A method and system for identifying the constitution of a patient with copd

Abstract

The present application relates to the field of constitution identification, and in particular to a COPD patient constitution identification method and system, wherein scale data is obtained through a COPD disease associated constitution screening scale; meanwhile, COPD associated clinical index data and tongue image data are obtained; after standardization processing of the three types of data, constitution model operation is performed to calculate a comprehensive score, the constitution type is determined according to the comprehensive score, the scale questions are selected from the most closely associated items in each constitution sub-scale, the number of questions is compressed from 60 in the general version to 12, the identification time is effectively shortened, the accessibility and compliance of COPD patients are significantly improved, and the application is suitable for the actual use scenarios of COPD patients, especially moderate and severe patients; the three types of heterogeneous data sources are fused for identification, the COPD disease characteristics are preset with weights, the subjective feeling and objective index, and the organic integration of traditional Chinese medicine diagnosis and western medicine detection are realized, and the accuracy and objectivity of identification are significantly improved.

Description

Technical Field

[0001] This invention relates to the field of constitution identification, and in particular to a method and system for identifying the constitution of COPD patients. Background Technology

[0002] Chronic obstructive pulmonary disease (COPD) is characterized by persistent airflow limitation, recurrent coughing, wheezing, and sputum production, and decreased lung function. It has a prolonged course, is prone to acute exacerbations, has many complications, and results in a poor quality of life. Western medicine mainly focuses on airway dilation, anti-infection, oxygen therapy, and respiratory rehabilitation, lacking an individualized and holistic approach to treatment. In contrast, traditional Chinese medicine (TCM) constitution identification can grasp the core imbalances of patients from the perspective of both innate endowment and acquired disorders, which is a key link in achieving individualized prevention and treatment, reducing acute exacerbations, slowing disease progression, and improving prognosis.

[0003] Traditional Chinese medicine (TCM) constitution theory is a core component of the "prevention of disease" theory. Professor Wang Qi's team established a classification standard for nine TCM constitutions (balanced constitution, qi deficiency constitution, yang deficiency constitution, yin deficiency constitution, phlegm-dampness constitution, damp-heat constitution, blood stasis constitution, qi stagnation constitution, and special constitution) and the "Classification and Determination of TCM Constitutions" scale, which has been widely adopted nationwide and has become an important tool for TCM health management. In 2025, the National Standardization Management Committee released an updated national standard, GB / T46939-2025, "Classification and Determination of TCM Constitutions," which innovated the description of constitution characteristics based on evidence and further optimized the determination method.

[0004] Existing body constitution identification technologies are mainly divided into two categories: (1) Paper / electronic scale method Based on the "Classification and Determination of Traditional Chinese Medicine Constitutions" standard, a 60-question (or simplified 30-question) constitution scale is used for self-assessment or peer assessment, and the constitution type is determined by the scores of each constitution subscale. This method is currently the most widely used in clinical practice, but it has the following problems: ① The number of questions is large and the completion time is long (the full version takes 15-20 minutes and the simplified version takes 8-10 minutes), which is not suitable for most COPD patients who have difficulty breathing and cannot answer questions for a long time; ② This is a general scale and has not been optimized for the specific pathogenesis characteristics of COPD patients (deficiency of lung qi as the root cause and phlegm turbidity and blood stasis as the symptoms); ③ The scale scores are not directly related to COPD clinical indicators (such as FEV1%Pred, CAT score, and frequency of acute exacerbations), and cannot reflect the correspondence between physical condition and disease severity. ④ The lack of objective indicators such as tongue appearance makes the identification results highly subjective.

[0005] (2) Intelligent physical constitution identification system In recent years, with the development of artificial intelligence technology, various intelligent physical constitution identification systems have emerged, such as: Professor Wang Qi's team at Beijing University of Chinese Medicine has 36 invention patents related to constitution identification, covering constitution identification scales, constitution-disease association databases, and constitution conditioning programs, but all of them are general designs and have not been specifically optimized for COPD.

[0006] The invention patent with application number 202511407370.1 and title "Intelligent Monitoring System for Traditional Chinese Medicine Constitution Identification" identifies constitution through questionnaires and tongue images, but it is designed for general health management scenarios and does not establish a mapping relationship between constitution type and COPD clinical phenotype. Summary of the Invention

[0007] To address the aforementioned problems, this invention proposes a method and system for identifying the constitution of COPD patients. By integrating three types of data sources—body weight scale data, COPD-specific clinical indicator data, and tongue image data—a COPD-specific constitution identification model is constructed. This model enables rapid (≤5 minutes), accurate, and objective identification of the TCM constitution of COPD patients and automatically outputs the constitution risk level, providing a foundation for subsequent personalized intervention.

[0008] To achieve the above objectives, the present invention is implemented according to the following technical solution: This invention provides a method for identifying the physical constitution of COPD patients, comprising the following steps: Step S1: Obtain scale data through the COPD pathogenesis-related constitution screening scale; the COPD pathogenesis-related constitution screening scale is based on the statistical analysis results of the constitution distribution of COPD patients and selects the high-incidence constitution types, including items from the standard constitution scale of traditional Chinese medicine that are differentially associated with the pathogenesis of COPD in each high-incidence constitution type; obtain clinical indicator data and tongue image data associated with COPD; Step S2: Standardize the scale data and clinical indicator data to obtain scale scores and clinical indicator scores respectively, and preprocess and standardize the tongue image data to obtain tongue image scores. Step S3: Calculate the overall score by combining the scale score, clinical indicator score, and tongue appearance score using the following formula: Overall score = Scale score × W1 + Clinical indicator score × W2 + Tongue appearance score × W3 Wherein, W1 is the weight of the scale score; W2 is the weight of the clinical indicator score; and W3 is the weight of the tongue appearance score. Step S4: Determine the body type based on the overall score according to the following rules: a. If a certain physical constitution score is greater than or equal to the first threshold and is the highest among all the physical constitutions of the patient, then it is determined to be that physical constitution type; b. If a certain physical condition's overall score is greater than or equal to the first threshold, and the difference between it and the highest-scoring physical condition is less than the second threshold, then it is determined to be a mixed physical condition. c. If all comprehensive physical fitness scores are less than the third threshold, then it is determined that the individual does not belong to any of the high-incidence physical fitness types. d. If the third threshold ≤ the highest score constitution < the first threshold, then it is determined to be a constitution tendency; The first threshold is a high score threshold set based on the total score of the comprehensive score; the second threshold is a difference range set based on the total score of the comprehensive score; and the third threshold is a low score threshold set based on the total score of the comprehensive score.

[0009] In the above technical solution, the common constitution types in step S1 include Qi deficiency constitution, Yang deficiency constitution, phlegm-dampness constitution and blood stasis constitution. The number of questions is two for each common constitution type, and each question is assigned a graded score.

[0010] In the above technical solution, the standardization processing method for the scale data in step S2 includes: Step S21: Add the scores of the two questions for each body type to obtain the first raw score for that body type; Step S22: Convert the first raw score into a scale score using the following formula: .

[0011] In the above technical solution, in step S2, the standardization processing method for clinical indicator data is to convert it into a percentage score based on the correlation between clinical indicators and physical condition.

[0012] In the above technical solution, step S2, the preprocessing includes unifying the resolution, lighting conditions and shooting angle of the image, automatically segmenting the tongue region, and extracting feature parameters such as tongue color, tongue coating thickness, tongue coating color, tongue body size, and presence or absence of teeth marks; the standardization processing of the tongue image data involves comparing the feature parameters with standard tongue image templates for each high-incidence constitution type to obtain a matching score, and then calculating the tongue image score through weighted average.

[0013] The above technical solution also includes step S5, in which step S4 is executed when it is determined that the constitution does not belong to any of the high-incidence constitution types: obtaining backup scale data through the COPD pathogenesis-related constitution screening backup scale; the COPD pathogenesis-related constitution screening backup scale is a low-incidence constitution type selected based on the statistical analysis results of the constitution distribution of COPD patients, including selecting items from the TCM constitution standard scale that are differentially related to the COPD pathogenesis in each low-incidence constitution type; and determining whether the constitution type is Yin deficiency constitution or balanced constitution based on the backup scale data.

[0014] The above technical solution also includes step S6, which is executed when the constitution is determined to be balanced in step S5: obtaining verification scale data through a verification scale; the verification scale is selected from the standard TCM constitution scale, and items that are correlated with the pathogenesis of COPD are selected from the balanced constitution; each item of the verification scale is set with a graded score. If all items are scored out of full marks and the clinical indicators are all normal, then the constitution is confirmed as balanced; otherwise, the constitution with the second highest score is taken as the main constitution.

[0015] The above technical solution also includes step S7: After obtaining the body type, assess the confidence level according to the following assessment rules: e. If the scale score is ≥70 points, the clinical indicator score is ≥60 points, and the tongue appearance score is ≥60 points, then it is considered to have high confidence. f. If the scale score is ≥60 points and the clinical indicator score is ≥40 points, then it is of medium confidence level; g. If the scale score is ≥60 points, but the clinical indicator score is <40 points, then the confidence level is low.

[0016] The above technical solution also includes step S8: After obtaining the body type, based on the mapping relationship database between body type and COPD clinical phenotype, check whether the clinical indicator data conforms to the mapping relationship. If not, then step S1 to S8 are reviewed.

[0017] Another aspect of the present invention provides a COPD patient constitution identification system for implementing any of the above-mentioned COPD patient constitution identification methods, comprising: The storage module is used to store the COPD pathogenesis-related constitution screening scale, the COPD pathogenesis-related constitution screening backup scale, the confirmation scale, the standard tongue image template database, and the mapping relationship database between constitution type and COPD clinical phenotype; The data acquisition module is used to collect patients' scale data, clinical indicator data, tongue image data, backup scale data, and confirmation scale data; The data processing module, connected to the data acquisition module, is used to perform standardized processing of scale data and clinical indicator data, as well as preprocessing and standardized processing of tongue image data. The physical fitness model calculation module, connected to the data processing module, is used to calculate the comprehensive score according to the constructed comprehensive score model. The constitution type determination module is connected to the constitution model calculation module and is used to determine the constitution type according to the determination rules. The confidence level assessment module, connected to the body type determination module, is used to assess the confidence level according to the assessment rules. The reverse verification module, connected to the body type determination module, is used to check whether the clinical indicator data conforms to the mapping relationship; The display module is used to output and display the constitution type and corresponding pathogenesis analysis text, confidence level, scores of each item, data feature description, risk warning, and intervention suggestions.

[0018] Compared with the prior art, the present invention has the following beneficial effects: This invention designs a rapid scale for identifying correlations based on the body constitution distribution characteristics of COPD patients. The scale items are carefully selected from the subscales of body constitution that are most closely related to the pathogenesis of COPD, resulting in stronger disease-specific identification power. The number of items is reduced from 60 in the general version to 12, and the identification time is shortened from 15-20 minutes to 3-5 minutes, significantly improving accessibility and compliance for COPD patients. It is suitable for the actual use scenarios of COPD patients, especially those with moderate to severe conditions. The invention integrates three heterogeneous data sources: body constitution scale data, COPD clinical indicator data, and tongue image data. By pre-setting weights for COPD disease characteristics, it achieves the organic integration of subjective feelings and objective indicators, as well as traditional Chinese medicine diagnosis and Western medicine testing, significantly improving the accuracy and objectivity of identification and reducing the subjective bias of a single data source. The overall identification method is compatible with the 2025 version of the national standard GB / T46939-2025, thereby improving the effectiveness of body constitution identification for COPD patients. Attached Figure Description

[0019] 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.

[0020] Figure 1 This is a logical framework diagram of the COPD patient constitution identification method of the present invention; Figure 2 This is a schematic diagram of the composition and structure of the COPD patient constitution identification system of the present invention. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are some embodiments of the present invention, but not all embodiments.

[0022] According to a method for identifying the physical constitution of COPD patients according to the present invention, such as Figure 1 As shown, this embodiment includes the following steps: Step S1: Obtain scale data through the COPD pathogenesis-related constitution screening scale; the COPD pathogenesis-related constitution screening scale is based on the statistical analysis results of the constitution distribution of COPD patients and selects the high-incidence constitution types, including items from the standard constitution scale of traditional Chinese medicine that are differentially associated with the pathogenesis of COPD in each high-incidence constitution type; obtain clinical indicator data and tongue image data associated with COPD; The TCM constitution standard scale is based on the 2025 national standard GB / T 46939-2025 "Classification and Determination of TCM Constitution". Based on meta-analysis data, among COPD patients, Qi deficiency accounts for 28%, Yang deficiency accounts for 21%, Phlegm-dampness accounts for 14%, and Blood stasis accounts for 9%, totaling 72%. Yin deficiency and balanced constitution each account for about 8%. This invention selects Qi deficiency, Yang deficiency, Phlegm-dampness, and Blood stasis as the four most common constitution types. The COPD pathogenesis-related constitution screening scale screens these four constitution types and excludes constitution types with a very low proportion (<3%) in COPD patients (such as damp-heat constitution, Qi stagnation constitution, and special constitution). To shorten the identification time, this invention selects items from the TCM constitution standard scale that are most closely related to the pathogenesis of COPD, i.e., the items with the strongest discriminative power, from each high-incidence constitution type. For example, two items are selected for each constitution, significantly reducing the number of items in the scale from the general version of 60. Each item is assigned a graded score, such as using a Likert 5-point scale, i.e., 0 = never / never, 1 = rarely, 2 = sometimes, 3 = often, 4 = always. Items related to the pathogenesis of COPD have been extensively documented in existing research, such as the core items related to cough, wheezing, shortness of breath, aversion to cold, and excessive phlegm extracted from the core items in Appendix B of the local standard "TCM Health Management Guidelines for Chronic Obstructive Pulmonary Disease DB44 / T 2576—2024" regarding the assessment of TCM health management for COPD. Therefore, the selection of items can be obtained from existing literature or standards during implementation.

[0023] For example, the specific questions for the four most common constitution types—Qi deficiency, Yang deficiency, Phlegm-dampness, and Blood stasis—are shown in Table 1: Table 1

[0024] Studies have shown that different constitution types are significantly associated with clinical indicators of COPD. The clinical indicators of Qi deficiency constitution are reduced FEV1%Pred, CAT score ≥10, and AE ≥2 times / year; the clinical indicators of Yang deficiency constitution are mMRC ≥2, FEV1%Pred <50%, and low SpO2; the clinical indicators of phlegm-dampness constitution are increased sputum volume and high BMI; and the clinical indicators of blood stasis constitution are purplish lips and elevated pulmonary artery pressure. Therefore, the above clinical indicators can be used as objective verification dimensions for constitution identification to improve the accuracy of identification.

[0025] COPD-related clinical indicator data can be obtained through system collection from the hospital's HIS system or through manual entry. Specific clinical indicator data are shown in Table 2: Table 2

[0026] Tongue appearance, as a core component of "inspection" in Traditional Chinese Medicine (TCM), is a key objective tool for identifying body constitution. COPD patients with different constitutions typically exhibit characteristic tongue appearances: Qi deficiency constitution: pale red tongue, thin white coating, swollen tongue with teeth marks. Yang deficiency constitution: pale and swollen tongue, white and slippery tongue coating, large tongue body with teeth marks. Phlegm-dampness constitution: pale red or pale white tongue, white or yellow greasy coating, and large tongue body. Blood stasis constitution: Dark purple tongue or with ecchymosis, thin white or thin yellow coating. Therefore, incorporating tongue image data into the identification process can significantly improve the objectivity and accuracy of identification.

[0027] Tongue image data can be obtained by taking pictures using a mobile camera or a dedicated tongue image acquisition device.

[0028] Step S2: Standardize the scale data and clinical indicator data to obtain scale scores and clinical indicator scores respectively, and preprocess and standardize the tongue image data to obtain tongue image scores. Specifically, the standardization methods for scale data include: Step S21: Add the scores of the two questions for each body type to obtain the first original score for that body type; in this embodiment of the invention, the first original score is out of 8 points; Step S22: Convert the first raw score into a scale score using the following formula: .

[0029] The standardization of clinical indicator data involves converting it into a percentage score based on the correlation between clinical indicators and physical condition. Specifically, this includes: Step S23: Assign different percentage scores to different levels according to the grading standards of each clinical indicator; Step S24: Convert the clinical indicator data into corresponding scores; Step S25: Calculate the clinical indicator scores associated with each constitution based on the correlation between each constitution and clinical indicators, where the clinical indicator score is the weighted average of the scores of each associated clinical indicator.

[0030] Specific examples are as follows: Different percentage scores are assigned to different levels based on international grading standards for each clinical indicator. For example, the severity scores of clinical indicators are graded according to the following international standards: FEV1%Pred according to the GOLD 2025 pulmonary function classification (GOLD I: ≥80%, GOLD II: 50-79%, GOLD III: 30-49%, GOLD IV: <30%); CAT score according to the clinical grading standards of the COPD assessment test (low symptom burden <10 points, moderate symptom burden 10-20 points, high symptom burden >20 points); mMRC grading according to the modified UK Medical Research Council Dyspnea Scale 0-4; BMI according to the WHO body mass index grading standards (normal <24, overweight 24-27.9, obese ≥28); number of acute exacerbations according to the GOLD frequent acute exacerbation phenotype; sputum volume according to the clinical semi-quantitative grading; SpO2 according to the clinical grading of resting hypoxia; and cyanosis of the lips according to the grading of signs observed in traditional Chinese medicine. Specific scoring standards are illustrated in Table 3. Table 3

[0031] The correlation between various physical constitutions and clinical indicators is shown in Table 4: Table 4

[0032] Taking a patient as an example, the clinical indicators were: FEV1%Pred = 55%, CAT = 12, AE = 2, mMRC = 2, SpO2 = 92%, moderate sputum volume, BMI = 21.5, and no lips. Upon reviewing the scoring table, the clinical indicator data were converted into scores as follows: FEV1%Pred = 55% → 50, CAT = 12 → 30, AE = 2 → 65, mMRC = 2 → 40, SpO2 = 92% → 30, moderate sputum volume → 50, BMI = 21.5 → 0, no lips → 0. Assuming equal weighting for each clinical indicator, the scores for each constitution's clinical indicators are calculated as follows: Qi deficiency constitution: (50+30+65) / 3=48.3 points; Yang deficiency quality: (50+40+30) / 3=40.0 points; Phlegm-dampness constitution: (50+0) / 2=25 points; Blood stasis constitution: (30+0) / 2=15 points.

[0033] The preprocessing of tongue image data includes standardizing image resolution, lighting conditions, and shooting angle; automatically segmenting the tongue region; and extracting feature parameters such as tongue color, tongue coating thickness, tongue coating color, tongue slenderness, and the presence or absence of teeth marks. The technical implementation principle is as follows: Tongue images are standardized to a standard resolution using size scaling and proportion-preserving algorithms to eliminate pixel differences in the shooting device; illumination normalization is achieved using white balance correction and multi-scale Retinex algorithms to eliminate interference from ambient light, color temperature, and shadows; shooting angle correction is completed through perspective transformation or keypoint alignment to ensure distortion-free tongue morphology; automatic segmentation of the tongue region is achieved using color space thresholding, morphological operations, or deep learning semantic segmentation models to extract effective regions of interest; within the segmented region, tongue color and tongue coating color features are extracted through color space mean statistics; tongue coating thickness is quantified through texture analysis and area ratio; tongue slenderness is determined through geometric parameter calculation; and the presence or absence of teeth marks is detected through contour curvature analysis. This technique is a standard technique in the field of digital tongue diagnosis and will not be elaborated upon in detail in this invention.

[0034] The standardization processing of the tongue image data involves comparing the feature parameters with standard tongue templates for each high-incidence constitution type to obtain a matching score, and then calculating the tongue image score through weighted average.

[0035] The methods for achieving a matching score for tongue color features through the system include, but are not limited to: converting RGB format tongue image data acquired by the device into a CIE (Characteristic Image of the Tongue) that is perceived as uniform by the human eye. Standard color space, pre-constructed standard tongue color for four types of high-risk constitutions. The value template, based on color difference calculation and normalization in a three-dimensional color space, quantifies the matching score between the tongue color to be tested and various constitution templates, achieving a standardized constitution matching score for tongue color characteristics; specifically including: Step S26: Acquire the original tongue image of the patient using an image acquisition device. The original tongue image is in RGB color format. Extract the RGB mean data of the effective area of ​​the tongue image. Convert the RGB mean data to CIE XYZ tristimulus values ​​using a preset standard color conversion matrix. Then, convert the XYZ tristimulus values ​​to CIE values ​​using the CIE standard conversion algorithm. The specific conversion process for color space parameters is illustrated below: The conversion relationships between X, Y, Z and R, G, B are as follows:

[0036] CIE The conversion formulas between color space and X, Y, Z axes are as follows:

[0037] in:

[0038] In the formula, Lightness; It is the colorimetric index, which reflects the color components of color. The proportions change from green to red. The components are a gradient from blue to yellow; X, Y, and Z are the tristimulus values ​​of the color; X0, Y0, and Z0 are the tristimulus values ​​on a fully diffuse surface under CIE standard illumination conditions.

[0039] Step S27: The system presets the standard tongue color for each body type. Value templates are shown in Table 5: Table 5

[0040] Calculate the color difference between the tested tongue color and the standard tongue color of each body constitution template using the following formula:

[0041] The numerical value is negatively correlated with the degree of tongue color matching. The smaller the value, the closer the color characteristics of the tongue color being tested are to the standard tongue color of the corresponding body type, and the higher the matching degree. The larger the value, the more significantly the tongue color features deviate from the standard template, and the lower the matching degree.

[0042] Step S28: Calculate the tongue color score using the following formula: Tongue color score = max(0, 100 – ×k) Where k=100 / , It represents the maximum color difference between any two body type templates. This formula, through normalization, calculates a standardized tongue color score within the 0-100 range, enabling adaptive extreme value constraints: when the tongue color being tested perfectly matches the body type standard template... =0, the tongue color score is out of 100; when the deviation of the tested tongue color reaches the maximum color difference threshold. At that time, the tongue color score was 0, effectively avoiding negative score abnormal data and ensuring that the scoring results remained stable within the valid range. Based on the pre-set scoring rules for phlegm-dampness constitution, the tongue color score was calculated using a normalization method and then adapted to its 80-point full score benchmark to complete the calibration and correction of the phlegm-dampness constitution tongue color score.

[0043] A specific example of the standardized calculation process for tongue color scores: The tongue color of a patient under test was obtained after image acquisition and color conversion. The measured parameters are (72, 12, 12). The standard template parameters for Qi deficiency constitution (70, 12, 10) are selected for matching and scoring calculation. ① Calculate color difference: ; ②Calculation =21.19, calculate the normalization coefficient k=100 / 21.19=4.72; ③ Calculate the tongue color score: 100 - 2.83 × 4.72 = 86.64 points. The final matching score of the patient's tongue color with the Qi deficiency constitution is 86.64 points, which accurately quantifies the degree of matching between the two.

[0044] The method used in the industry to obtain the matching score of tongue coating and tongue body thickness features is to extract the feature vectors of each dimension from the tongue image to be tested and the standard tongue image template, calculate the similarity of the corresponding feature vectors according to the body constitution to obtain the matching degree, and then standardize and convert it into a percentage score. This technology is a conventional implementation method and will not be described in detail in this application.

[0045] Examples of matching scores for tongue color, tongue coating, and tongue body thickness are shown in Tables 6, 8, and 10: Table 6 Tongue Color Matching Rating Table

[0046] Each column of "full marks" corresponds to the standard tongue color for that constitution: Qi deficiency constitution (full marks 100) is pale red, Yang deficiency constitution (full marks 100) is pale and swollen, phlegm-dampness constitution (full marks 80) is pale red, and blood stasis constitution (full marks 100) is dark red. The patient's tongue color is compared to the standard tongue color; the greater the deviation, the lower the score. The characteristic tongue appearance of phlegm-dampness constitution is manifested in two dimensions: tongue coating (thick and greasy coating → 100 points) and tongue body (swollen → 100 points). The discriminative power of tongue color is relatively weak; therefore, in this table, the full score for tongue color for phlegm-dampness constitution is 80 points.

[0047] Taking a certain patient as an example, whose tongue color is pale red, their tongue color score is shown in Table 7: Table 7

[0048] Table 8. Moss Weevil Matching Score Table

[0049] Among them, the criteria for Qi deficiency constitution are a thin white tongue coating, the criteria for Yang deficiency constitution are a white and slippery tongue coating, and the criteria for phlegm-dampness constitution are a thick / greasy tongue coating. Blood stasis constitution is characterized by a thin white tongue coating, but the correlation is not strong (the highest value is only 60).

[0050] Taking a certain patient as an example, whose tongue coating is thin and white, the tongue coating score is shown in Table 9: Table 9

[0051] Table 10 Tongue Size Matching Rating Table

[0052] Among them: the standard for Qi deficiency constitution is large size with teeth marks, the standard for Yang deficiency constitution is large size with teeth marks, the standard for phlegm-dampness constitution is large size (regardless of teeth marks), and the standard for blood stasis constitution is moderate size (no teeth marks).

[0053] Taking a certain patient as an example, whose tongue is swollen with teeth marks, the score for tongue slenderness is shown in Table 11: Table 11

[0054] Taking a weighted average set to an equal-weighted average as an example, where each item has a weight of 1 / 3, the tongue image score is calculated using the following formula: Tongue appearance score = (tongue color score + tongue coating score + tongue body size score) / 3 Step S3: Calculate the overall score by combining the scale score, clinical indicator score, and tongue appearance score using the following formula: Overall score = Scale score × W1 + Clinical indicator score × W2 + Tongue appearance score × W3 Wherein, W1 is the weight of the scale score; W2 is the weight of the clinical indicator score; and W3 is the weight of the tongue appearance score. A three-layer weighted fusion algorithm was adopted to integrate scale scores, clinical indicator scores, and tongue appearance scores according to preset weights. The weight settings were optimized based on COPD disease characteristics. Preferably, the body mass index is the gold standard for constitution identification, and the patient's subjective feelings are the core of constitution identification, so W1 was set to 60%. Since there is a clear correlation between constitution and clinical indicators in COPD patients (e.g., the lowest FEV1%Pred for Qi deficiency constitution and the highest mMRC for Yang deficiency constitution), clinical indicators can serve as a strong objective verification for constitution identification, so W2 was set to 30%. The 30% weight ensures the corrective effect of objective indicators while avoiding over-reliance on Western medicine indicators and deviating from the essence of TCM constitution. Tongue appearance is an objective observation indicator, but it is greatly affected by differences in shooting conditions and equipment, so a lower weight was set as an auxiliary verification, so W3 was set to 10%.

[0055] Step S4: Based on the scale score, clinical indicator score, and tongue appearance score, all using a 100-point scale, the first threshold is set to 60 points; the second threshold is set to 10 points; and the third threshold is set to 40 points. The constitution type is determined according to the following rules based on the comprehensive score: a. If a certain physical constitution has a comprehensive score of ≥60 points and is the highest among all the physical constitutions of the patient, then it is determined to be that physical constitution type; b. If a certain physical condition score is ≥60 points and the difference between it and the highest-scoring physical condition is <10 points, it is determined to be a mixed physical condition. c. If the overall score of all physical conditions is less than 40 points, then it is determined that it does not belong to any of the common physical condition types mentioned. d. If 40 points ≤ highest score constitution < 60 points, then it is determined that the constitution is inclined to that constitution; This invention designs a rapid scale for identifying correlations based on the body constitution distribution characteristics of COPD patients. The scale items are carefully selected from the subscales of body constitution that are most closely related to the pathogenesis of COPD, resulting in stronger disease-specific identification power. The number of items is reduced from 60 in the general version to 12, and the identification time is shortened from 15-20 minutes to 3-5 minutes, significantly improving accessibility and compliance for COPD patients. It is suitable for the actual use scenarios of COPD patients, especially those with moderate to severe conditions. The invention integrates three heterogeneous data sources: body constitution scale data, COPD clinical indicator data, and tongue image data. By pre-setting weights for COPD disease characteristics, it achieves the organic integration of subjective feelings and objective indicators, as well as traditional Chinese medicine diagnosis and Western medicine testing, significantly improving the accuracy and objectivity of identification and reducing the subjective bias of single data sources. The overall identification method is compatible with the 2025 version of the national standard GB / T 46939-2025, thereby improving the effectiveness of body constitution identification for COPD patients.

[0056] In a preferred embodiment of the present invention, when it is determined that the constitution does not belong to any of the high-incidence constitution types, step S5 is executed: Reserve scale data is obtained through the COPD pathogenesis-related constitution screening reserve scale; the COPD pathogenesis-related constitution screening reserve scale is a low-incidence constitution type selected based on statistical analysis results of the constitution distribution of COPD patients, including items selected from the TCM constitution standard scale that are differentially related to the COPD pathogenesis in each low-incidence constitution type; the constitution type is determined to be Yin deficiency or balanced according to the reserve scale data. Low-incidence constitution types include Yin deficiency and balanced constitution. Correspondingly, the COPD pathogenesis-related constitution screening reserve scale contains items that are differentially related to the COPD pathogenesis in either of the two constitution types. Preferably, the two items with the closest correlation and strongest discriminative power are selected, for example, items related to Yin deficiency constitution, as follows: Q9: Do you experience dry mouth and throat, and hot palms and soles? Q10: Do you experience night sweats or sweating during sleep? Each question is also graded, such as using a Likert 5-level rating system, where 0 = never / never, 1 = rarely, 2 = sometimes, 3 = often, and 4 = always.

[0057] Methods for determining whether someone has a Yin deficiency constitution or a balanced constitution include: Step S51: Add the scores of the two questions together to obtain the second original score; Step S52: Convert the second raw score into the alternative scale score using the following formula:

[0058] Step S53: If the score on the backup scale is ≥50, the individual is determined to have a Yin deficiency constitution; if the score on the backup scale is <50, the individual is determined to have a balanced constitution tendency.

[0059] As a preferred embodiment of the present invention, when the tendency of the constitution to be balanced is determined in step S5, step S6 is executed: obtaining verification scale data through a verification scale; the verification scale is selected from the standard scale of traditional Chinese medicine constitution, and items that are correlated with the pathogenesis of COPD are selected from the balanced constitution; each item of the verification scale is set with a graded score. If all items are scored to the maximum and the clinical indicators are normal, the constitution is confirmed to be balanced; otherwise, the constitution with the second highest score is taken as the main constitution.

[0060] Preferably, the confirmation scale has two items, as shown in the following example: Q11: Are you usually energetic and not easily fatigued? Q12: Do you sleep well, have normal bowel movements, and have an outgoing personality? Each item is assigned a graded score, such as a Likert 5-point scale: 0 = never / never, 1 = rarely, 2 = sometimes, 3 = often, 4 = always. If both items are scored out of 4, and the clinical indicators are all normal, such as FEV1%Pred≥80%, CAT<10, and mMRC≤1, then the individual is considered to have a balanced temperament.

[0061] As a preferred embodiment of the present invention, step S7 is further included: after obtaining the body type, the confidence level is assessed according to the following assessment rules: e. If the scale score is ≥70, the clinical indicator score is ≥60, and the tongue appearance score is ≥60, then the confidence level is high; in this case, all three data sources point to the same constitution, and the result is reliable. f. If the scale score is ≥60 and the clinical indicator score is ≥40, then it is of medium confidence level; in this case, the scale score and the clinical indicator score are basically consistent, and the tongue appearance score can be used as an auxiliary reference.

[0062] g. If the scale score is ≥60 points, but the clinical indicator score is <40 points, it is considered low confidence. In this case, the subjective feeling and the objective indicator are inconsistent, and manual review is required.

[0063] As a preferred embodiment of the present invention, it further includes step S8: after obtaining the body type, based on the mapping relationship database between body type and COPD clinical phenotype, check whether the clinical indicator data conforms to the mapping relationship. If not, then steps S1 to S8 are reviewed.

[0064] This invention constructs a mapping database between body constitution type and COPD clinical phenotype based on evidence-based literature, as exemplarily shown in Table 12, but not as a specific limitation: Table 12

[0065] The clinical indicator data is used to verify the accuracy of the determined body constitution type. If the clinical indicator data contradicts the body constitution type, the determination process needs to be reviewed. This step is designed to identify misjudgments caused by scale completion errors or changes in the patient's condition. Examples, but not limited to, are listed below: If the individual is diagnosed with Qi deficiency, but FEV1%Pred≥80% and CAT<10, the reverse verification result will be "Scale score does not match clinical indicators, review is recommended"; If the individual is diagnosed with Yang deficiency constitution, but SpO2 ≥ 95% and mMRC ≤ 1, the reverse verification result is "Clinical indicators do not support the diagnosis of Yang deficiency constitution, and there may be a concurrent constitution".

[0066] As a preferred embodiment of the present invention, step S9 is further included: after obtaining the body type, assessing the risk level according to the risk assessment rules. h. If the constitution type is determined to be Qi deficiency, the risk level is extremely high; this constitution type accounts for the highest proportion of COPD patients, and lung Qi deficiency is the fundamental pathogenesis, leading to repeated acute exacerbations due to weak defensive Qi. i. If the constitution type is determined to be Yang deficiency, the risk level is high risk; this constitution type is due to prolonged Qi deficiency affecting Yang, resulting in the decline of Yang Qi and the overflow of Yang Qi, which is associated with the risk of death within three years. j. If the constitution type is determined to be phlegm-dampness constitution or yin deficiency constitution, the risk level is medium to high risk; phlegm-dampness constitution has a higher risk of acute exacerbation due to phlegm-dampness obstructing the lungs, which is significantly related to the amount of sputum; yin deficiency constitution is more common in patients who use glucocorticoids for a long time, and the yin deficiency and internal heat increase the difficulty of infection control. k. If the constitution type is determined to be blood stasis constitution, the risk level is medium risk; this constitution type is more common in moderate to severe patients and has a high risk of hypoxemia. 1. If the constitution type is determined to be balanced, the risk level is low; the disease is relatively mild for this constitution type, and prevention and maintenance are the main focus.

[0067] According to a COPD patient constitution identification system of the present invention, in one embodiment, the concept is the same as that of the COPD patient constitution identification method described above, and the system architecture is implemented in the form of a B / S architecture or a WeChat mini-program, such as... Figure 2 As shown, it includes: The storage module is used to store the COPD pathogenesis-related constitution screening scale, the COPD pathogenesis-related constitution screening backup scale, the confirmation scale, the standard tongue image template database, and the mapping relationship database between constitution type and COPD clinical phenotype. The storage module is a data storage carrier, which is implemented by combining non-volatile storage media with a database solid-state architecture. It can be deployed using any one or more storage media such as solid-state drives, flash memory, and cloud databases. It has the basic hardware and software functions for persistent data storage, fast retrieval, encrypted retention, and data updates. The data acquisition module is used to collect patients' scale data, clinical indicator data, tongue image data, backup scale data, and confirmation scale data. The data acquisition module is the data input terminal of the system and establishes a data transmission link with the subsequent data processing module. The hardware implementation consists of a human-computer interaction input terminal, a clinical data interface, an image acquisition device, and a data transmission bus. The software layer is configured with data protocol parsing, data verification, and data caching programs. The data processing module, connected to the data acquisition module, is used to perform standardization processing of scale data and clinical indicator data, as well as preprocessing and standardization processing of tongue image data. Its hardware can achieve high-speed data operation through embedded processors, image processing chips, and floating-point arithmetic units, while the software can be configured with data standardization algorithm programs, image preprocessing algorithm programs, and feature parameter extraction programs. This module is the core data preprocessing and standardization operation unit of the system, used to standardize various types of raw data collected, eliminate differences in the dimension, format, and scale of the raw data, and output standardized score data that can be used for model calculation. The physical fitness model calculation module, connected to the data processing module, is used to calculate the comprehensive score according to the constructed comprehensive score model. The hardware can realize batch data calculation through the core main control chip and parallel computing processing unit, and the software can have a built-in fixed multi-weight comprehensive score calculation model. This module is the core calculation unit of the system for quantitatively assessing the patient's physical fitness attributes. It is used to complete the comprehensive quantitative scoring of each physical fitness type of the patient based on the standardized multi-dimensional score data and through the preset comprehensive score model. The constitution type determination module, connected to the constitution model calculation module, is used to determine the constitution type according to the determination rules. The hardware implements the logical discrimination calculation through the main control processor, and the software has built-in multi-level threshold determination program, mixed constitution discrimination program, low-incidence constitution supplementary determination and balanced constitution verification program. This module is the core result determination unit of the system. Based on the comprehensive score of each constitution output by the model, combined with multiple preset threshold determination rules, it completes the accurate classification determination of the patient's constitution type, covering four basic determination scenarios: single constitution, mixed constitution, constitution tendency, and non-high-incidence constitution. At the same time, it links the backup determination and confirmation process to realize the full-scenario constitution identification.

[0068] The confidence level assessment module, connected to the constitution type determination module, is used to assess the confidence level according to the assessment rules. The hardware is based on the main control processor to perform threshold comparison calculations, and the software is configured with a three-level confidence level grading judgment logic program. After the patient's constitution type is determined, this module is used to grade and assess the credibility of the constitution identification results based on multi-dimensional standardized score data, providing a reliable basis for clinical reference. The reverse verification module, connected to the constitution type determination module, is used to check whether the clinical indicator data conforms to the mapping relationship. The hardware completes data matching and comparison through the main control processor, and the software has a built-in clinical phenotype mapping comparison program and a full-process review trigger program. This module is used to complete the secondary reverse verification after the initial determination of constitution type and confidence assessment, to ensure the consistency between the identification results and the clinical pathogenesis of COPD and improve the system's identification accuracy. The display module is used to output and display the constitution type and corresponding pathogenesis analysis text, confidence level, scores of each item, data feature description, risk warning, and intervention suggestions. The display module is the system's result visualization output terminal. The hardware can consist of an LCD display terminal, a touch screen, or a host computer display interface, and the software can be configured with result rendering, text generation, and data visualization layout programs. This module is used to summarize, organize, and visualize the full-dimensional results data of this COPD patient constitution identification, providing medical staff and patients with intuitive and comprehensive identification results reference.

[0069] In one specific implementation, the system automatically generates and outputs a pathogenesis analysis text based on the determined constitution type: Qi deficiency constitution: "Deficiency of lung qi, weak defensive qi, and insufficient ancestral qi make one prone to repeated external pathogenic factors that induce acute exacerbations. This is the fundamental pathogenesis of COPD. If not intervened in time, it can eventually affect the spleen and kidneys, and may progress to yang deficiency constitution." Yang deficiency constitution: "Prolonged qi deficiency affects yang, leading to a decline in the yang qi of the lungs, spleen, and kidneys, resulting in a failure of warming and nourishing functions and an inability to transform dampness. The disease has spread from the lungs to the spleen and kidneys, indicating a more severe condition, requiring stronger interventions to warm yang and consolidate the foundation." Phlegm-dampness constitution: "When the spleen fails to function properly, phlegm and dampness are generated internally, which stagnate in the lungs and obstruct the airways. Phlegm-dampness is a manifestation of COPD, and it is necessary to strengthen the spleen, remove dampness, resolve phlegm, and relieve asthma." Blood stasis constitution: "Prolonged illness affects the collaterals, causing qi stagnation and blood stasis, resulting in poor blood circulation and obstructed blood flow. Blood stasis obstructs the lungs, affecting gas exchange, requiring the promotion of blood circulation, removal of blood stasis, and relieving of asthma." Yin deficiency constitution: "Lung and kidney yin deficiency, internal heat disturbance, insufficient body fluids, dry mouth and throat. Yin deficiency with internal heat easily leads to difficulty in controlling infection, requiring nourishing yin and moistening the lungs, nourishing yin and generating body fluids." In one specific implementation, the data feature description includes characteristic tongue appearance descriptions such as tongue color, tongue coating, tongue shape, pulse, etc., corresponding TCM syndrome types such as "lung and spleen qi deficiency syndrome" and "lung and kidney yang deficiency syndrome", etc., and the clinical indicator matching degree, that is, the matching status of each indicator with the constitution; the risk warning includes the risk level and the existing constitution transformation risk, such as "qi deficiency constitution, there is a risk of transformation to yang deficiency constitution, it is recommended to strengthen qi and consolidate the exterior intervention".

[0070] Example 1 The patient is a 65-year-old male with stable COPD and a Global Initiative for Chronic Obstructive Lung Disease (GOLD) class II.

[0071] Enter data: Clinical indicators: FEV1%Pred 55%, CAT score 12, mMRC grade 2, SpO2 92%, 2 adverse events (AEs), BMI 21.5 Scale completion: Q1=4 (always fatigued and short of breath), Q2=3 (frequent colds), Q3=1 (rarely chills), Q4=0 (no nocturia), Q5=2 (sometimes excessive phlegm), Q6=1 (rarely sticky mouth), Q7=1 (rarely chest tightness), Q8=1 (rarely cyanosis). Tongue image: Pale red tongue with a thin white coating, enlarged tongue body with teeth marks. Calculation process: Scale score: Qi deficiency constitution: Q1(4) + Q2(3) = 7 points → 100 points = 7 / 8 × 100 = 87.5 points Yang deficiency quality: Q3(1)+Q4(0)=1 point→100 point system=1 / 8×100=12.5 points Phlegm-dampness constitution: Q5(2) + Q6(1) = 3 points → percentage system = 3 / 8 × 100 = 37.5 points Blood stasis constitution: Q7(1) + Q8(1) = 2 points → percentage system = 2 / 8 × 100 = 25.0 points Clinical indicator scores (lookup table, equal weighting of clinical indicators): Qi deficiency constitution: FEV1%Pred(55%→50)+CAT(12→30)+AE(2→65) / 3=48.33 points Yang deficiency constitution: FEV1%Pred (55%→50) + mMRC (2→40) + SpO2 (92%→30) / 3 = 40 points Phlegm-dampness constitution: Phlegm volume (moderate → 50) + BMI (21.5 → 0) / 2 = 25 points Blood stasis constitution: Lips (no cyanosis → 0) + SpO2 (92% → 30) / 2 = 15 points Tongue image score (equal weights for tongue image features): Qi deficiency constitution: Pale red tongue (100) + thin white tongue (100) + large teeth marks (100) → 100 points Yang deficiency constitution: Pale red (60) + Thin white (60) + Large teeth marks (100) → 73.33 points Phlegm-dampness constitution: Pale red (80) + Thin white (30) + Large (100) → 70 points Blood stasis constitution: pale red (20) + thin white (60) + large teeth marks (60) → 46.67 points Overall Score (Three-Layer Integration): Qi deficiency constitution: 87.5×0.6+48.33×0.3+100×0.1=52.5+14.5+10=77 points Yang deficiency quality: 12.5×0.6+40×0.3+73.33×0.1=7.5+12+7.3=26.8 points Phlegm-dampness constitution: 37.5×0.6+25×0.3+70×0.1=22.5+7.5+7=37 points Blood stasis constitution: 25.0×0.6+15×0.3+47×0.1=15+4.5+4.7=24.2 points Judgment result: Main constitution type: Qi deficiency constitution (overall score 77 points, highest and ≥60 points) Risk level: Extremely high risk Confidence level: High confidence (scale 87.5 ≥ 70, clinical 48.33 ≥ 40, tongue appearance 100 ≥ 60) Pathogenesis analysis: "Deficiency of lung qi, weak defensive qi, and insufficient ancestral qi make it easy for repeated external pathogens to induce acute exacerbations. This is the fundamental pathogenesis of COPD. If not intervened in time, it will eventually affect the spleen and kidneys, and may progress to a yang deficiency constitution." Reverse validation of clinical indicators: FEV1%Pred55% (decreased), CAT score 12 (≥10), AEs 2 times / year (≥2) – consistent with the clinical phenotype of Qi deficiency constitution, validation passed. Risk warning for transformation: "There is a risk of transformation from Qi deficiency to Yang deficiency. It is recommended to strengthen Qi-tonifying and surface-strengthening interventions and monitor the degree of cold intolerance and nocturia." Example 2 The patient is a 72-year-old female, one week after being discharged from the hospital following an acute exacerbation of chronic obstructive pulmonary disease (AECOPD), classified as GOLD III.

[0072] Enter data: Clinical indicators: FEV1%Pred 38%, CAT score 22, mMRC grade 3, SpO2 88%, 3 adverse events (AEs), BMI 20.8 Scale completion: Q1=3 (frequent fatigue), Q2=2 (sometimes colds), Q3=4 (always feeling cold), Q4=4 (always urinating at night), Q5=1 (rarely producing phlegm), Q6=0 (no sticky mouth), Q7=1 (rarely experiencing chest tightness), Q8=1 (rarely experiencing cyanosis). Tongue image: Pale and swollen tongue with a white and slippery coating, large tongue body with teeth marks. Calculation process: Scale score: Qi deficiency constitution: Q1(3) + Q2(2) = 5 points → 62.5 points Yang deficiency quality: Q3(4)+Q4(4)=8 points→100 points Phlegm-dampness constitution: Q5(1) + Q6(0) = 1 point → 12.5 points Blood stasis constitution: Q7(1) + Q8(1) = 2 points → 25.0 points Clinical indicator scores (lookup table, equal weighting of clinical indicators): Qi deficiency constitution: FEV1%Pred (38%→75)+CAT (22→80)+AE (3→100) / 3→85 points Yang deficiency constitution: FEV1%Pred (38%→75) + mMRC (3→70) + SpO2 (88%→65) / 3→70 points Phlegm-dampness constitution: Phlegm volume (small amount → 0) + BMI (20.8 → 0) / 2 → 0 points Blood stasis constitution: Lips (absent → 0) + SpO2 (88% → 65) / 2 → 32.5 points Tongue image score (equal weights for tongue image features): Qi deficiency constitution: pale red (60) + thin white (60) + large teeth marks (100) → 73.33 points Yang deficiency constitution: Pale and plump (100) + smooth and white skin (100) + large teeth marks (100) → 100 points Phlegm-dampness constitution: Pale white (80) + Slippery white (100) + Large and swollen (100) → 93.33 points Blood stasis constitution: Pale red (20) + White and smooth (50) + Normal (60) → 43.33 points Overall Score: Qi deficiency constitution: 62.5×0.6+85×0.3+73.33×0.1=37.5+25.5+7.3=70.3 points Yang deficiency quality: 100×0.6+70×0.3+100×0.1=60+21+10=91 points Phlegm-dampness constitution: 12.5×0.6+0×0.3+93×0.1=7.5+0+9.3=16.8 points Blood stasis constitution: 25.0×0.6+32.5×0.3+43.33×0.1=15+9.75+4.3=29.05 points Judgment result: Body constitution type: Yang deficiency constitution (overall score 91 points) Concurrent constitution: The overall score for Qi deficiency was 70.3 points, which is 20.7 points lower than the highest score (>10 points), therefore it is not considered a concurrent constitution. However, the Qi deficiency score was ≥60 points, indicating "prolonged Qi deficiency affecting Yang" in the pathogenesis analysis. Risk level: High risk Confidence level: High confidence level Pathogenesis analysis: "Prolonged Qi deficiency affects Yang, leading to a decline in the Yang Qi of the lungs, spleen, and kidneys, resulting in impaired warming function and failure to transform dampness. The disease location has expanded from the lungs to the spleen and kidneys, indicating a more severe condition. Yang deficiency and water retention are associated with a 3-year mortality risk, necessitating interventions to strengthen Yang and consolidate the body's foundation." Reverse validation of clinical indicators: mMRC grade 3 (≥2), SpO2 88% (low) - consistent with the clinical phenotype of Yang deficiency constitution, validation passed.

[0073] The above description is merely a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for identifying the physical constitution of COPD patients, characterized in that, Includes the following steps: Step S1: Obtain scale data through the COPD pathogenesis-related constitution screening scale; the COPD pathogenesis-related constitution screening scale is based on the statistical analysis results of the constitution distribution of COPD patients and selects the high-incidence constitution types, including items from the standard constitution scale of traditional Chinese medicine that are differentially associated with the pathogenesis of COPD in each high-incidence constitution type; obtain clinical indicator data and tongue image data associated with COPD; Step S2: Standardize the scale data and clinical indicator data to obtain scale scores and clinical indicator scores respectively, and preprocess and standardize the tongue image data to obtain tongue image scores. Step S3: Calculate the overall score by combining the scale score, clinical indicator score, and tongue appearance score using the following formula: Overall score = Scale score × W1 + Clinical indicator score × W2 + Tongue appearance score × W3 Wherein, W1 is the weight of the scale score; W2 is the weight of the clinical indicator score; and W3 is the weight of the tongue appearance score. Step S4: Determine the body type based on the overall score according to the following rules: a. If a certain physical constitution score is greater than or equal to the first threshold and is the highest among all the physical constitutions of the patient, then it is determined to be that physical constitution type; b. If a certain physical condition's overall score is greater than or equal to the first threshold, and the difference between it and the highest-scoring physical condition is less than the second threshold, then it is determined to be a mixed physical condition. c. If all comprehensive physical fitness scores are less than the third threshold, then it is determined that the individual does not belong to any of the high-incidence physical fitness types. d. If the third threshold ≤ the highest score constitution < the first threshold, then it is determined to be a constitution tendency; The first threshold is a high score threshold set based on the total score of the comprehensive score; the second threshold is a difference range set based on the total score of the comprehensive score; and the third threshold is a low score threshold set based on the total score of the comprehensive score.

2. The method for identifying the physical constitution of COPD patients according to claim 1, characterized in that, In step S1, the common constitution types include Qi deficiency, Yang deficiency, phlegm-dampness, and blood stasis. There are two questions for each common constitution type, and each question is assigned a graded score.

3. The method for identifying the physical constitution of COPD patients according to claim 1, characterized in that, In step S2, the standardization processing method for the scale data includes: Step S21: Add the scores of the two questions for each body type to obtain the first raw score for that body type; Step S22: Convert the first raw score into a scale score using the following formula: 。 4. The method for identifying the physical constitution of COPD patients according to claim 3, characterized in that, In step S2, the standardization method for clinical indicator data is to convert it into a percentage score based on the correlation between clinical indicators and physical condition.

5. The method for identifying the physical constitution of COPD patients according to claim 4, characterized in that, In step S2, the preprocessing includes standardizing the image resolution, lighting conditions and shooting angle, automatically segmenting the tongue region, and extracting feature parameters such as tongue color, tongue coating thickness, tongue coating color, tongue body size, and presence or absence of teeth marks. The standardization processing of the tongue image data involves comparing the feature parameters with standard tongue image templates for each high-incidence constitution type to obtain a matching score, and then calculating the tongue image score through weighted average.

6. The method for identifying the physical constitution of COPD patients according to claim 5, characterized in that, The method also includes step S5, in which step S4 is performed when it is determined that the constitution does not belong to any of the high-incidence constitution types: obtaining backup scale data through the COPD pathogenesis-related constitution screening backup scale; the COPD pathogenesis-related constitution screening backup scale is a low-incidence constitution type selected based on the statistical analysis results of the constitution distribution of COPD patients, including selecting items from the TCM constitution standard scale that are differentially related to the COPD pathogenesis in each low-incidence constitution type; and determining whether the constitution type is Yin deficiency constitution or balanced constitution based on the backup scale data.

7. The method for identifying the physical constitution of COPD patients according to claim 6, characterized in that, It also includes step S6, which is executed when the constitution is determined to be balanced in step S5: obtaining verification scale data through a verification scale; the verification scale is selected from the standard TCM constitution scale, and items that are correlated with the pathogenesis of COPD are selected from the balanced constitution; each item of the verification scale is set with a graded score. If all items are scored out of full marks and the clinical indicators are all normal, the constitution is confirmed as balanced; otherwise, the constitution with the second highest score is taken as the main constitution.

8. The method for identifying the physical constitution of COPD patients according to claim 7, characterized in that, It also includes step S7: After obtaining the constitution type, assess the confidence level according to the following assessment rules: e. If the scale score is ≥70 points, the clinical indicator score is ≥60 points, and the tongue appearance score is ≥60 points, then it is considered to have high confidence. f. If the scale score is ≥60 points and the clinical indicator score is ≥40 points, then it is of medium confidence level; g. If the scale score is ≥60 points, but the clinical indicator score is <40 points, then the confidence level is low.

9. A method for identifying the physical constitution of COPD patients according to claim 8, characterized in that, It also includes step S8: After obtaining the body type, based on the mapping relationship database between body type and COPD clinical phenotype, check whether the clinical indicator data conforms to the mapping relationship. If not, then steps S1 to S8 are reviewed.

10. A COPD patient constitution identification system, characterized in that, The method for identifying the physical condition of COPD patients according to any one of claims 1 to 9 includes: The storage module is used to store the COPD pathogenesis-related constitution screening scale, the COPD pathogenesis-related constitution screening backup scale, the confirmation scale, the standard tongue image template database, and the mapping relationship database between constitution type and COPD clinical phenotype; The data acquisition module is used to collect patients' scale data, clinical indicator data, tongue image data, backup scale data, and confirmation scale data; The data processing module, connected to the data acquisition module, is used to perform standardized processing of scale data and clinical indicator data, as well as preprocessing and standardized processing of tongue image data. The physical fitness model calculation module, connected to the data processing module, is used to calculate the comprehensive score according to the constructed comprehensive score model. The constitution type determination module is connected to the constitution model calculation module and is used to determine the constitution type according to the determination rules. The confidence level assessment module, connected to the body type determination module, is used to assess the confidence level according to the assessment rules. The reverse verification module, connected to the body type determination module, is used to check whether the clinical indicator data conforms to the mapping relationship; The display module is used to output and display the constitution type and corresponding pathogenesis analysis text, confidence level, scores of each item, data feature description, risk warning, and intervention suggestions.

Images