AI health steward data intelligent management method and system

Abstract

The invention provides an AI health steward data intelligent management method and system, and relates to the technical field of data intelligent management. A user health model is constructed, data filling is performed on the user health model, monitoring block division is performed on the user health model, model monitoring blocks are obtained, and data health coefficients of the model monitoring blocks are calculated. Obtaining block health judgment information according to the data health coefficient; historical associated block information and a newly added abnormal block of the user health model are obtained, associated blocks are extracted, abnormal associated blocks are obtained, associated health coefficients are calculated, associated anomaly judgment is carried out, and block associated anomaly judgment information is obtained; the block cumulative change coefficient is obtained and compared with a preset block cumulative change threshold value, the associated collection monitoring frequency is adjusted, and comprehensive, accurate and timely evaluation management and early warning of the health state of the user are achieved.

Description

Technical Field

[0001] The present invention provides an AI health steward data intelligent management method and system, which relates to the technical field of data intelligent management, specifically to the technical field of AI health steward data intelligent management. Background Art

[0002] Traditional health management mainly relies on regular physical examinations and doctor diagnoses, but this method has problems such as long time intervals, discontinuous data, and lagging health status feedback. Existing user health management systems usually collect users' health data in real-time or at regular intervals through various health collection devices (such as wearable devices, intelligent medical devices, etc.), such as heart rate, blood pressure, blood sugar, sleep quality, etc. In actual applications, these systems face problems such as complex data processing, inaccurate health status assessment, and imperfect warning mechanisms. Traditional health data management methods are difficult to monitor and manage users' health data in a targeted and relevant manner, and it is difficult to accurately identify and analyze changes in health data. Summary of the Invention

[0003] The present invention provides an AI health steward data intelligent management method and system to solve the above problems:

[0004] An AI health steward data intelligent management method and system proposed by the present invention, the method includes

[0005] S1. Construct a user health model, collect health data, perform data processing and classification to obtain health classification data, and fill the user health model with the health classification data to obtain health model filled data;

[0006] S2. Divide the monitoring blocks of the user health model to obtain model monitoring blocks, calculate the data health coefficients of the model monitoring blocks, and obtain block health determination information according to the data health coefficients;

[0007] S3. Obtain historical associated block information and newly added abnormal blocks of the user health model, extract associated blocks to obtain abnormal associated blocks, calculate associated health coefficients, perform associated abnormal determination, and obtain block associated abnormal determination information;

[0008] S4. Perform block warning and block association warning according to the determination information, obtain the block cumulative change coefficient, compare it with the preset block cumulative change threshold, and adjust the associated collection monitoring frequency.

[0009] Further, the S1 includes:

[0010] Obtain the user's historical health information, and generate a user health model according to the user's historical health information;

[0011] Collect health data through a health collection device to obtain device health collection data;

[0012] Integrate the device health collection data through a preset relay device node to obtain health integration data;

[0013] Preprocess the health integration data to obtain preprocessed health integration data;

[0014] Obtain a preset health data category, and classify the preprocessed health integration data according to the health data category to obtain health classification data;

[0015] Fill the user health model with new data through the health classification data to obtain health model filled data.

[0016] Further, the S2 includes:

[0017] Divide the monitoring blocks of the user health model according to the health model filled data to obtain model monitoring blocks;

[0018] Calculate the data health coefficient of the model monitoring blocks through the health model filled data;

[0019] Compare the data health coefficient with a preset data health threshold to obtain a data health comparison result;

[0020] Determine the regional data health of the model monitoring blocks corresponding to the data health coefficient according to the data health comparison result to obtain block data health determination information.

[0021] Further, the S3 includes:

[0022] Extract historical health associated block information according to the user's historical health information to obtain historical associated block information;

[0023] Obtain new block health determination information, and according to the new block health determination information, obtain the new abnormal blocks of the user health model;

[0024] Extract the associated blocks of the new abnormal blocks according to the historical associated block information to obtain associated extraction blocks;

[0025] Establish an abnormal association relationship for the associated extraction blocks to obtain abnormal associated blocks;

[0026] Calculate the associated health coefficient of the abnormal associated blocks according to the data health coefficient;

[0027] Compare the associated health coefficient with a preset associated health threshold to obtain an associated health comparison result;

[0028] Perform an associated anomaly determination on the abnormal associated block according to the associated health comparison result, and obtain the block association anomaly determination information.

[0029] Further, the S4 includes:

[0030] Perform block warning according to the block data health determination information;

[0031] Perform block association warning according to the block association anomaly determination information;

[0032] Obtain the block cumulative change coefficient of the data health coefficient of the model monitoring block, compare the block cumulative change coefficient with the preset block cumulative change threshold, and obtain the block cumulative comparison result;

[0033] Adjust the block collection monitoring frequency according to the block cumulative comparison result;

[0034] Obtain the associated cumulative change coefficient of the data health coefficient of the abnormal associated block, compare the associated cumulative change coefficient with the preset associated cumulative change threshold, and obtain the associated cumulative comparison result;

[0035] Adjust the associated collection monitoring frequency according to the associated cumulative comparison result.

[0036] Further, the system includes:

[0037] A model filling module, which is used to construct a user health model, collect health data, perform data processing and classification, obtain health classification data, and fill the user health model with the health classification data to obtain health model filling data;

[0038] A block health determination module, which is used to divide the monitoring blocks of the user health model to obtain the model monitoring blocks, calculate the data health coefficient of the model monitoring blocks, and obtain the block health determination information according to the data health coefficient;

[0039] An associated health determination module, which is used to obtain the historical associated block information and the newly added abnormal blocks of the user health model, extract the associated blocks, obtain the abnormal associated blocks, calculate the associated health coefficient, perform an associated anomaly determination, and obtain the block association anomaly determination information;

[0040] A warning adjustment module, which is used to perform block warning and block association warning according to the determination information, obtain the block cumulative change coefficient, compare it with the preset block cumulative change threshold, and adjust the associated collection monitoring frequency.

[0041] Further, the model filling module includes:

[0042] A model construction module, configured to obtain the user's historical health information and generate a user health model according to the user's historical health information;

[0043] A data collection and integration module, configured to collect health data through a health collection device to obtain device health collection data;

[0044] Integrate the device health collection data through a preset relay device node to obtain health integration data;

[0045] Preprocess the health integration data to obtain preprocessed health integration data;

[0046] A data classification and filling module, configured to obtain preset health data categories, classify the preprocessed health integration data through the health data categories to obtain health classification data;

[0047] Fill the user health model with new data through the health classification data to obtain health model filled data.

[0048] Further, the block health determination module includes:

[0049] A block division module, configured to divide monitoring blocks of the user health model according to the health model filled data to obtain model monitoring blocks;

[0050] A block health analysis module, configured to calculate the data health coefficient of the model monitoring blocks through the health model filled data;

[0051] Compare the data health coefficient with a preset data health threshold to obtain a data health comparison result;

[0052] Perform regional data health determination on the model monitoring blocks corresponding to the data health coefficient according to the data health comparison result to obtain block data health determination information.

[0053] Further, the associated health determination module includes:

[0054] An associated block division module, configured to extract historical health associated block information according to the user's historical health information to obtain historical associated block information;

[0055] Obtain new block health determination information, and according to the new block health determination information, obtain new abnormal blocks of the user health model;

[0056] Extract associated blocks for the new abnormal blocks according to the historical associated block information to obtain associated extraction blocks;

[0057] Establish an abnormal association relationship for the associated extraction blocks to obtain abnormal associated blocks;

[0058] An associated block analysis module, configured to calculate an associated health coefficient of an abnormal associated block according to a data health coefficient;

[0059] Compare the associated health coefficient with a preset associated health threshold to obtain an associated health comparison result;

[0060] Perform an associated anomaly determination on the abnormal associated block according to the associated health comparison result, and obtain block association anomaly determination information.

[0061] Further, the warning adjustment module includes:

[0062] A warning module, configured to perform block warning according to block data health determination information;

[0063] Perform block association warning according to block association anomaly determination information;

[0064] A change monitoring module, configured to obtain a block cumulative change coefficient of the data health coefficient of the model monitoring block, compare the block cumulative change coefficient with a preset block cumulative change threshold to obtain a block cumulative comparison result;

[0065] Adjust the block collection monitoring frequency according to the block cumulative comparison result;

[0066] A comparison adjustment module, configured to obtain an associated cumulative change coefficient of the data health coefficient of the abnormal associated block, compare the associated cumulative change coefficient with a preset associated cumulative change threshold to obtain an associated cumulative comparison result;

[0067] Adjust the associated collection monitoring frequency according to the associated cumulative comparison result.

[0068] Advantages of the present invention: By constructing a user health model, personalized health management is realized, and customized suggestions and warnings can be provided for the health status of specific areas of users. Through the method of the present invention, targeted monitoring of health data of specific areas and associated areas of users can be carried out, which can not only realize targeted management of health data, but also realize flexible management of the relevance of construction data. Through continuous monitoring and data analysis, potential health data problems can be detected early. By automatically adjusting the monitoring frequency, dynamic collection and analysis of health data are realized, and the efficiency and accuracy of health management are improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] Figure 1 It is a schematic diagram of an AI health steward data intelligent management method. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0070] 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 only for the purpose of illustrating and explaining the present invention, and are not intended to limit the present invention.

[0071] In one embodiment of the present invention, a method and system for intelligent management of AI health steward data are proposed. The method includes

[0072] S1. Construct a user health model, collect health data, perform data processing and classification to obtain health classification data, and fill the user health model with the health classification data to obtain health model filled data;

[0073] S2. Divide the monitoring blocks of the user health model to obtain model monitoring blocks, calculate the data health coefficient of the model monitoring blocks, and obtain block health determination information according to the data health coefficient;

[0074] S3. Obtain historical associated block information and new abnormal blocks of the user health model, extract associated blocks to obtain abnormal associated blocks, calculate the associated health coefficient, perform associated abnormal determination, and obtain block associated abnormal determination information;

[0075] S4. Perform block warning and block association warning according to the determination information, obtain the block cumulative change coefficient, compare it with the preset block cumulative change threshold, and adjust the associated collection monitoring frequency.

[0076] The working principle of the above technical solution is as follows: Based on the user's basic information (such as age, gender, height, and weight, etc.) and existing health data (such as physical examination reports, daily health monitoring data, etc.), a basic user health model is constructed. The user health model includes a three-dimensional electronic model of the human body containing the above basic information and existing health data. Through channels such as wearable devices and health APPs, health data of the user are continuously collected, including but not limited to heart rate, blood pressure, blood sugar, sleep quality, etc. The collected health data are processed such as cleaning, denoising, and standardization to improve the accuracy and comparability of the data. The processed data are classified according to different health indicators, such as cardiovascular health, endocrine health, nervous system health, etc. (classified according to the human body system area) to form health classification data. According to the health classification data, the user health model is filled to update the health status of the model and form health model filling data. The user health model is divided into different monitoring blocks, and each block represents a specific health area. Calculate the data health coefficient of each monitoring block, which reflects the overall health status of the health data within the block. According to the data health coefficient, obtain the block health determination information to judge whether there are health data problems in each block. Extract historical associated block information, that is, the blocks that were associated with or affected each other with the current block in the past. Compare the current health model with historical data to identify newly added abnormal blocks. Conduct abnormal determination on the associated blocks, calculate the associated health coefficient, and determine whether there is an associated abnormality. According to the block health determination information and the associated abnormal determination information, trigger corresponding block warnings and block association warnings. Calculate the block cumulative change coefficient, which reflects the changing trend of the block health status over time. Compare the block cumulative change coefficient with a preset block cumulative change threshold, and adjust the associated acquisition monitoring frequency according to the comparison result to more accurately track the user's health status.

[0077] The technical effects of the above technical solution are as follows: By constructing a user health model, personalized health management is achieved, and customized suggestions and warnings can be provided for the health status of specific areas of the user. Through the method of the present invention, targeted monitoring of health data can be carried out for specific areas and associated areas of the user, which can not only achieve targeted management of health data but also achieve flexible management of the relevance of health data. Through continuous monitoring and data analysis, potential health data problems can be detected early. By automatically adjusting the monitoring frequency, dynamic acquisition and analysis of health data are realized, and the efficiency and accuracy of health management are improved.

[0078] In one embodiment of the present invention, S1 includes:

[0079] Obtain the user's historical health information and generate a user health model according to the user's historical health information;

[0080] Collect health data through a health collection device to obtain device health collection data;

[0081] Integrate the device health collection data through a preset relay device node to obtain health integration data;

[0082] Preprocess the health integration data to obtain preprocessed health integration data;

[0083] Obtain a preset health data category, and classify the preprocessed health integration data according to the health data category to obtain health classification data;

[0084] Fill the user health model with new data through the health classification data to obtain health model filled data.

[0085] The working principle of the above technical solution is as follows: Health information is collected from various sources such as the user's historical health records, physical examination reports, and medical diagnoses. This information is integrated and analyzed to form an initial user health model. The model contains key data such as the user's physiological indicators, disease history, allergy history, and family medical history. The user wears or uses health collection devices (such as smart bracelets, blood pressure monitors, blood glucose meters, etc.) to monitor and record health data in real time. These devices transmit the collected data to the system regularly or on demand. Data from different health collection devices is received through preset relay device nodes. These nodes may be located in the user's home, medical institutions, or cloud servers, and are responsible for summarizing and preliminarily processing the data. Operations such as deduplication and timestamp alignment may be performed during the integration process to ensure the accuracy and consistency of the data. The integrated health data contains noise, missing values, or outliers and needs to be preprocessed. The preprocessing steps include data cleaning (removing invalid or incorrect data), data imputation (filling in missing values), data transformation (such as standardization, normalization), etc. The system classifies the preprocessed data according to preset health data categories (such as heart rate, blood pressure, blood glucose, sleep quality, etc.). These categories reflect different health indicators and concerns. The classified health data is used to update and fill the user health model. The new data may reflect changes in the user's health status or newly emerging health problems. Through continuous data filling, the user health model can maintain real-time and accuracy.

[0086] The technical effects of the above technical solution are as follows: By integrating the historical and real-time health data of users, the system can provide users with personalized health management solutions and suggestions. Through the preset data integration and preprocessing steps of the relay device nodes, the system realizes the unified management and standardized processing of health data, improving the availability and accuracy of the data. The real-time collection and classification of health data enable the system to promptly detect health abnormalities of users and trigger an early warning mechanism to remind users or medical institutions to take corresponding measures. As time goes by, the data in the user health model accumulates continuously, and the system can conduct trend analysis on this data to help users understand the changes and development trends of their own health conditions. Through real-time monitoring and early warning, the system can guide users to rationally use medical resources, reduce unnecessary medical examinations and hospital visits, thereby reducing medical costs and improving medical efficiency.

[0087] In one embodiment of the present invention, S2 includes:

[0088] Dividing the monitoring blocks of the user health model according to the data filled in the health model to obtain model monitoring blocks;

[0089] Calculating the data health coefficient of the model monitoring block through the data filled in the health model;

[0090] The calculation formula of the data health coefficient is:

[0091]

[0092] Where QK is the data health coefficient of the model monitoring block, X is the data point filled in the health model of the model monitoring block, u is the mean value of the historical data points filled in the health model of the model monitoring block, e is the standard deviation, is the degree of deviation of the data point from the mean value. If X is within [a, b], the abnormal coefficient is 0 (indicating that the data point is within the preset range and there is no abnormality). If X is outside [a, b], then according to calculate the data health coefficient, is the factor of the degree of exceeding the range, BJ is the range boundary, is half of the range width. The "range boundary" can be a or b, specifically depending on whether X is lower than a or higher than b;

[0093] Comparing the data health coefficient with a preset data health threshold to obtain a data health comparison result;

[0094] Judging the regional data health of the model monitoring block corresponding to the data health coefficient according to the data health comparison result to obtain block data health judgment information.

[0095] When the data health coefficient is greater than the preset data health threshold, it is determined that the regional data health is an abnormal block. Otherwise, it is a normal block.

[0096] The working principle of the above technical solution is: fill in data based on the user health model and divide it into multiple monitoring blocks. These blocks correspond to different physiological systems (such as cardiovascular system, respiratory system, digestive system, etc.); the purpose of dividing the monitoring blocks is to monitor and manage the user's health status more finely and ensure that each key area receives sufficient attention. For each monitoring block, the system uses the relevant indicator values ​​in the health model to fill in the data and calculate a data health coefficient. Used to standardize data points, indicating the degree to which the data point deviates from the mean. The out-of-range degree factor is an adjustment factor used to further amplify or reduce the abnormal coefficient to reflect the degree to which the data point exceeds the preset range. If X is within [a, b], the abnormal coefficient is 0 (indicating that the data point is within the preset range and there is no abnormality). If X is outside [a, b], the abnormal coefficient is calculated based on the z-score and the out-of-range degree factor. The "range boundary" can be a or b, depending on whether X is lower than a or higher than b. This coefficient is a quantitative indicator used to reflect the current health status and risk level of the block. The calculated data health coefficient is compared with the preset data health thresholds. These thresholds are derived from medical standards, expert advice, or historical data analysis to distinguish between different states such as health, sub-health, and disease. The results of the comparison will determine whether the monitoring block needs further attention or intervention measures. Based on the results of the data health comparison, the system makes a regional data health judgment for each monitoring block. This includes determining the health status of the block (such as normal, abnormal, etc.) and possible causes or risk factors.

[0097] The technical effect of the above technical solution is: by dividing the health status into multiple monitoring blocks, the system can provide more refined health management services and give personalized suggestions and measures for the specific situation of each block. The calculation and comparison of the data health coefficient enables the system to detect health abnormalities in time and trigger the early warning mechanism before the problem worsens. The system can provide users with health risk assessment services based on the data health coefficient and judgment information. Through regular monitoring and feedback, the system can motivate users to pay attention to their own health. Through accurate health monitoring and early warning, the system can guide users to make rational use of medical resources, reduce unnecessary medical examinations and visits, thereby reducing medical costs and improving medical efficiency. At the same time, for users who really need medical attention, the system can provide timely guidance and support.

[0098] In one embodiment of the present invention, S3 includes:

[0099] Extract historical health-related block information based on the user's historical health information to obtain historical related block information;

[0100] Obtain the health determination information of the newly added block, and based on the health determination information of the newly added block, obtain the newly added abnormal block of the user health model;

[0101] Extract the associated blocks of the newly added abnormal block according to the historical associated block information to obtain the associated extracted blocks;

[0102] Establish an abnormal association relationship for the associated extracted blocks to obtain abnormal associated blocks; the abnormal associated blocks include information such as the association relationships of multiple abnormal blocks.

[0103] Calculate the association health coefficient of the abnormal associated blocks according to data such as the data health coefficient;

[0104] The calculation formula of the association health coefficient is:

[0105]

[0106] Among them, GK is the association health coefficient, n is the total number of blocks of the abnormal associated blocks, QK i is the data health coefficient of the i-th abnormal block in the abnormal associated blocks, W i is the preset weight value of the i-th abnormal block in the abnormal associated blocks, and α is a preset adjustment factor used to refine or correct the calculation of the association health coefficient;

[0107] Compare the association health coefficient with the preset association health threshold to obtain the association health comparison result;

[0108] Perform an association abnormality determination on the abnormal associated blocks according to the association health comparison result to obtain the block association abnormality determination information.

[0109] When the association health coefficient is greater than the preset association health threshold, determine that the abnormal associated block is an associated abnormal block; otherwise, it is an associated normal block. The associated abnormal blocks include the abnormal associated blocks determined to be abnormally associated.

[0110] The working principle of the above technical solution is as follows: Based on the user's historical health information, block information associated with the user's health status is identified and extracted. These historically associated block information may include past health problems, medical history, long-term trends of physiological indicators, etc. The purpose of extraction is to establish a historical baseline of the user's health status. The latest block health determination information is obtained, which reflects the latest status of the user's current health condition. By comparing the newly added block health determination information with the historical baseline, the system identifies the newly added abnormal blocks. For each newly added abnormal block, the system uses the historically associated block information to further extract other blocks associated with it. These associated blocks may be related to the newly added abnormal block physiologically, functionally, or in terms of the disease development path. The system establishes abnormal association relationships based on the extracted associated blocks to form abnormal associated blocks. The relationships between these blocks reflect the potential and complex interactions in the user's health condition. The system combines the data health coefficient, historical health information, and other relevant data to calculate the associated health coefficient for each abnormal associated block. This coefficient is used to quantify the overall health condition or risk level of the abnormal associated block. The calculated associated health coefficient is compared with a preset associated health threshold. Based on the comparison result, the system makes an associated abnormality determination for the abnormal associated block to determine whether there are health problems or risks. According to the result of the associated abnormality determination, the system generates block associated abnormality determination information. These information include the detailed information of the abnormal associated block, possible causes, recommended intervention measures, etc.

[0111] The technical effects of the above technical solution are as follows: By combining historical health information and newly added block health determination information, the system can more accurately identify abnormal changes in the user's health condition, improving the accuracy and comprehensiveness of health monitoring. According to the individual differences and historical health conditions of the user, personalized health management suggestions and intervention measures are provided, enhancing the pertinence and effectiveness of health management. By establishing abnormal association relationships and calculating associated health coefficients, the system can timely detect potential risks in the user's health condition and take necessary intervention measures before the risks deteriorate, reducing the occurrence probability of health risks. Through precise health monitoring and abnormal determination, the system can guide users to rationally utilize medical resources, reduce unnecessary medical examinations and hospital visits, lower medical costs, and improve the utilization efficiency of medical resources. By providing timely, accurate, and personalized health management services, the system can enhance the user's health awareness and health management level, thereby enhancing the user's experience and satisfaction.

[0112] In one embodiment of the present invention, S4 includes:

[0113] Conduct block warning according to the block data health determination information;

[0114] Conduct block association warning according to the block association abnormal determination information;

[0115] Obtain the block cumulative change coefficient of the data health coefficient of the model monitoring block, compare the block cumulative change coefficient with a preset block cumulative change threshold, and obtain a block cumulative comparison result;

[0116] Adjust the block acquisition monitoring frequency according to the block cumulative comparison result;

[0117] Obtain the associated cumulative change coefficient of the data health coefficient of the abnormally associated block, compare the associated cumulative change coefficient with a preset associated cumulative change threshold, and obtain an associated cumulative comparison result;

[0118] Adjust the associated acquisition monitoring frequency according to the associated cumulative comparison result.

[0119] The working principle of the above technical solution is as follows: According to the block data health determination information, evaluate the data health status of each block. If it is found that the block data health status is lower than the preset health standard, trigger the block warning mechanism to notify relevant personnel or systems to take corresponding measures. The system further analyzes the association relationship between blocks, and identifies whether there are abnormal block associations through the block association abnormality determination information. If it is found that there are abnormal associations between blocks, trigger the block association warning to facilitate timely handling of these association abnormalities. Obtain the block cumulative change coefficient of the data health coefficient of the model monitoring block, and this coefficient reflects the change trend of the block data health status over time. Compare the block cumulative change coefficient with the preset block cumulative change threshold to obtain the block cumulative comparison result. According to the block cumulative comparison result, the system dynamically adjusts the acquisition monitoring frequency of the block. If the block data health status changes greatly, it may be necessary to increase the monitoring frequency to more accurately track the data changes; conversely, if the data is stable, the monitoring frequency can be reduced to save resources. The system also obtains the associated cumulative change coefficient of the data health coefficient of the abnormally associated block, and this coefficient reflects the change trend of the data health status of the associated block over time. Compare the associated cumulative change coefficient with the preset associated cumulative change threshold to obtain the associated cumulative comparison result. According to the associated cumulative comparison result, the system dynamically adjusts the acquisition monitoring frequency of the associated block. If the data health status of the associated block changes greatly, it may be necessary to increase the monitoring frequency of the associated block to better understand and handle the association abnormalities; conversely, if the associated data is stable, the monitoring frequency can be reduced.

[0120] The technical effects of the above technical solution are as follows: By analyzing and monitoring the health status and correlation of block data in real time, the system can timely detect potential problems and trigger warnings, thereby improving the accuracy and timeliness of warnings. The system can dynamically adjust the acquisition and monitoring frequency according to the changing trend of the health status of blocks and related blocks, thereby optimizing resource allocation while ensuring data quality and reducing unnecessary monitoring overhead. By timely detecting and handling health and correlation anomalies of block data, the system can avoid system instability or crashes caused by data errors or inconsistencies, thereby improving the overall stability and reliability of the system. The real-time warning and dynamic adjustment functions provided by the system provide timely and accurate data support for relevant personnel, thereby improving decision-making efficiency and accuracy.

[0121] In an embodiment of the present invention, the system includes:

[0122] A model filling module, configured to construct a user health model, collect health data, perform data processing and classification, obtain health classification data, and fill the user health model with the health classification data to obtain health model filling data;

[0123] A block health determination module, configured to divide the monitored blocks of the user health model to obtain model monitored blocks, calculate the data health coefficient of the model monitored blocks, and obtain block health determination information according to the data health coefficient;

[0124] An associated health determination module, configured to obtain historical associated block information and newly added abnormal blocks of the user health model, extract associated blocks, obtain abnormally associated blocks, calculate an associated health coefficient, perform associated anomaly determination, and obtain block associated anomaly determination information;

[0125] A warning adjustment module, configured to perform block warnings and block association warnings according to the determination information, obtain a block cumulative change coefficient, compare it with a preset block cumulative change threshold, and adjust the associated acquisition and monitoring frequency.

[0126] The working principle of the above technical solution is as follows: Based on the user's basic information (such as age, gender, height, and weight, etc.) and existing health data (such as physical examination reports, daily health monitoring data, etc.), a basic user health model is constructed. The user health model includes a three-dimensional electronic model of the human body containing the above basic information and existing health data. Through channels such as wearable devices and health APPs, health data of the user is continuously collected, including but not limited to heart rate, blood pressure, blood sugar, sleep quality, etc. The collected health data is processed such as cleaning, denoising, and standardization to improve the accuracy and comparability of the data. The processed data is classified according to different health indicators, such as cardiovascular health, endocrine health, nervous system health, etc. (classified according to the human body system area) to form health classification data. According to the health classification data, the user health model is filled to update the health status of the model and form health model filling data. The user health model is divided into different monitoring blocks, and each block represents a specific health area. The data health coefficient of each monitoring block is calculated, and this coefficient reflects the overall health status of the health data within the block. According to the data health coefficient, block health determination information is obtained to determine whether there are health data problems in each block. Historical associated block information is extracted, that is, blocks that were associated with or influenced the current block in the past. The current health model is compared with historical data to identify newly added abnormal blocks. Abnormal determination is performed on the associated blocks, the associated health coefficient is calculated, and it is determined whether there is an associated abnormality. According to the block health determination information and the associated abnormality determination information, corresponding block warnings and block association warnings are triggered. The block cumulative change coefficient is calculated, and this coefficient reflects the change trend of the block health status over time. The block cumulative change coefficient is compared with a preset block cumulative change threshold, and according to the comparison result, the associated collection monitoring frequency is adjusted to more accurately track the user's health status.

[0127] The technical effects of the above technical solution are as follows: By constructing a user health model, personalized health management is achieved, and customized suggestions and warnings can be provided for the health status of specific areas of the user. Through the method of the present invention, targeted monitoring of health data for specific areas and associated areas of the user can be carried out, which can not only achieve targeted management of health data but also achieve flexible management of the relevance of health data. Through continuous monitoring and data analysis, potential health data problems can be detected early. By automatically adjusting the monitoring frequency, dynamic collection and analysis of health data are achieved, improving the efficiency and accuracy of health management.

[0128] In an embodiment of the present invention, the model filling module includes:

[0129] A model construction module, configured to obtain the user's historical health information and generate a user health model according to the user's historical health information;

[0130] A data acquisition and integration module for collecting health data through a health collection device to obtain device health collection data;

[0131] Integrating the device health collection data through a preset relay device node to obtain health integration data;

[0132] Preprocessing the health integration data to obtain preprocessed health integration data;

[0133] A data classification and filling module for obtaining preset health data categories, classifying the preprocessed health integration data according to the health data categories to obtain health classification data;

[0134] Filling new data into the user health model through the health classification data to obtain health model filling data.

[0135] The working principle of the above technical solution is as follows: Health information is collected from various sources such as the user's historical health records, physical examination reports, and medical diagnoses. This information is integrated and analyzed to form an initial user health model. The model contains key data such as the user's physiological indicators, disease history, allergy history, and family medical history. The user monitors and records health data in real time through a worn or used health collection device (such as a smart bracelet, blood pressure monitor, blood glucose meter, etc.). These devices transmit the collected data to the system regularly or on demand. Data from different health collection devices is received through preset relay device nodes. These nodes may be located in the user's home, medical institution, or cloud server and are responsible for aggregating and preliminarily processing the data. Operations such as deduplication and timestamp alignment may be performed during the integration process to ensure the accuracy and consistency of the data. The integrated health data contains noise, missing values, or outliers and needs to be preprocessed. The preprocessing steps include data cleaning (removing invalid or incorrect data), data imputation (filling in missing values), data transformation (such as standardization, normalization), etc. The system classifies the preprocessed data according to preset health data categories (such as heart rate, blood pressure, blood glucose, sleep quality, etc.). These categories reflect different health indicators and concerns. The classified health data is used to update and fill the user health model. The new data may reflect changes in the user's health status or newly emerging health problems. Through continuous data filling, the user health model can maintain real-time performance and accuracy.

[0136] The technical effects of the above technical solution are as follows: By integrating the historical and real-time health data of users, the system can provide personalized health management solutions and suggestions for users. Through the preset data integration and preprocessing steps of the relay device nodes, the system realizes the unified management and standardized processing of health data, improving the availability and accuracy of the data. The real-time collection and classification of health data enable the system to promptly detect health abnormalities of users and trigger an early warning mechanism to remind users or medical institutions to take corresponding measures. As time goes by, the data in the user health model accumulates continuously, and the system can perform trend analysis on this data to help users understand the changes and development trends of their own health conditions. Through real-time monitoring and early warning, the system can guide users to rationally use medical resources, reduce unnecessary medical examinations and hospital visits, thereby reducing medical costs and improving medical efficiency.

[0137] In one embodiment of the present invention, the block health determination module includes:

[0138] A block division module, configured to monitor and divide the user health model according to the data filled in the health model to obtain a model monitoring block;

[0139] A block health analysis module, configured to calculate the data health coefficient of the model monitoring block through the data filled in the health model;

[0140] The calculation formula of the data health coefficient is:

[0141]

[0142] Where QK is the data health coefficient of the model monitoring block, X is the data point filled in the health model of the model monitoring block, u is the mean value of the historical data points filled in the health model of the model monitoring block, e is the standard deviation, is the degree of deviation of the data point from the mean value. If X is within [a, b], the abnormal coefficient is 0 (indicating that the data point is within the preset range and there is no abnormality). If X is outside [a, b], then according to calculate the data health coefficient, is the factor of the degree of exceeding the range, BJ is the range boundary, is half of the range width. The range boundary can be a or b, specifically depending on whether X is lower than a or higher than b;

[0143] Compare the data health coefficient with a preset data health threshold to obtain a data health comparison result;

[0144] Perform regional data health determination on the model monitoring block corresponding to the data health coefficient according to the data health comparison result to obtain block data health determination information.

[0145] The working principle of the above technical solution is as follows: Based on the user health model, data is filled and divided into multiple monitoring blocks. These blocks correspond to different physiological systems (such as the cardiovascular system, respiratory system, digestive system, etc.); the purpose of dividing the monitoring blocks is to more finely monitor and manage the user's health status to ensure that sufficient attention is paid to each key area. For each monitoring block, the system uses the relevant index values in the data filled by the health model to calculate a data health coefficient. This coefficient is a quantitative index used to reflect the current health status and risk level of this block. The calculated data health coefficient is compared with a preset data health threshold. These thresholds are obtained based on medical standards, expert suggestions, or historical data analysis and are used to distinguish different states such as health, sub-health, and disease. The comparison result will determine whether this monitoring block requires further attention or intervention measures. According to the result of the data health comparison, the system makes a regional data health determination for each monitoring block. This includes determining the health status of the block (such as normal, warning, dangerous, etc.) and possible causes or risk factors.

[0146] The technical effects of the above technical solution are as follows: By dividing the health status into multiple monitoring blocks, the system can provide more refined health management services and give personalized suggestions and measures according to the specific conditions of each block. The calculation and comparison of the data health coefficient enable the system to detect health abnormalities in a timely manner and trigger an early warning mechanism before the problem deteriorates. The system can provide a health risk assessment service for users based on the data health coefficient and determination information. Through regular monitoring and feedback, the system can encourage users to pay attention to their own health. Through accurate health monitoring and early warning, the system can guide users to make reasonable use of medical resources, reduce unnecessary medical examinations and hospital visits, thereby reducing medical costs and improving medical efficiency. At the same time, for users who truly need medical attention, the system can provide timely guidance and support.

[0147] In one embodiment of the present invention, the associated health determination module includes:

[0148] An associated block division module, which is used to extract historical health associated block information according to the user's historical health information to obtain historical associated block information;

[0149] Obtain new block health determination information, and according to the new block health determination information, obtain the new abnormal blocks of the user health model;

[0150] Extract associated blocks for the new abnormal blocks according to the historical associated block information to obtain associated extraction blocks;

[0151] Establish an abnormal association relationship for the associated extraction blocks to obtain abnormal associated blocks;

[0152] The associated block analysis module is used to calculate the associated health coefficient of the abnormal associated block according to data such as the data health coefficient;

[0153] The calculation formula of the associated health coefficient is as follows:

[0154]

[0155] where GK is the associated health coefficient, n is the total number of blocks in the abnormal associated block, QK i is the data health coefficient of the i-th abnormal block in the abnormal associated block, and W i is the preset weight value of the i-th abnormal block in the abnormal associated block, and α is a preset adjustment factor used to refine or correct the calculation of the associated health coefficient;

[0156] Compare the associated health coefficient with the preset associated health threshold to obtain an associated health comparison result;

[0157] Perform an associated abnormality determination on the abnormal associated block according to the associated health comparison result, and obtain block associated abnormality determination information.

[0158] The working principle of the above technical solution is as follows: According to the user's historical health information, identify and extract block information associated with the user's health status. These historical associated block information may include past health problems, medical history, long-term trends of physiological indicators, etc. The purpose of extraction is to establish a historical baseline of the user's health status. Obtain the latest block health determination information, which reflects the latest status of the user's current health status. By comparing the newly added block health determination information with the historical baseline, the system identifies the newly added abnormal blocks. For each newly added abnormal block, the system uses the historical associated block information to further extract other blocks associated with it. These associated blocks may be related to the newly added abnormal blocks in terms of physiology, function, or disease development path. The system establishes an abnormal association relationship based on the extracted associated blocks to form an abnormal associated block. The relationship between these blocks reflects the potential and complex interactions in the user's health status. The system combines the data health coefficient, historical health information, and other relevant data to calculate the associated health coefficient of each abnormal associated block. This coefficient is used to quantify the overall health status or risk level of the abnormal associated block. Compare the calculated associated health coefficient with the preset associated health threshold. According to the comparison result, the system performs an associated abnormality determination on the abnormal associated block to determine whether there are health problems or risks. According to the associated abnormality determination result, the system generates block associated abnormality determination information. These information include the detailed information of the abnormal associated block, possible reasons, recommended intervention measures, etc.

[0159] The technical effects of the above technical solution are as follows: By combining historical health information and newly added block health determination information, the system can more accurately identify abnormal changes in the user's health status, improving the accuracy and comprehensiveness of health monitoring. According to the individual differences and historical health conditions of users, personalized health management suggestions and intervention measures are provided, enhancing the pertinence and effectiveness of health management. By establishing abnormal association relationships and calculating associated health coefficients, the system can timely detect potential risks in the user's health status and take necessary intervention measures before the risks deteriorate, reducing the occurrence probability of health risks. Through precise health monitoring and abnormal determination, the system can guide users to rationally utilize medical resources, reduce unnecessary medical examinations and clinic visits, lower medical costs, and improve the utilization efficiency of medical resources. By providing timely, accurate, and personalized health management services, the system can enhance users' health awareness and health management level, thereby enhancing users' experience and satisfaction.

[0160] In one embodiment of the present invention, the warning and adjustment module includes:

[0161] A warning module for performing block warning according to the block data health determination information;

[0162] Performing block association warning according to the block association abnormal determination information;

[0163] A change monitoring module for obtaining the block cumulative change coefficient of the data health coefficient of the model monitoring block, comparing the block cumulative change coefficient with a preset block cumulative change threshold to obtain a block cumulative comparison result;

[0164] Adjusting the block collection monitoring frequency according to the block cumulative comparison result;

[0165] A comparison and adjustment module for obtaining the associated cumulative change coefficient of the data health coefficient of the abnormal association block, comparing the associated cumulative change coefficient with a preset associated cumulative change threshold to obtain an associated cumulative comparison result;

[0166] Adjusting the associated collection monitoring frequency according to the associated cumulative comparison result.

[0167] The working principle of the above technical solution is as follows: Based on the block data health determination information, the data health status of each block is evaluated. If it is found that the data health status of a block is lower than the preset health standard, the block warning mechanism is triggered to notify relevant personnel or systems to take corresponding measures. The system further analyzes the association relationship between blocks, and uses the block association anomaly determination information to identify whether there are abnormal block associations. If abnormal associations are found between blocks, a block association warning is triggered to promptly handle these association anomalies. The block cumulative change coefficient of the data health coefficient of the block monitored by the model is obtained. This coefficient reflects the change trend of the block data health status over time. The block cumulative change coefficient is compared with the preset block cumulative change threshold to obtain the block cumulative comparison result. According to the block cumulative comparison result, the system dynamically adjusts the acquisition and monitoring frequency of the block. If the change in the block data health status is large, it may be necessary to increase the monitoring frequency to more accurately track the data changes; conversely, if the data is stable, the monitoring frequency can be reduced to save resources. The system also obtains the association cumulative change coefficient of the data health coefficient of the abnormally associated blocks. This coefficient reflects the change trend of the data health status of the associated blocks over time. The association cumulative change coefficient is compared with the preset association cumulative change threshold to obtain the association cumulative comparison result. According to the association cumulative comparison result, the system dynamically adjusts the acquisition and monitoring frequency of the associated blocks. If the data health status of the associated blocks changes significantly, it may be necessary to increase the monitoring frequency of the associated blocks to better understand and handle the association anomalies; conversely, if the associated data is stable, the monitoring frequency can be reduced.

[0168] The technical effects of the above technical solution are as follows: By analyzing and monitoring the block data health status and association relationship in real time, the system can promptly discover potential problems and trigger warnings, thereby improving the accuracy and timeliness of the warnings. The system can dynamically adjust the acquisition and monitoring frequency according to the change trends of the data health status of the blocks and the associated blocks, thereby optimizing resource allocation while ensuring data quality and reducing unnecessary monitoring overhead. By promptly discovering and handling block data health and association anomalies, the system can avoid system instability or crashes caused by data errors or inconsistencies, thereby improving the overall stability and reliability of the system. The real-time warning and dynamic adjustment functions provided by the system provide timely and accurate data support for relevant personnel, thereby enhancing the decision-making efficiency and accuracy.

[0169] Obviously, those skilled in the art can make various modifications and deformations to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and deformations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention also intends to include these modifications and deformations.

Claims

1. An AI health manager data intelligent management method, characterized in that: The method comprises S1. Build a user health model, collect health data, process and classify the data, obtain health classification data, fill the user health model with data according to the health classification data, and obtain health model filling data; S2. Divide the user health model into monitoring blocks, obtain model monitoring blocks, calculate the data health coefficient of the model monitoring blocks, and obtain block health determination information based on the data health coefficient; S3, obtain the historical associated block information and the newly added abnormal block of the user health model, extract the associated block, obtain the abnormal associated block, calculate the associated health coefficient, perform associated abnormality determination, and obtain the block associated abnormality determination information; S4. Perform block warning and block-related warning according to the judgment information, obtain the block cumulative change coefficient, compare it with the preset block cumulative change threshold, and adjust the associated acquisition monitoring frequency.

2. According to claim 1, an AI health manager data intelligent management method is characterized in that: The S1 includes: Acquire historical health information of the user, and generate a user health model according to the historical health information of the user; Collect health data through health collection equipment to obtain equipment health collection data; Integrate the device health collection data through a preset relay device node to obtain health integration data; Preprocessing the health integrated data to obtain preprocessed health integrated data; Obtaining a preset health data category, classifying the preprocessed health integrated data according to the health data category, and obtaining health classification data; The user health model is populated with new data through health classification data to obtain health model populated data.

3. According to claim 1, an AI health manager data intelligent management method is characterized in that: The S2 includes: Divide the user health model into monitoring blocks according to the health model filling data to obtain model monitoring blocks; Fill the data into the health model to calculate the data health coefficient of the model monitoring block; Comparing the data health coefficient with a preset data health threshold to obtain a data health comparison result; According to the data health comparison results, regional data health judgment is performed on the model monitoring block corresponding to the data health coefficient to obtain block data health judgment information.

4. According to claim 1, an AI health manager data intelligent management method is characterized in that: The S3 includes: Extract historical health-related block information based on the user's historical health information to obtain historical related block information; Obtain the health determination information of the newly added blocks, and obtain the newly added abnormal blocks of the user health model based on the health determination information of the newly added blocks; Extract associated blocks from newly added abnormal blocks according to historical associated block information to obtain associated extracted blocks; Establishing an abnormal association relationship for the associated extraction block to obtain an abnormal associated block; Calculate the associated health coefficient of the abnormal associated block according to the data health coefficient; Comparing the associated health coefficient with a preset associated health threshold to obtain an associated health comparison result; According to the association health comparison result, an association abnormality determination is performed on the abnormal associated block, and block association abnormality determination information is obtained.

5. According to claim 1, an AI health manager data intelligent management method is characterized in that: The S4 includes: Issue block warnings based on block data health determination information; Issue block association warning based on block association abnormality determination information; Obtaining a block cumulative change coefficient of a data health coefficient of a model monitoring block, comparing the block cumulative change coefficient with a preset block cumulative change threshold, and obtaining a block cumulative comparison result; Adjusting the block acquisition monitoring frequency according to the block cumulative comparison result; Obtaining a correlation cumulative change coefficient of a data health coefficient of an abnormally associated block, comparing the correlation cumulative change coefficient with a preset correlation cumulative change threshold, and obtaining a correlation cumulative comparison result; The association collection monitoring frequency is adjusted according to the association accumulation comparison result.

6. An AI health steward data intelligent management system, characterized in that: The system comprises: The model filling module is used to build a user health model, collect health data, process and classify the data, obtain health classification data, fill the user health model with data according to the health classification data, and obtain health model filling data; The block health determination module is used to divide the user health model into monitoring blocks, obtain the model monitoring blocks, calculate the data health coefficient of the model monitoring blocks, and obtain the block health determination information according to the data health coefficient; The associated health judgment module is used to obtain the historical associated block information and the newly added abnormal blocks of the user health model, extract the associated blocks, obtain the abnormal associated blocks, calculate the associated health coefficient, perform associated abnormal judgment, and block associated abnormal judgment information; The early warning adjustment module is used to perform block early warning and block-related early warning according to the judgment information, obtain the block cumulative change coefficient, compare it with the preset block cumulative change threshold, and adjust the associated collection and monitoring frequency.

7. According to claim 6, an AI health manager data intelligent management system is characterized in that: The model filling module includes: A model building module, used to obtain the user's historical health information and generate a user health model based on the user's historical health information; The data collection integration module is used to collect health data through health collection equipment to obtain equipment health collection data; Integrate the device health collection data through a preset relay device node to obtain health integration data; Preprocessing the health integrated data to obtain preprocessed health integrated data; A data classification filling module is used to obtain preset health data categories, classify the preprocessed health integrated data according to the health data categories, and obtain health classification data; The user health model is populated with new data through health classification data to obtain health model populated data.

8. According to claim 6, an AI health manager data intelligent management system is characterized in that: The block health determination module includes: A block division module, used to divide the user health model into monitoring blocks according to the health model filling data to obtain model monitoring blocks; The block health analysis module is used to fill the data calculation model with the health model to monitor the data health coefficient of the block; Comparing the data health coefficient with a preset data health threshold to obtain a data health comparison result; According to the data health comparison results, regional data health judgment is performed on the model monitoring block corresponding to the data health coefficient to obtain block data health judgment information.

9. According to claim 6, an AI health manager data intelligent management system is characterized in that: The associated health determination module includes: The associated block division module is used to extract the historical health associated block information according to the user's historical health information to obtain the historical associated block information; Obtain the health determination information of the newly added blocks, and obtain the newly added abnormal blocks of the user health model based on the health determination information of the newly added blocks; Extract associated blocks from newly added abnormal blocks according to historical associated block information to obtain associated extracted blocks; Establishing an abnormal association relationship for the associated extraction block to obtain an abnormal associated block; The associated block analysis module is used to calculate the associated health coefficient of the abnormal associated block according to the data health coefficient; Comparing the associated health coefficient with a preset associated health threshold to obtain an associated health comparison result; According to the association health comparison result, an association abnormality determination is performed on the abnormal associated block, and block association abnormality determination information is obtained.

10. According to claim 6, an AI health manager data intelligent management system is characterized in that: The early warning adjustment module includes: The early warning module is used to issue early warnings based on the health judgment information of block data; Issue block association warning based on block association abnormality determination information; A change monitoring module is used to obtain a block cumulative change coefficient of a data health coefficient of a model monitoring block, compare the block cumulative change coefficient with a preset block cumulative change threshold, and obtain a block cumulative comparison result; Adjusting the block acquisition monitoring frequency according to the block cumulative comparison result; A comparison and adjustment module, used for obtaining the correlation cumulative change coefficient of the data health coefficient of the abnormal correlation block, comparing the correlation cumulative change coefficient with a preset correlation cumulative change threshold, and obtaining a correlation cumulative comparison result; The association collection monitoring frequency is adjusted according to the association accumulation comparison result.

Images