Health management system and method based on traditional Chinese medicine and blood sugar monitoring

Abstract

The invention relates to the technical field of health management, and particularly discloses a health management system and method based on traditional Chinese medicine and blood glucose monitoring, and the method comprises an analysis module which obtains historical data of a plurality of users, and determines a plurality of basic clustering sets and a plurality of habit clustering sets; the construction module is used for determining direct influence data of each habit label of each basic label and constructing a prediction influence model of each viscera of each habit label of each basic label; the determining module is used for acquiring multi-source data of a to-be-managed user, acquiring blood glucose monitoring data and determining a user basic label and a user habit label of the to-be-managed user; and the management module is used for determining a prediction signal vector and health management recommendation data of each viscera of the to-be-managed user. According to the method, the viscera prediction model can be constructed in a targeted manner, precise prediction of viscera states and personalized traditional Chinese medicine intervention recommendation are realized, the pertinence, suitability and scientificity of health management are improved, and objectivity of blood sugar monitoring and systematicness of traditional Chinese medicine conditioning are both considered.

Description

Technical Field

[0001] This invention relates to the field of health management technology, and in particular to a health management system and method based on traditional Chinese medicine and blood glucose monitoring. Background Technology

[0002] In traditional health management, TCM (Traditional Chinese Medicine) treatments rely heavily on physicians' experience and diagnosis, which is highly subjective and lacks quantitative data support. Blood glucose monitoring is often used solely for blood sugar control, without deep integration with organ health and TCM treatment. User segmentation is also crude, making personalized intervention difficult. Currently, health management is gradually incorporating sensor technology and data mining, but most solutions still rely on a single user segmentation dimension, have highly generalized models, and lack specific predictive models for groups with different health conditions and habits. The integration of TCM with multi-dimensional quantitative data is insufficient, and the accuracy of organ status prediction and TCM recommendations needs improvement.

[0003] Therefore, this invention proposes a health management system and method based on traditional Chinese medicine and blood glucose monitoring. Summary of the Invention

[0004] This invention provides a health management system and method based on Traditional Chinese Medicine (TCM) and blood glucose monitoring. It acquires historical user data from multiple users, determines multiple basic cluster sets and multiple habit cluster sets within each basic cluster set, identifies the direct impact data of each habit tag for each basic tag, constructs a predictive impact model for each organ (organ) for each habit tag of each basic tag, acquires user management data, environmental data, and user emotion data for the user to be managed, collects blood glucose monitoring data for the user to be managed, determines the user's basic tags and user habit tags, determines the predictive signal vector for each organ of the user to be managed, and determines the health management recommendation data for the user to be managed. It can specifically construct organ prediction models under various tag combinations, integrate TCM management logic with multi-dimensional real-time data, achieve accurate prediction of organ status and personalized TCM intervention recommendations, improve the pertinence, adaptability, and scientific rigor of health management, and balance the objectivity of blood glucose monitoring with the systematic nature of TCM conditioning.

[0005] This invention provides a health management system based on traditional Chinese medicine and blood glucose monitoring, comprising: Analysis module: Acquires user history data from multiple historical users, analyzes the user history data of each historical user, and determines multiple basic cluster sets and multiple habit cluster sets for each basic cluster set; Building Module: Based on the user history data of all historical users and the habit clustering sets of all basic clustering sets, determine the direct impact data of each habit tag of each basic tag, and build a predictive impact model for each organ of each habit tag of each basic tag; Determine the module: acquire the managed user data, environmental data, and user emotion data of the users to be managed; collect the blood glucose monitoring data of the users to be managed; and determine the user basic tags and user habit tags of the users to be managed based on the managed user data and all habit clusters of all basic cluster sets. Management Module: Based on the managed user data, user basic tags, user habit tags, prediction influence models of all organs of all habit tags of all basic tags, and traditional Chinese medicine management model, determine the prediction signal vector of each organ of the managed user, and determine the health management recommendation data of the managed user.

[0006] Preferably, a health management system based on traditional Chinese medicine and blood glucose monitoring includes user historical data such as user basic data, user habit data, historical blood glucose monitoring data, and historical blood glucose impact data. User habit data includes user sleep vector, user diet vector, and user exercise vector. Historical blood glucose monitoring data includes blood glucose trend graphs for multiple blood glucose event cycles, cycle length, historical blood glucose monitoring data within a specified time period before the blood glucose event cycle, environmental feature vectors, emotion tag event set, recording time of each emotion tag in the emotion tag event set, and emotion intensity. Historical blood glucose impact data includes physiological signal vectors of multiple organs in each blood glucose event cycle.

[0007] Preferably, a health management system based on traditional Chinese medicine and blood glucose monitoring includes an analysis module comprising: Basic Feature Vector Unit: Extract features from the basic user data in the historical user data of each historical user to determine the basic feature vector of each historical user; Basic cluster set unit: Based on the basic feature vectors of all historical users, the first cluster analysis is performed on all historical users to determine multiple basic cluster sets and the basic label and basic cluster center vector of each basic cluster set. The basic cluster set includes multiple historical users. Habit Clustering Unit: Based on the user sleep vector, user diet vector, and user exercise vector in the user history data of all historical users in the basic clustering set of each basic label, a second clustering analysis is performed on all users in each basic clustering set to determine multiple habit clustering sets for each basic label, as well as the habit label, sleep cluster center vector, diet cluster center vector, and exercise cluster center vector for each habit clustering set. The habit clustering set includes multiple historical users.

[0008] Preferably, a health management system based on traditional Chinese medicine and blood glucose monitoring includes the following modules: Blood glucose event vector unit: Extract features from the blood glucose trend map of each blood glucose event period in the historical blood glucose monitoring data of each historical user's historical user data. Based on the period length extracted for each blood glucose event period and all features, determine the blood glucose event vector of each blood glucose event period in the historical blood glucose monitoring data of each historical user's historical user data. Blood glucose emotion tag set unit: Based on the set of emotion tag events within a specified time period before the blood glucose event period in the historical blood glucose monitoring data of all historical users, the blood glucose emotion tag set is determined. Directly impacting data units: Based on the set of blood glucose emotion tags, the habit clustering set of each habit tag of each basic tag, the historical blood glucose monitoring data of all historical users, and the historical blood glucose impact data, the direct impact data of each habit tag of each basic tag is determined; Predictive Impact Model Unit: The predictive impact model takes the historical blood glucose monitoring data and direct impact data of all historical users in the habit cluster set of each habit label of each basic label as input, and the physiological signal vector of each organ in the historical blood glucose impact data of all historical users in the habit cluster set of each habit label of each basic label as output, and constructs the predictive impact model of each organ of each habit label of each basic label.

[0009] Preferably, a health management system based on traditional Chinese medicine and blood glucose monitoring directly affects data units, including: The emotional impact value sub-unit is calculated based on the environmental feature vectors of all blood glucose event cycles and the specified time period before the blood glucose event cycles in the historical blood glucose monitoring data of each historical user in the habit cluster set of each habit label of each basic label, the set of emotional label events, the recording time of each emotional label in the emotional label event set, and the emotional intensity. It calculates the emotional impact value of each historical user in each blood glucose event cycle on each emotional label in the blood glucose emotional label set in the habit cluster set of each habit label of each basic label. First direct impact value and second direct impact value subunits: Based on the blood glucose emotion tag set, the blood glucose event vector of all historical users in the historical blood glucose monitoring data of all blood glucose event cycles in the habit cluster set of each habit tag of each basic tag, and the emotion impact value of each historical user in each blood glucose event cycle on each emotion tag in the blood glucose emotion tag set of each habit tag of each basic tag, calculate the first direct impact value of each emotion tag in the blood glucose emotion tag set of each habit tag of each basic tag on each blood glucose feature in the blood glucose event vector, and calculate the second direct impact value of each environmental feature in the environmental feature vector of each habit tag of each basic tag on each blood glucose feature in the blood glucose event vector; The third direct impact value subunit: For each habitual label of each basic label, the correlation analysis is performed on the blood glucose event vector of all historical blood glucose monitoring data of all historical users in the habitual cluster set of each habitual label of each basic label, and the physiological signal vector of each organ in the historical blood glucose impact data of all blood glucose event cycles, to determine the third direct impact value of each blood glucose feature in the blood glucose event vector of each habitual label of each basic label on each signal feature in the physiological signal vector of each organ. Directly Influencing Data Subunits: Based on the first direct influence value of all emotion tags in the blood glucose emotion tag set of each basic tag on all blood glucose features in the blood glucose event vector, the second direct influence value of all environmental features in the environmental feature vector on all blood glucose features in the blood glucose event vector, and the third direct influence value of all blood glucose features in the blood glucose event vector on all signal features in the physiological signal vector of each organ, the direct influence data of each habit tag of each basic tag is determined.

[0010] Preferably, a health management system based on traditional Chinese medicine and blood glucose monitoring includes the following modules: User data management unit: Acquires user data of users to be managed, including user basic data, sleep habit vector, diet habit vector, exercise habit vector, blood sugar treatment data, and organ condition data for each organ. Blood sugar treatment data includes at least one or more drugs or supplements. Acquisition Unit: Acquires environmental data of the user to be managed based on the intelligent environmental sensor group, and acquires user emotion data of the user to be managed within a specified time period based on the emotion sensor group. The user emotion set includes at least one and more emotion tags and the recording time of each emotion tag. Blood glucose monitoring data unit: Based on microneedle collection of blood glucose monitoring data of the user to be managed, the blood glucose monitoring data of the user to be managed is extracted according to a specified time period to determine the real-time monitoring data of the user to be managed.

[0011] Preferably, a health management system based on traditional Chinese medicine and blood glucose monitoring includes a management module comprising: User Basic Tag Unit: The basic cluster center vector of the user basic data in the managed user data of the user to be managed and the basic cluster set of all basic tags, to determine the user basic tags of the user to be managed; User habit tag unit: Based on the sleep habit vector, diet habit vector, exercise habit vector of the user to be managed, and the sleep clustering vector, diet clustering vector, and exercise clustering vector of each habit tag of the habit cluster set corresponding to the basic tag of the user to be managed, the user habit tag of the user to be managed is determined; Predictive signal vector unit: Input the environmental data, user emotion data, blood glucose treatment data and real-time monitoring data of the user to be managed into the predictive influence model of each organ corresponding to the user's basic tags and user habit tags, and determine the predictive signal vector of each organ of the user to be managed based on the predictive influence model of each organ. Health management recommendation data unit: Input the organ condition data and prediction signal vector of all organs in the management user data of the user to be managed into the TCM management model, and determine the health management recommendation data of the user to be managed based on the TCM management model. The health management recommendation data includes the organ influence set and multiple health recommendations for each organ in the organ influence set.

[0012] This invention provides a health management method based on traditional Chinese medicine and blood glucose monitoring, for implementing any one of the health management systems based on traditional Chinese medicine and blood glucose monitoring in Examples 1 to 7, comprising: S1: Obtain user history data from multiple historical users, analyze the user history data of each historical user, and determine multiple basic cluster sets and multiple habit cluster sets for each basic cluster set; S2: Based on the user history data of all historical users and the habit clustering sets of all basic clustering sets, determine the direct impact data of each habit tag of each basic tag, and construct the predictive impact model of each organ of each habit tag of each basic tag; S3: Obtain the managed user data, environmental data, and user emotion data of the users to be managed; collect the blood glucose monitoring data of the users to be managed; and determine the user basic tags and user habit tags of the users to be managed based on the managed user data and all habit clustering sets of all basic clustering sets. S4: Based on the management user data of the user to be managed, the user's basic tags, the user's habit tags, the predictive influence model of all organs of all habit tags of all basic tags, and the TCM management model, determine the predictive signal vector of each organ of the user to be managed, and determine the health management recommendation data of the user to be managed.

[0013] The beneficial effects of this invention compared to existing technologies are as follows: It acquires historical user data from multiple users, determines multiple basic cluster sets and multiple habit cluster sets within each basic cluster set, determines the direct impact data of each habit tag for each basic tag, constructs a predictive impact model for each organ for each habit tag of each basic tag, acquires management user data, environmental data, and user emotion data for the user to be managed, collects blood glucose monitoring data for the user to be managed, determines the user's basic tags and user habit tags, determines the predictive signal vector for each organ of the user to be managed, and determines the health management recommendation data for the user to be managed. It can specifically construct organ prediction models under various tag combinations, integrate traditional Chinese medicine management logic with multi-dimensional real-time data, achieve accurate prediction of organ status and personalized traditional Chinese medicine intervention recommendations, improve the pertinence, adaptability, and scientific nature of health management, and balance the objectivity of blood glucose monitoring with the systematic nature of traditional Chinese medicine conditioning.

[0014] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in this application.

[0015] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0016] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings: Figure 1 This is a schematic diagram of a health management system based on traditional Chinese medicine and blood glucose monitoring in an embodiment of the present invention; Figure 2 This is a flowchart of a health management method based on traditional Chinese medicine and blood glucose monitoring, as described in an embodiment of the present invention. Detailed Implementation

[0017] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention. Example 1:

[0018] This invention provides a health management system based on traditional Chinese medicine and blood glucose monitoring, with reference to... Figure 1 ,include: Analysis module: Acquires user history data from multiple historical users, analyzes the user history data of each historical user, and determines multiple basic cluster sets and multiple habit cluster sets for each basic cluster set; Building Module: Based on the user history data of all historical users and the habit clustering sets of all basic clustering sets, determine the direct impact data of each habit tag of each basic tag, and build a predictive impact model for each organ of each habit tag of each basic tag; Determine the module: acquire the managed user data, environmental data, and user emotion data of the users to be managed; collect the blood glucose monitoring data of the users to be managed; and determine the user basic tags and user habit tags of the users to be managed based on the managed user data and all habit clusters of all basic cluster sets. Management Module: Based on the managed user data, user basic tags, user habit tags, prediction influence models of all organs of all habit tags of all basic tags, and traditional Chinese medicine management model, determine the prediction signal vector of each organ of the managed user, and determine the health management recommendation data of the managed user.

[0019] In this embodiment, historical user data from multiple historical users is acquired. This data encompasses multi-dimensional information such as basic user data, user habit data, historical blood glucose monitoring data, and historical blood glucose impact data, comprehensively reflecting the basic characteristics, lifestyle habits, blood glucose changes, and organ physiological states of historical users. Subsequently, this data for each historical user is systematically analyzed. By extracting basic features and clustering them, multiple basic cluster sets are determined. Each basic cluster set contains multiple historical users with similar basic features. Within each basic cluster set, further clustering is performed based on user habit data, resulting in multiple habit cluster sets. This ensures that users within the same set not only have similar basic features but also highly consistent lifestyle habits, achieving a progressive segmentation of the user group from basic features to habits.

[0020] In this embodiment, by combining all basic cluster sets and habit cluster sets, the intrinsic relationship between factors such as emotions, environment, blood glucose, and organ physiological signals in different groups is analyzed in depth. The direct impact data of each habit label corresponding to each basic label is quantified. This data covers the influence strength of emotion labels and environmental characteristics on blood glucose characteristics, as well as the influence strength of blood glucose characteristics on organ signal characteristics. Based on this, a separate predictive impact model is constructed for each organ corresponding to each combination of basic and habit labels, ensuring that each model can accurately capture the influence patterns of specific organs in the corresponding group.

[0021] In this embodiment, a comprehensive collection of key data from the users to be managed is acquired. This includes user management data encompassing basic data, habit vectors, treatment data, and organ condition data, reflecting the user's basic health status. Environmental data captures the characteristics of the user's external environment. User emotion data records emotional changes over a specified period. Simultaneously, blood glucose monitoring data from the users to be managed is collected using specialized equipment to ensure data real-time performance and completeness. Then, using the user management data, feature comparisons are performed on all identified basic cluster sets and habit cluster sets to find the basic cluster set that best matches the user's basic characteristics, determining their basic user label. Next, within all habit cluster sets corresponding to this basic label, similarity matching is performed based on the user's habit vector to determine their user habit label, thereby clarifying the specific subgroup to which the user to be managed belongs.

[0022] In this embodiment, the established user base tags and user habit tags of the managed users are integrated. These data, along with the managed users' environmental data, emotional data, blood glucose monitoring data, etc., are input into the predictive impact model for each organ corresponding to the tag combination. Based on pre-trained group-specific patterns, the model outputs a predictive signal vector for each organ of the managed user. This vector accurately reflects the future physiological state of each organ. Then, the current organ status data of the managed user, along with the predictive signal vectors, are input into a Traditional Chinese Medicine (TCM) management model. This model, based on the holistic concept and syndrome differentiation and treatment principles of TCM, combines the organ imbalance state and predictive trends to comprehensively analyze and determine the recommended health management data for the managed user.

[0023] The beneficial effects of the above technologies are as follows: They acquire historical user data from multiple users, determine multiple basic cluster sets and multiple habit cluster sets within each basic cluster set, determine the direct impact data of each habit tag for each basic tag, construct a predictive impact model for each organ for each habit tag of each basic tag, acquire management user data, environmental data, and user emotion data for users to be managed, collect blood glucose monitoring data for users to be managed, determine the user's basic tags and user habit tags for users to be managed, determine the predictive signal vector for each organ for users to be managed, and determine the health management recommendation data for users to be managed. Targeted organ prediction models can be constructed under various tag combinations, integrating TCM management logic with multi-dimensional real-time data to achieve accurate prediction of organ status and personalized TCM intervention recommendations, improving the targeting, adaptability, and scientific rigor of health management, while balancing the objectivity of blood glucose monitoring with the systematic nature of TCM conditioning. Example 2:

[0024] Based on Example 1, a health management system based on traditional Chinese medicine and blood glucose monitoring is provided. The user's historical data includes user basic data, user habit data, historical blood glucose monitoring data, and historical blood glucose impact data. The user habit data includes user sleep vector, user diet vector, and user exercise vector. The historical blood glucose monitoring data includes blood glucose trend graphs for multiple blood glucose event cycles, cycle length, historical blood glucose monitoring data within a specified time period before the blood glucose event cycle, environmental feature vectors, emotion tag event set, recording time of each emotion tag in the emotion tag event set, and emotion intensity. The historical blood glucose impact data includes physiological signal vectors of multiple organs in each blood glucose event cycle.

[0025] In this embodiment, user historical data includes multiple types of specific information. Basic user data comprises the fundamental information of historical users. User habit data consists of user sleep vectors, user diet vectors, and user exercise vectors, reflecting the user's characteristics in sleep, diet, and exercise, respectively. Historical blood glucose monitoring data is rich in content, including blood glucose trend graphs and cycle lengths for multiple blood glucose event cycles. It also includes historical blood glucose monitoring data within a specified time range prior to each blood glucose event cycle. Furthermore, it covers environmental feature vectors, emotion-labeled event sets, the recording time of each emotion label within the emotion-labeled event set, and the emotion intensity. This data records dynamic information related to the user's blood glucose from different perspectives. Historical blood glucose impact data consists of physiological signal vectors of multiple organs in each blood glucose event cycle, used to demonstrate the impact of blood glucose changes on the physiological state of these organs.

[0026] In this embodiment, the duration of the blood glucose event cycle value is the total time span from when blood glucose begins to deviate from the benchmark to when it completely falls back to the benchmark.

[0027] In this embodiment, the length of the specified time period preceding the blood glucose event period can be 60 minutes.

[0028] In this embodiment, the user's basic data includes at least gender, age, height, weight, basic physical constitution such as yin deficiency constitution, yang deficiency constitution and other TCM syndrome differentiation types, past chronic disease history such as whether or not prediabetes has been diagnosed, etc., to reflect the user's basic physiological and health background characteristics.

[0029] In this embodiment, the user sleep vector is a multi-dimensional feature vector describing the sleep state from the user habit data. The features include sleep duration, sleep onset time, deep sleep percentage, number of awakenings, sleep quality, etc.

[0030] In this embodiment, the user's diet vector is a multi-dimensional feature vector describing the dietary status in the user's habit data. The features include daily staple food intake, such as the grams of staple foods like rice and noodles, protein content, frequency of high-sugar foods, regularity of three meals, and proportion of dietary taste preferences.

[0031] In this embodiment, the user motion vector is a multi-dimensional feature vector describing the motion state in the user habit data. The features include exercise duration such as the average number of hours per exercise session, exercise intensity, exercise frequency, exercise type such as running, swimming, Tai Chi, and exercise time distribution such as the frequency ratio of exercise in the morning, afternoon, and evening.

[0032] In this embodiment, the environmental feature vector describes a multi-dimensional feature vector of the user's environmental state, including features such as temperature, humidity, activity environment type (e.g., indoor, outdoor, office), air quality index, and season.

[0033] In this embodiment, the emotional tag event set is a set of tags that record the user's emotional state within a specified time period in historical blood glucose monitoring data, such as excited, anxious, frightened, angry, depressed, etc., which are tags that affect blood glucose. Each tag corresponds to an emotional recording event, which is used to reflect the user's emotional changes within that period.

[0034] In this embodiment, the recording time of each emotion tag in the emotion tag event set refers to the specific moment when each emotion tag in the emotion tag event set is recorded.

[0035] In this embodiment, the internal organs are the heart, liver, spleen, lungs, kidneys, gallbladder, stomach, small intestine, large intestine, bladder, triple burner, and pericardium.

[0036] In this embodiment, for example, the physiological signal vector of the liver includes liver area skin temperature, ocular surface infrared temperature, ulnar pulse amplitude, exhaled nitric oxide, etc., and the physiological signal vector of the heart includes QTc interval, cardiac surface skin temperature, sublingual vein RGB contrast, etc.

[0037] The beneficial effects of the above technologies are: obtaining user history data from multiple historical users can provide data support for determining multiple basic cluster sets and multiple habit cluster sets for each basic cluster set. Example 3:

[0038] Based on Example 2, a health management system based on traditional Chinese medicine and blood glucose monitoring includes an analysis module, comprising: Basic Feature Vector Unit: Extract features from the basic user data in the historical user data of each historical user to determine the basic feature vector of each historical user; Basic cluster set unit: Based on the basic feature vectors of all historical users, the first cluster analysis is performed on all historical users to determine multiple basic cluster sets and the basic label and basic cluster center vector of each basic cluster set. The basic cluster set includes multiple historical users. Habit Clustering Unit: Based on the user sleep vector, user diet vector, and user exercise vector in the user history data of all historical users in the basic clustering set of each basic label, a second clustering analysis is performed on all users in each basic clustering set to determine multiple habit clustering sets for each basic label, as well as the habit label, sleep cluster center vector, diet cluster center vector, and exercise cluster center vector for each habit clustering set. The habit clustering set includes multiple historical users.

[0039] In this embodiment, the basic feature vector unit performs feature extraction on the user basic data contained in the user history data of each historical user, and arranges the contents contained in the basic data of all users in a specific order, so that each historical user corresponds to a unique basic feature vector, and each dimension of the vector represents a basic feature that has been standardized.

[0040] In this embodiment, the core principle of the basic clustering set unit is to divide users into groups based on the similarity of their basic feature vectors, grouping users with similar basic features into one category to form a clearly defined basic user group. The first clustering analysis is a clustering operation based on the basic feature vectors of all historical users. Its core logic is to calculate the similarity or distance between different users' basic feature vectors, grouping users with high similarity and close proximity into the same group. The basic clustering set consists of multiple user groups obtained through the first clustering analysis. Users within each set have highly similar basic features, while the differences in basic features between different sets are significant. The basic label is a unique identifier assigned to each basic clustering set. The basic cluster center vector represents the core feature of each basic clustering set. It is calculated by taking the mean or mode of the corresponding dimensions of the basic feature vectors of all users within the set, ultimately forming a vector that reflects the average level of the group's basic features.

[0041] In this embodiment, the habit clustering unit further explores differences in users' lifestyle habits within the already divided basic clusters, achieving a refined and progressive segmentation of user groups and making group characteristics more closely reflect users' actual living conditions. While users within the same basic cluster may have similar basic characteristics, they may have significant differences in their lifestyle habits, which directly affect blood sugar changes and organ health. Therefore, a second clustering analysis is necessary. The habit clustering set is a set of multiple subgroups divided within each basic cluster through the second clustering analysis. Users within each subgroup have highly similar lifestyle habits. The habit label is a unique identifier assigned to each habit clustering set, used to distinguish subgroups with different lifestyle habits within the same basic group. The sleep core vector represents the core feature of each habit clustering set in the sleep habit dimension, obtained by averaging the sleep vectors of all users within the set, reflecting the average sleep characteristics of the group, such as average sleep duration and average deep sleep percentage. The dietary clustering vector represents the core feature of the set in the dimension of dietary habits. It is obtained by averaging the dietary vectors of users within the set along the corresponding dimension, reflecting the average dietary characteristics of the group, such as average staple food intake and average frequency of high-sugar foods. The exercise clustering vector represents the core feature of the set in the dimension of exercise habits. It is obtained by averaging the exercise vectors of users within the set along the corresponding dimension, reflecting the average exercise characteristics of the group, such as average exercise frequency and average exercise duration. This two-layer clustering approach ensures the consistency of the group's basic characteristics while refining the differences in lifestyle habits, providing a precise foundation for subsequent construction of customized models for user groups with different combinations of basic characteristics and habits.

[0042] The beneficial effects of the above technologies are as follows: by analyzing the user history data of each historical user, multiple basic cluster sets and multiple habit cluster sets of each basic cluster set can be determined, which can form a refined user segmentation system, provide accurate group segmentation basis for modeling, and improve the adaptability of the model to user characteristics. Example 4:

[0043] Based on Example 3, a health management system based on traditional Chinese medicine and blood glucose monitoring is constructed, including the following modules: Blood glucose event vector unit: Extract features from the blood glucose trend map of each blood glucose event period in the historical blood glucose monitoring data of each historical user's historical user data. Based on the period length extracted for each blood glucose event period and all features, determine the blood glucose event vector of each blood glucose event period in the historical blood glucose monitoring data of each historical user's historical user data. Blood glucose emotion tag set unit: Based on the set of emotion tag events within a specified time period before the blood glucose event period in the historical blood glucose monitoring data of all historical users, the blood glucose emotion tag set is determined. Directly impacting data units: Based on the set of blood glucose emotion tags, the habit clustering set of each habit tag of each basic tag, the historical blood glucose monitoring data of all historical users, and the historical blood glucose impact data, the direct impact data of each habit tag of each basic tag is determined; Predictive Impact Model Unit: The predictive impact model takes the historical blood glucose monitoring data and direct impact data of all historical users in the habit cluster set of each habit label of each basic label as input, and the physiological signal vector of each organ in the historical blood glucose impact data of all historical users in the habit cluster set of each habit label of each basic label as output, and constructs the predictive impact model of each organ of each habit label of each basic label.

[0044] In this embodiment, the blood glucose event vector unit transforms the unstructured blood glucose trend graph into quantitative features that can be used for data analysis and modeling. Simultaneously, it combines the time attribute of the blood glucose event cycle to form a structured data carrier that comprehensively reflects the blood glucose fluctuation characteristics of a single blood glucose event cycle. The blood glucose trend graph is a dynamic trajectory of blood glucose values ​​changing over time within a single blood glucose event cycle, containing multiple key dynamic features such as the rate of increase, peak size, rate of decrease, and fluctuation amplitude. Through feature extraction, these implicit features can be transformed into quantifiable numerical indicators. The blood glucose event vector is a multi-dimensional vector formed after systematically quantifying and integrating the blood glucose change features of a single blood glucose event cycle. It not only includes all the dynamic features extracted from the blood glucose trend graph but also incorporates the time feature of the cycle length of the blood glucose event cycle.

[0045] In this embodiment, the blood glucose emotion tag set is determined by extracting and deduplicating all emotion tags from all emotion tag event sets within a specified time range before each blood glucose event period of a historical user.

[0046] In this embodiment, the predictive influence model unit constructs a targeted, dedicated predictive model based on the characteristic differences of different user groups, achieving accurate mapping from input data to organ physiological signal vectors. Since user groups under different combinations of basic and habitual labels exhibit significant differences in basic characteristics and lifestyle habits, the correlation patterns between their emotions, environment, blood glucose, and organ signals also differ. Therefore, it is necessary to construct a separate model for each organ for each group. The predictive influence model is a prediction model specifically designed for each organ corresponding to each combination of basic and habitual labels. Its input data includes two parts: one part is the user group's historical blood glucose monitoring data, encompassing multi-dimensional raw data such as blood glucose event vectors and environmental feature vectors; the other part is the previously determined direct influence data, including the quantitative influence relationships between various factors. The combination design of the input data ensures that the model can simultaneously utilize the raw data and the mined influence relationships. The model output is the physiological signal vector of each organ for the user group within the corresponding blood glucose event cycle. These vectors comprehensively reflect the physiological state of the organs. By training the model with the input and output data of a specific group, the model can learn the unique correlation patterns of that group, thereby achieving accurate prediction of organ physiological signal vectors for new users of the same type.

[0047] In this embodiment, the predictive impact model constructs a dedicated mapping model for each user segment corresponding to the combination of basic and habitual labels, and for each organ within that segment. It uses supervised learning logic to construct a dedicated mapping model, taking historical blood glucose monitoring data and direct impact data of users in that segment as joint inputs. This model provides complete scenario support based on the original data and focuses on core correlations using prior impact data. The corresponding organ physiological signal vectors of all blood glucose event cycles in that segment are used as supervisory labels. Through iterative optimization, the model minimizes the error between the predicted and actual values, accurately capturing the nonlinear chain relationship between multiple factors and specific organ physiological signals in the segmented segment, and finally forming a dedicated predictive model adapted to that organ in that segment.

[0048] The beneficial effects of the above technologies are as follows: Based on the user history data of all historical users and the habit clustering sets of all basic clustering sets, the direct impact data of each habit tag of each basic tag is determined, and a predictive impact model of each organ of each habit tag of each basic tag is constructed. This can achieve refined group modeling, improve the pertinence and accuracy of organ physiological signal prediction, and provide deeply adapted model support for personalized health management. Example 5:

[0049] Based on Example 4, a health management system based on traditional Chinese medicine and blood glucose monitoring directly affects the data unit, including: The emotional impact value sub-unit is calculated based on the environmental feature vectors of all blood glucose event cycles and the specified time period before the blood glucose event cycles in the historical blood glucose monitoring data of each historical user in the habit cluster set of each habit label of each basic label, the set of emotional label events, the recording time of each emotional label in the emotional label event set, and the emotional intensity. It calculates the emotional impact value of each historical user in each blood glucose event cycle on each emotional label in the blood glucose emotional label set in the habit cluster set of each habit label of each basic label. First direct impact value and second direct impact value subunits: Based on the blood glucose emotion tag set, the blood glucose event vector of all historical users in the historical blood glucose monitoring data of all blood glucose event cycles in the habit cluster set of each habit tag of each basic tag, and the emotion impact value of each historical user in each blood glucose event cycle on each emotion tag in the blood glucose emotion tag set of each habit tag of each basic tag, calculate the first direct impact value of each emotion tag in the blood glucose emotion tag set of each habit tag of each basic tag on each blood glucose feature in the blood glucose event vector, and calculate the second direct impact value of each environmental feature in the environmental feature vector of each habit tag of each basic tag on each blood glucose feature in the blood glucose event vector; The third direct impact value subunit: For each habitual label of each basic label, the correlation analysis is performed on the blood glucose event vector of all historical blood glucose monitoring data of all historical users in the habitual cluster set of each habitual label of each basic label, and the physiological signal vector of each organ in the historical blood glucose impact data of all blood glucose event cycles, to determine the third direct impact value of each blood glucose feature in the blood glucose event vector of each habitual label of each basic label on each signal feature in the physiological signal vector of each organ. Directly Influencing Data Subunits: Based on the first direct influence value of all emotion tags in the blood glucose emotion tag set of each basic tag on all blood glucose features in the blood glucose event vector, the second direct influence value of all environmental features in the environmental feature vector on all blood glucose features in the blood glucose event vector, and the third direct influence value of all blood glucose features in the blood glucose event vector on all signal features in the physiological signal vector of each organ, the direct influence data of each habit tag of each basic tag is determined.

[0050] In this embodiment, the emotion impact value subunit is as follows: Based on the set of blood glucose emotion tags and the habit clustering set of each habit tag of each basic tag, the environmental feature vector within the specified time period range before the blood glucose event cycle in the historical blood glucose monitoring data of each historical user, the set of emotion tag events, the recording time of each emotion tag in the set of emotion tag events, and the emotion intensity, the emotional impact value of each historical user in each blood glucose event cycle on each emotion tag in the set of blood glucose emotion tags is calculated in the habit clustering set of each habit tag of each basic tag. The calculation formula is expressed as follows: ; in, Let ijkmN1 represent the nth emotion tag in the emotion tag event set within the specified time period before the m-th blood glucose event period of the historical blood glucose monitoring data of the k-th historical user in the habit cluster set of the j-th habit tag of the i-th basic tag. Let ijkmN1 represent the number of emotion tags in the emotion tag event set within the specified time period before the m-th blood glucose event period of the historical blood glucose monitoring data of the k-th historical user in the habit cluster set of the j-th habit tag of the i-th basic tag. This represents the recording time and intensity of the nth emotion tag in the set of emotion tag events within the habit clustering set of the jth habit tag of the jth habit tag of the jth historical user in the historical blood glucose monitoring data of the kth historical user, including the mth blood glucose event period and the specified time period before the blood glucose event period. This represents the p-th emotion label in the blood glucose emotion label set. Let represent the set of emotion tag events within the habit clustering set of the j-th habit tag of the j-th basic tag, which is the historical blood glucose monitoring data of the k-th historical user, including the m-th blood glucose event period and the specified time period before the blood glucose event period. Based on the first indicator function of the p-th emotion tag in the blood glucose emotion tag set, SL represents the period length of the specified time period before the blood glucose event period. This represents the start and end times of the m-th blood glucose event cycle in the historical blood glucose monitoring data of the k-th historical user within the habit cluster set of the j-th habit tag of the i-th basic tag. This represents the emotional impact sub-value of the nth emotional label in the emotional label event set within the habit clustering set of the jth habitual label of the jth historical user in the historical blood glucose monitoring data of the kth historical user, including the mth blood glucose event period and the specified time period before the blood glucose event period. This represents all emotion tags in the emotion tag event set within the habit clustering set of the j-th habit tag of the i-th basic tag, based on the emotion impact value of the p-th emotion tag in the blood glucose emotion tag set, within the m-th blood glucose event period and the specified time period before the blood glucose event period in the historical blood glucose monitoring data of the k-th historical user.

[0051] In this embodiment, the first direct impact value and the second direct impact value subunits are as follows: Based on the blood glucose emotion tag set and the habit clustering set of each habit tag of each basic tag, the blood glucose event vector of all historical blood glucose monitoring data of all historical users in all blood glucose event cycles, the environmental feature vector within a specified time period before the blood glucose event cycle, the emotion tag event set, the recording time and emotion intensity of each emotion tag in the emotion tag event set, the first direct impact value of each emotion tag in the blood glucose emotion tag set of each habit tag of each basic tag on each blood glucose feature in the blood glucose event vector is calculated, and the second direct impact value of each environmental feature in the environmental feature vector of each habit tag of each basic tag on each blood glucose feature in the blood glucose event vector is calculated. The calculation formula is expressed as follows: ; ; in, This represents the first direct influence value of the p-th emotion label in the set of blood glucose emotion labels (i-th basic label, j-th habit label) on the o-th blood glucose feature in the blood glucose event vector. This represents the second direct influence value of the q-th environmental feature value in the environmental feature vector of the j-th habitual label of the i-th basic label on the o-th blood glucose feature in the blood glucose event vector. Let Nij represent the q-th environmental feature value in the environmental feature vector within the m-th blood glucose event period and the specified time period preceding the blood glucose event period in the habit cluster set of the j-th habitual tag of the i-th basic tag. Let NijNk represent the number of blood glucose event periods in the historical blood glucose monitoring data of the k-th historical user in the habit cluster set of the j-th habitual tag of the i-th basic tag. Let NijN3 represent the number of historical users in the habit cluster set of the j-th habitual tag of the i-th basic tag. Let N4 represent the number of emotion tags in the blood glucose emotion tag set. This represents the o-th blood glucose feature value in the blood glucose event vector of the m-th blood glucose event period of the k-th historical user in the habit cluster set of the j-th habitual tag of the i-th basic tag. It represents the o-th average blood glucose feature value in the blood glucose event vector of all blood glucose event cycles of the k-th historical user in the habit cluster set of the j-th habit label of the i-th basic label.

[0052] In this embodiment, Indicates the value of emotional impact. The weights of the emotional impact values ​​of all emotional labels in the blood glucose emotional label set.

[0053] In this embodiment, the third direct impact value subunit focuses on the relevant data of all historical users in the habit cluster set of each habitual tag for each basic tag. Specifically, it analyzes the blood glucose event vector of all blood glucose event cycles in the historical blood glucose monitoring data, and the physiological signal vector of each organ in all blood glucose event cycles in the historical blood glucose impact data. By performing Pearson correlation analysis on these two types of vector data, the corresponding influence relationship between blood glucose features in the blood glucose event vector and signal features in the organ physiological signal vector is explored. Then, the third direct impact value of each blood glucose feature in the blood glucose event vector of each habitual tag for each basic tag on each organ is determined. This value quantifies the degree of direct influence of blood glucose features on organ signal features.

[0054] In this embodiment, the direct impact data subunit integrates the three types of core impact values ​​calculated previously to form the direct impact data corresponding to each habitual tag of each basic tag.

[0055] The beneficial effects of the above technologies are as follows: Based on the set of blood glucose emotion tags, the set of habit clusters for each habit tag of each basic tag, the historical blood glucose monitoring data and historical blood glucose impact data of all historical users in the habit cluster set of each basic tag, the direct impact data of each habit tag of each basic tag can be determined. This can realize the refined and group-based quantification of the impact of emotions and environment on blood glucose and blood glucose on organ signals, providing more accurate impact relationship support for subsequent modeling, and improving the adaptability and prediction accuracy of the underlying data for personalized health management. Example 6:

[0056] Based on Example 4, a health management system based on traditional Chinese medicine and blood glucose monitoring is defined, including the following modules: User data management unit: Acquires user data of users to be managed, including user basic data, sleep habit vector, diet habit vector, exercise habit vector, blood sugar treatment data, and organ condition data for each organ. Blood sugar treatment data includes at least one or more drugs or supplements. Acquisition Unit: Acquires environmental data of the user to be managed based on the intelligent environmental sensor group, and acquires user emotion data of the user to be managed within a specified time period based on the emotion sensor group. The user emotion set includes at least one and more emotion tags and the recording time of each emotion tag. Blood glucose monitoring data unit: Based on microneedle collection of blood glucose monitoring data of the user to be managed, the blood glucose monitoring data of the user to be managed is extracted according to a specified time period to determine the real-time monitoring data of the user to be managed.

[0057] In this embodiment, the user data management unit acquires the managed user data of the users to be managed. This data system covers multiple key dimensions and can comprehensively reflect the health-related status of the users to be managed. Among them, the basic user data is the basic personal information of the users to be managed; the sleep habit vector is specifically used to record and reflect the characteristics and patterns of the users to be managed in terms of sleep, including sleep duration, sleep onset time, deep sleep ratio, and other related features; the dietary habit vector focuses on the user's dietary situation, covering key information such as daily dietary structure, staple food intake, frequency of consumption of high-sugar and high-fat foods, and regularity of three meals; the exercise habit vector records the user's exercise-related data, including core features such as exercise frequency, exercise duration, exercise intensity, and exercise type; blood glucose treatment data is treatment information directly related to the user's blood glucose control, explicitly including at least one or more drugs or supplements that the user is currently using, which directly affect the user's blood glucose changes and health status; the organ condition data for each organ records the specific status of each organ of the users to be managed.

[0058] In this embodiment, for example, the data on the condition of the spleen includes the level of spleen function such as normal, slightly weak, moderately disordered, and severely imbalanced; carbohydrate digestion and absorption rate; frequency of spleen and stomach function-related symptoms such as the number of times abdominal distension occurs, duration of postprandial fullness, frequency of loss of appetite, baseline values ​​of spleen-related physiological signals such as the average amplitude of spleen physiological signals, the baseline range of fluctuation frequency, the ability to regulate function such as the time it takes for spleen signals to recover to the baseline after eating high-carbohydrate foods, and the state of spleen and stomach qi movement such as normal, insufficient upward movement, and poor downward movement.

[0059] In this embodiment, the acquisition unit achieves real-time data collection through two types of specialized sensor groups. The environmental intelligent sensor group is a combination of devices specifically designed to sense the environmental state of the user under management. It can continuously capture environmental data around the user, including key environmental indicators such as ambient temperature, humidity, light intensity, and air quality. These environmental factors indirectly affect the user's blood sugar level and organ function. The emotion sensor group focuses on collecting emotion-related data of the user under management. The collection scope is limited to a specified time period. The acquired user emotion data contains at least one emotion tag, and the recording time corresponding to each emotion tag is accurately recorded. In this way, the user's emotional change trajectory within a specific time period can be accurately tracked, capturing the time nodes and specific types of emotional fluctuations, providing continuous and accurate emotion data support for analyzing the impact of emotions on blood sugar and organ status.

[0060] In this embodiment, the emotion sensor group can be a smartwatch, smart bracelet, portable EEG headband / strap, etc.

[0061] In this embodiment, the blood glucose monitoring data unit uses microneedle technology as the core method for blood glucose collection. Microneedle technology is minimally invasive, safe, and comfortable, enabling accurate collection of blood glucose monitoring data from the user under management while minimizing user discomfort, ensuring the authenticity and reliability of the blood glucose data. After completing the collection of blood glucose monitoring data, the unit further processes the collected raw blood glucose data, extracting relevant data within a specified time period according to actual needs, forming real-time monitoring data for the user under management. This real-time monitoring data can dynamically reflect the user's blood glucose changes during that time period, including key information such as the fluctuation range, peak value, and trough value of blood glucose levels.

[0062] The beneficial effects of the above technologies are as follows: acquiring management user data, environmental data, and user emotion data of users to be managed, and collecting blood glucose monitoring data of users to be managed can achieve comprehensiveness, real-time performance, and comfort in data acquisition, providing high-quality, multi-dimensional data support for subsequent personalized health management, and improving the adaptability and accuracy of management plans. Example 7:

[0063] Based on Example 6, a health management system based on traditional Chinese medicine and blood glucose monitoring includes a management module comprising: User Basic Tag Unit: The basic cluster center vector of the user basic data in the managed user data of the user to be managed and the basic cluster set of all basic tags, to determine the user basic tags of the user to be managed; User habit tag unit: Based on the sleep habit vector, diet habit vector, exercise habit vector of the user to be managed, and the sleep clustering vector, diet clustering vector, and exercise clustering vector of each habit tag of the habit cluster set corresponding to the basic tag of the user to be managed, the user habit tag of the user to be managed is determined; Predictive signal vector unit: Input the environmental data, user emotion data, blood glucose treatment data and real-time monitoring data of the user to be managed into the predictive influence model of each organ corresponding to the user's basic tags and user habit tags, and determine the predictive signal vector of each organ of the user to be managed based on the predictive influence model of each organ. Health management recommendation data unit: Input the organ condition data and prediction signal vector of all organs in the management user data of the user to be managed into the TCM management model, and determine the health management recommendation data of the user to be managed based on the TCM management model. The health management recommendation data includes the organ influence set and multiple health recommendations for each organ in the organ influence set.

[0064] In this embodiment, the user basic tagging unit operates around the basic data of the user to be managed and the basic clustering information of historical users. This unit acquires the user basic data from the managed user data of the user to be managed, and simultaneously retrieves the basic cluster center vectors of all basic tags corresponding to their respective basic cluster sets. By comparing and analyzing the user basic data of the user to be managed with these basic cluster center vectors, the unit finds the basic tag corresponding to the most matching basic cluster center vector and determines it as the user basic tag for the user to be managed.

[0065] In this embodiment, the user habit tagging unit, based on the user base tags of the user to be managed determined by the user base tagging unit, first locks all habit cluster sets under the base tags corresponding to the user base tags. Next, it obtains the sleep habit vector, diet habit vector, and exercise habit vector of the user to be managed, and simultaneously extracts the sleep cluster center vector, diet cluster center vector, and exercise cluster center vector of each of the above habit cluster sets. By comparing the similarity of the three types of habit vectors of the user to be managed with the three types of cluster center vectors of the corresponding habit cluster sets, the habit cluster set with the highest comprehensive similarity is found, and its corresponding habit tag is the user habit tag of the user to be managed.

[0066] In this embodiment, the unit's environmental data, user emotion data, blood glucose treatment data, and real-time monitoring data are input into the predictive influence model of each organ corresponding to the user's basic tags and user habit tags. The predictive influence model of each organ will perform calculations based on the input multi-dimensional data and finally output the predictive signal vector of each organ of the user to be managed. This vector can reflect the predicted physiological state of each organ under the current data conditions.

[0067] In this embodiment, the health management recommendation data unit first acquires the organ condition data of all organs from the managed user data of the user to be managed. This data represents the current actual state of the user's organs. Then, after feature extraction of the organ condition data, it is input into the TCM management model along with the predicted signal vector for each organ output by the prediction signal vector unit. The TCM management model, based on TCM theory, comprehensively analyzes the actual condition and predicted state of the organs to determine the potential impact on each organ, thereby generating health management recommendation data for the user to be managed. This health management recommendation data includes a set of organ impacts, used to identify the names of multiple organs that are significantly affected. Simultaneously, for each organ in the organ impact set, multiple specific health recommendations are provided, forming comprehensive and personalized health management suggestions that align with TCM theory.

[0068] In this embodiment, the core principle of the TCM management model is based on the holistic concept and syndrome differentiation and treatment of TCM, integrating the theory of organ function correlation, the regulation rules of qi, blood and body fluids, and the syndrome differentiation system of cold, heat and deficiency. It comprehensively adapts and analyzes the organ condition data of the user to be managed with the predicted signal vector. First, by comparing the normal threshold of organ function, the standard of qi and blood circulation, and the characteristics of pathological states in TCM theory, it identifies the type of imbalance of each organ, such as qi deficiency, yin deficiency, phlegm dampness, and blood stasis, and the core influencing factors, such as emotional disorders, hot and humid environment, and organ burden caused by blood sugar fluctuations, forming an organ shadow... The system combines various approaches, including dietary therapy (such as food combinations to strengthen the spleen and remove dampness), herbal medicine (such as herbal formulas and herbal teas), meridian health maintenance (such as acupoint massage and moxibustion), and lifestyle adjustments (such as sleep regulation and emotional regulation). It generates multiple targeted health recommendations for each organ, ensuring that the recommendations align with the actual imbalances and predicted trends of the organs while also conforming to the TCM intervention logic of addressing both the root cause and symptoms, and supporting the body's resistance to pathogens. This achieves a deep integration of personalized organ health management with TCM theory.

[0069] The beneficial effects of the above technologies are as follows: Based on the managed user data, basic user tags, user habit tags, a predictive influence model for all organs of all basic tags and habit tags, and a traditional Chinese medicine management model, the predictive signal vector for each organ of the managed user is determined, and the health management recommendation data for the managed user is determined. This enables more precise user tagging, contextualized organ prediction, and TCM-adapted health recommendations, improving the personalization and targeting of health management and providing users with organ health recommendation solutions tailored to their individual characteristics. Example 8:

[0070] This invention provides a health management method based on traditional Chinese medicine and blood glucose monitoring, used to implement any one of the health management systems based on traditional Chinese medicine and blood glucose monitoring in Examples 1 to 7, with reference to... Figure 2 ,include: S1: Obtain user history data from multiple historical users, analyze the user history data of each historical user, and determine multiple basic cluster sets and multiple habit cluster sets for each basic cluster set; S2: Based on the user history data of all historical users and the habit clustering sets of all basic clustering sets, determine the direct impact data of each habit tag of each basic tag, and construct the predictive impact model of each organ of each habit tag of each basic tag; S3: Obtain the managed user data, environmental data, and user emotion data of the users to be managed; collect the blood glucose monitoring data of the users to be managed; and determine the user basic tags and user habit tags of the users to be managed based on the managed user data and all habit clustering sets of all basic clustering sets. S4: Based on the management user data of the user to be managed, the user's basic tags, the user's habit tags, the predictive influence model of all organs of all habit tags of all basic tags, and the TCM management model, determine the predictive signal vector of each organ of the user to be managed, and determine the health management recommendation data of the user to be managed.

[0071] The beneficial effects of the above technologies are as follows: They acquire historical user data from multiple users, determine multiple basic cluster sets and multiple habit cluster sets within each basic cluster set, determine the direct impact data of each habit tag for each basic tag, construct a predictive impact model for each organ for each habit tag of each basic tag, acquire management user data, environmental data, and user emotion data for users to be managed, collect blood glucose monitoring data for users to be managed, determine the user's basic tags and user habit tags for users to be managed, determine the predictive signal vector for each organ for users to be managed, and determine the health management recommendation data for users to be managed. Targeted organ prediction models can be constructed under various tag combinations, integrating TCM management logic with multi-dimensional real-time data to achieve accurate prediction of organ status and personalized TCM intervention recommendations, improving the targeting, adaptability, and scientific rigor of health management, while balancing the objectivity of blood glucose monitoring with the systematic nature of TCM conditioning.

[0072] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A health management system based on traditional Chinese medicine and blood glucose monitoring, characterized in that, Comprise: Analysis module: obtain a plurality of historical user's user history data, analyze each historical user's user history data, determine a plurality of basic cluster sets, a plurality of habit cluster sets of each basic cluster set; Construction module: based on all historical user's user history data, all habit cluster sets of all basic cluster sets, determine the direct influence data of each habit label of each basic label, and construct the prediction influence model of each Zangfu of each habit label of each basic label; Determination module: obtain the management user data, environment data, user emotion data of the to-be-managed user, collect the blood glucose monitoring data of the to-be-managed user, based on the management user data of the to-be-managed user, all habit cluster sets of all basic cluster sets, determine the user basic label, user habit label of the to-be-managed user; Management module: based on the management user data of the to-be-managed user, user basic label, user habit label, prediction influence model of all Zangfu of all habit labels of all basic labels, TCM management model, determine the prediction signal vector of each Zangfu of the to-be-managed user, and determine the health management recommendation data of the to-be-managed user.

2. The health management system based on traditional Chinese medicine and blood glucose monitoring according to claim 1, characterized in that, User history data includes user basic data, user habit data, historical blood glucose monitoring data, historical blood glucose influence data, user habit data includes user sleep vector, user diet vector, user exercise vector, historical blood glucose monitoring data includes blood glucose trend chart of multiple blood glucose event periods, period length, blood glucose event period and historical blood glucose monitoring data in a specified time period before blood glucose event period, environment feature vector, emotion label event set, record time of each emotion label in emotion label event set, emotion intensity, historical blood glucose influence data includes physiological signal vector of multiple Zangfu of each blood glucose event period.

3. The health management system based on traditional Chinese medicine and blood glucose monitoring according to claim 2, characterized in that, Analysis module, comprising: Basic feature vector unit: feature extraction is carried out on the user basic data in the user history data of each historical user, and the basic feature vector of each historical user is determined; Basic cluster set unit: based on the basic feature vector of all historical users, first clustering analysis is carried out on all historical users, a plurality of basic cluster sets and the basic label and basic cluster center vector of each basic cluster set are determined, wherein the basic cluster set comprises a plurality of historical users; Habit cluster set unit: based on the user sleep vector, user diet vector and user exercise vector in the user habit data in the user history data of all historical users in the basic cluster set of each basic label, second clustering analysis is carried out on all users in each basic cluster set, a plurality of habit cluster sets of each basic label and the habit label, sleep cluster center vector, diet cluster center vector and exercise cluster center vector of each habit cluster set are determined, wherein the habit cluster set comprises a plurality of historical users.

4. The health management system based on traditional Chinese medicine and blood glucose monitoring according to claim 3, characterized in that, Construction module, comprising: The blood glucose event vector unit: feature extraction is performed on the blood glucose trend chart of each blood glucose event period in the historical blood glucose monitoring data in the historical user data of each historical user, and a blood glucose event vector of each blood glucose event period in the historical blood glucose monitoring data in the historical user data of each historical user is determined based on the period length and all the features extracted from each blood glucose event period; The blood glucose emotion label set unit: based on all the blood glucose event periods in the historical blood glucose monitoring data in the user historical data of all the historical users and the emotion label event set in a specified time period range before the blood glucose event period, a blood glucose emotion label set is determined; The direct influence data unit: based on the blood glucose emotion label set, the historical blood glucose monitoring data of all the historical users in the habit cluster set of each habit label of each basic label, and the historical blood glucose influence data, direct influence data of each habit label of each basic label is determined; The prediction influence model unit: all the historical blood glucose monitoring data of all the historical users in the habit cluster set of each habit label of each basic label and the direct influence data are taken as the input of the prediction influence model, and the physiological signal vector of each Zangfu of all the blood glucose event periods in the historical blood glucose influence data of all the historical users in the habit cluster set of each habit label of each basic label is taken as the output of the prediction influence model, and a prediction influence model of each Zangfu of each habit label of each basic label is constructed.

5. The health management system based on traditional Chinese medicine and blood glucose monitoring according to claim 4, characterized in that, The direct influence data unit includes: The emotion influence value sub-unit: based on the blood glucose emotion label set, the environment feature vector in a specified time period range before each blood glucose event period in the historical blood glucose monitoring data of each historical user in the habit cluster set of each habit label of each basic label, the emotion label event set, the record time of each emotion label in the emotion label event set, and the emotion intensity, the emotion influence value of each blood glucose event period of each historical user in the habit cluster set of each habit label of each basic label on each emotion label in the blood glucose emotion label set is calculated; The first and second direct influence value sub-unit: based on the blood glucose emotion label set, the blood glucose event vector of all the blood glucose event periods in the historical blood glucose monitoring data of all the historical users in the habit cluster set of each habit label of each basic label, the emotion influence value of each blood glucose event period of each historical user in the habit cluster set of each habit label of each basic label on each emotion label in the blood glucose emotion label set, the first direct influence value of each blood glucose feature in the blood glucose event vector by each emotion label in the blood glucose emotion label set of each habit label of each basic label is calculated, and the second direct influence value of each environment feature in the environment feature vector of each habit label of each basic label on each blood glucose feature in the blood glucose event vector is calculated; The third direct influence value subunit: performing correlation analysis on the blood glucose event vector of all blood glucose event periods in the historical blood glucose monitoring data of all historical users in the habit cluster set of each habit tag of each basic tag and the physiological signal vector of each zang organ of all blood glucose event periods in the historical blood glucose influence data, to determine the third direct influence value of each blood glucose feature in the blood glucose event vector of each habit tag of each basic tag on each signal feature in the physiological signal vector of each zang organ; The direct influence data subunit: based on the first direct influence value of all emotional tags in the blood glucose emotional tag set of each habit tag of each basic tag on all blood glucose features in the blood glucose event vector, the second direct influence value of all environmental features in the environmental feature vector on all blood glucose features in the blood glucose event vector, and the third direct influence value of all blood glucose features in the blood glucose event vector on all signal features in the physiological signal vector of each zang organ, determine the direct influence data of each habit tag of each basic tag.

6. The health management system based on traditional Chinese medicine and blood glucose monitoring according to claim 4, characterized in that, The determination module comprises: The management user data unit: obtaining the management user data of the user to be managed, wherein the management user data comprises user basic data, sleep habit vector, diet habit vector, exercise habit vector, blood glucose treatment data, and zang organ condition data of each zang organ, and the blood glucose treatment data comprises at least one or more drugs or supplements; The acquisition unit: based on the environmental intelligent sensor group, acquiring the environmental data of the user to be managed, and based on the emotional sensor group, acquiring the user emotional data of the user to be managed within a specified time period, wherein the user emotion set comprises at least one or more emotional tags and the recording time of each emotional tag; The blood glucose monitoring data unit: based on the microneedle, collecting the blood glucose monitoring data of the user to be managed, and based on the specified time period, extracting the blood glucose monitoring data of the user to be managed to determine the real-time monitoring data of the user to be managed.

7. The health management system based on traditional Chinese medicine and blood glucose monitoring according to claim 6, characterized in that, The management module comprises: The user basic tag unit: determining the user basic tag of the user to be managed based on the user basic data in the management user data of the user to be managed and the basic cluster set of the basic cluster center vector of all basic tags; The user habit tag unit: determining the user habit tag of the user to be managed based on the sleep habit vector, the diet habit vector, the exercise habit vector of the user to be managed, and the sleep cluster center vector, the diet cluster center vector, and the exercise cluster center vector of the habit cluster set of each habit tag of the basic tag corresponding to the user basic tag of the user to be managed; The prediction signal vector unit: inputting the environmental data, the user emotional data, the blood glucose treatment data, and the real-time monitoring data of the user to be managed into the prediction influence model of each zang organ corresponding to the user basic tag and the user habit tag of the user to be managed, and determining the prediction signal vector of each zang organ of the user to be managed based on the prediction influence model of each zang organ; Health management recommendation data unit: input all zangfu condition data and prediction signal vector of the management user data of the user to be managed into the traditional Chinese medicine management model, determine the health management recommendation data of the user to be managed based on the traditional Chinese medicine management model, wherein the health management recommendation data includes a zangfu influence set and multiple health recommendations of each zangfu in the zangfu influence set.

8. A health management method based on traditional Chinese medicine and blood glucose monitoring, characterized in that, A health management system based on traditional Chinese medicine and blood glucose monitoring according to any one of claims 1-7, comprising: S1: obtaining user historical data of a plurality of historical users, analyzing the user historical data of each historical user, determining a plurality of basic cluster sets and a plurality of habit cluster sets of each basic cluster set; S2: determining direct influence data of each habit label of each basic label based on the user historical data of all historical users and the habit cluster sets of all basic cluster sets, and constructing a prediction influence model of each zangfu of each habit label of each basic label; S3: obtaining management user data, environment data and user emotion data of a user to be managed, collecting blood glucose monitoring data of the user to be managed, determining user basic labels and user habit labels of the user to be managed based on the management user data of the user to be managed and the habit cluster sets of all basic cluster sets; S4: determining a prediction signal vector of each zangfu of the user to be managed based on the management user data of the user to be managed, the user basic labels, the user habit labels, the prediction influence model of all zangfu of all habit labels of all basic labels, and the traditional Chinese medicine management model, and determining health management recommendation data of the user to be managed.

Images