Tongue picture characteristic change rule analysis method based on rehabilitation process of chronic disease patient

Abstract

The invention discloses a tongue picture characteristic change rule analysis method based on the rehabilitation process of a chronic disease patient, and relates to the technical field of health big data analysis, and the method comprises the steps of tongue picture data standardized collection, tongue picture data recording system establishment, tongue picture characteristic data analysis, result visualization and report generation. According to the invention, through a standardized tongue picture data acquisition mechanism, the quality and consistency of the tongue picture data are ensured, and through controlling illumination conditions, background colors, diet restrictions and posture requirements, the influence of external factors on the tongue picture data is reduced, and the accuracy and reliability of the data are improved. By comprehensively evaluating definition, noisy points, contrast ratio and color accuracy verification parameters, it is ensured that collected tongue picture data reaches a preset quality standard, and a high-quality data basis is provided for subsequent analysis; meanwhile, a comprehensive tongue picture data recording system is established, and tongue picture data are effectively stored and managed.

Description

Technical Field

[0001] The present invention relates to the technical field of health big data analysis, and specifically to a method for analyzing the changing rules of tongue image characteristics during the rehabilitation process of chronic disease patients. Background Art

[0002] With the deepening of rehabilitation treatment for patients with chronic diseases, the demand for monitoring and analysis of physiological changes during the rehabilitation process is increasing. As one of the important means of diagnosis in traditional Chinese medicine, the color, shape, degree of moistness and other characteristics of tongue are closely related to the patient's health status. However, traditional tongue analysis methods mainly rely on the doctor's subjective experience and immediate observation, lack of systematic quantitative analysis and long-term tracking, and it is difficult to fully reflect the subtle changes in the patient's rehabilitation process. In order to improve the accuracy and objectivity of tongue analysis, it is necessary to develop an analysis method based on the law of changes in tongue characteristics during the rehabilitation process of chronic disease patients;

[0003] The rehabilitation process of patients with chronic diseases is a long and complex process, which involves a variety of physiological and pathological changes. Tongue image is an important window to reflect these changes, and the change pattern of its characteristics has important reference value for evaluating the patient's recovery status. Traditional tongue image analysis methods mainly rely on the doctor's subjective judgment and lack quantitative standards and systematic analysis, which to a certain extent limits the in-depth mining and application of tongue image information. Therefore, it is necessary to develop a tongue image characteristic change law analysis technology based on objective data and scientific analysis methods to improve the accuracy and reliability of tongue image analysis.

[0004] In order to solve the above defects, a technical solution is now provided. Summary of the invention

[0005] The purpose of the present invention is to solve the problems of strong subjectivity, lack of quantification and long-term tracking in existing tongue image analysis methods, and to propose an analysis method based on the changing rules of tongue image characteristics during the rehabilitation process of chronic disease patients.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] The method for analyzing the changing rules of tongue features during the rehabilitation process of chronic disease patients includes the following steps:

[0008] S1. Standardized collection of tongue image data: Use professional equipment to collect tongue image data of chronic disease patients through standardized mechanisms of lighting, background, diet, and posture, and verify the collected tongue image data;

[0009] S2. Establish a tongue image data recording system: enter the collected data into a system with storage, query and update functions, and manage the data by patient number, time and disease type;

[0010] S3. Tongue feature data analysis: trend and correlation analysis of tongue data of the same patient at different times, drawing trend graphs and judging feature changes;

[0011] S4. Result visualization and report generation: Visualize the analysis results and generate a report containing patient information, changes in tongue characteristics and correlation analysis.

[0012] Furthermore, the process of the standardization mechanism in S1 is as follows:

[0013] Lighting conditions: Use lighting equipment with constant light intensity and color temperature to ensure uniform and stable lighting during each collection to avoid deviations in tongue color due to differences in lighting. Specifically, use LED or LEP standard light sources, set fixed brightness and color temperature values, and a natural light color temperature of 5300-7000K to simulate natural lighting conditions; control the lighting angle so that the light is irradiated from the front at an angle of 42-47° above to show the color, shape, and coating characteristics of the tongue;

[0014] Background color: Choose a single, non-interfering background color, including white or gray. The background material should be light-absorbing to avoid reflection affecting the tongue image collection effect. Use a matte background cloth or background board to ensure the authenticity and accuracy of the tongue image.

[0015] Dietary restrictions: Before the tongue image is collected, remind the patient to avoid consuming food and drinks that may affect the tongue image, including spicy food, coffee, and strong tea; stipulate the fasting time before collection, specifically fasting within 1-2 hours before collection, to reduce the impact of food residues on the tongue image. At the same time, require the patient to keep the oral cavity clean to avoid food residues adhering to the tongue surface, and rinse the mouth with mouthwash or purified water for more than three times;

[0016] Posture requirements: prompt the patient to take a uniform sitting or lying posture to ensure that the tongue is in a natural and relaxed state during the collection process. Specifically, the patient sits on a chair, tilts the head forward, sticks out the tongue, and makes the tongue surface parallel to the ground; require the patient to remain quiet during the collection process, avoid talking and swallowing, so as not to affect the stability of the tongue image. At the same time, remind the patient not to stick out the tongue forcefully to avoid deformation of the tongue.

[0017] Furthermore, the specific operation steps for verifying the collected tongue image data in S1 are as follows:

[0018] Comprehensive evaluation is performed through the verification parameters of tongue image data to determine whether it is qualified. The verification parameters include:

[0019] Clarity parameters: including edge sharpness and texture clarity. Edge sharpness is measured by calculating the gradient value of the tongue edge and tongue coating edge in the tongue image. Texture clarity reflects the clarity of the texture by calculating the contrast parameter of the gray level co-occurrence matrix and quantifying it by 100%. The obtained edge sharpness and texture clarity are normalized and weightedly summed to obtain the clarity evaluation value, which is used as the standard for measuring clarity parameters.

[0020] Noise parameters: including noise density and noise intensity variance. The noise density is obtained by counting the number of noise points in the image and dividing it by the total number of pixels in the image. The noise intensity variance is obtained by calculating the intensity variance of the noise points in the image. The noise density and noise intensity variance are normalized, summed and multiplied by the correction factor to obtain the noise judgment value, which is used as the standard for measuring noise parameters.

[0021] Contrast parameter: by calculating the contrast value of the tongue in color or grayscale, calculating the grayscale difference for quantification, and then evaluating the overall contrast, by calculating the grayscale histogram distribution range of the image and quantifying it from 1 to 100, recording it as the model evaluation value, normalizing the obtained contrast value and model evaluation value, respectively, using the contrast value and the model evaluation value as two sides, and the preset coefficient as the other side to establish a triangle, calculate the area of ​​the triangle, and record it as the pair parameter value, and use this pair parameter value as the standard for measuring the contrast parameter;

[0022] Color accuracy parameters: including color deviation and color saturation uniformity. The color deviation is calculated by comparing the color of the collected tongue image with the standard color sample. The color saturation uniformity is calculated by calculating the color saturation standard deviation of multiple areas in the image. The product of the obtained color deviation and the color saturation standard deviation is recorded as the color judgment value, which is used as the standard for measuring color accuracy parameters.

[0023] The obtained clarity evaluation value, noise judgment value, reference value and color judgment value are normalized and then input into the following formula: To get the comprehensive analysis value ZP, where P, Z, D, S are the clarity value, noise value, reference value and color value respectively. are the preset weight coefficients of the clear evaluation value, noise judgment value, reference value and color judgment value, and the obtained comprehensive analysis value ZP is used as the comprehensive evaluation standard of the tongue image data verification parameter;

[0024] The obtained comprehensive analysis value ZP is compared with the preset comprehensive analysis threshold. When the comprehensive analysis value ZP is greater than the preset comprehensive analysis threshold, the tongue image data is judged to be up to standard. Otherwise, the tongue image data is judged to be not up to standard, and the patient's tongue image data is recollected until the tongue image data meets the standard.

[0025] Furthermore, the tongue image data of chronic disease patients collected in S1 includes the color, shape, and moistness characteristics of the tongue. For each feature, a specific quantitative standard is preset. The specific process is as follows:

[0026] Tongue color quantification: Select the internationally accepted color standard system, including RGB color mode or HSV color mode, and determine the range of each tongue color in the color standard system based on the traditional description of tongue color in traditional Chinese medicine; continuously adjust and refine these ranges through observation and analysis of different samples; use color measuring instruments to measure the color of the collected tongue images and convert them into values ​​in the corresponding color standard system. At the same time, standardize the measurement position and range, and select the middle area of ​​the tongue for measurement;

[0027] Tongue shape quantification: Use image measurement software or tools to measure the length and width of the tongue on the tongue image. The longest distance from the tip to the root of the tongue is taken as the tongue length, and the distance from the widest part of the two sides of the tongue is taken as the tongue width. To improve the accuracy of the measurement, set a ruler or reference line on the tongue image acquisition device to make accurate measurements on the image. At the same time, the measurement method and unit should be unified.

[0028] For the measurement of tongue thickness, the side image of the tongue is obtained by side shooting. Then, the thickness of the tongue is measured on the side image, that is, the distance from the back of the tongue to the belly of the tongue. Similarly, image measurement software or tools are needed to measure and ensure the accuracy and consistency of the measurement. The measurement error can be reduced by taking the average value of multiple measurements.

[0029] According to the shape characteristics of the tongue, it is divided into different types, including normal type, fat type, and thin type. Each shape type is defined and described, and corresponding quantitative standards are formulated. Specifically, a fat tongue is defined as a tongue width greater than the preset normal range, and there are tooth marks on the edge of the tongue; a thin tongue is defined as a tongue width less than the preset normal range, and the tongue surface is dry. Through the classification and quantification of the tongue shape, the morphological changes of the tongue can be more intuitively understood;

[0030] Quantification of moistness: Analyze the surface texture and reflection characteristics of the tongue coating through image processing technology, and analyze the changes in optical properties; Specifically: Obtain the surface roughness index of the tongue coating through image processing technology and reflectivity , the degree of moisturizing is quantified by the formula , ,in The coefficients are determined according to the actual situation. By adjusting these coefficients, the formula output Values ​​range from 0 to 1, indicating changes in moisture from dry to wet.

[0031] Furthermore, the specific operation steps for data storage in S2 are as follows:

[0032] The tongue image data is stored safely and effectively, and the stored data is classified and managed, including classification according to patient number, collection time, and disease type. The process is as follows:

[0033] First, the catalog categories are divided according to the disease type, including cardiovascular disease and respiratory disease. Under each disease type catalog, the year is further subdivided into sub-catalogs, and then folders are created according to the month. This hierarchical structure is used to initially sort out the data for easy management and search;

[0034] For each patient's data, the patient number is used as the name of an independent folder, and all tongue image data files and their attached information files related to the patient are stored in the patient number folder;

[0035] Create the following data tables: patient information table, tongue image collection schedule, and tongue image characteristic data table; the patient information table records the patient number, name, age, gender, contact information, and basic information of disease type diagnosis results; the tongue image collection schedule records the specific time and collection sequence number of each tongue image collection, and determines the order of collection by the collection sequence number; the tongue image characteristic data table records the specific characteristic values ​​of each tongue image collected, including the RGB value of the color, the measurement value of the thickness of the tongue coating, and the quantitative value of the degree of moistness, and is linked to the patient information table and the tongue image collection schedule through foreign keys to ensure data integrity and traceability;

[0036] For tongue images, the lossless compression format PNG is used to ensure that the image quality is not damaged. When storing the image, the image acquisition device parameters and illumination condition parameter metadata information are recorded and stored in the corresponding patient folder;

[0037] For the patient's basic information and quantitative description text data of tongue characteristics, standardized text formats are used, including comma-separated value format or XML format, and the stored data is backed up regularly, using off-site backup and multi-version backup strategies to prevent data loss due to hardware failure, human error and other accidents.

[0038] Furthermore, the specific operation steps of S3 are as follows:

[0039] Screening out the tongue image data of patients who need to be analyzed for trend from the tongue image data recording system, including tongue image data collected from patients at different time points and corresponding quantitative feature data;

[0040] Sort the tongue image data in the order of collection time, establish a time axis, and associate the tongue image characteristic data at each time point with the corresponding time point to ensure the consistency of the time interval so as to better observe the trend changes;

[0041] The previously determined tongue coating thickness quantification method is used to measure the tongue coating thickness of the tongue image at each time point. The edge of the tongue coating is determined using image processing software or an algorithm, and then the thickness of the tongue coating is measured. The measurement results are recorded as the quantified value of the tongue coating thickness. For other tongue features that need to be analyzed, they are also measured and quantified according to the corresponding quantification standards to ensure that the quantification method of each feature remains consistent at different time points, so as to conduct accurate trend analysis.

[0042] The data was smoothed by using the moving average method, taking the average value of the data in the preset time window as the smoothed value, and using linear regression analysis and polynomial regression analysis methods to fit the trend curve of tongue characteristics changing over time;

[0043] By analyzing the slope and intercept of the trend curve, determine whether the changing trend of the tongue image characteristics is gradually increasing, gradually decreasing, or remaining stable; draw a trend graph of the tongue image characteristics changing over time; with time as the horizontal axis and the tongue image characteristic quantification value as the vertical axis, mark the tongue image characteristic data at each time point on the graph and connect them into a curve, and use chart drawing software for visual display;

[0044] Based on the shape and change trend of the trend chart, explain how the tongue characteristics change over time. If the thickness of the tongue coating gradually becomes thinner with the recovery time, it indicates that the patient's condition is improving; if the thickness of the tongue coating remains stable or gradually increases, further analyze the reasons, including poor treatment effect and bad living habits;

[0045] The changing trend of tongue characteristics is correlated with the patient's recovery process, and combined with the patient's clinical symptoms and biochemical index information, a comprehensive judgment is made as to whether the changes in tongue characteristics are consistent with the recovery situation;

[0046] If abnormal data points or sudden trend changes are found during the trend analysis process, further investigation is needed into the causes, which may be errors in the collection process or sudden changes in the patient's condition. By reviewing the tongue data, asking the patient about their condition, and checking the treatment records, the cause of the abnormality can be determined and appropriate measures can be taken to deal with it.

[0047] Furthermore, the specific process of determining whether the change trend of the tongue image characteristic is gradually increasing, gradually decreasing, or remaining stable in S3 is as follows:

[0048] First, determine the type of trend curve. When the trend curve belongs to a linear regression curve, that is, y=ax+b, the slope a represents the change in the dependent variable, i.e., the quantitative value of the tongue image characteristics, when the independent variable, i.e., time, increases by one unit. The intercept b is the value of the dependent variable when the independent variable is 0.

[0049] When the trend curve belongs to the polynomial regression curve, that is, y=ax²+bx+c, the first-order derivative is 2ax+b. In the process of the curve changing, when the first-order derivative is greater than 0, the curve rises; when the first-order derivative is less than 0, the curve falls; when the first-order derivative is 0, it is the extreme point of the curve, and the intercept c is also the value of the dependent variable when the independent variable is 0;

[0050] If the slope a of the linear regression curve is a positive number, it means that the quantified value of the tongue image characteristic is gradually increasing over time. For the polynomial regression curve, if the first-order derivative is a positive number above the preset time within the observed time range, it indicates that the tongue image characteristic is gradually increasing.

[0051] If the slope a of the linear regression curve is negative, the quantified value of the tongue image characteristic gradually decreases over time. For the polynomial regression curve, if its first-order derivative is negative over a preset time, the tongue image characteristic shows a gradually decreasing trend.

[0052] When the difference between the slope a of the linear regression curve and 0 is less than the preset threshold, it indicates that the tongue image characteristics remain stable during this period. For the polynomial regression curve, if the difference between its first-order derivative and 0 within the observed time range is less than the preset threshold, the tongue image characteristics are judged to be stable.

[0053] Furthermore, the specific process of visually displaying the analysis results in S4 is as follows:

[0054] For the changing trend of tongue image characteristics, when a single characteristic changes over time, a line graph is used, with time as the horizontal axis and the tongue image characteristic quantification value as the vertical axis to draw a line to show the continuous change process of the data;

[0055] When comparing different tongue features or showing the differences in tongue features of different patient groups at the same time point, choose a bar chart to display, compare the average RGB values ​​of the tongue coating color of patients with different disease types at the same recovery stage, and intuitively present the differences through the height of different colored columns;

[0056] When showing the relationship between tongue features and multiple clinical symptoms or biochemical indicators, a heat map is used, with tongue features, clinical symptoms, and biochemical indicators as variables, and different colors are used to indicate the strength of the correlation between the variables. The warmer the color, the stronger the positive correlation, and the colder the color, the stronger the negative correlation or no correlation.

[0057] When drawing a chart, add a specific title to summarize the content of the chart; label the coordinate axis name, indicate the time unit or feature category on the horizontal axis, indicate the quantitative unit on the vertical axis, and add a legend to the chart to explain the meaning of different lines, colors or columns;

[0058] Adjust the color matching of the chart to ensure that the data is clear and readable, avoid colors that are too glaring or similar and difficult to distinguish, and choose a color scheme that is consistent with medical and health themes, including blue to represent the normal range and red to represent abnormal values.

[0059] Furthermore, the steps for generating the report in S4 are as follows:

[0060] Extract the patient's basic information from the tongue image data recording system, including name, age, gender, disease diagnosis, and treatment history, and summarize the results of tongue image characteristic analysis, including quantitative data of tongue image characteristics at different time points, conclusions of trend analysis, and correlation coefficients and significance levels with clinical symptoms and biochemical indicators in correlation analysis;

[0061] The report begins with the patient's basic information section, which presents the patient's background information in the form of tables or paragraphs, followed by a section on the trend analysis of tongue characteristics, which describes the changes in each characteristic and explains them with visual charts to explain the physiological or pathological significance behind the trend changes;

[0062] In the correlation analysis results section, a table or chart showing the correlation between tongue features and clinical symptoms and biochemical indicators is listed. The relevant indicators are discussed and their potential value for patient condition assessment and rehabilitation program formulation is analyzed. A summary and suggestions are given at the end of the report. The comprehensive tongue analysis results provide doctors with an overall evaluation of the patient's rehabilitation status and recommendations for further treatment or monitoring.

[0063] After the first draft of the report is generated, it will be manually reviewed to check the accuracy of the data, the consistency of the charts and text descriptions, and the logical coherence. It will be reviewed by relevant medical experts or data analysts, and feedback will be collected and modified to ensure that the report content is accurate, clear, and valuable, providing a scientific basis for doctors to formulate personalized rehabilitation plans.

[0064] Compared with the prior art, the present invention has the following beneficial effects:

[0065] (1) The present invention ensures the quality and consistency of tongue image data through a standardized tongue image data collection mechanism. It reduces the impact of external factors on tongue image data by controlling lighting conditions, background color, dietary restrictions and posture requirements, thereby improving the accuracy and reliability of data. It ensures that the collected tongue image data meets the preset quality standards by comprehensively evaluating verification parameters such as clarity, noise, contrast and color accuracy, thus providing a high-quality data foundation for subsequent analysis.

[0066] (2) The present invention establishes a comprehensive tongue image data recording system to effectively store and manage tongue image data. The system can classify and manage data according to patient number, collection time and disease type, which is convenient for doctors and researchers to query and analyze. By creating a patient information table, tongue image collection schedule and tongue image characteristic data table, the integrity and traceability of the data are achieved. The image and text data are stored in a lossless compression format and a standardized text format, which ensures the quality of the data and the feasibility of long-term preservation.

[0067] (3) The present invention, through trend analysis and correlation analysis, can reveal the changing patterns of tongue characteristics over time and correlate them with the patient's rehabilitation process, which not only improves the scientificity and objectivity of tongue analysis, but also provides an important reference basis for doctors to formulate personalized rehabilitation plans. Through result visualization and report generation, complex data analysis results are easy to understand and apply, thereby improving the doctor's evaluation efficiency of the patient's rehabilitation status and the treatment effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] In order to facilitate understanding by those skilled in the art, the present invention is further described below in conjunction with the accompanying drawings;

[0069] Figure 1 The figure is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0070] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0071] It should be understood that the terms "include" and "comprising" used in the specification and claims of the present disclosure indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or collections thereof.

[0072] It should also be understood that the terms used in this disclosure are only for the purpose of describing specific embodiments and are not intended to limit the disclosure. As used in this disclosure and claims, the singular forms of "a", "an", and "the" are intended to include the plural forms unless the context clearly indicates otherwise. It should also be further understood that the term "and / or" used in this disclosure and claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes these combinations.

[0073] like Figure 1 As shown, the analysis method based on the changing rules of tongue features in the rehabilitation process of chronic disease patients includes the following steps:

[0074] Step 1: Standardized collection of tongue image data

[0075] Use professional tongue image collection equipment to collect tongue images of chronic disease patients regularly. During the collection process, a standardized mechanism is used to ensure that the collected tongue image data are comparable. The standardized mechanism includes:

[0076] Lighting conditions: Use lighting equipment with constant light intensity and color temperature to ensure uniform and stable lighting during each collection to avoid deviations in tongue color due to differences in lighting. Specifically, use LED or LEP standard light sources, set fixed brightness and color temperature values, and a natural light color temperature of 5300-7000K to simulate natural lighting conditions; control the lighting angle so that the light is irradiated from the front at an angle of 42-47° above to show the color, shape, and coating characteristics of the tongue;

[0077] Background color: Choose a single, non-interfering background color, including white or gray, to highlight the characteristics of the tongue. A unified background color can reduce the interference caused by background differences on tongue image analysis, making the tongue image clearer and more recognizable. The background material should be light-absorbing to avoid reflections that affect the collection of tongue images. Specifically, use a matte background cloth or background board to ensure the authenticity and accuracy of the tongue image;

[0078] Dietary restrictions: Before collecting tongue images, remind patients to avoid consuming foods and drinks that may affect tongue images, including spicy foods, coffee, and strong tea. These foods and drinks may cause changes in the color and thickness of the tongue coating, affecting the accuracy of the tongue image. The fasting time before collection is specified, specifically fasting within 1-2 hours before collection to reduce the impact of food residues on the tongue image. At the same time, patients are required to keep their mouths clean to avoid food residues adhering to the tongue surface, and rinse their mouths more than three times with mouthwash or purified water;

[0079] Posture requirements: prompt patients to take a uniform sitting or lying posture to ensure that the tongue is in a natural and relaxed state during the collection process. For example, patients can sit in a comfortable chair, tilt their heads slightly forward, and stick out their tongues so that the tongue surface is parallel to the ground. Require patients to remain quiet during the collection process and avoid talking or swallowing to avoid affecting the stability of the tongue image. At the same time, remind patients not to use excessive force to stick out their tongues to avoid deformation of the tongue.

[0080] After the tongue image data is collected, verification is performed, and a comprehensive evaluation is performed based on the verification parameters of the tongue image data to determine whether it is qualified. The verification parameters include:

[0081] Clarity parameters: including edge sharpness and texture clarity. Edge sharpness is measured by calculating the gradient value of the tongue edge and tongue coating edge in the tongue image. Texture clarity reflects the clarity of the texture by calculating the contrast of the grayscale co-occurrence matrix and other parameters, and quantified by 100%. The edge sharpness and texture clarity obtained are normalized and weighted summed to obtain the clarity evaluation value, and this clarity evaluation value is used as the standard for measuring clarity parameters. Noise parameters: including noise density and noise intensity variance. Noise density is obtained by counting the number of noise points in the image and dividing it by the total number of pixels in the image. The noise intensity variance is obtained by calculating the intensity variance of the noise points in the image. The noise density and noise intensity variance are normalized, summed and multiplied by the correction factor to obtain the noise judgment value, and this noise judgment value is used as the standard for measuring noise parameters. The correction factor is set between 0.92-1.21. Contrast parameters :By calculating the contrast value of the tongue in color or grayscale, calculating the grayscale difference for quantification, and then evaluating the overall contrast, by calculating the grayscale histogram distribution range of the image, and quantifying it through 1-100, recorded as the model evaluation value, the obtained contrast value and model evaluation value are normalized, and a triangle is established with the contrast value and the model evaluation value as two sides, and the preset coefficient as the other side, and the area of ​​the triangle is calculated and recorded as the pair of parameter values, and the pair of parameter values ​​is used as the standard for measuring the contrast parameters; Color accuracy parameters: including color deviation and color saturation uniformity. The color deviation is calculated by comparing the color of the collected tongue image with the standard color sample, and the color deviation is calculated. The color saturation uniformity is calculated by calculating the color saturation standard deviation of multiple areas in the image; the product of the obtained color deviation and the color saturation standard deviation is recorded as the color judgment value, and the color judgment value is used as the standard for measuring the color accuracy parameters;

[0082] The obtained clarity evaluation value, noise judgment value, reference value and color judgment value are normalized and then input into the following formula: To get the comprehensive analysis value ZP, where P, Z, D, S are the clarity value, noise value, reference value and color value respectively. The preset weight coefficients are the clarity evaluation value, noise judgment value, parameter value and color judgment value respectively, and the obtained comprehensive analysis value ZP is used as the comprehensive evaluation standard of the tongue image data verification parameters; the obtained comprehensive analysis value ZP is compared with the preset comprehensive analysis threshold value, when the comprehensive analysis value ZP is greater than the preset comprehensive analysis threshold value, the tongue image data is judged to be up to standard, otherwise, the tongue image data is judged to be not up to standard, and the patient's tongue image data is recollected until the tongue image data meets the standard.

[0083] The collected content includes the color, shape, and moistness of the tongue. For each feature, a specific quantitative standard is preset, specifically:

[0084] Tongue color quantification: Select the internationally accepted color standard system, including RGB (red, green, blue) color mode or HSV (hue, saturation, value) color mode, and determine the approximate range of each tongue color in the color standard system based on the traditional description of tongue color in traditional Chinese medicine, such as light red, red, crimson, and purple. It should be noted that the light red tongue corresponds to an RGB value range of about (240, 180, 180), and the red tongue corresponds to an RGB value range of about (255, 100, 100). Through a large number of sample observations and analyses, these ranges are continuously adjusted and refined. Use professional color measurement instruments, such as color analyzers or spectrophotometers, to measure the color of the collected tongue images. These instruments can accurately measure the color value of the tongue and convert it into the corresponding value in the color standard system. During measurement, the accuracy and stability of the instrument should be ensured to avoid inaccurate color measurement results due to instrument errors. At the same time, the measurement position and range should be standardized. Generally, a relatively uniform area in the middle of the tongue is selected for measurement.

[0085] Quantification of tongue shape: On the tongue image, use image measurement software or tools to measure the length and width of the tongue. The longest distance from the tip to the root of the tongue is taken as the tongue length, and the distance from the widest part of the two sides of the tongue is taken as the tongue width. To improve the accuracy of the measurement, a ruler or reference line can be set on the tongue image acquisition device to make accurate measurements on the image. At the same time, the measurement method and unit should be unified, such as measuring in millimeters;

[0086] For the measurement of tongue thickness, the side image of the tongue is obtained by side shooting. Then, the thickness of the tongue is measured on the side image, that is, the distance from the back of the tongue to the belly of the tongue. Similarly, image measurement software or tools are needed to measure and ensure the accuracy and consistency of the measurement. The measurement error can be reduced by taking the average value of multiple measurements. According to the shape characteristics of the tongue, it is divided into different types, including normal type, fat type, and thin type. Each shape type is defined and described, and corresponding quantitative standards are formulated. For example, a fat tongue can be defined as a tongue with a width significantly larger than the normal range and tooth marks on the edge of the tongue; a thin tongue can be defined as a tongue with a width significantly smaller than the normal range and a dry tongue surface. By classifying and quantifying the shape of the tongue, the morphological changes of the tongue can be more intuitively understood.

[0087] Quantification of the degree of moistness and dryness: The surface texture and reflective properties of the tongue coating are analyzed through image processing technology. The surface of a moist tongue coating is relatively smooth and the reflected light is relatively uniform; while the surface of a dry tongue coating may appear rough and the reflected light is unevenly distributed. By analyzing the changes in these optical properties, the surface roughness index of the tongue coating is obtained through image processing technology. (larger values ​​indicate rougher, i.e. drier) and reflectivity (The smaller the value, the drier it is), a comprehensive formula is used to quantify the degree of moisturizing , ,in The coefficients are determined according to the actual situation. By adjusting these coefficients, the output of the formula can be Values ​​range from 0 to 1, representing the degree of wetness from dry (0) to wet (1).

[0088] Step 2: Establish a tongue image data recording system

[0089] The collected tongue data is entered into a specially designed tongue data recording system, which has the following functions:

[0090] Data storage: Safely and effectively store a large amount of tongue image data, and classify and manage the data, including classification by patient number, collection time, and disease type. The specific process is as follows:

[0091] First, the directory is divided into large categories according to the disease type, such as cardiovascular disease, respiratory disease, etc. Under each disease type directory, it is further subdivided by year. For example, there is a 2024 subdirectory under the cardiovascular disease category, and folders are created according to the month in the year subdirectory. Such a hierarchical structure helps to initially sort out the data and facilitate management and search; for each patient's data, the patient number is used as the name of an independent folder, and all tongue image data files and their attached information files related to the patient are stored in the patient number folder;

[0092] Create the following data tables: patient information table, tongue image collection schedule, and tongue image characteristic data table; the patient information table records the patient number, name, age, gender, contact information, and basic information of disease type diagnosis results; the tongue image collection schedule records the specific time and collection sequence number of each tongue image collection, and the collection sequence number can be used to determine the order of collection; the tongue image characteristic data table records the specific characteristic values ​​of each tongue image collected, including the RGB value of the color, the measurement value of the thickness of the tongue coating, and the quantitative value of the degree of moistness, and is associated with the patient information table and the tongue image collection schedule through foreign keys to ensure data integrity and traceability;

[0093] For tongue images, lossless compression formats, such as PNG format, are used to ensure that image quality is not impaired. When storing images, metadata information such as image acquisition device parameters and lighting condition parameters must also be recorded and stored in the corresponding patient folder. These metadata are crucial for the accuracy and comparability of subsequent analysis of tongue data. For text data such as basic patient information and quantitative descriptions of tongue characteristics, standardized text formats such as CSV (comma separated values) format or XML format are used. The CSV format is simple and easy to read, and is convenient for batch processing and import and export of data. The XML format has good structure and scalability, and is suitable for storing complex hierarchical data structure information. The stored data should be backed up regularly, and off-site backup and multi-version backup strategies can be used to prevent data loss due to hardware failure, human error or other unexpected situations.

[0094] Data query: It is convenient for doctors and researchers to quickly query the tongue image data of specific patients or specific time periods; Data update: As the patient's recovery process progresses, the tongue image data is updated in a timely manner.

[0095] Step 3: Tongue feature data analysis

[0096] Trend analysis: Analyze the tongue data of the same patient at different time points and draw a trend chart of the tongue characteristics changing over time. The specific process is as follows:

[0097] Filter the tongue data of specific patients who need trend analysis from the tongue data recording system. Ensure the integrity and accuracy of the data, including the tongue images collected from patients at different time points and the corresponding quantitative feature data; sort the tongue data in the order of collection time. Establish a timeline and associate the tongue feature data at each time point with the corresponding time point. Ensure the consistency of time intervals to better observe trend changes;

[0098] Use the previously determined tongue coating thickness quantification method to measure the tongue coating thickness of the tongue image at each time point. Image processing software or algorithms, such as edge detection algorithms, can be used to determine the edge of the tongue coating and then measure the thickness of the tongue coating. Record the measurement results as the quantitative value of the tongue coating thickness. For other tongue features that need to be analyzed, such as tongue color, shape, tongue coating color, degree of moistness, etc., they are also measured and quantified according to the corresponding quantitative standards. Ensure that the quantification method of each feature remains consistent at different time points to facilitate accurate trend analysis;

[0099] Smoothing the data, including moving average method, exponential smoothing method, etc. For example, using the moving average method, taking the average value of the data in a certain time window as the smoothed value, using linear regression analysis, polynomial regression analysis and other methods to fit the trend curve of tongue characteristics changing over time;

[0100] By analyzing the slope and intercept of the trend curve, determine whether the changing trend of the tongue image characteristics is gradually increasing, gradually decreasing, or remaining stable; draw a trend graph of the tongue image characteristics changing over time; with time as the horizontal axis and the tongue image characteristic quantitative value as the vertical axis, mark the tongue image characteristic data at each time point on the graph and connect them into a curve. Use chart drawing software, such as Excel, Python's Matplotlib library, etc., for visual display. Specifically:

[0101] First, determine the type of trend curve. When the trend curve belongs to a linear regression curve, that is, y=ax+b, the slope a represents the change in the dependent variable (tongue characteristic quantification value) when the independent variable (time) increases by one unit, and the intercept b is the value of the dependent variable when the independent variable is 0; when the trend curve belongs to a polynomial regression curve, that is, y=ax²+bx+c, the first-order derivative is 2ax+b. During the change of the curve, when the first-order derivative is greater than 0, the curve rises; when the first-order derivative is less than 0, the curve falls; when the first-order derivative is 0, it is the extreme point of the curve, and the intercept c is also the value of the dependent variable when the independent variable is 0;

[0102] If the slope a of the linear regression curve is a positive number, it means that with the passage of time, the quantitative value of the tongue image characteristics is gradually increasing. For the polynomial regression curve, if its first-order derivative is a positive number above the preset time within the observation time range, it indicates that the tongue image characteristics are gradually increasing; if the slope a of the linear regression curve is a negative number, the quantitative value of the tongue image characteristics is gradually decreasing with time. For the polynomial regression curve, if its first-order derivative is a negative number above the preset time, the tongue image characteristics are gradually decreasing; when the difference between the slope a of the linear regression curve and 0 is less than the preset threshold, it indicates that the tongue image characteristics remain basically stable during this period of time. For the polynomial regression curve, if the difference between its first-order derivative and 0 within the observation time range is less than the preset threshold, and the fluctuation is small, it is considered that the tongue image characteristics are relatively stable.

[0103] According to the shape and change trend of the trend chart, explain the changes of tongue characteristics over time. If the thickness of the tongue coating gradually becomes thinner with the recovery time, it indicates that the patient's condition is improving; if the thickness of the tongue coating remains stable or gradually increases, it is necessary to further analyze the reasons, including poor treatment effect, bad living habits, etc.; correlate the change trend of tongue characteristics with the patient's recovery process, and combine the patient's clinical symptoms, biochemical indicators and other information to comprehensively judge whether the changes in tongue characteristics are consistent with the recovery situation. For example, if the patient's clinical symptoms have improved and the tongue characteristics also show a positive change trend, then it can be considered that the tongue characteristics have a certain correlation with the recovery process;

[0104] If abnormal data points or sudden trend changes are found during trend analysis, further investigation is needed. The causes may be caused by errors in the collection process, sudden changes in the patient's condition, or other factors. By reviewing tongue data, asking patients about their conditions, checking treatment records, etc., the cause of the abnormality can be determined and appropriate measures can be taken to deal with it.

[0105] Step 4: Result visualization and report generation

[0106] The tongue feature analysis results are visualized in an intuitive way, such as drawing bar graphs, line graphs, heat maps, etc. At the same time, a detailed analysis report is generated, which includes the patient's basic information, tongue feature change trends, correlation analysis results, etc., to provide a scientific basis for doctors to formulate personalized rehabilitation plans; the specific process is as follows:

[0107] For the changing trend of tongue features, when a single feature changes over time, a line graph is selected, with time as the horizontal axis and the tongue feature quantification value as the vertical axis to draw a line, clearly showing the continuous change process of the data; such as the changing trend of tongue coating thickness with the number of treatment weeks, the ups and downs of the line can intuitively show how the thickness gradually changes; when comparing different tongue features or showing the difference in tongue features of different patient groups at the same time point, a bar graph is selected for display, such as comparing the average RGB value of the tongue coating color of patients with different disease types at the same recovery stage, and visually showing the difference through the height of different colored columns;

[0108] When showing the complex relationship between tongue characteristics and multiple clinical symptoms or biochemical indicators, a heat map is used, with tongue characteristics, clinical symptoms, biochemical indicators, etc. as variables, and different colors are used to indicate the strength of the correlation between the variables. The warmer the color, the stronger the positive correlation, and the colder the color, the stronger the negative correlation or no correlation. When drawing a chart, add a specific title to summarize the content of the chart, such as "Trend chart of patient A's tongue color RGB value changing with recovery time"; label the coordinate axis name, indicate the time unit or feature category on the horizontal axis, and the quantitative unit on the vertical axis. Add a legend to the chart to explain the meaning of different lines, colors or columns. Adjust the color matching of the chart to ensure that the data is clear and readable, avoid colors that are too glaring or similar and difficult to distinguish, and choose some color schemes that are in line with medical and health themes, such as blue for the normal range and red for abnormal values.

[0109] The steps to generate a report are as follows:

[0110] Extract the patient's basic information from the tongue image data recording system, including name, age, gender, disease diagnosis, and treatment history, and summarize the results of the tongue image characteristic analysis, such as the quantitative data of tongue image characteristics at different time points, the conclusion of trend analysis (trend descriptions such as rising, falling, and stable), and the correlation coefficient and significance level with clinical symptoms and biochemical indicators in the correlation analysis. The report begins with the patient's basic information section, which presents the patient's main background information in the form of tables or paragraphs, followed by a section on the analysis of tongue image characteristic change trends, which details the changes in each characteristic and explains them in combination with visual charts to explain the possible physiological or pathological significance behind the trend changes.

[0111] In the correlation analysis results section, a table or chart showing the correlation between tongue characteristics and clinical symptoms and biochemical indicators is listed. The significantly correlated indicators are discussed in detail, and their potential value for patient condition assessment and rehabilitation program formulation is analyzed. A summary and suggestions are given at the end of the report. The comprehensive tongue analysis results provide doctors with an overall evaluation of the patient's recovery and suggestions on the direction of further treatment or monitoring. After the first draft of the report is generated, a manual review is conducted to check the accuracy of the data, the consistency of the charts and text descriptions, and the logical coherence. It is specifically reviewed by relevant medical experts or data analysts, and feedback is collected and modified to ensure that the report content is accurate, clear, and valuable, and can truly provide a scientific basis for doctors to formulate personalized rehabilitation programs.

[0112] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A method for analyzing the changing rules of tongue features in the rehabilitation process of chronic disease patients, characterized by: The following steps are involved: S1. Standardized collection of tongue image data: Use professional equipment to collect tongue image data of chronic disease patients through standardized mechanisms of lighting, background, diet, and posture, and verify the collected tongue image data; S2. Establish a tongue image data recording system: enter the collected data into a system with storage, query and update functions, and manage the data by patient number, time and disease type; S3. Tongue feature data analysis: trend and correlation analysis of tongue data of the same patient at different times, drawing trend graphs and judging feature changes; S4. Result visualization and report generation: Visualize the analysis results and generate a report containing patient information, changes in tongue characteristics and correlation analysis.

2. The method for analyzing the changing rules of tongue features in the rehabilitation process of chronic disease patients according to claim 1 is characterized in that: The process of the standardization mechanism in S1 is as follows: Lighting conditions: Use lighting equipment with constant light intensity and color temperature to ensure uniform and stable lighting during each collection to avoid deviations in tongue color due to differences in lighting. Specifically, use LED or LEP standard light sources, set fixed brightness and color temperature values, and a natural light color temperature of 5300-7000K to simulate natural lighting conditions; control the lighting angle so that the light is irradiated from the front at an angle of 42-47° above to show the color, shape, and coating characteristics of the tongue; Background color: Choose a single, non-interfering background color, including white or gray. The background material should be light-absorbing to avoid reflection affecting the tongue image collection effect. Use a matte background cloth or background board to ensure the authenticity and accuracy of the tongue image. Dietary restrictions: Before the tongue image is collected, remind the patient to avoid consuming food and drinks that may affect the tongue image, including spicy food, coffee, and strong tea; stipulate the fasting time before collection, specifically fasting within 1-2 hours before collection, to reduce the impact of food residues on the tongue image. At the same time, require the patient to keep the oral cavity clean to avoid food residues adhering to the tongue surface, and rinse the mouth with mouthwash or purified water for more than three times; Posture requirements: prompt the patient to take a uniform sitting or lying posture to ensure that the tongue is in a natural and relaxed state during the collection process. Specifically, the patient sits on a chair, tilts the head forward, sticks out the tongue, and makes the tongue surface parallel to the ground; require the patient to remain quiet during the collection process, avoid talking and swallowing, so as not to affect the stability of the tongue image. At the same time, remind the patient not to stick out the tongue forcefully to avoid deformation of the tongue.

3. The method for analyzing the changing rules of tongue features in the rehabilitation process of chronic disease patients according to claim 1 is characterized in that: The specific operation steps for verifying the collected tongue image data in S1 are as follows: Comprehensive evaluation is performed through the verification parameters of tongue image data to determine whether it is qualified. The verification parameters include: Clarity parameters: including edge sharpness and texture clarity. Edge sharpness is measured by calculating the gradient value of the tongue edge and tongue coating edge in the tongue image. Texture clarity reflects the clarity of the texture by calculating the contrast parameter of the gray level co-occurrence matrix and quantifying it by 100%. The obtained edge sharpness and texture clarity are normalized and weightedly summed to obtain the clarity evaluation value, which is used as the standard for measuring clarity parameters. Noise parameters: including noise density and noise intensity variance. The noise density is obtained by counting the number of noise points in the image and dividing it by the total number of pixels in the image. The noise intensity variance is obtained by calculating the intensity variance of the noise points in the image. The noise density and noise intensity variance are normalized, summed and multiplied by the correction factor to obtain the noise judgment value, which is used as the standard for measuring noise parameters. Contrast parameter: by calculating the contrast value of the tongue in color or grayscale, calculating the grayscale difference for quantification, and then evaluating the overall contrast, by calculating the grayscale histogram distribution range of the image and quantifying it from 1 to 100, recording it as the model evaluation value, normalizing the obtained contrast value and model evaluation value, respectively, using the contrast value and the model evaluation value as two sides, and the preset coefficient as the other side to establish a triangle, calculate the area of ​​the triangle, and record it as the pair parameter value, and use this pair parameter value as the standard for measuring the contrast parameter; Color accuracy parameters: including color deviation and color saturation uniformity. The color deviation is calculated by comparing the color of the collected tongue image with the standard color sample. The color saturation uniformity is calculated by calculating the color saturation standard deviation of multiple areas in the image. The product of the obtained color deviation and the color saturation standard deviation is recorded as the color judgment value, which is used as the standard for measuring color accuracy parameters. The obtained clarity evaluation value, noise judgment value, reference value and color judgment value are normalized and then input into the following formula: To get the comprehensive analysis value ZP, where P, Z, D, S are the clarity value, noise value, reference value and color value respectively. are the preset weight coefficients of the clear evaluation value, noise judgment value, reference value and color judgment value, and the obtained comprehensive analysis value ZP is used as the comprehensive evaluation standard of the tongue image data verification parameter; The obtained comprehensive analysis value ZP is compared with the preset comprehensive analysis threshold. When the comprehensive analysis value ZP is greater than the preset comprehensive analysis threshold, the tongue image data is judged to be up to standard. Otherwise, the tongue image data is judged to be not up to standard, and the patient's tongue image data is recollected until the tongue image data meets the standard.

4. The method for analyzing the changing rules of tongue features in the rehabilitation process of chronic disease patients according to claim 1 is characterized in that: The tongue image data collected in S1 includes the color, shape, and moistness of the tongue. For each feature, a specific quantitative standard is preset. The specific process is as follows: Tongue color quantification: Select the internationally accepted color standard system, including RGB color mode or HSV color mode, and determine the range of each tongue color in the color standard system based on the traditional description of tongue color in traditional Chinese medicine; continuously adjust and refine these ranges through observation and analysis of different samples; use color measuring instruments to measure the color of the collected tongue images and convert them into values ​​in the corresponding color standard system. At the same time, standardize the measurement position and range, and select the middle area of ​​the tongue for measurement; Tongue shape quantification: Use image measurement software or tools to measure the length and width of the tongue on the tongue image. The longest distance from the tip to the root of the tongue is taken as the tongue length, and the distance from the widest part of the two sides of the tongue is taken as the tongue width. To improve the accuracy of the measurement, set a ruler or reference line on the tongue image acquisition device to make accurate measurements on the image. At the same time, the measurement method and unit should be unified. For the measurement of tongue thickness, the side image of the tongue is obtained by side shooting. Then, the thickness of the tongue is measured on the side image, that is, the distance from the back of the tongue to the belly of the tongue. Similarly, image measurement software or tools are needed to measure and ensure the accuracy and consistency of the measurement. The measurement error can be reduced by taking the average value of multiple measurements. According to the shape characteristics of the tongue, it is divided into different types, including normal type, fat type, and thin type. Each shape type is defined and described, and corresponding quantitative standards are formulated. Specifically, a fat tongue is defined as a tongue width greater than the preset normal range, and there are tooth marks on the edge of the tongue; a thin tongue is defined as a tongue width less than the preset normal range, and the tongue surface is dry. Through the classification and quantification of the tongue shape, the morphological changes of the tongue can be more intuitively understood; Quantification of moistness: Analyze the surface texture and reflection characteristics of the tongue coating through image processing technology, and analyze the changes in optical properties; Specifically: Obtain the surface roughness index of the tongue coating through image processing technology and reflectivity , the degree of moisturizing is quantified by the formula , ,in The coefficients are determined according to the actual situation. By adjusting these coefficients, the formula output The value is between 0 and 1, which is used to indicate the change of moistness from dry to wet.

5. The method for analyzing the changing rules of tongue features in the rehabilitation process of chronic disease patients according to claim 1 is characterized in that: The specific steps for data storage in S2 are as follows: The tongue image data is stored safely and effectively, and the stored data is classified and managed, including classification according to patient number, collection time, and disease type. The process is as follows: First, the catalog categories are divided according to the disease type, including cardiovascular disease and respiratory disease. Under each disease type catalog, the year is further subdivided into sub-catalogs, and then folders are created according to the month. This hierarchical structure is used to initially sort out the data for easy management and search; For each patient's data, the patient number is used as the name of an independent folder, and all tongue image data files and their attached information files related to the patient are stored in the patient number folder; Create the following data tables: patient information table, tongue image collection schedule, and tongue image characteristic data table; The patient information table records the basic information of the patient number, name, age, gender, contact information, and disease type diagnosis results; the tongue image collection schedule records the specific time and collection sequence number of each tongue image collection, and the collection sequence number is used to determine the order of collection; the tongue image feature data table records the specific feature values ​​of each tongue image collection, including the RGB value of the color, the measurement value of the thickness of the tongue coating, and the quantitative value of the degree of moistness, and is linked to the patient information table and the tongue image collection schedule through foreign keys to ensure the integrity and traceability of the data; For tongue images, the lossless compression format PNG is used to ensure that the image quality is not damaged. When storing the image, the image acquisition device parameters and illumination condition parameter metadata information are recorded and stored in the corresponding patient folder; For the patient's basic information and quantitative description text data of tongue characteristics, standardized text formats are used, including comma-separated value format or XML format, and the stored data is backed up regularly, using off-site backup and multi-version backup strategies to prevent data loss due to hardware failure, human error and other accidents.

6. The method for analyzing the changing rules of tongue features in the rehabilitation process of chronic disease patients according to claim 1 is characterized in that: The specific operation steps of S3 are as follows: Screening out the tongue image data of patients who need to be analyzed for trend from the tongue image data recording system, including tongue image data collected from patients at different time points and corresponding quantitative feature data; Sort the tongue image data in the order of collection time, establish a time axis, and associate the tongue image characteristic data at each time point with the corresponding time point to ensure the consistency of the time interval so as to better observe the trend changes; The previously determined tongue coating thickness quantification method is used to measure the tongue coating thickness of the tongue image at each time point. The edge of the tongue coating is determined using image processing software or an algorithm, and then the thickness of the tongue coating is measured. The measurement results are recorded as the quantified value of the tongue coating thickness. For other tongue features that need to be analyzed, they are also measured and quantified according to the corresponding quantification standards to ensure that the quantification method of each feature remains consistent at different time points, so as to conduct accurate trend analysis. The data was smoothed by using the moving average method, taking the average value of the data in the preset time window as the smoothed value, and using linear regression analysis and polynomial regression analysis methods to fit the trend curve of tongue characteristics changing over time; By analyzing the slope and intercept of the trend curve, determine whether the changing trend of the tongue image characteristics is gradually increasing, gradually decreasing, or remaining stable; draw a trend graph of the tongue image characteristics changing over time; with time as the horizontal axis and the tongue image characteristic quantification value as the vertical axis, mark the tongue image characteristic data at each time point on the graph and connect them into a curve, and use chart drawing software for visual display; Based on the shape and change trend of the trend chart, explain how the tongue characteristics change over time. If the thickness of the tongue coating gradually becomes thinner with the recovery time, it indicates that the patient's condition is improving; if the thickness of the tongue coating remains stable or gradually increases, further analyze the reasons, including poor treatment effect and bad living habits; The changing trend of tongue characteristics is correlated with the patient's recovery process, and combined with the patient's clinical symptoms and biochemical index information, a comprehensive judgment is made as to whether the changes in tongue characteristics are consistent with the recovery situation; If abnormal data points or sudden trend changes are found during the trend analysis process, further investigation is needed into the causes, which may be errors in the collection process or sudden changes in the patient's condition. By reviewing the tongue data, asking the patient about their condition, and checking the treatment records, the cause of the abnormality can be determined and appropriate measures can be taken to deal with it.

7. The method for analyzing the changing rules of tongue features in the rehabilitation process of chronic disease patients according to claim 6 is characterized in that: The specific process of determining whether the changing trend of the tongue image characteristics is gradually increasing, gradually decreasing, or remaining stable in S3 is as follows: First, determine the type of trend curve. When the trend curve belongs to a linear regression curve, that is, y=ax+b, the slope a represents the change in the dependent variable, i.e., the quantitative value of the tongue image characteristics, when the independent variable, i.e., time, increases by one unit. The intercept b is the value of the dependent variable when the independent variable is 0. When the trend curve belongs to a polynomial regression curve, that is, y=ax²+bx+c, the first-order derivative is 2ax+b. During the change of the curve, when the first-order derivative is greater than 0, the curve rises; when the first-order derivative is less than 0, the curve falls. When the first-order derivative is 0, it is the extreme point of the curve, and the intercept c is also the value of the dependent variable when the independent variable is 0; If the slope a of the linear regression curve is a positive number, it means that the quantified value of the tongue image characteristic is gradually increasing over time. For the polynomial regression curve, if the first-order derivative is a positive number above the preset time within the observed time range, it indicates that the tongue image characteristic is gradually increasing. If the slope a of the linear regression curve is negative, the quantified value of the tongue image characteristic gradually decreases over time. For the polynomial regression curve, if its first-order derivative is negative over a preset time, the tongue image characteristic shows a gradually decreasing trend. When the difference between the slope a of the linear regression curve and 0 is less than the preset threshold, it indicates that the tongue image characteristics remain stable during this period. For the polynomial regression curve, if the difference between its first-order derivative and 0 within the observed time range is less than the preset threshold, the tongue image characteristics are judged to be stable.

8. The method for analyzing the changing rules of tongue features in the rehabilitation process of chronic disease patients according to claim 1 is characterized in that: The specific process of visual display of the analysis results in S4 is as follows: For the changing trend of tongue image characteristics, when a single characteristic changes over time, a line graph is used, with time as the horizontal axis and the tongue image characteristic quantification value as the vertical axis to draw a line to show the continuous change process of the data; When comparing different tongue features or showing the differences in tongue features of different patient groups at the same time point, choose a bar chart to display, compare the average RGB values ​​of the tongue coating color of patients with different disease types at the same recovery stage, and intuitively present the differences through the height of different colored columns; When showing the relationship between tongue features and multiple clinical symptoms or biochemical indicators, a heat map is used, with tongue features, clinical symptoms, and biochemical indicators as variables, and different colors are used to indicate the strength of the correlation between the variables. The warmer the color, the stronger the positive correlation, and the colder the color, the stronger the negative correlation or no correlation. When drawing a chart, add a specific title to summarize the content of the chart; label the coordinate axis name, indicate the time unit or feature category on the horizontal axis, indicate the quantitative unit on the vertical axis, and add a legend to the chart to explain the meaning of different lines, colors or columns; Adjust the color matching of the chart to ensure that the data is clear and readable, avoid colors that are too glaring or similar and difficult to distinguish, and choose a color scheme that is consistent with medical and health themes, including blue to represent the normal range and red to represent abnormal values.

9. The method for analyzing the changing rules of tongue features in the rehabilitation process of chronic disease patients according to claim 1 is characterized in that: The steps for generating the report in S4 are as follows: Extract the patient's basic information from the tongue image data recording system, including name, age, gender, disease diagnosis, and treatment history, and summarize the results of tongue image characteristic analysis, including quantitative data of tongue image characteristics at different time points, conclusions of trend analysis, and correlation coefficients and significance levels with clinical symptoms and biochemical indicators in correlation analysis; The report begins with the patient's basic information section, which presents the patient's background information in the form of tables or paragraphs, followed by a section on the trend analysis of tongue characteristics, which describes the changes in each characteristic and explains them with visual charts to explain the physiological or pathological significance behind the trend changes; In the correlation analysis results section, a table or chart showing the correlation between tongue features and clinical symptoms and biochemical indicators is listed. The relevant indicators are discussed and their potential value for patient condition assessment and rehabilitation program formulation is analyzed. A summary and suggestions are given at the end of the report. The comprehensive tongue analysis results provide doctors with an overall evaluation of the patient's rehabilitation status and recommendations for further treatment or monitoring. After the first draft of the report is generated, it will be manually reviewed to check the accuracy of the data, the consistency of the charts and text descriptions, and the logical coherence. It will be reviewed by relevant medical experts or data analysts, and feedback will be collected and modified to ensure that the report content is accurate, clear, and valuable, providing a scientific basis for doctors to formulate personalized rehabilitation plans.

Images