, pulse, and facial features, significantly improving the accuracy and efficiency of constitution identification. These innovative methods and technologies have jointly promoted the scientific and modern process of TCM constitution identification.

[0005] However, in the practice of TCM constitution identification, although it contains profound medical wisdom and wide application value, it still faces a series of problems and drawbacks. These challenges are mainly concentrated in the following aspects:

[0006] 1. Subjective limitations of the scale method: As one of the important means of TCM constitution identification, the scale method is based on individual self-reporting. This method inevitably introduces greater subjectivity and uncertainty. Different individuals may have deviations in their understanding and description of the same symptom or experience. Coupled with the influence of memory bias and emotional factors, it is difficult for the scale results to fully and objectively reflect the individual's physical condition. In addition, the design and interpretation of the scale also requires professional knowledge, and it is more likely to produce errors when used by non-professionals, affecting the accuracy and consistency of the results.

[0007] 2. The experience dependence of the doctor's four diagnostic methods: As a traditional diagnostic method of traditional Chinese medicine, the essence of the doctor's four diagnostic methods is that the doctor comprehensively judges the patient's physical condition through four methods: observation, smell, questioning, and palpation. This method is highly dependent on the doctor's personal experience, skills, and intuition, resulting in different judgments between different doctors due to differences in experience. This subjectivity not only affects the standardization and objectivity of diagnosis, but also increases the risk of misdiagnosis and missed diagnosis. At the same time, it is even more difficult for inexperienced doctors to accurately master and apply the four diagnostic methods.

[0008] 3. Lack of data standardization: In the field of TCM constitution identification, due to the lack of unified data standards and standardized operating procedures, there are differences in data collection, processing and analysis between different research or medical institutions. This difference not only limits the comparability and integration of data, but also affects the universality and reliability of research results. In addition, the lack of data standardization may also lead to waste of resources and duplication of work, hindering the in-depth development of TCM constitution identification research. Summary of the invention

[0009] The purpose of the present invention is to provide a method for intelligent identification of TCM constitution based on routine test data, aiming to optimize and accelerate the process of TCM constitution identification, and to achieve accurate prediction and analysis of individual physical characteristics through intelligent means.

[0010] To achieve the above object, the present invention provides a method for intelligent identification of TCM constitution based on conventional test data, comprising the following steps:

[0011] S1. Collect basic personal information and routine test data of each participant;

[0012] S2, marking the participant information data obtained in step S1;

[0013] S3, training the intelligent prediction model of TCM syndrome;

[0014] S4. Evaluate the feasibility of TCM syndrome identification methods based on routine test data;

[0015] S5. Based on the mapping relationship between TCM syndrome types and constitution, TCM constitution prediction is achieved.

[0016] Preferably, in step S1, collecting data information includes:

[0017] S11. Comprehensively record the basic information of each participant, including age, gender, and key information points of medical history;

[0018] S12. Obtain the participants’ physiological status, pathological changes, clinical manifestations, and detailed medical history information;

[0019] S13. Collect routine test data of participants, including blood routine, urine routine, biochemical examination and stool routine test at multiple levels.

[0020] Preferably, in step S2, the participant information data obtained in step S1 is annotated, and the specific process is as follows:

[0021] S21. Senior TCM experts will conduct detailed physical examinations on the subjects and accurately determine their TCM physical category;

[0022] S22. Use the identification results of TCM experts to accurately label the data;

[0023] S23. Combine with other data labeling methods to further expand and improve the richness and completeness of the data set; data labeling methods include scale-based self-assessment and machine learning-assisted labeling.

[0024] Preferably, in step S21, the constitution categories include: Qi deficiency constitution, blood stasis constitution, Yang deficiency constitution, Yin deficiency constitution, damp-heat constitution, phlegm-damp constitution, Qi stagnation constitution, special constitution, and peaceful constitution;

[0025] Syndrome types include: heart qi deficiency syndrome, qi deficiency and blood stasis syndrome, qi deficiency syndrome, lung qi deficiency syndrome, kidney qi deficiency syndrome, spleen qi deficiency syndrome, heart blood stasis syndrome, gastrointestinal blood stasis syndrome, blood heat syndrome, blood stasis syndrome, limb blood stasis syndrome, qi stagnation and blood stasis syndrome, heart yang deficiency syndrome, kidney yang deficiency syndrome, qi and yin deficiency syndrome, liver and kidney yin deficiency syndrome, lung yin deficiency syndrome, damp-heat downward movement syndrome, damp-heat accumulation in the lungs syndrome, damp-heat accumulation in the spleen syndrome, phlegm-damp internal obstruction syndrome, phlegm-damp obstruction syndrome, phlegm-damp obstruction in the lungs syndrome, lung and spleen qi deficiency and phlegm-damp syndrome, liver qi stagnation syndrome, spleen and stomach qi stagnation syndrome, allergic constitution, and qi and blood harmony.

[0026] Preferably, in step S3, the TCM syndrome type intelligent prediction model is trained, and the specific process is as follows:

[0027] S31. In view of the diversity and coexistence of TCM syndromes, an intelligent prediction model for TCM syndromes is constructed based on a deep neural network. The model takes routine test data as input and multiple labels of TCM syndromes as output, aiming to capture the complex relationship between input features and multiple output labels through deep learning technology.

[0028] S32. The intelligent prediction model of TCM syndrome type consists of an input layer, a hidden layer, and an output layer. The activation function of the output layer is set to sigmoid to meet the needs of multi-label classification tasks and ensure that the output value of each label is between 0 and 1. The Adam optimizer and the binary cross entropy loss function are used to quantify the difference between the model prediction results and the true labels and guide the learning process of the model.

[0029] S33. During the training process, the accuracy and generalization ability of the model are improved by adjusting model parameters, optimizing algorithms and regularization terms. At the same time, strategies such as cross-validation are used to evaluate the performance of the model to avoid overfitting and underfitting.

[0030] Preferably, in step S4, the feasibility assessment is comprehensively evaluated using multiple evaluation indicators, including top five accuracy and accuracy;

[0031] The top five accuracy rates are calculated as follows:

[0032]

[0033] The accuracy calculation method is as follows:

[0034]

[0035] Preferably, in step S5, TCM constitution prediction is implemented according to the mapping relationship between TCM syndrome type and constitution, and the specific process is as follows:

[0036] S51. Based on TCM theory, a mapping relationship network between TCM syndrome types and constitutions is constructed. Combined with the TCM syndrome type prediction model, the TCM syndrome type prediction results of the test samples are converted into corresponding constitution types, thereby achieving the intelligent prediction goal of TCM constitution identification;

[0037] S52. Use a combination of ring chart and radar chart to display the results of TCM constitution and syndrome type identification.

[0038] Therefore, the present invention adopts the above-mentioned TCM constitution intelligent identification method based on conventional test data, and the beneficial effects are as follows:

[0039] (1) The present invention uses routine test data as the basis for TCM constitution identification, effectively weakening the subjective influence of traditional scale method and doctor's four diagnostic methods, thereby providing more objective, reliable and consistent identification results, and setting a new standard for TCM constitution research.

[0040] (2) The present invention realizes the automatic and efficient processing and analysis of massive data by introducing advanced machine learning algorithms, greatly shortening the time period for physical constitution identification while ensuring the accuracy of the results, providing strong support for users to quickly obtain personalized physical constitution classification information.

[0041] (3) The present invention strictly follows the standardized routine test data specifications to ensure the uniformity, consistency and comparability of the data, which is not only conducive to improving data quality, but also promotes the seamless integration and comparative analysis of data between different studies and applications, and promotes the in-depth development of TCM constitution research.

[0042] The innovation and practicality of this invention are not limited to the field of TCM constitution identification. Its technical framework and data processing methods can also be flexibly applied to multiple related fields such as TCM diagnosis and treatment, showing broad development prospects and market potential, and making important contributions to the modernization and intelligent development of TCM.

[0043] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 This is a flow chart of the implementation of a method for intelligent identification of TCM constitution based on conventional test data of the present invention;

[0045] Figure 2 It is the result diagram of TCM constitution and syndrome type identification according to the present invention. DETAILED DESCRIPTION

[0046] The technical solution of the present invention is further described below through the accompanying drawings and embodiments. Figure 1 As shown, the present invention provides a method for intelligent identification of TCM constitution based on conventional test data, comprising the following steps:

[0047] S1. Collect basic personal information and routine test data of each participant.

[0048] S11. Comprehensively record the basic information of each participant, including age, gender, medical history and other key information points.

[0049] S12. Obtain the participants' physiological status, pathological changes, clinical manifestations and detailed medical history information.

[0050] S13. Collect routine test data of participants, including blood routine, urine routine, biochemical examination, stool routine and other test levels.

[0051] S2. Annotate the participant information data obtained in step S1.

[0052] S21. Experienced TCM experts will conduct detailed physical examinations on the subjects and accurately determine their TCM physical category.

[0053] Among them, the constitution categories include: Qi deficiency constitution, blood stasis constitution, Yang deficiency constitution, Yin deficiency constitution, damp-heat constitution, phlegm-dampness constitution, Qi stagnation constitution, special constitution, and peaceful constitution.

[0054] Syndrome types include: heart qi deficiency syndrome, qi deficiency and blood stasis syndrome, qi deficiency syndrome, lung qi deficiency syndrome, kidney qi deficiency syndrome, spleen qi deficiency syndrome, heart blood stasis syndrome, gastrointestinal blood stasis syndrome, blood heat syndrome, blood stasis syndrome, limb blood stasis syndrome, qi stagnation and blood stasis syndrome, heart yang deficiency syndrome, kidney yang deficiency syndrome, qi and yin deficiency syndrome, liver and kidney yin deficiency syndrome, lung yin deficiency syndrome, damp-heat downward movement syndrome, damp-heat accumulation in the lungs syndrome, damp-heat accumulation in the spleen syndrome, phlegm-damp internal obstruction syndrome, phlegm-damp obstruction syndrome, phlegm-damp obstruction in the lungs syndrome, lung and spleen qi deficiency and phlegm-damp syndrome, liver qi stagnation syndrome, spleen and stomach qi stagnation syndrome, allergic constitution, and qi and blood harmony.

[0055] S22. Use the identification results of Chinese medicine experts to accurately label the data to ensure high accuracy and uniformity of the data.

[0056] S23. Combine with other data annotation methods, such as scale-based self-assessment and machine learning-assisted annotation, to further expand and improve the richness and completeness of the dataset.

[0057] S3. Train the intelligent prediction model of TCM syndrome types.

[0058] S31. In view of the diversity and coexistence of TCM syndrome types, an intelligent prediction model for TCM syndrome types is constructed based on deep neural networks. The model takes routine test data as input and multiple labels of TCM syndrome types as output, aiming to capture the complex relationship between input features and multiple output labels through deep learning technology.

[0059] S32. The intelligent prediction model of TCM syndrome type consists of an input layer, a hidden layer, and an output layer; the activation function of the output layer is set to sigmoid to meet the needs of multi-label classification tasks and ensure that the output value of each label is between 0 and 1. The Adam optimizer and the binary cross entropy loss function are used to quantify the difference between the model prediction results and the true labels and guide the model's learning process.

[0060] S33. During the training process, the accuracy and generalization ability of the model are improved by adjusting model parameters, optimizing algorithms and regularization terms. At the same time, strategies such as cross-validation are used to evaluate the performance of the model to avoid overfitting and underfitting.

[0061] S4. Evaluate the feasibility of TCM syndrome identification methods based on routine test data.

[0062] The feasibility assessment uses a variety of evaluation indicators for comprehensive evaluation, including the top five accuracy and accuracy, that is, the Top-1 accuracy.

[0063] Top-5 accuracy measures how accurate the model is when provided with a small number of candidate predictions, as shown below:

[0064]

[0065] Accuracy is one of the most intuitive performance indicators, as shown below:

[0066]

[0067] S5. Based on the mapping relationship between TCM syndrome types and constitution, TCM constitution prediction is achieved.

[0068] S51. Based on TCM theory, a mapping relationship network between TCM syndrome types and constitutions is constructed. Combined with the TCM syndrome type prediction model, the TCM syndrome type prediction results of the test samples are converted into corresponding constitution types, thereby achieving the intelligent prediction goal of TCM constitution identification.

[0069] S52. Use a combination of ring chart and radar chart to display the results of TCM constitution and syndrome type identification.

[0070] like Figure 2 As shown in the figure, in the donut chart, each TCM constitution type is assigned a specific color and arc range. The length of the arc directly reflects the predicted probability of the constitution type, that is, the longer the arc, the higher the possibility of the constitution type. The radar chart shows the prediction results of TCM syndrome types in the form of pointers. Each pointer represents a specific TCM syndrome type. The direction of the pointer accurately points to the corresponding syndrome type, and the length of the pointer intuitively indicates the predicted probability of the syndrome type.

[0071] Example

[0072] This embodiment provides a method for intelligent identification of TCM constitution based on routine blood data, comprising the following steps:

[0073] S1. Collect basic personal information and blood test data of each participant.

[0074] S11. Comprehensively record the basic information of each participant, including age, gender, medical history and other key information points.

[0075] S12. Obtain the participants' physiological status, pathological changes, clinical manifestations and detailed medical history information.

[0076] S13. Collect the participants’ blood routine test data.

[0077] S2. Annotate the participant information data obtained in step S1.

[0078] S21. Senior TCM experts will conduct detailed constitution identification of the participants and accurately determine their TCM constitution category and syndrome category. Among them, the constitution categories include: Qi deficiency, blood stasis, Yang deficiency, Yin deficiency, damp-heat, phlegm-dampness, Qi stagnation, special constitution, and peaceful constitution; syndrome categories include: heart qi deficiency, qi deficiency and blood stasis, qi deficiency, lung qi deficiency, kidney qi deficiency, spleen qi deficiency, heart blood stasis, gastrointestinal blood stasis, blood heat syndrome, blood stasis, limb blood stasis, qi stagnation and blood stasis, heart yang deficiency, kidney yang deficiency, qi and yin deficiency, liver and kidney yin deficiency, lung yin deficiency, damp-heat downward injection, damp-heat accumulation in the lung, damp-heat accumulation in the spleen, phlegm-dampness internal obstruction, phlegm-dampness obstruction, phlegm-dampness obstruction in the lung, lung and spleen qi deficiency and phlegm-dampness syndrome, liver qi stagnation, spleen and stomach qi stagnation, allergic constitution, and qi and blood harmony.

[0079] S22. Use the identification results of Chinese medicine experts to accurately label the data to ensure high accuracy and uniformity of the data.

[0080] S23. Combine with other data annotation methods, such as scale-based self-report data and machine learning-assisted annotation, to further expand and improve the richness and completeness of the dataset.

[0081] S3. Train the intelligent prediction model of TCM syndrome types.

[0082] S31, TCM syndrome type intelligent prediction model consists of 1 input layer, 4 fully connected layers, 1 Dropout layer and 1 28-dimensional output layer. The number of neurons in the fully connected layer is 288, 416, 224 and 416 respectively. Among them, the model takes 24-dimensional blood routine test data as input and multivariate labels of TCM syndrome type as output.

[0083] S32. Use BinaryCrossentropy as the loss function, set the training cycle (epoch) to 200, and add an early stopping mechanism to quantify the difference between the model prediction results and the true labels and guide the model's learning process.

[0084] S4. Evaluate the feasibility of TCM syndrome identification based on routine test data. The evaluation indicators include top five accuracy and accuracy (also called Top-1 accuracy). The specific results are shown in Table 1.

[0085] Table 1 Model performance list

[0086] Model Name Top five accuracy Accuracy Intelligent Identification Model of TCM Syndrome Types 78％ 35％

[0087] S5. Based on the mapping relationship between TCM syndrome types and constitution, TCM constitution prediction is achieved.

[0088] S51. As shown in Table 2, a mapping relationship network between TCM syndrome types and constitutions is constructed based on TCM theory. Combined with the TCM syndrome type prediction model, the TCM syndrome type prediction results of the test samples are converted into corresponding constitution types, thereby achieving the intelligent prediction goal of TCM constitution identification.

[0089] S52. Use a combination of a ring chart and a radar chart to display the results of TCM constitution and syndrome identification, such as Figure 2 shown.

[0090] Table 2 Mapping relationship between TCM syndrome types and constitution

[0091]

[0092]

[0093] Therefore, the present invention adopts the above-mentioned intelligent identification method of TCM constitution based on routine test data, takes routine test data as the basis for TCM constitution identification, effectively weakens the subjective influence of traditional scale method and doctor's four diagnostic methods, thereby providing more objective, reliable and consistent identification results, and establishing a new standard for TCM constitution research; by introducing advanced machine learning algorithms, it realizes automatic and efficient processing and analysis of massive data, greatly shortens the time period of constitution identification, and ensures the accuracy of the results, providing strong support for users to quickly obtain personalized constitution classification information; the present invention strictly follows the standardized routine test data specifications to ensure the uniformity, consistency and comparability of the data, which is not only conducive to improving data quality, but also promotes seamless integration and comparative analysis of data between different studies and applications, and promotes the in-depth development of TCM constitution research.

[0094] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solution of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solution to deviate from the spirit and scope of the technical solution of the present invention.

Claims

1. A method for intelligent identification of TCM constitution based on routine test data, characterized in that: The following steps are involved: S1. Collect basic personal information and routine test data of each participant; S2, marking the participant information data obtained in step S1; S3, training the intelligent prediction model of TCM syndrome types; S4. Evaluate the feasibility of TCM syndrome identification methods based on routine test data; S5. Based on the mapping relationship between TCM syndrome types and constitution, TCM constitution prediction is achieved.

2. According to claim 1, a method for intelligent identification of TCM constitution based on routine test data, characterized in that: In step S1, the collected data information includes: S11. Comprehensively record the basic information of each participant, including age, gender, and key information points of medical history; S12. Obtain the participants’ physiological status, pathological changes, clinical manifestations, and detailed medical history information; S13. Collect routine test data of participants, including blood routine, urine routine, biochemical examination and stool routine test at multiple levels.

3. The method for intelligent identification of TCM constitution based on routine test data according to claim 1, characterized in that: In step S2, the participant information data obtained in step S1 is labeled. The specific process is as follows: S21. Senior TCM experts will conduct detailed physical examinations on the subjects and accurately determine their TCM physical category; S22. Use the identification results of TCM experts to accurately label the data; S23. Combine with other data labeling methods to further expand and improve the richness and completeness of the data set; data labeling methods include scale-based self-assessment and machine learning-assisted labeling.

4. The method for intelligent identification of TCM constitution based on routine test data according to claim 3, characterized in that: In step S21, the constitution categories include: Qi deficiency constitution, blood stasis constitution, Yang deficiency constitution, Yin deficiency constitution, damp-heat constitution, phlegm-damp constitution, Qi stagnation constitution, special constitution, and peaceful constitution; Syndrome types include: heart qi deficiency syndrome, qi deficiency and blood stasis syndrome, qi deficiency syndrome, lung qi deficiency syndrome, kidney qi deficiency syndrome, spleen qi deficiency syndrome, heart blood stasis syndrome, gastrointestinal blood stasis syndrome, blood heat syndrome, blood stasis syndrome, limb blood stasis syndrome, qi stagnation and blood stasis syndrome, heart yang deficiency syndrome, kidney yang deficiency syndrome, qi and yin deficiency syndrome, liver and kidney yin deficiency syndrome, lung yin deficiency syndrome, damp-heat downward movement syndrome, damp-heat accumulation in the lungs syndrome, damp-heat accumulation in the spleen syndrome, phlegm-damp internal obstruction syndrome, phlegm-damp obstruction syndrome, phlegm-damp obstruction in the lungs syndrome, lung and spleen qi deficiency and phlegm-damp syndrome, liver qi stagnation syndrome, spleen and stomach qi stagnation syndrome, allergic constitution, and qi and blood harmony.

5. The method for intelligent identification of TCM constitution based on routine test data according to claim 1, characterized in that: In step S3, the TCM syndrome type intelligent prediction model is trained. The specific process is as follows: S31. In view of the diversity and coexistence of TCM syndromes, an intelligent prediction model for TCM syndromes is constructed based on a deep neural network. The model takes routine test data as input and multiple labels of TCM syndromes as output, aiming to capture the complex relationship between input features and multiple output labels through deep learning technology. S32. The intelligent prediction model of TCM syndrome type consists of an input layer, a hidden layer, and an output layer. The activation function of the output layer is set to sigmoid to meet the needs of multi-label classification tasks and ensure that the output value of each label is between 0 and 1. The Adam optimizer and the binary cross entropy loss function are used to quantify the difference between the model prediction results and the true labels and guide the learning process of the model. S33. During the training process, the accuracy and generalization ability of the model are improved by adjusting model parameters, optimizing algorithms and regularization terms. At the same time, strategies such as cross-validation are used to evaluate the performance of the model to avoid overfitting and underfitting.

6. The method for intelligent identification of TCM constitution based on routine test data according to claim 1, characterized in that: In step S4, the feasibility assessment is comprehensively evaluated using a variety of evaluation indicators, including top five accuracy and accuracy; The top five accuracy rates are calculated as follows: The accuracy calculation method is as follows:

7. The method for intelligent identification of TCM constitution based on routine test data according to claim 1, characterized in that: In step S5, according to the mapping relationship between TCM syndrome type and constitution, TCM constitution prediction is realized, and the specific process is as follows: S51. Based on TCM theory, a mapping relationship network between TCM syndrome types and constitutions is constructed. Combined with the TCM syndrome type prediction model, the TCM syndrome type prediction results of the test samples are converted into corresponding constitution types, thereby achieving the intelligent prediction goal of TCM constitution identification; S52. Use a combination of ring chart and radar chart to display the results of TCM constitution and syndrome type identification.

Images