Intelligent health care method and device fusing medical state recognition

Abstract

The application discloses an intelligent syndrome differentiation reasoning method and device fusing a medical body-use law, and belongs to the technical field of intelligent syndrome differentiation and wisdom medical treatment of traditional Chinese medicine. The application aims at the industry pain points that the existing intelligent syndrome differentiation technology of traditional Chinese medicine is separated from the original theory of traditional Chinese medicine, the reasoning logic is mechanized, the complex pathogenesis recognition capability is weak, and dynamic syndrome differentiation is absent, constructs a three-dimensional integrated syndrome differentiation rule library based on the core body-use syndrome differentiation law of traditional Chinese medicine, collects standardized multi-dimensional diagnosis and treatment data, extracts three-dimensional features of damage of viscera ontology, abnormal function and dynamic conversion of pathogenesis in layers, adopts a multi-level reasoning mechanism of layered progression and linkage correction, realizes full-dimensional syndrome differentiation reasoning from the ontology pathogenesis and functional syndrome type to dynamic conversion, and completes result verification and optimization through the core logic of traditional Chinese medicine.

Description

Technical Field

[0001] This invention relates to the fields of intelligent health care, traditional Chinese medicine constitution identification, and artificial intelligence. Specifically, it relates to an intelligent health care method and device that integrates ten medical states identification, which is suitable for accurate assessment of the physical condition of middle-aged and elderly people and sub-healthy people, intelligent generation of personalized health care plans, and closed-loop health care intervention scenarios. Background Technology

[0002] With the aging population and the upgrading of national health needs, the health and wellness industry is gradually developing towards intelligence, precision, and personalization. Traditional health and wellness models mostly adopt general health and wellness programs, carrying out health conditioning, rest guidance, dietary matching, and exercise intervention according to uniform standards. They lack precise identification of individual differences and dynamic states of physical condition, resulting in problems such as weak intervention targeting, low adaptability, and poor conditioning effects.

[0003] Currently, most mainstream health and wellness technologies combine basic constitution identification, age, and symptoms as single-dimensional data to formulate health and wellness plans, resulting in limited identification dimensions and insufficient assessment precision. Traditional Chinese medicine constitution identification often uses a nine-constitution classification system, which only achieves basic constitution classification and cannot cover the comprehensive dynamic changes in the body's Qi and blood, Yin and Yang, internal organs, cold and heat, deficiency and excess, making it difficult to accurately reflect the body's real-time health and wellness needs. The Medical Ten-State Identification, as a new precision medical identification system, covers ten core physiological and pathological states of the human body: balanced, deficient, excessive, cold, hot, damp, dry, stagnant, stagnant, and toxic. It can comprehensively and meticulously depict the body's constitution and health imbalances, overcoming the shortcomings of traditional constitution identification in terms of incomplete dimensions and insufficient precision.

[0004] Existing intelligent health and wellness systems generally face technological barriers: First, they fail to deeply integrate the medical ten-state identification system with intelligent health and wellness technology, making it impossible to achieve a systematic and hierarchical assessment of the human body's multidimensional health status; second, the generation of health and wellness plans relies heavily on fixed algorithm models, making it impossible to iterate and optimize in real time based on the dynamic changes in the ten-state identification results, and thus difficult to adapt to the dynamic evolution of human constitution; third, they lack real-time monitoring, feedback, and closed-loop control mechanisms for the effects of health and wellness interventions, resulting in a crude intervention process, unquantifiable effects, and the inability to continuously optimize.

[0005] In summary, existing intelligent healthcare technologies suffer from technical deficiencies such as limited identification dimensions, low accuracy, insufficient personalization of solutions, and lack of closed-loop control mechanisms. These deficiencies fail to meet the current demands for precise, dynamic, and intelligent healthcare. Therefore, there is an urgent need for an intelligent healthcare technology solution that integrates ten medical state identification methods to address these issues. Summary of the Invention

[0006] Purpose of the invention

[0007] The purpose of this invention is to address the shortcomings of existing technologies by providing an intelligent health care method and device that integrates medical ten-state identification. This addresses the problems of low accuracy in physical constitution identification, incomplete dimensions, strong universality of health care plans, lack of personalization, and absence of dynamic closed-loop regulation in existing intelligent health care technologies. It achieves accurate identification based on the ten states of human health, intelligent generation of personalized health care plans, dynamic intervention, and closed-loop optimization of effects, significantly improving the accuracy, adaptability, and effectiveness of intelligent health care.

[0008] Technical solution

[0009] To achieve the above objectives, the first aspect of this invention provides an intelligent health and wellness method integrating ten medical states of identification, comprising the following steps:

[0010] S1. Multi-source health data collection: Through smart terminals, wearable devices, and manual input ports, basic vital signs data, lifestyle behavior data, physical fitness questionnaire data, medical history data, and organ status data of users are collected to construct a multi-dimensional health dataset for users. Among them, basic vital signs data include heart rate, blood pressure, blood oxygen, body temperature, sleep duration, and number of steps; lifestyle behavior data includes dietary structure, work and rest patterns, exercise habits, and emotional state; physical fitness questionnaire data is special survey data adapted to the ten states of medical identification; medical history data includes past medical history, chronic disease status, and allergy history; organ status data includes data on the functional imbalance of the five internal organs and six bowels.

[0011] S2. Medical Ten-State Intelligent Identification: Construct a medical ten-state identification algorithm model. Input a multi-dimensional health dataset into the identification algorithm model, and quantify and score ten health states of the user: balanced state, empty state, full state, cold state, hot state, damp state, dry state, stagnant state, depressed state, and toxic state. Generate a quantitative health report of the user in ten states. Specifically, it includes five sub-steps: data preprocessing, feature extraction, state weight assignment, quantitative scoring, and abnormal state marking, so as to achieve accurate positioning and hierarchical classification of the human body's health imbalance state.

[0012] S3. Personalized Health and Wellness Plan Intelligent Matching: A pre-set ten-state health and wellness knowledge base is set up, which stores dietary health and wellness rules, exercise health and wellness rules, work and rest regulation rules, emotional guidance rules, meridian regulation rules, and contraindications corresponding to the ten health states. Based on the scores of each state, abnormal state types, and degree of imbalance in the user's ten-state health quantitative report, a personalized health and wellness plan that is adapted to the user's individual state is generated through big data matching and intelligent algorithm deduction.

[0013] S4. Health and Wellness Intervention and Dynamic Monitoring: Push personalized health and wellness plans to user terminals to guide users in carrying out daily health and wellness interventions; collect users' vital signs, behavioral data and status feedback data in real time during the health and wellness process, and dynamically track the changing trends of users' ten states of health status.

[0014] S5. Closed-loop optimization and control: Regularly review the user's ten-state health status, compare the results of previous identifications, and quantitatively evaluate the effect of health care intervention; dynamically adjust the intervention intensity, content ratio, and conditioning focus of the health care plan according to the improvement, stagnation, or deterioration of the health status, and realize closed-loop intelligent health care control of "identification-intervention-monitoring-review-optimization".

[0015] Furthermore, the medical ten-state identification algorithm model described in step S2 is a multi-dimensional fusion identification model that integrates random forest, neural network and fuzzy comprehensive evaluation. It is trained and optimized through massive clinical ten-state identification sample data, and sets a quantitative scoring range of 0-100 points for each of the ten health states. Based on the scoring results, it is divided into four levels: normal health, mild imbalance, moderate imbalance and severe imbalance.

[0016] Furthermore, the ten-state health and wellness knowledge base mentioned in step S3 adopts a hierarchical and classified storage structure. It presets differentiated health and wellness intervention strategies for single-state imbalance, multi-state compound imbalance, and imbalance scenarios of varying degrees of severity. It supports intelligent weight allocation and scheme combination under the condition of multiple state superposition, avoiding the conflict problem of different health and wellness intervention measures.

[0017] Furthermore, the specific logic of closed-loop optimization and control in step S5 is as follows: when the user's abnormal state score improves and the imbalance level decreases, maintain the existing health care plan and gradually reduce the intensity of intervention; when the user's health status does not change significantly, optimize the content ratio and execution frequency of the health care plan; when the user's abnormal state worsens and the imbalance level increases, upgrade the intervention plan, add special conditioning measures and push health warning prompts.

[0018] The second aspect of the present invention provides an intelligent health care device that integrates medical ten-state identification to implement the above-mentioned intelligent health care method, including a data acquisition module, a data processing module, a ten-state identification module, a health care plan generation module, an interactive push module, a dynamic monitoring module, a closed-loop optimization module, and a storage module.

[0019] The data acquisition module is used to collect multidimensional health data from users, including a vital signs acquisition unit, a behavior acquisition unit, and an information entry unit. The vital signs acquisition unit connects to wearable smart devices to acquire users' physiological vital signs data in real time; the behavior acquisition unit is used to collect data on users' daily diet, exercise, and rest behavior; and the information entry unit supports manual entry of medical history, questionnaires, and special health status data.

[0020] The data processing module communicates with the data acquisition module and is used to clean, deduplicate, and normalize the collected raw health data, remove abnormal data, fill in missing data, standardize the data format, and generate a standard health data sequence that can be used for model identification.

[0021] The ten-state identification module communicates with the data processing module and has a built-in trained medical ten-state identification algorithm model. It is used to receive standard health data sequences, complete the quantitative scoring and level classification of ten health states of users, and generate a standardized ten-state health assessment report.

[0022] The health and wellness plan generation module communicates with the ten-state identification module and the storage module respectively. It is used to call the ten-state health and wellness knowledge base in the storage module and intelligently match, combine and generate personalized health and wellness plans based on the type, quantity and level of imbalance in the ten-state health assessment report.

[0023] The interactive push module communicates with the health and wellness plan generation module and includes a user terminal interaction unit, which is used to push health and wellness plans, health reports, intervention reminders and early warning information to users, while receiving feedback from users on health and wellness implementation and status.

[0024] The dynamic monitoring module communicates with the data acquisition module and the ten-state identification module respectively. It is used to track the changes in health data during the user's health care intervention process in real time, dynamically monitor the evolution trend of ten health states, and record health care implementation data and effect data.

[0025] The closed-loop optimization module communicates with the dynamic monitoring module and the health and wellness plan generation module respectively. It is used to periodically review the user's health status, quantify the effect of health and wellness intervention, and dynamically iterate and optimize the health and wellness plan based on the monitoring and review results to achieve closed-loop intelligent regulation.

[0026] The storage module is used to store user health data, parameters of the ten-state identification model, the ten-state health and wellness knowledge base, historical health and wellness programs and intervention effect data, providing data support for model iteration and program optimization.

[0027] Beneficial effects

[0028] Compared with the prior art, the present invention has the following significant advantages:

[0029] 1. This invention is the first to deeply integrate the medical ten-state identification system with intelligent health care technology, breaking through the technical bottleneck of the traditional constitution identification which has a single dimension and insufficient precision. Through the comprehensive quantitative assessment of ten core human body states, namely balance, deficiency, excess, cold, heat, dampness, dryness, stagnation, depression and toxicity, it accurately covers various health care scenarios of physiological, pathological, emotional and organ imbalance, and realizes the refined and systematic identification of health status, providing reliable data support for precision health care.

[0030] 2. This invention adopts multi-source health data fusion collection and AI intelligent identification algorithm, combined with big data health and wellness knowledge base, which can intelligently generate personalized and differentiated health and wellness plans for users' single or multiple health imbalances, completely changing the shortcomings of the generalization and homogenization of traditional health and wellness plans, and greatly improving the pertinence and adaptability of health and wellness intervention.

[0031] 3. This invention constructs a full-process intelligent health and wellness system consisting of "data collection - ten-state identification - scheme generation - dynamic monitoring - closed-loop optimization". It can track the dynamic changes in the user's health status in real time, continuously iterate and optimize the scheme based on the effect of health and wellness intervention, realize the dynamic and precise control of health and wellness intervention, and solve the problems of traditional static health and wellness intervention, which cannot quantify the effect and cannot be continuously optimized.

[0032] 4. The device of this invention has a clear modular design and rigorous logic, and can be adapted to various smart terminals and wearable devices. It has strong compatibility and convenient deployment, and can be widely used in multiple scenarios such as community health care, home-based elderly care, sub-health conditioning, and chronic disease health care. It has a high degree of intelligence, strong practicality, and extremely high industrialization and promotion value. Detailed Implementation

[0033] The present invention will be further described in detail below with reference to specific embodiments. These embodiments are implemented based on the technical solution of the present invention, providing detailed implementation methods and specific operating procedures. However, the scope of protection of the present invention is not limited to the following embodiments.

[0034] Example 1: Implementation of Smart Rehabilitation for Users with Mild Hydration and Depression

[0035] This embodiment targets a 38-year-old sub-healthy adult user, employing the intelligent health care method and device integrating ten medical states identification as described in this invention to conduct intelligent health care intervention. The specific steps are as follows:

[0036] The first step is to collect multi-source health data. Wearable wristbands are used to collect users' heart rate, blood pressure, sleep, and exercise data; mobile questionnaires are used to collect users' behavioral data such as a diet high in oil, sedentary lifestyle, emotional depression, and irregular work and rest, as well as physical manifestation data such as physical fatigue and emotional distress, to build a multi-dimensional health dataset for users.

[0037] The second step is the intelligent identification of ten medical states. The pre-processed standard health data is input into the ten-state identification algorithm model to complete the quantitative scoring of ten states. The identification results are: 72 points for the damp state (mild imbalance), 68 points for the stagnation state (mild imbalance), and the remaining states are all above 90 points (normal health). A ten-state health report is generated for the user, which is marked as a mild imbalance state of dampness and stagnation.

[0038] The third step is the generation of personalized health and wellness plans. This involves accessing the Ten-State Health and Wellness Knowledge Base to match plans for mild dampness imbalance (e.g., strengthening the spleen and removing dampness, a light diet, moderate aerobic exercise, and meridian regulation); and for mild depression imbalance (e.g., emotional regulation, liver soothing, gentle exercise, and regular sleep patterns). The plan avoids contraindications such as cold, oily, and raw foods, prolonged sitting, and emotional suppression, intelligently combining these elements to generate a unique health and wellness plan. Specific plans include: a light diet daily, increased intake of dampness-removing foods; 30 minutes of jogging or yoga daily; going to bed before 11 PM every night; 15 minutes of daily emotional regulation meditation; and gentle liver and spleen meridian massage twice a week.

[0039] The fourth step is dynamic health and wellness intervention and monitoring. Health and wellness plans and daily reminders are pushed to users via mobile devices. Data such as sleep duration, steps taken, and sleep patterns are monitored in real time, recording the user's daily health and wellness implementation and simultaneously tracking feedback on the user's physical and emotional state.

[0040] The fifth step is closed-loop optimization and regulation. After 30 days of continuous intervention, the user's health was reassessed using the ten-state health system. The reassessment results showed a score of 85 for the dampness state and 82 for the depression state, indicating a significant improvement in the imbalance. The system determined that the health intervention was effective and maintained the existing health plan, appropriately reducing the intensity of exercise and the frequency of meridian conditioning to further consolidate the health effects. At the same time, the user's health status was continuously and dynamically monitored to achieve long-term stable health conditioning.

[0041] Example 2: Implementation of intelligent health care for middle-aged and elderly users with moderate imbalances in deficiency and stagnation states.

[0042] This embodiment is designed for middle-aged and elderly users with chronic diseases aged 62 and above. The specific implementation process is as follows:

[0043] First, we collected multidimensional health data from users, including basic physical characteristics, chronic diseases in the elderly, daily routines and diet, limb condition, and Qi and blood status, and completed data preprocessing and standardization. Using a ten-state identification model, we quantified the scores, determining that the user had a deficiency state of 55 points (moderate imbalance) and a stagnation state of 52 points (moderate imbalance), indicating core health issues of Qi and blood deficiency and meridian blockage.

[0044] Secondly, based on the characteristics of moderate deficiency and stagnation imbalance, the system matches a health and wellness program suitable for the physical condition of middle-aged and elderly people, abandoning high-intensity intervention measures and adopting a gentle conditioning mode: in terms of diet, it focuses on the combination of ingredients that replenish qi and blood and promote blood circulation and remove blood stasis; in terms of exercise, it is suitable for gentle aerobic exercises such as Tai Chi and slow walking, with 20 minutes of exercise per day; in terms of work and rest, it strengthens early to bed and early to rise and midday rest; it adds special health and wellness measures for meridian dredging and qi and blood conditioning, and avoids taboos such as fatigue, cold, and high-fat diet.

[0045] Finally, long-term dynamic intervention and monthly review were carried out. After the first month of intervention, the user's scores for deficiency and stagnation improved to 68 and 65 points respectively, and the moderate imbalance turned into mild imbalance. The system automatically optimized the plan, gradually adjusted the conditioning ratio, and continuously strengthened the intervention of qi and blood conditioning and meridian unblocking. After 3 months of closed-loop optimization intervention, the user's two imbalance states basically returned to normal health, and the health care effect was significant.

[0046] Device Operation and Implementation Instructions

[0047] When the intelligent health and wellness device of this invention is in operation, the various modules work collaboratively and in a closed-loop manner: the data acquisition module completes multi-dimensional health data collection in real time, the data processing module completes data standardization preprocessing, the ten-state identification module achieves accurate ten-state assessment based on the trained AI model, the health and wellness plan generation module intelligently matches personalized plans based on the knowledge base, the interactive push module completes information interaction with the user, the dynamic monitoring module tracks the health and wellness process data throughout the process, the closed-loop optimization module continuously iterates the plan based on the intervention effect, and the storage module retains various types of data throughout the process, ensuring the automated, intelligent, and precise operation of the entire intelligent health and wellness process.

Claims

1. An intelligent dialectical reasoning method integrating the principles of medical application, characterized in that, Includes the following steps: S1. Construct a TCM body-function-transformation syndrome differentiation knowledge system and build a three-dimensional syndrome differentiation rule base; the three-dimensional syndrome differentiation rule base includes body layer rules, function layer rules, and transformation layer rules, which respectively correspond to the determination of damage to the form and substance of the viscera, the determination of abnormality in the function and transformation of the viscera, and the determination of the dynamic transformation of yin and yang in the pathogenesis. S2. Collect multidimensional diagnostic and treatment data and complete standardized preprocessing; the diagnostic and treatment data includes chief symptoms, concurrent signs, tongue and pulse information, constitution information and medical history data, which are normalized, quantified and noise-reducing to obtain standardized structured syndrome differentiation input data. S3. Using hierarchical feature analysis, construct a three-dimensional dialectical feature matrix; The body layer damage characteristics, the function abnormality characteristics of the function layer, and the pathogenesis and outcome characteristics of the transformation layer are extracted respectively, and the body damage characteristic value, function stagnation characteristic value, and transformation change characteristic value are quantified to form a structured three-dimensional dialectical characteristic matrix. S4. Integrate the principles of body, function, and transformation to carry out multi-level linkage dialectical reasoning; follow the core logic of traditional Chinese medicine that takes body as the foundation, function as the image, and transformation as change, and adopt a layered and progressive + linkage correction mechanism to complete the determination of the pathogenesis of the body, the matching of functional syndromes, and the correction of dynamic outcomes in sequence, so as to achieve accurate reasoning of basic syndromes and complex combined syndromes. S5. Verification and Integration of Diagnostic Results: Based on the core principle of TCM theory of body-function, the reasoning results are logically verified, contradictions and erroneous conclusions are eliminated, and a complete diagnostic report is generated, including the core syndrome type, pathogenesis, points of imbalance between body and function, disease progression trend, and treatment recommendations.

2. The intelligent dialectical reasoning method integrating the principles of medical application as described in claim 1, characterized in that, The body layer rules mentioned in step S1 include the criteria for judging the deficiency, stagnation, damage, and insufficiency of the five viscera and six bowels, qi, blood, body fluids, and meridians; the function layer rules include the criteria for judging the dysfunction of the ascending and descending qi of the viscera and bowels, the circulation of qi and blood, the metabolism of water and fluids, and the smooth flow of the meridians; the transformation layer rules include the dynamic criteria for judging the imbalance of yin and yang, the advance and retreat of pathogenic factors, the transformation of deficiency and excess, the evolution of cold and heat, and the severity and prognosis of the disease.

3. The intelligent dialectical reasoning method integrating the principles of medical application as described in claim 1, characterized in that, The standardization preprocessing described in step S2 specifically includes: normalizing and correcting colloquial and dialectal symptoms and signs based on a standard TCM terminology dictionary; quantifying and assigning values ​​to tongue appearance, pulse appearance, and constitution level; removing duplicate data, invalid noise data, and irrelevant medical history information to generate a structured diagnostic input dataset.

4. The intelligent dialectical reasoning method integrating the principles of medical application as described in claim 1, characterized in that, The body layer features described in step S3 focus on pathological changes in the human body's physical structure, identifying characteristics of organ deficiency, qi and blood insufficiency, and meridian obstruction; the function layer features focus on pathological changes in organ function, identifying characteristics of qi stagnation, impaired transportation and transformation, and abnormal distribution; the transformation layer features focus on the dynamic evolution of pathogenesis, identifying dynamic pathological characteristics of deficiency-excess transformation, cold-heat complexion, and yin-yang waxing and waning.

5. The intelligent dialectical reasoning method integrating the principles of medical application as described in claim 1, characterized in that, The multi-level linkage dialectical reasoning in step S4 is as follows: First, the basic pathogenesis of the viscera is determined based on the characteristics and rules of the body layer. Then, the basic syndrome corresponding to the external functional abnormality is matched by the characteristics and rules of the application layer. Finally, the dynamic transformation relationship of the pathogenesis is corrected by the characteristics and rules of the transformation layer, and the complex mixed syndrome of deficiency and excess and cold and heat is corrected.

6. The intelligent dialectical reasoning method integrating the principles of medical application as described in claim 1, characterized in that, The logical verification described in step S5 is based on the core principles of the unity of body and function, mutual transformation of body and function, and balance of yin and yang. It verifies the matching between the symptom and the pathogenesis, the correlation between physical damage and functional stagnation, the rationality of the disease outcome, and corrects logically contradictory reasoning results.

7. An intelligent dialectical reasoning device integrating the principles of medical application, characterized in that, The reasoning method described in any one of claims 1-6 includes a knowledge system construction module, a data preprocessing module, a three-dimensional feature analysis module, a holistic dialectical reasoning module, a result verification and output module, and a data storage module. The knowledge system construction module is used to build a three-dimensional TCM diagnostic rule base that includes the body layer, function layer, and transformation layer, and to solidify the core logic and judgment criteria of TCM body-function-transformation diagnostics. The data preprocessing module is used to collect multidimensional diagnostic and treatment data, complete data normalization, quantification, and noise reduction preprocessing, and output standardized structured dialectical input data. The three-dimensional feature analysis module is used to extract the three-dimensional dialectical features of volume, function, and transformation, quantify the feature weights, and construct a three-dimensional dialectical feature matrix. The embodiment-based dialectical reasoning module is used to perform multi-level dialectical reasoning with hierarchical progression and linkage correction to complete the accurate determination of basic syndrome types and complex pathogenesis. The result verification output module is used to verify the logical compliance of the dialectical results and integrate and generate a full-dimensional intelligent dialectical report. The data storage module is used to store the diagnostic rule base, diagnosis and treatment data, feature data, model parameters and diagnostic results, and supports data iteration and query calls.