A myopia prevention and control health management system integrating traditional Chinese medicine intervention strategies

Abstract

The application belongs to the technical field of medical information, and discloses a myopia prevention and control health management system fusing traditional Chinese medicine intervention strategies. The system integrates a multispectral tongue image collection device, a pulse pressure sensor array, a voiceprint collection microphone and a standardized inquiry interface, and is matched with an AI follow-up function to collect full-dimension data of observation, smell, inquiry and incision. A special model is used to automatically generate traditional Chinese medicine syndrome types and syndrome scores. Meanwhile, a deep learning multi-modal dynamic fusion model is constructed, a layered architecture is used to extract features, and data weights are dynamically allocated. The evaluation logic is optimized for differences between teenagers and people with weak constitutions. Traditional Chinese medicine syndromes are deeply associated with Western ophthalmic indexes, eye behavior and environmental data to accurately determine myopia risk levels and development trends. The system automatically generates an integrated scheme of traditional Chinese medicine tea, AR acupoint massage, ear acupoint bean pressing, traditional Chinese medicine fumigation, meridian and collateral regulation and other schemes based on syndrome differentiation and clinical knowledge base.

Description

Technical Field

[0001] This invention belongs to the field of medical information technology, specifically a myopia prevention and control health management system that integrates traditional Chinese medicine intervention strategies. Background Technology

[0002] With the increasing demand for myopia prevention and control, the application scenarios of existing myopia prevention and control health management systems are becoming more and more widespread. However, their core defects are also gradually becoming apparent, mainly reflected in the following technical issues: Traditional Chinese medicine believes that the occurrence and development of myopia are closely related to physical factors such as deficiency of liver and kidney, insufficient qi and blood, and weakness of spleen and stomach. At the same time, it is also influenced by lifestyle factors such as eye use habits, work and rest rhythm, diet and emotions. However, the existing system cannot quantify and evaluate the above-mentioned TCM factors and provide targeted interventions. Furthermore, it has not achieved deep integration of TCM and Western medicine data, and is mostly a simple superposition, lacking the ability to perform dynamic correlation analysis.

[0003] Traditional Chinese medicine (TCM) has a variety of effective methods for myopia prevention and control, such as acupuncture, massage, auricular acupressure, herbal fumigation, and dietary therapy. However, these methods are highly dependent on the personal experience of TCM practitioners and are difficult to promote on a large scale and with precision through existing systems. Moreover, existing related patents mostly focus on the digitization of single TCM intervention methods and have not achieved the synergistic integration and adaptive adaptation of multiple TCM intervention methods.

[0004] Most systems can only provide static health advice and cannot dynamically adjust intervention plans according to the user's real-time status. They also lack a multi-dimensional quantitative evaluation system for the effectiveness of TCM interventions, resulting in poor user compliance and difficulty in sustaining prevention and control effects. Summary of the Invention

[0005] The purpose of this invention is to provide a myopia prevention and control health management system that integrates traditional Chinese medicine intervention strategies, so as to solve one or more problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a myopia prevention and control health management system integrating traditional Chinese medicine intervention strategies, comprising the following modules: Furthermore, the four diagnostic methods module integrates a multispectral tongue image acquisition device, a pulse pressure sensor array, a voiceprint acquisition microphone, and a standardized consultation interface to achieve standardized and quantitative acquisition and intelligent diagnostic integration of data from the four diagnostic methods of traditional Chinese medicine (inspection, auscultation, inquiry, and palpation). Tongue image acquisition uses multispectral imaging technology to automatically identify features such as tongue texture, tongue coating, tongue shape, and tongue color, and matches them with TCM syndrome correlation features through a deep learning model. Pulse image acquisition uses a pressure sensor array to obtain parameters such as waveform, frequency, and rhythm of the pulse in the cun, guan, and chi positions. Combined with the pulse feature database, it realizes intelligent judgment of pulse type. Voiceprint acquisition analyzes the strength, clarity, and other features of the user's voice to help judge the body's constitution. The consultation interface is based on the TCM myopia diagnosis scale, guiding users to complete standardized symptom self-assessment. It also incorporates an AI follow-up question function, which supplements targeted consultation questions based on the user's initial answers. The collected raw data is converted into structured TCM syndrome data. The AI ​​diagnosis model integrates and analyzes the four diagnostic methods data to automatically generate TCM syndrome types and syndrome scores, which are then simultaneously transmitted to the risk assessment module.

[0007] The AI-assisted follow-up questioning system has clearly defined termination conditions. When the system determines that the completeness of the four diagnostic methods data meets the standard requirements, the core symptom information is not missing, and the syndrome type is clearly distinguishable, the follow-up questioning will automatically stop and enter the data processing stage. If the user still cannot provide effective information after continuous follow-up questioning, the system will combine the existing data to make reasonable supplements and mark the data source.

[0008] Furthermore, the risk assessment module constructs a multimodal dynamic fusion model based on deep learning, which integrates the TCM syndrome differentiation results and syndrome scores transmitted by the four diagnostic methods module with Western medical psychological indicator data, behavioral habit data, and environmental data. A layered fusion architecture is adopted. The bottom layer performs dedicated feature extraction on different modal data and designs a dedicated feature extraction algorithm for TCM syndrome data. The middle layer realizes cross-modal association and dynamic weight allocation of features of each modality. The weight of each data is adjusted in real time according to the user's age, disease course and physical condition. The weight of data such as axial length and eye use habits is increased in adolescents, and the weight of TCM syndrome data is increased in people with weak constitution. The system employs a multi-dimensional comprehensive assessment rule for individuals with weak constitutions. It combines four core pieces of information: TCM syndrome score, self-assessment of physical constitution, chronic disease history, and daily physical condition. Individuals with weak constitutions are accurately identified as having a weak constitution if they meet any two or more of the following four criteria: syndrome score above 70, self-assessment showing obvious Qi / blood / yin / yang deficiency, chronic weakness or a history of chronic spleen, stomach, liver, or kidney diseases, or easy fatigue during daily activities. Once the system identifies this state, it automatically switches to a dedicated assessment logic, strengthening the analysis weight of TCM constitution and syndrome data while weakening the influence of single Western medical indicators.

[0009] The weights of multimodal features are not fixed. When the user's age changes, myopia progresses, TCM constitution changes, or eye use behavior or living environment changes significantly, the system will automatically trigger a recalculation and reassignment of the weights.

[0010] The top-level output integrates myopia risk level, development trend prediction and risk factor analysis, combined with TCM syndrome differentiation to identify myopia-related pathogenesis, clarify the correlation between each syndrome and myopia risk, and push the assessment results to the intervention plan module simultaneously.

[0011] Furthermore, the intervention program module is based on the TCM syndrome differentiation and treatment theory and clinical knowledge base, and combines the myopia risk level, TCM syndrome type and risk factor analysis pushed by the risk assessment module to automatically generate an individualized comprehensive intervention program, which includes a variety of TCM intervention methods. Based on the syndrome differentiation, traditional Chinese medicine teas or medicinal diets are formulated and the ingredients are dynamically adjusted according to the user's dietary preferences. Massage guidance is provided for acupoints such as Jingming, Zanzhu, and Fengchi. Acupoints are located using AR technology, and standardized operation videos are pushed simultaneously. The selection of acupoints and operation methods for ear acupressure are clarified. The frequency and intensity of acupressure are adjusted according to the user's physical condition and sensitivity. The time and frequency of moxibustion and traditional Chinese medicine fumigation for the eyes are optimized based on environmental temperature and humidity data. The work and rest conditioning plan is set based on the theory of meridian flow in traditional Chinese medicine and is adapted to the user's daily work and rest habits. The system matches corresponding ingredients and medicinal materials based on the results of TCM syndrome differentiation. For liver and kidney deficiency syndrome, it prioritizes the selection of ingredient combinations that nourish the liver and kidneys; for qi and blood deficiency syndrome, it prioritizes the selection of ingredient combinations that replenish qi and nourish blood; and for spleen and stomach weakness syndrome, it prioritizes the selection of ingredient combinations that strengthen the spleen and nourish the stomach. At the same time, it takes into account the user's dietary restrictions, taste preferences, and allergy history to remove unsuitable ingredients and automatically replace them with alternative ingredients with similar effects.

[0012] Based on the meridian circulation patterns of the meridians, the system matches the active and rest periods of the meridians related to the liver, kidneys, spleen, stomach, and eyes, and plans the best time nodes for waking up in the morning, using the eyes during the day, eye care, relaxing at night, and falling asleep. At the same time, it collects personalized information such as the user's daily routine, study and work schedule, and family life habits, and makes flexible fine-tuning of the routine nodes without disrupting the user's original life rhythm.

[0013] The system is equipped with a multi-level dynamic adjustment trigger mechanism. The optimization process will be initiated immediately when any of the following conditions are met: TCM syndrome score does not decrease or even increases for two consecutive weeks, visual acuity and refractive error show a deteriorating trend for two consecutive weeks, axial length growth rate exceeds the normal range, intervention execution rate is consistently below 60%, user's constitution syndrome type changes significantly, or new eye discomfort symptoms appear. After triggering, the system automatically locates the core cause of poor effect and, combined with the latest four diagnostic methods data and behavioral data, quickly completes iterative optimization of intervention methods, operating parameters, and conditioning plans.

[0014] A knowledge graph technology is used to build a knowledge base for myopia intervention in traditional Chinese medicine, which supports dynamic updates and optimization of intervention plans. The system is configured with a collaborative scheduling algorithm for intervention methods, which automatically adjusts the execution order and duration of each intervention method based on the user's intervention implementation and effect feedback. The system also verifies the compliance of the intervention plan and pushes the generated plan to the behavior correction module and the remote supervision module simultaneously.

[0015] Furthermore, the behavior correction module receives individualized intervention plans pushed by the intervention plan module, collects behavioral data such as user screen time, screen distance, and screen posture through smart devices, uses AI algorithms to predict potential poor screen behavior, issues reminders in advance, and explains the harm of the behavior and its impact on physical constitution in conjunction with traditional Chinese medicine theory. Based on the user's TCM constitution, myopia syndrome type, and real-time behavioral data, we provide individualized dietary conditioning suggestions and set personalized exercise duration and difficulty based on the user's physical condition. It has the function of analyzing the correlation between lifestyle habits and TCM constitution, and regularly generates habit correction reports, which are simultaneously pushed to the remote supervision module and the effect evaluation module.

[0016] Furthermore, the remote supervision module establishes a remote communication platform that links users, TCM doctors, parents, and school teachers. It receives user behavior data pushed by the behavior correction module and individualized intervention plans pushed by the intervention plan module. It supports users to upload photos, videos, and logs of intervention implementation. The system automatically verifies the standardization of intervention implementation and reports any abnormalities to TCM doctors. Traditional Chinese medicine practitioners can view user intervention records, four diagnostic methods data, and risk assessment results online, provide guidance and correction for user operations, adjust intervention plans when user symptoms or physical condition change, and synchronize them to the intervention plan module and behavior correction module. Parents can view their child's intervention implementation and prevention progress in real time, receive reminder messages, and assist in supervising intervention implementation. It supports school teachers in viewing myopia prevention and control data of student groups, conducting centralized education on common problems, establishing an incentive mechanism for intervention implementation, and transmitting relevant supervision data to the effect evaluation module simultaneously.

[0017] Furthermore, the effect evaluation module establishes a comprehensive evaluation system that includes four dimensions: improvement of TCM syndromes, changes in Western medicine psychological indicators, improvement in quality of life, and intervention compliance. It receives intervention execution data pushed by the remote supervision module, TCM syndrome data pushed by the four diagnostic methods module, initial risk data pushed by the risk assessment module, and behavioral data pushed by the behavior correction module. Based on the data from the four diagnostic methods and AI-based diagnosis submitted by users on a regular basis, the changes in syndrome scores are calculated and the trend of syndrome improvement is analyzed. The user's visual acuity, refractive error, axial length and other Western medical indicators are tracked. The correlation between the intervention effect and the changes in indicators is judged by combining the data from traditional Chinese and Western medicine. The impact of myopia on the user's learning, life and psychological state is assessed by using a standardized scale. The user's intervention execution record is combined to quantitatively assess compliance and its impact on the intervention effect. The system uses visualization technology to display the curves of intervention effect changes and the evaluation results of various dimensions, automatically generates evaluation reports, clarifies the advantages and disadvantages of the program, identifies the causes of poor intervention effect through AI algorithms and pushes targeted optimization suggestions, and the evaluation results are simultaneously fed back to the intervention program module and the intelligent education module.

[0018] Furthermore, the intelligent education module constructs a knowledge graph of TCM myopia prevention and control, which includes the TCM etiology and pathogenesis of myopia, syndrome differentiation, intervention methods, prevention and health care, and common misconceptions. It incorporates a TCM myopia prevention and control clinical case library, updates research results and clinical experience in real time, and receives evaluation results pushed by the effect evaluation module and TCM syndrome data pushed by the four diagnostic methods module. The system combines user health profiles and personalized intervention plans pushed by the intervention plan module, pushes personalized education content through natural language processing technology, supports intelligent question and answer via voice or text, and provides personalized answers by retrieving information through knowledge graphs and combining them with the user's specific situation. We use questionnaires and interactive Q&A to understand users' grasp of the educational content, dynamically adjust the educational content and push frequency, and synchronously feed the educational data back to the behavior correction module and intervention program module.

[0019] The beneficial effects of this invention are as follows: 1. This invention integrates a multispectral tongue image acquisition device, a pulse pressure sensor array, a voiceprint acquisition microphone, and a standardized consultation interface. It is equipped with an AI follow-up questioning function to collect data from all dimensions of observation, auscultation, inquiry, and palpation. Through a dedicated model, it automatically generates TCM syndrome types and syndrome scores. At the same time, it constructs a deep learning multimodal dynamic fusion model, adopts a hierarchical architecture to extract features and dynamically allocate data weights, and optimizes the assessment logic differently for adolescents and people with weak constitutions. It deeply correlates TCM syndromes with Western ophthalmological indicators, eye use behavior, and environmental data to accurately determine the risk level and development trend of myopia.

[0020] 2. Based on syndrome differentiation and treatment and a clinical knowledge base, the system of this invention automatically generates integrated solutions such as Chinese herbal tea, AR acupoint massage, ear acupressure, Chinese herbal fumigation, and meridian-based lifestyle regulation. It dynamically adjusts the intensity and method of intervention based on the user's constitution, dietary preferences, and environmental data. Based on knowledge graphs and collaborative scheduling algorithms, the system optimizes the solutions in real time, links with the behavior correction module, matches exclusive dietary therapy and traditional exercise suggestions according to Chinese medicine syndrome types, and provides early warnings of poor eye use behaviors and interprets the harms in conjunction with Chinese medicine theory. This achieves the synergistic integration and adaptive adaptation of multiple Chinese medicine intervention methods.

[0021] 3. The remote supervision module of this invention establishes a four-party linkage platform involving users, TCM doctors, parents, and schools, automatically verifying the standardization of intervention implementation, providing online guidance from doctors and adjusting the plan in real time, and significantly improving the implementation rate through home-school collaborative supervision and incentive mechanisms; the effect evaluation module quantifies the evaluation from four dimensions: improvement of TCM syndromes, changes in Western medical psychological indicators, improvement in quality of life, and intervention compliance, generating visual reports and targeted optimization suggestions; the intelligent education module pushes personalized science popularization content based on knowledge graphs. Attached Figure Description

[0022] Figure 1 This is a flowchart illustrating the overall workflow of the system of the present invention; Figure 2 This is a flowchart of the four diagnostic methods for syndrome differentiation in this invention. Detailed Implementation

[0023] 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 only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0024] like Figures 1 to 2 As shown, this embodiment of the invention provides a myopia prevention and control health management system that integrates traditional Chinese medicine intervention strategies, including the following modules: In this embodiment of the invention, the four diagnostic methods module integrates a multispectral tongue image acquisition device, a pulse pressure sensor array, a voiceprint acquisition microphone, and a standardized consultation interface to achieve standardized and quantitative acquisition and intelligent diagnostic integration of data from the four diagnostic methods of traditional Chinese medicine (inspection, auscultation, inquiry, and palpation). The system employs a real-time encrypted data transmission mechanism between its modules. Data collected from the acquisition end is processed in a standardized manner and then synchronized to the corresponding functional modules in real time. Information such as assessment results, intervention plans, and execution data are all transmitted in encrypted form and stored securely locally. The system also supports bidirectional data interaction and real-time updates between modules.

[0025] Tongue image acquisition uses multispectral imaging technology to automatically identify features such as tongue texture, tongue coating, tongue shape, and tongue color, and matches them with TCM syndrome correlation features through a deep learning model. Pulse image acquisition uses a pressure sensor array to obtain parameters such as waveform, frequency, and rhythm of the pulse in the cun, guan, and chi positions. Combined with the pulse feature database, it realizes intelligent judgment of pulse type. Voiceprint acquisition analyzes the strength, clarity, and other features of the user's voice to help judge the state of Qi and blood. The multispectral tongue image acquisition device has a fixed acquisition distance of 30cm and uses a wide-band imaging of 450nm-950nm. It automatically completes illumination compensation, image noise reduction, and tongue segmentation preprocessing. The pulse pressure sensor array uses an 8-channel flexible sensor with an acquisition pressure range of 0-30kPa. It automatically completes pulse signal filtering, baseline correction, and feature point extraction. The voiceprint acquisition uses a noise-reducing microphone with a sampling rate of 44.1kHz, automatically shielding environmental noise and extracting sound intensity, timbre, and rhythm features. All acquired data are normalized according to the TCM four diagnostic methods data standards and converted into structured data in a unified format.

[0026] The consultation interface is based on the TCM myopia diagnosis scale, guiding users to complete standardized symptom self-assessment. It also incorporates an AI follow-up question function to supplement targeted consultation questions based on the user's initial answer, thereby improving the accuracy of the consultation. The collected raw data is converted into structured TCM syndrome data, and the four diagnostic methods are integrated and analyzed through the AI ​​diagnosis model to automatically generate TCM syndrome types and syndrome scores, which are then simultaneously transmitted to the risk assessment module.

[0027] The AI ​​diagnostic model adopts a multi-branch CNN and Transformer fusion architecture. The tongue diagnosis branch outputs 256-dimensional quantitative features of tongue quality, tongue coating, tongue shape, and tongue color through a multispectral feature extraction network. The pulse diagnosis branch outputs 128-dimensional waveform, frequency, and rhythm features of the cun, guan, and chi positions through a temporal feature encoding network. The voiceprint diagnosis branch outputs 64-dimensional features of sound intensity, clarity, and rhythm through an audio feature extraction network. The symptom inquiry branch outputs 80-dimensional standardized symptom quantification features through a semantic encoding network. The features of the four branches are weighted and aggregated by a cross-modal fusion layer and then input into the classification head and regression head. The classification head outputs core TCM syndrome types such as liver and kidney deficiency, qi and blood deficiency, and spleen and stomach weakness. The regression head outputs syndrome scores of 0-100. The training set contains 120,000 cases of myopia diagnosis data from adolescents annotated by associate chief physicians or above, and the validation set accounts for 15%. The AdamW optimizer is used with an initial learning rate of 1×10⁻⁶. -4 Batch size 32, training 60 epochs, early stopping strategy to prevent overfitting; The AI ​​follow-up questioning algorithm is built on a Bayesian probabilistic network and a symptom association map. It takes a symptom completeness threshold of 85% as the trigger condition, calculates the probability of missing symptoms and the correlation between syndrome types, and automatically outputs the top 3 highly relevant follow-up questions for missing symptoms.

[0028] The syndrome score adopts a multi-dimensional weighted scoring method, which comprehensively calculates the quantitative results of four dimensions: tongue appearance by observation, pulse by palpation, voice by auscultation, and symptoms by inquiry. It fully reflects the degree of physical imbalance, the severity of myopia-related syndromes, and the functional status of meridians and internal organs. The score ranges from 0 to 100 points. The higher the score, the greater the impact of TCM pathogenesis on the occurrence and development of myopia and the more severe the physical imbalance.

[0029] The formula for weighted calculation of syndrome differentiation in Traditional Chinese Medicine is as follows: The TCM syndrome score represents a quantitative score indicating the degree of imbalance in a user's TCM constitution and the severity of myopia-related pathogenesis. This represents the standardized quantitative value of the i-th TCM diagnostic dimension, encompassing the standardized processing results of each dimension of the four diagnostic methods: observation, auscultation, inquiry, and palpation. This represents the contribution weight of the i-th TCM syndrome differentiation dimension, which is automatically allocated by the AI ​​syndrome differentiation model, reflecting the influence of each dimension on syndrome determination. The value range of the syndrome score is indicated. The lower the value, the milder the physical imbalance, and the higher the value, the more severe the physical imbalance.

[0030] In this embodiment of the invention, the risk assessment module constructs a multimodal dynamic fusion model based on deep learning, which integrates the TCM syndrome differentiation results and syndrome scores transmitted by the four diagnostic methods module with Western medicine physiological indicators, behavioral habit data, and environmental data. The Western medicine optometry data includes visual acuity, refractive error, axial length, corneal curvature, etc.; the behavioral habit data includes eye use duration, eye use distance, outdoor activity time, etc.; and the environmental data includes light intensity, color temperature, etc. A layered fusion architecture is adopted. The bottom layer performs dedicated feature extraction on different modal data and designs a dedicated feature extraction algorithm for TCM syndrome data. The middle layer realizes cross-modal association and dynamic weight allocation of features of each modality. The weight of each data is adjusted in real time according to the user's age, disease course and physical condition. The weight of data such as axial length and eye use habits is increased in adolescents, and the weight of TCM syndrome data is increased in people with weak constitution. The multimodal dynamic fusion model adopts a three-layer heterogeneous fusion architecture. The bottom layer is a modality-specific feature extraction layer: TCM syndrome data is extracted with 64-dimensional core pathogenesis features through a fully connected neural network and syndrome feature mapping algorithm; Western medicine ophthalmology indicators are extracted with 32-dimensional physiological features such as visual acuity, refractive error, axial length, and corneal curvature through a 1D-CNN temporal feature extraction network; and eye use behavior and environmental data are extracted with 64-dimensional dynamic features such as duration, distance, posture, illumination, and color temperature through an LSTM temporal coding network. The middle layer is a cross-modal attention fusion layer, which uses an 8-head self-attention mechanism to realize the semantic association and information interaction of features of different modalities. The dynamic weight allocation network calculates the weights of each modal feature to realize the real-time adaptive adjustment of the feature weights. The top layer is the risk decision output layer, which inputs 256-dimensional fused features into a fully connected classification and regression network, and simultaneously outputs 5 levels of myopia risk, the slope of myopia development trend, and 3 core risk factors. At the same time, it links with traditional Chinese medicine syndrome types to clarify the correspondence between myopia pathogenesis and risk. Level 1: score < 20 points; Level 2: 20-40 points; Level 3: 40-60 points; Level 4: 60-80 points; Level 5: > 80 points.

[0031] The intervention method collaborative scheduling algorithm is built based on the deep reinforcement learning DQN model. It takes the user intervention execution cycle as the decision unit. The state space includes four core state features: intervention execution rate, syndrome integral change value, visual acuity improvement value, and refractive error change value. The action space covers the execution order, single duration, and daily frequency combination strategy of five types of intervention methods: Chinese herbal tea, AR acupoint massage, ear acupressure, Chinese herbal fumigation, and meridian flow therapy. The model employs an experience replay mechanism to optimize training, with a learning rate of 5×10. -4 The initial exploration rate was 0.12, and the strategy iteration and plan scheduling were completed every 7 days, automatically adapting to changes in user physical condition, intervention implementation and prevention and control effects.

[0032] Feature weight formula for multimodal dynamic fusion model: This represents the dynamic weighting of the i-th modality data in myopia risk assessment, used to precisely adjust the contribution ratio of each dimension of data to the final assessment result; This represents the feature importance calculated after the i-th modality data is extracted from the underlying features, reflecting the core value of this modality data in determining myopia risk; This represents the user's physical fitness compatibility coefficient, which is dynamically assigned based on information such as the user's age, physical condition, and TCM syndrome type, adapting to the assessment logic of different groups of people. This represents the correlation coefficient of myopia course, calculated by combining the user's myopia progression rate and course duration, reflecting the impact of the disease stage on risk assessment.

[0033] Reward function for intervention-based coordinated scheduling algorithm: The score represents the overall benefit of the intervention program. The score directly reflects the degree of optimization of the intervention strategy and the actual prevention and control effect. The user intervention task execution rate represents the actual percentage of users who complete daily TCM intervention tasks, reflecting the completeness of intervention execution. It represents the improvement rate of TCM syndromes, used to quantify the degree of relief of TCM constitution imbalance and myopia-related pathogenesis after intervention; The improvement value of the user's vision is the actual change in the user's vision indicators before and after the intervention, reflecting the prevention and control effect at the Western medicine level; This represents the intervention compliance improvement value, used to quantify the degree of active cooperation and behavioral improvement of users in implementing the intervention plan.

[0034] The top-level output integrates myopia risk level, development trend prediction and risk factor analysis, combined with TCM syndrome differentiation to identify myopia-related pathogenesis, clarify the correlation between each syndrome and myopia risk, and push the assessment results to the intervention plan module simultaneously.

[0035] The system classifies myopia risk into five levels, combining traditional Chinese medicine (TCM) syndrome differentiation, Western ophthalmological indicators, eye-use behavior, and environmental data for comprehensive judgment. Level 1 risk is a low-risk state for myopia, with all indicators within the normal range; Level 2 risk is a slightly low-risk state for myopia, with minor issues in eye-use behavior; Level 3 risk is a moderate-risk state for myopia, with faster axial length development or mild TCM constitution imbalance; Level 4 risk is a slightly high-risk state for myopia, with a declining vision trend and significant constitution imbalance; Level 5 risk is a high-risk state for myopia, with rapid increase in refractive error, abnormal axial length growth, and a high TCM syndrome score, which can directly predict the occurrence or rapid progression of myopia.

[0036] In this embodiment of the invention, the intervention plan module is based on the theory of syndrome differentiation and treatment in traditional Chinese medicine and a clinical knowledge base. Combined with the analysis of myopia risk level, TCM syndrome type and risk factors pushed by the risk assessment module, it automatically generates an individualized comprehensive intervention plan, which includes a variety of TCM intervention methods. Based on the syndrome differentiation, traditional Chinese medicine teas or medicinal diets are formulated and the ingredients are dynamically adjusted according to the user's dietary preferences. Massage guidance is provided for acupoints such as Jingming, Zanzhu, and Fengchi. Acupoints are located using AR technology, and standardized operation videos are pushed simultaneously. The selection of acupoints and operation methods for ear acupressure are clarified. The frequency and intensity of acupressure are adjusted according to the user's physical condition and sensitivity. The time and frequency of moxibustion and traditional Chinese medicine fumigation for the eyes are optimized based on environmental temperature and humidity data. The work and rest conditioning plan is set based on the theory of meridian flow in traditional Chinese medicine and is adapted to the user's daily work and rest habits. The AR acupoint positioning system uses a mobile phone camera to collect real-time images of the user's face and neck. Based on the key point recognition technology of human facial bones, it establishes a spatial coordinate system and automatically matches the standard positions of eye-related acupoints such as Jingming, Zanzhu, and Fengchi. The precise positions of the acupoints are displayed on the screen with dynamic icons superimposed on them. At the same time, the standard pressure, direction, and frequency of acupoint massage are displayed simultaneously. Users can adjust their operating posture in real time by comparing with the screen, thereby completing acupoint positioning and standardized operation.

[0037] Traditional Chinese medicine fumigation automatically adjusts the steam temperature and fumigation time according to the ambient temperature and humidity. In high-temperature environments, the steam temperature is reduced and the duration of each session is shortened, while in low-temperature environments, the steam temperature is appropriately increased and the duration of each session is extended. For eye moxibustion, the distance and temperature of the moxibustion are strictly controlled to avoid eye burns, and the frequency of moxibustion is adjusted according to the user's physical condition and sensitivity.

[0038] A knowledge graph technology is used to build a knowledge base for myopia intervention in traditional Chinese medicine, which supports dynamic updates and optimization of intervention plans. The system is configured with a collaborative scheduling algorithm for intervention methods, which automatically adjusts the execution order and duration of each intervention method based on the user's intervention implementation and effect feedback. The system also verifies the compliance of the intervention plan and pushes the generated plan to the behavior correction module and the remote supervision module simultaneously.

[0039] The compliance verification of intervention plans covers four dimensions: contraindications of Chinese herbal tea combinations, safe range of acupoint operation, limits on fumigation temperature and duration, and rationality of work and rest schedule. The system has a built-in TCM clinical safety standard library, which automatically screens for problems such as conflicts in the combination of Chinese herbal ingredients, risks of acupoint operation, and excessive intervention intensity. Non-compliant content is automatically replaced or deleted to ensure that the generated intervention plan meets the TCM clinical safety standards.

[0040] The coordinated scheduling of intervention measures adopts an update method that combines fixed-cycle iteration with real-time triggering. Under normal circumstances, the intervention strategy is optimized once every fixed cycle. If special circumstances occur, such as a sudden drop in the user's execution rate, rapid changes in physical condition, or abnormal fluctuations in the prevention and control effect, the system will immediately skip the normal cycle and start adjusting the scheduling strategy in real time.

[0041] In this embodiment of the invention, the behavior correction module receives an individualized intervention plan pushed by the intervention plan module, collects behavioral data such as user eye usage time, eye distance, and eye posture through smart devices, uses AI algorithms to predict possible bad eye behaviors of users, issues reminders in advance, and explains the harm of the behavior and its impact on physical constitution in conjunction with traditional Chinese medicine theory. The AI ​​algorithm is constructed using the LightGBM gradient boosting tree model. The input features include 16 dimensions of real-time behavioral and environmental data such as eye use duration, eye use distance, eye use posture, ambient light, color temperature, continuous eye use interval, and outdoor activity duration. Core predictive factors are selected by ranking features by importance. The model training set contains 600,000 labeled adolescent eye use behavior data. Parameters are optimized using 5-fold cross-validation. The system automatically triggers a warning when the probability of poor eye use behavior is ≥0.82. At the same time, it combines the theory of traditional Chinese medicine constitution to link poor eye use behavior with pathogenesis such as liver and kidney deficiency and qi and blood depletion, and explains the harm of the behavior in layman's terms.

[0042] The system clearly defines the rules for judging poor eye use behavior. Continuous use of eyes for more than the prescribed time, using eyes at too close a distance, using eyes with a distorted posture, using eyes in excessively strong or dim light, and looking at electronic screens at close range for a long time are all judged as poor eye use behavior. The system collects behavioral data in real time and compares it with the judgment rules to predict the poor behavior that is about to occur, and issues timely reminders and informs you of the corrective methods.

[0043] Based on users' TCM constitution, myopia syndrome type, and real-time behavioral data, we provide individualized dietary conditioning suggestions. For users with liver and kidney deficiency, we recommend foods such as goji berries and yams. For users with qi and blood deficiency, we recommend foods such as red dates and longan. We also recommend traditional Chinese medicine exercises such as Baduanjin and Tai Chi. We set personalized exercise duration and difficulty based on users' physical condition. The system automatically matches the exercise version of Baduanjin and Tai Chi to the user's age, physical condition and fitness level. Simplified movements are recommended for teenagers and low-intensity, gentle movements are recommended for people with weak constitutions. At the same time, the system dynamically adjusts the exercise duration and difficulty based on the user's daily exercise completion.

[0044] The diet and exercise plan will be automatically adjusted according to the changes in the user's TCM syndrome, improvement in physical condition, and progress of the intervention cycle. When the syndrome changes, the appropriate food and exercise type will be changed simultaneously. After the physical condition improves, the exercise intensity will be gradually increased to ensure that the conditioning plan is always consistent with the user's constitution and physical condition.

[0045] It has the function of analyzing the correlation between lifestyle habits and TCM constitution, and regularly generates habit correction reports, which are simultaneously pushed to the remote supervision module and the effect evaluation module.

[0046] The system delves into the intrinsic connection between eye-use behavior and TCM constitution, focusing on four core dimensions: prolonged close-range eye use depletes Qi and blood, staying up late damages liver and kidney essence and blood, prolonged sitting and lack of movement affects spleen and stomach function, and poor posture leads to meridian stagnation. It establishes a clear correspondence between each type of poor eye-use behavior and the corresponding constitution imbalance and myopia pathogenesis; the correction report clearly marks the specific impact of the behavior on constitution and myopia development.

[0047] In this embodiment of the invention, the remote supervision module establishes a remote communication platform that links users, TCM doctors, parents, and school teachers. It receives user behavior data pushed by the behavior correction module and individualized intervention plans pushed by the intervention plan module. It supports users to upload photos, videos, and logs of the intervention implementation. The system automatically verifies the standardization of the intervention implementation, including the standardization of massage movements and the frequency of drinking Chinese herbal teas, and feeds back any abnormalities to the TCM doctor. The standardization check of intervention execution covers five dimensions: operation duration, operation frequency, operation posture, ingredients used, and acupoint location. The system compares the user's execution data against the standardized intervention plan. Insufficient massage duration, insufficient frequency of ear acupoint pressing, acupoint location deviation, irregular tea drinking, and excessive fumigation temperature are all judged as non-standard execution. The system automatically records abnormal data and simultaneously pushes it to TCM doctors for guidance and correction.

[0048] Traditional Chinese medicine practitioners can view user intervention records, four diagnostic methods data, and risk assessment results online, provide guidance and correction for user operations, adjust intervention plans when user symptoms or physical condition change, and synchronize them to the intervention plan module and behavior correction module. Parents can view their child's intervention implementation and prevention progress in real time, receive reminder messages, and assist in supervising intervention implementation. The remote communication platform adopts a hierarchical access control mechanism. Users can view their personal intervention plans, implementation records, and effect reports. Traditional Chinese medicine practitioners have full access to view data and modify plans. Parents can view their children's intervention implementation data and reminder information. School teachers can view myopia prevention and control statistics for the student population. The system establishes an intervention implementation incentive mechanism. Users can earn points by completing daily intervention tasks. These points can be redeemed for rewards such as personalized intervention guidance and health education materials, continuously improving users' enthusiasm for intervention implementation.

[0049] It supports school teachers in viewing myopia prevention and control data of student groups, conducting centralized education on common problems, establishing an incentive mechanism for intervention implementation, and improving user compliance through points, badges, and other means. Relevant supervision data is simultaneously transmitted to the effect evaluation module.

[0050] The home-school collaborative supervision adopts a full-scenario data interconnection mode. Parents receive daily intervention task reminders, home execution data, and changes in eye condition in real time, assisting in home intervention supervision and operation guidance. Teachers obtain statistics on the overall eye use behavior of students in the class, on-campus eye care data, and the distribution of myopia risk in the group, and conduct collective eye care education and on-site behavior correction at school. Data from both home and school ends are synchronized to the system platform in real time, and the incentive mechanism is interconnected across terminals.

[0051] In this embodiment of the invention, the effect evaluation module establishes a comprehensive evaluation system that includes four dimensions: improvement of TCM syndrome, change of Western medicine psychological indicators, improvement of quality of life and intervention compliance. It receives intervention execution data pushed by the remote supervision module, TCM syndrome data pushed by the four diagnostic methods module, initial risk data pushed by the risk assessment module, and behavioral data pushed by the behavior correction module. The TCM syndrome improvement dimension uses the decrease in syndrome score and the relief of syndrome bias as quantitative indicators; the Western medicine pathological indicator change dimension uses the improvement in visual acuity, the decrease in refractive error, and the slowdown in the rate of axial length growth as quantitative indicators; the quality of life improvement dimension uses the reduction in the impact of myopia on learning and life and the relief of eye discomfort symptoms as quantitative indicators; and the intervention compliance dimension uses the daily intervention task completion rate and the completeness of the program implementation as quantitative indicators.

[0052] Based on the data from the four diagnostic methods and AI-based diagnosis submitted by users on a regular basis, the changes in syndrome scores are calculated and the trend of syndrome improvement is analyzed. The user's visual acuity, refractive error, axial length and other Western medical indicators are tracked. The correlation between the intervention effect and the changes in indicators is judged by combining the data from traditional Chinese and Western medicine. The impact of myopia on the user's learning, life and psychological state is assessed by using a standardized scale. The user's intervention execution record is combined to quantitatively assess compliance and its impact on the intervention effect. The correlation analysis of TCM and Western medicine data focuses on the correspondence between the improvement of TCM syndromes and the changes in Western medicine physiological indicators. It focuses on analyzing the synchronous changes in the rate of axial length growth and the degree of visual stability after the relief of physical imbalance, and clarifies the positive influence path of TCM intervention on Western medicine ophthalmological indicators.

[0053] Formula for calculating the improvement rate of TCM syndromes: This indicates the rate of improvement in TCM syndromes during the intervention period, directly reflecting the conditioning effect of TCM intervention on constitution and pathogenesis; The initial symptom score before intervention is implemented is a basic reference value for evaluating the effectiveness of the intervention; This represents the current symptom score after the intervention, used to compare and calculate the actual improvement effect brought about by the intervention.

[0054] The system uses visualization technology to display the curves of intervention effect changes and the evaluation results of various dimensions, automatically generates evaluation reports, clarifies the advantages and disadvantages of the program, identifies the causes of poor intervention effect through AI algorithms and pushes targeted optimization suggestions, and the evaluation results are simultaneously fed back to the intervention program module and the intelligent education module.

[0055] The AI ​​algorithm is built on a multi-feature decision tree and random forest fusion model. The input features cover 18 dimensions of data, including changes in syndrome scores, visual acuity, refractive error, axial length, intervention execution rate, dietary and exercise compliance, and work-rest compliance rate. The threshold for judging poor effect is that the myopia prevention and control effect improvement rate is <5% within the intervention period. The model classifies and outputs five core causes: insufficient intervention execution, misjudgment of TCM syndrome type, inappropriate intervention plan, rebound of bad eye behavior, and no improvement in physical condition. Based on the cause type, personalized optimization suggestions are automatically matched.

[0056] The visualization system uses line graphs to show the cyclical trends of TCM syndrome scores, visual acuity, refractive error, and axial length; bar charts to compare the degree of improvement in each dimension before and after intervention; and radar charts to present the comprehensive evaluation results of TCM syndromes, physiological indicators, quality of life, and intervention compliance. All charts can be viewed by day, week, or month, and the data is updated in real time.

[0057] In this embodiment of the invention, the intelligent education module constructs a knowledge graph of TCM myopia prevention and control, which includes the TCM etiology and pathogenesis of myopia, syndrome differentiation, intervention methods, prevention and health care, and common misconceptions. It incorporates a TCM myopia prevention and control clinical case library, updates research results and clinical experience in real time, and receives evaluation results pushed by the effect evaluation module and TCM syndrome data pushed by the four diagnostic methods module. The system combines user health profiles and personalized intervention plans pushed by the intervention plan module. It also pushes personalized educational content through natural language processing technology. For users with liver and kidney deficiency, the system focuses on pushing knowledge on liver and kidney conditioning. For teenagers, the system focuses on pushing knowledge on developing good eye habits. It supports intelligent question and answer via voice or text, and provides personalized answers by retrieving information through knowledge graphs and combining it with the user's specific situation. The user health profile includes TCM syndrome type, myopia risk level, age, lifestyle habits, etc. The push algorithm is built upon a fusion strategy of collaborative filtering and knowledge graph rule matching. It constructs a user health profile based on their TCM syndrome type, myopia risk level, age, physical condition, and intervention plan type. The algorithm then uses the knowledge graph to correlate the matching degree between educational content and the user profile, with a content matching degree ≥ 0.88 serving as the push threshold. Simultaneously, the push strategy is dynamically adjusted based on the user's mastery of the educational content: core knowledge is pushed once daily when mastery is < 60%, once every two days when mastery is 60%-89%, and advanced knowledge twice a week when mastery is ≥ 90%. The intelligent question-answering module quickly retrieves TCM-related knowledge on myopia prevention and control through knowledge graph entity links and semantic similarity calculations, and outputs answers based on the user's individual circumstances.

[0058] After a user enters a text or voice question, the system first performs semantic analysis and keyword extraction on the question, then matches it with the corresponding knowledge points in the TCM myopia prevention and control knowledge graph, and combines the user's individual information such as TCM syndrome type and myopia risk level to generate an easy-to-understand personalized answer; questions that cannot be answered directly are automatically recorded and pushed to the TCM doctor's backend, where professional doctors will supplement the reply and then synchronize it with the user.

[0059] By using questionnaires and interactive Q&A to understand users' grasp of the educational content, we can dynamically adjust the educational content and push frequency to help users and related personnel master the knowledge of TCM myopia prevention and control, and assist in improving the overall prevention and control effect. The educational data is synchronously fed back to the behavior correction module and intervention program module.

[0060] The TCM myopia prevention and control knowledge graph takes TCM syndrome types, intervention methods, prevention and health care, and common misconceptions as core entities, establishes the relationship between entities, and incorporates data resources such as clinical cases, research results, and expert experience. The knowledge graph adopts a regular update mechanism, synchronizing the latest TCM myopia prevention and control clinical research and technological achievements every month.

[0061] Users' knowledge mastery is comprehensively evaluated based on three aspects: answer accuracy, interactive Q&A matching degree, and compliance of practical behavior. The system classifies the mastery level according to the evaluation results and matches differentiated education content to different levels to avoid repeatedly pushing knowledge that has already been mastered, and to focus on strengthening popular science guidance for weak areas.

[0062] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0063] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A myopia prevention and control health management system integrating traditional Chinese medicine intervention strategies, characterized in that, It includes modules for diagnosis based on four diagnostic methods, risk assessment, intervention plan, behavior correction, remote supervision, effect evaluation, and intelligent education. The four diagnostic methods module standardizes and quantifies the data collected from the four diagnostic methods of traditional Chinese medicine (inspection, auscultation and olfaction, inquiry and palpation), and uses an AI diagnostic model to integrate and analyze the data to automatically generate TCM syndrome types and syndrome scores. The risk assessment module constructs a multimodal dynamic fusion model based on deep learning, which integrates data related to traditional Chinese medicine, data on Western medical psychology indicators, data on user eye behavior, and data related to the environment. It outputs a comprehensive myopia risk level, a prediction of development trends, and an analysis of risk factors, clarifying the correlation between TCM syndrome types and myopia risk. The intervention program module is based on the TCM syndrome differentiation and treatment theory and clinical knowledge base, combined with the myopia risk assessment results, to automatically generate individualized comprehensive TCM intervention programs and verify the compliance of program implementation; The behavior correction module collects user eye-related behavior data, uses algorithms to predict poor eye-use behaviors and issue reminders, combines the user's TCM constitution and syndrome type to provide individualized dietary and exercise conditioning suggestions, and generates a habit correction report. The remote supervision module establishes a multi-party collaborative remote communication platform to verify and guide the implementation of TCM myopia intervention for users, dynamically adjust and optimize intervention plans, and link multiple stakeholders to participate in intervention supervision. The performance evaluation module establishes a multi-dimensional comprehensive evaluation system, dynamically evaluates relevant data, and generates visual reports and optimization suggestions. The intelligent education module constructs a knowledge graph of myopia prevention and control using traditional Chinese medicine, pushes personalized health education content to users, provides intelligent question-and-answer services, and dynamically adjusts education strategies based on the user's understanding.

2. The myopia prevention and control health management system integrating traditional Chinese medicine intervention strategies according to claim 1, characterized in that, The four diagnostic methods module integrates a multispectral tongue image acquisition device, a pulse pressure sensor array, a voiceprint acquisition microphone, and a standardized consultation interface. The tongue image acquisition uses multispectral imaging technology to automatically identify the characteristics of the tongue body, tongue coating, tongue shape, and tongue color. The pulse image acquisition uses a pressure sensor array to obtain the waveform, frequency, and rhythm parameters of the pulse at the cun, guan, and chi positions. The voiceprint acquisition analyzes the strength and clarity of the user's voice. The consultation interface is based on the TCM myopia diagnosis scale to guide users to complete a standardized symptom self-assessment.

3. The myopia prevention and control health management system integrating traditional Chinese medicine intervention strategies according to claim 2, characterized in that, The four diagnostic methods module extracts features from each of the four diagnostic dimensions (inspection, auscultation and olfaction, inquiry, and palpation) using specialized feature extraction technology. These features are then input into the AI ​​diagnostic model to complete multi-dimensional data fusion analysis. The module also integrates an AI follow-up questioning function, which supplements targeted diagnostic questions based on the user's initial response. The collected raw data is then converted into structured TCM syndrome data. The AI ​​diagnostic model performs fusion analysis on the four diagnostic data, automatically generating TCM syndrome types and syndrome scores, which are then simultaneously transmitted to the risk assessment module.

4. The myopia prevention and control health management system integrating traditional Chinese medicine intervention strategies according to claim 3, characterized in that, The risk assessment module constructs a multimodal dynamic fusion model based on deep learning, which integrates the TCM syndrome differentiation results and syndrome scores transmitted by the four diagnostic methods module with Western medical psychological indicators, behavioral habit data and environmental data. The architecture adopts a layered fusion structure. The bottom layer is designed with a special feature extraction algorithm for TCM syndrome data to complete the extraction of modal features. The middle layer realizes cross-modal association and dynamic weight allocation of features of each modality, and adjusts the weight of each data in real time according to the user's age, disease course and physical condition. The top-level output integrates myopia risk level, development trend prediction and risk factor analysis, combined with TCM syndrome differentiation to identify myopia-related pathogenesis, clarify the correlation between each syndrome and myopia risk, and push the assessment results to the intervention plan module simultaneously.

5. The myopia prevention and control health management system integrating traditional Chinese medicine intervention strategies according to claim 4, characterized in that, The intervention program module is based on the TCM syndrome differentiation and treatment theory and clinical knowledge base. Combined with the myopia risk level, TCM syndrome type and risk factor analysis pushed by the risk assessment module, it automatically generates an individualized comprehensive intervention program that includes multiple TCM intervention methods. Based on the syndrome differentiation, Chinese herbal teas or medicinal diets are formulated and the ingredients are dynamically adjusted according to the user's dietary preferences; massage guidance is provided for acupoints including Jingming, Zanzhu, and Fengchi, and acupoints are located using AR technology; the selection of acupoints and operation methods for ear acupressure are clearly defined; Based on the theory of meridian flow in traditional Chinese medicine, a daily routine adjustment plan is designed and adapted to the user's daily routine to achieve individualized adaptation; a knowledge graph technology is used to build a knowledge base for myopia intervention in traditional Chinese medicine, which supports dynamic updates and optimization of the intervention plan.

6. The myopia prevention and control health management system integrating traditional Chinese medicine intervention strategies according to claim 5, characterized in that, The intervention plan module is configured with an intervention method coordination scheduling algorithm, which automatically adjusts the execution order and duration of each intervention method based on the user's intervention execution status and effect feedback; it performs compliance verification on the intervention plan, and pushes the generated plan to the behavior correction module and the remote supervision module simultaneously.

7. The myopia prevention and control health management system integrating traditional Chinese medicine intervention strategies according to claim 6, characterized in that, The behavior correction module receives individualized intervention plans pushed by the intervention plan module, collects behavioral data including user eye usage time, eye distance, and eye posture through smart devices, uses AI algorithms to predict possible bad eye behaviors of users, issues reminders in advance, and explains the harm of behaviors and their impact on the constitution in combination with the theory of myopia constitution in traditional Chinese medicine. Based on the user's TCM constitution, myopia pattern, and real-time behavioral data, it provides individualized dietary conditioning suggestions and sets personalized exercise duration and difficulty based on the user's physical condition; it has a function to analyze the correlation between lifestyle habits and TCM constitution, and regularly generates habit correction reports, which are simultaneously pushed to the remote supervision module and the effect evaluation module.

8. The myopia prevention and control health management system integrating traditional Chinese medicine intervention strategies according to claim 7, characterized in that, The remote supervision module establishes a remote communication platform that connects users, TCM doctors, parents, and school teachers. It receives user behavior data pushed by the behavior correction module and individualized intervention plans pushed by the intervention plan module. It supports users to upload photos, videos, and logs of intervention implementation. The system automatically verifies the standardization of intervention implementation from dimensions including operation duration, posture, and frequency, and feeds back any abnormalities to TCM doctors. Traditional Chinese medicine practitioners can view user intervention records, four diagnostic methods data, and risk assessment results online, provide guidance and correction for user operations, and adjust the intervention plan when user symptoms or physical condition change and synchronize it to the intervention plan module and behavior correction module. It allows parents and school teachers to view intervention data according to their respective permissions, establishes an incentive mechanism for intervention implementation, and synchronously transmits relevant supervision data to the effect evaluation module.

9. The myopia prevention and control health management system integrating traditional Chinese medicine intervention strategies according to claim 8, characterized in that, The effect evaluation module establishes a comprehensive evaluation system that includes four dimensions: improvement of TCM syndrome, changes in Western medicine psychological indicators, improvement in quality of life, and intervention compliance. It receives intervention execution data pushed by the remote supervision module, TCM syndrome data pushed by the four diagnostic methods module, initial myopia risk data pushed by the risk assessment module, and behavioral data pushed by the behavior correction module. Based on the four diagnostic methods data submitted by users regularly and the results of AI diagnosis, the changes in syndrome scores are calculated and the trend of syndrome improvement is analyzed; Western medicine physiological indicators, including users' visual acuity, refractive error, and axial length, are tracked, and the correlation between the intervention effect and the changes in indicators is judged by combining the data of traditional Chinese and Western medicine; the impact of myopia on users' learning, life and psychological state is assessed through standardized scales, and intervention compliance is quantitatively assessed by combining users' intervention implementation records. Visualization technology is used to display the curves of intervention effect changes and the evaluation results of each dimension. Evaluation reports are automatically generated and targeted optimization suggestions are pushed. The evaluation results are simultaneously fed back to the intervention plan module and the intelligent education module.

10. The myopia prevention and control health management system integrating traditional Chinese medicine intervention strategies according to claim 9, characterized in that, The intelligent education module constructs a TCM myopia prevention and control knowledge graph that includes the TCM etiology and pathogenesis of myopia, syndrome differentiation, intervention methods, prevention and health care, and common misconceptions. It also incorporates a TCM myopia prevention and control clinical case database and updates research results in real time. It receives evaluation results pushed by the effect evaluation module and TCM syndrome data pushed by the four diagnostic methods module, combines user health profiles and individualized intervention plans pushed by the intervention plan module, and pushes individualized education content; it supports intelligent question and answer via voice or text, retrieves information through knowledge graphs and provides individualized answers based on the user's specific situation; We use questionnaires and interactive Q&A to understand users' grasp of the educational content, dynamically adjust the educational content and push frequency, and synchronously feed the educational data back to the behavior correction module and intervention program module.

Images