Health state assessment and early warning method and system based on biosensor

Abstract

The invention provides a health state evaluation and early warning method and system based on a biosensor, and relates to the technical field of health evaluation and early warning, physiological information data is collected and processed through the biosensor, and physiological data collection and processing information is obtained; performing health feature processing analysis according to the physiological data acquisition processing information to obtain health feature processing analysis data, and performing health state assessment analysis according to the health feature processing analysis data to obtain health state assessment analysis data; and carrying out risk early warning analysis according to the health state assessment analysis data to obtain risk early warning analysis data, and carrying out differential early warning feedback according to the risk early warning analysis data to obtain assessment early warning information. According to the invention, end-to-end processing from physiological information acquisition to differential early warning feedback is realized.

Description

Technical Field

[0001] This invention proposes a method and system for health status assessment and early warning based on biosensors, which relates to the field of health assessment and early warning technology, specifically to the field of health status assessment and early warning technology based on biosensors. Background Technology

[0002] With the development of wearable monitoring devices and health management technologies, biosensor-based human health status monitoring has been widely applied in scenarios such as daily health monitoring, chronic disease management, and early warning of sudden illnesses. Traditional health monitoring methods often use a single sensor to collect physiological data and make simple comparisons based on fixed thresholds. This approach suffers from problems such as limited physiological information collection dimensions and low data utilization, making it difficult to comprehensively reflect the overall health status of the human body. Furthermore, existing methods generally employ fixed sampling frequencies and uniform judgment standards, failing to incorporate dynamic analysis based on the temporal changes in physiological signals. This makes them susceptible to interference from instantaneous fluctuations, resulting in a high rate of false positives in anomaly identification and insufficient accuracy in early warning. In addition, most health assessment systems do not fully consider individual differences such as age, medical history, and lifestyle habits, using generic assessment models and early warning rules. This fails to achieve personalized status judgments, leading to biased assessment results and limited early warning methods. Moreover, existing technologies lack systematic processing procedures in feature selection and multi-indicator fusion analysis, making it difficult to accurately identify and promptly warn of potential health risks, and thus failing to meet the needs for efficient, accurate, and personalized health status assessment and early warning. Summary of the Invention

[0003] This invention provides a method and system for health status assessment and early warning based on biosensors, in order to solve the above-mentioned problems: The present invention proposes a health status assessment and early warning method and system based on biosensors, wherein the method includes: S1. Physiological information data is collected and processed through biosensors to obtain physiological data collection and processing information; S2. Based on the physiological data collection and processing information, perform health characteristic processing and analysis to obtain health characteristic processing and analysis data. Based on the health characteristic processing and analysis data, perform health status assessment and analysis to obtain health status assessment and analysis data. S3. Conduct risk warning analysis based on health status assessment data to obtain risk warning analysis data, and provide differentiated warning feedback based on the risk warning analysis data to obtain assessment warning information.

[0004] Furthermore, the system includes: The data acquisition and processing module is used to acquire and process physiological information data through biosensors to obtain physiological data acquisition and processing information. The feature analysis module is used to perform health feature processing and analysis based on physiological data collection and processing information to obtain health feature processing and analysis data, and to perform health status assessment and analysis based on the health feature processing and analysis data to obtain health status assessment and analysis data. The status assessment module is used to perform risk warning analysis based on health status assessment data, obtain risk warning analysis data, provide differentiated warning feedback based on the risk warning analysis data, and obtain assessment warning information.

[0005] The beneficial effects of this invention are as follows: This method solves the technical problems of fragmented processes and disconnected links in traditional health monitoring methods, which prevent the realization of closed-loop monitoring throughout the entire process; it also overcomes the shortcomings of traditional methods, which can only monitor single physiological indicators and cannot complete the integrated processing from data collection to early warning feedback. It achieves fully automated and integrated processing of health status assessment and early warning, completing the entire operation from physiological data collection to early warning information push without manual intervention; it improves the continuity and completeness of health monitoring, ensuring that physiological data can be used efficiently and that health assessment results can be promptly transformed into early warning feedback; it reduces manual operation costs and human error, and avoids problems such as untimely monitoring and delayed early warning caused by process disconnections. Attached Figure Description

[0006] Figure 1 This is a schematic diagram of a health status assessment and early warning method based on biosensors. Detailed Implementation

[0007] 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.

[0008] In one embodiment of the present invention, the present invention proposes a health status assessment and early warning method and system based on biosensors, wherein the method includes: S1. Physiological information data is collected and processed through biosensors to obtain physiological data collection and processing information; S2. Based on the physiological data collection and processing information, perform health characteristic processing and analysis to obtain health characteristic processing and analysis data. Based on the health characteristic processing and analysis data, perform health status assessment and analysis to obtain health status assessment and analysis data. S3. Conduct risk warning analysis based on health status assessment data to obtain risk warning analysis data. Provide differentiated warning feedback based on this data to obtain assessment and warning information, such as... Figure 1 As shown.

[0009] The working principle and technical effects of the above technical solution are as follows: This method acquires multi-dimensional physiological information of the human body through biosensors to complete the initial data collection and preprocessing; then, based on the preprocessed physiological data, it extracts, analyzes and combines health characteristics to further complete the accurate assessment of the user's health status, clarifying the user's current health level and potential risks; finally, combined with the health status assessment results, it conducts risk warning analysis and implements differentiated warning feedback according to the risk level to ensure the pertinence and effectiveness of the warning information, while forming a complete closed-loop process to ensure smooth connection of each link and realize dynamic monitoring and timely warning of health status.

[0010] This method addresses the technical challenges of fragmented processes and disconnected stages in traditional health monitoring methods, which prevent the achievement of closed-loop monitoring throughout the entire process. It also overcomes the limitation of traditional methods, which can only monitor single physiological indicators and cannot achieve integrated processing from data collection to early warning feedback. This method automates and integrates the entire process of health status assessment and early warning, completing the entire operation from physiological data collection to early warning information delivery without human intervention. It improves the continuity and completeness of health monitoring, ensuring efficient use of physiological data and timely conversion of health assessment results into early warning feedback. Furthermore, it reduces manual operation costs and human error, avoiding problems such as untimely monitoring and delayed early warnings caused by process disconnects.

[0011] In one embodiment of the present invention, S1 includes: A biosensor is obtained by combining multiple preset acquisition sensors; Physiological data is acquired through biosensors. Physiological data collection is divided into time series to obtain time series physiological data. Based on the temporal physiological segmentation data, temporal physiological abnormality analysis was performed to obtain temporal physiological abnormality analysis data; Based on the analysis data of time-series physiological abnormalities, the frequency of physiological sensor acquisition is adjusted to obtain frequency adjustment information; Physiological data is updated, collected, and preprocessed based on frequency adjustment information to obtain physiological data acquisition and processing information.

[0012] The working principle and technical effects of the above technical solution are as follows: Based on the needs of health monitoring, this method combines various preset acquisition sensors (such as ECG sensors, blood oxygen sensors, body temperature sensors, biochemical sensors, etc.) to construct a multimodal biosensor network. This ensures comprehensive acquisition of physiological information from different dimensions of the human body, avoiding the limitations of single-sensor acquisition. The biosensor network is used to initially acquire human physiological data, obtaining raw physiological data. To facilitate subsequent anomaly analysis, the raw physiological data is divided into time-series segments according to preset time intervals, breaking down continuous physiological data into multiple time segments to obtain time-series physiological segmentation data. Based on the time-series physiological segmentation data, time-series physiological anomaly analysis is conducted to determine each... The system identifies any anomalies in the physiological data within a given timeframe, generating time-series physiological anomaly analysis data. Based on the anomaly analysis results, the acquisition frequency of the biosensors is dynamically adjusted. For timeframes containing abnormal data, the acquisition frequency is increased to obtain more accurate anomaly data; for timeframes containing normal data, the acquisition frequency is decreased to conserve energy, thus obtaining frequency adjustment information. Based on this frequency adjustment information, the physiological data is updated and collected. Simultaneously, both the updated and original physiological data undergo preprocessing (including noise reduction, outlier correction, and standardization) to remove interference factors, unify data formats, and obtain high-quality physiological data acquisition and processing information. This provides reliable data support for the subsequent health characteristic analysis in step S2.

[0013] This method addresses the technical problems of incomplete physiological data acquisition and inability to cover multi-dimensional physiological indicators caused by traditional single-sensor acquisition. It also overcomes the drawbacks of high energy consumption and insufficient accuracy in acquiring abnormal data due to the fixed acquisition frequency of traditional sensors. Furthermore, it solves the problem of noise and outliers in raw physiological data, which prevent direct use in subsequent analysis. It achieves comprehensive acquisition of multi-dimensional physiological information, ensuring the capture of state information from different physiological systems in the human body and improving the comprehensiveness of physiological data. Through dynamic frequency adjustment, it achieves a balance between acquisition accuracy and device energy consumption, ensuring accurate acquisition of abnormal data while reducing device energy consumption and extending sensor battery life. Preprocessing removes interfering factors from physiological data, improving its purity and reliability and avoiding the impact of interfering data on subsequent health characteristic analysis and health assessment. Simultaneously, the time-series segmentation method facilitates precise location of the occurrence period of abnormal data, improving the targeting of anomaly analysis.

[0014] In one embodiment of the present invention, the step of performing temporal physiological anomaly analysis based on temporal physiological segmentation data to obtain temporal physiological anomaly analysis data includes: Acquire preset time-series physiological standard data, and determine the preset time-series physiological standard curve based on the preset time-series physiological standard data; Determine the actual time-series physiological curves based on time-series physiological segmentation data; The time-series physiological standard curve is compared with the actual time-series physiological curve to obtain the time-series physiological comparison curve; The maximum and minimum values ​​of time-series physiological differences were obtained from the time-series physiological comparison curves. The difference between the maximum and minimum values ​​of temporal physiological differences is calculated to obtain temporal physiological fluctuation data; Compare the time-series physiological fluctuation data with the preset time-series physiological fluctuation threshold to obtain fluctuation comparison data; The time-series physiological fluctuation data is compared with a preset time-series time fluctuation threshold to obtain time comparison data; By combining fluctuation comparison data with time comparison data, time-series physiological anomalies are determined, resulting in time-series physiological anomaly analysis data. This process simultaneously satisfies [the relevant conditions].

[0015] The working principle and technical effects of the above technical solution are as follows: This method addresses the problems of inaccurate judgment of anomalies in time-series physiological data and susceptibility to transient fluctuations. It designs an anomaly analysis method based on curve comparison, fluctuation analysis, and dual threshold judgment. Pre-set time-series physiological standard data is acquired. This standard data combines clinical health standards and general health data, covering the normal physiological parameter ranges for different time segments. A pre-set time-series physiological standard curve is plotted based on this standard data, serving as the benchmark for anomaly judgment. Based on the time-series physiological data, an actual time-series physiological curve is plotted, intuitively reflecting the temporal change trend of the user's actual physiological parameters. The time-series physiological standard curve and the actual time-series physiological curve are compared point-by-point to obtain a time-series physiological comparison curve, which clearly shows the differences between the actual physiological data and the standard data. From the time-series physiological comparison curve, the maximum and minimum values ​​of the time-series physiological difference are extracted, corresponding to the maximum and minimum deviations between the actual data and the standard data, respectively. The values ​​of both are then calculated. The difference is used to obtain time-series physiological fluctuation data, which can quantify the fluctuation amplitude of actual physiological parameters. The time-series physiological fluctuation data is compared with a preset time-series physiological fluctuation threshold to determine whether the fluctuation amplitude exceeds the normal range, thus obtaining fluctuation comparison data. At the same time, the time-series physiological fluctuation data is compared with a preset time-series fluctuation threshold to determine whether the fluctuation duration exceeds the normal range, thus obtaining time comparison data. According to the judgment rule that both are satisfied, the fluctuation comparison data and the time comparison data are combined. That is, only when the fluctuation amplitude exceeds the fluctuation threshold and the fluctuation duration exceeds the time threshold is it judged as a physiological abnormality. Finally, time-series physiological abnormality analysis data containing abnormal time periods, abnormal fluctuation conditions, and abnormal judgment results are obtained, ensuring the rigor of abnormal judgment.

[0016] This method addresses the technical problems of traditional anomaly detection relying solely on a single threshold, which is susceptible to transient fluctuations, leading to high false positive and false negative rates. It also overcomes the shortcomings of traditional anomaly analysis, which cannot quantify the amplitude and duration of physiological parameter fluctuations, resulting in imprecise and inaccurate anomaly detection. This method achieves precise identification of time-series physiological anomalies, visually presenting deviations in physiological data through curve comparison. By combining dual thresholds for amplitude and duration, it effectively eliminates false positives caused by transient fluctuations, ensuring that only genuine anomalies (meeting both amplitude and duration limits) are identified. This improves the rigor and accuracy of anomaly analysis, enabling precise location of the time period and fluctuation patterns of anomalies. Furthermore, it reduces the false positive and false negative rates, preventing user panic and resource waste caused by false positives, and avoiding overlooked health risks due to false negatives.

[0017] In one embodiment of the present invention, S2 includes: Information feature extraction and temporal feature extraction are performed on physiological data acquisition and processing information to obtain information feature extracted data and temporal feature extracted data. The information feature extraction data and the time series feature extraction data are combined accordingly to obtain feature combination data; Perform feature-health correlation analysis based on feature combination data to obtain feature-health correlation analysis data; Based on the feature-based health association analysis data, health combination analysis is performed on the feature combination data to obtain health combination analysis data; Based on the health portfolio analysis data, user health status analysis is performed to obtain health status assessment analysis data.

[0018] The working principle and technical effects of the above technical solution are as follows: This method performs dual feature extraction on physiological data acquisition and processing information. On the one hand, it extracts information features (i.e., the core parameters of the physiological data itself, such as static features like heart rate amplitude, blood glucose concentration, and blood oxygen saturation) to obtain information feature extraction data. On the other hand, it extracts time-series features (i.e., dynamic features such as the changing trend and fluctuation pattern of physiological data over time) to obtain time-series feature extraction data. The information feature extraction data and the time-series feature extraction data are combined accordingly to associate static features with dynamic features, thus obtaining feature combination data and avoiding the limitations of single feature analysis. Based on the feature combination data, feature health correlation analysis is carried out to explore the correlation between features and health status, thus obtaining feature health correlation analysis data. According to the feature health correlation analysis data, health combination analysis is performed on the feature combination data. Features with high correlation are grouped, and the health status corresponding to each group of features is judged based on the health correlation, thus obtaining health combination analysis data. Based on the health combination analysis data and combined with preset health assessment standards, the overall health status of the user is comprehensively analyzed to determine the user's current health level (such as healthy, sub-healthy, high-risk, etc.), thus obtaining health status assessment analysis data.

[0019] This method addresses the technical problem of traditional health assessments relying solely on static features and failing to consider dynamic changes in physiological data, leading to biased and inaccurate results. It also overcomes the shortcomings of traditional feature analysis, which neglects the correlation between features and health status, resulting in low feature utilization efficiency. Furthermore, it addresses the lack of systematic approach in health assessments, hindering comprehensive multi-feature evaluation. By combining static and dynamic features, it enhances the comprehensiveness and richness of health features, ensuring a multi-dimensional reflection of the user's health status. Through feature-health correlation analysis, it uncovers the intrinsic connections between features and health status, improving feature utilization efficiency and avoiding interference from ineffective features. Through health combination analysis and overall health status analysis, it achieves a systematic and comprehensive assessment of the user's health status, improving the accuracy and scientific rigor of health assessments. Finally, it avoids the bias inherent in single-feature assessments, providing a more accurate reflection of the user's actual health condition.

[0020] In one embodiment of the present invention, the step of performing feature health association analysis based on feature combination data to obtain feature health association analysis data includes: Construct a health feature association database, and determine the association relationships of health features based on the health feature association database; Calculate the correlation degree of each feature relationship to obtain feature correlation degree data; The feature correlation data is compared with a preset feature correlation threshold to obtain the feature correlation comparison results; Based on the feature association comparison results, the features corresponding to the feature association relationships are combined to obtain key feature combination data; The association relationships of feature combinations are determined by combining key feature combination data with a health feature association database; Calculate the degree of association for each feature combination to obtain the combination association degree data; The combined correlation data is compared with the preset combined correlation threshold to obtain the combined correlation comparison results; The results of feature association comparison and combined association comparison are integrated to obtain feature health association analysis data.

[0021] The working principle and technical effects of the above technical solution are as follows: This method addresses the problems of unclear correlation between features and health status and insufficient rationality in feature combinations by designing a two-layer association analysis and screening method to construct accurate feature-health correlation relationships. A health feature association database is constructed, containing the association relationships, association rules, and clinical validation information between various physiological features (single features and feature combinations) and different health statuses (healthy, sub-healthy, high-risk, and disease-related statuses), providing data support for association analysis. Based on this health feature association database, the association relationship between each single physiological feature and health status is determined, i.e., the health feature association relationship. Using algorithms such as mutual information entropy, the correlation degree of each health feature association relationship is calculated to quantify the influence of a single feature on the health status judgment, obtaining feature correlation degree data. The feature correlation degree data is compared with a preset feature association threshold, and effective feature association relationships with correlation degrees higher than the threshold are screened out, obtaining the feature association comparison results. Based on the comparison results, features with qualified correlation degrees are grouped... The process involves: 1) obtaining key feature combination data; 2) eliminating invalid features with low relevance to ensure the effectiveness of feature combinations; 3) determining the correlation between feature combinations and health status, i.e., feature combination correlation, based on the key feature combination data and a health feature association database; 4) calculating the correlation degree of each feature combination correlation using a correlation degree calculation algorithm to obtain combination correlation degree data; 5) comparing the combination correlation degree data with a preset combination correlation threshold to filter out valid feature combination correlations that meet the correlation degree standard, obtaining combination correlation comparison results; and 6) integrating the feature correlation comparison results (single feature correlation) with the combination correlation comparison results (feature combination correlation), summarizing valid single feature correlations, valid feature combination correlations, and corresponding correlation degree information to obtain feature health correlation analysis data.

[0022] This method addresses the technical problems of unclear correlations between features and health status in traditional feature analysis, making it impossible to distinguish between effective and ineffective features. It also overcomes the shortcomings of traditional feature combinations, which lack scientific basis and rationality, leading to inaccurate subsequent assessment results. Furthermore, it addresses the issue that single-feature correlation analysis is insufficiently comprehensive and fails to reflect the synergistic effects of features on health status. It achieves a two-level correlation analysis of single features and feature combinations, considering both the impact of individual features on health status and the influence of feature synergies, thus improving the comprehensiveness of feature-health correlation analysis. Through dual threshold screening, ineffective features and ineffective feature combinations are eliminated, improving feature utilization efficiency and reducing the computational load of subsequent analyses. It clarifies the correlation and degree of correlation between features and health status, enhancing the scientific rigor and accuracy of the entire assessment process. It avoids interference from ineffective features in the assessment results, and through feature combination correlation analysis, it can capture health risks that cannot be reflected by single features, further improving the accuracy of health assessment.

[0023] In one embodiment of the present invention, the step of performing health combination analysis on feature combination data based on feature health association analysis data to obtain health combination analysis data includes: Based on the physiological data acquisition and processing information, the ratio of time-series physiological abnormality analysis data to time-series physiological normality analysis data is obtained from the characteristic health association analysis data to obtain the characteristic association health assessment coefficient. Based on the characteristic health association analysis data, health associations are grouped to obtain health association analysis data; The feature-related health assessment coefficient is compared with the preset feature-related health assessment threshold to obtain the associated health assessment comparison data. Based on the comparative data of associated health assessments, the health status of the feature combination data is determined, and health combination analysis data is obtained.

[0024] The working principle and technical effect of the above technical solution are as follows: This method addresses the problem that the correlation between feature combinations and health status is not close and the determination of health status is not accurate enough. By combining time-series abnormal data and health-related grouping, it achieves accurate health determination of feature combinations. Based on physiological data collection and processing information, time-series physiological abnormality analysis data and time-series physiological normality analysis data are extracted from the feature health association analysis data. The ratio between the two is calculated to obtain the feature-related health assessment coefficient. This coefficient can quantify the degree of health deviation corresponding to the feature combination; the larger the ratio, the more serious the health deviation. Based on the feature health association analysis data, the feature combination data is grouped according to the correlation between features and health status. Feature combinations related to the same health system (such as the cardiovascular system and metabolic system) are grouped together to obtain health association analysis data, realizing the classification management of feature combinations. The calculated feature-related health assessment coefficient is compared with the preset feature-related health assessment threshold to determine whether the coefficient exceeds the normal range, obtaining the associated health assessment comparison data and clarifying the degree of health deviation corresponding to the feature combination. Based on the associated health assessment comparison data and the health association grouping information, the health status of each group of feature combination data is judged to determine whether there is an abnormality in the health system corresponding to each group of feature combinations. The judgment results of all groups are summarized to obtain health combination analysis data containing the status of each health system and the degree of health deviation of the feature combination.

[0025] This method addresses the technical problem of traditional health combination analysis failing to incorporate time-series anomaly data, thus hindering the quantification of health deviations. It also resolves the shortcomings of inaccurate health status assessments due to chaotic feature combination grouping that fails to correspond to specific health systems. Furthermore, it addresses the disconnect between health combination analysis and health correlations, resulting in insufficient data for assessment. It achieves quantitative analysis of health deviations, intuitively reflecting the health status corresponding to feature combinations through the ratio of abnormal to normal data, thereby improving the accuracy of health assessments. Through health correlation grouping, it maps feature combinations to specific health systems, facilitating precise identification of systems with health abnormalities and enhancing the targeting of health analysis. Combined with correlated health assessment thresholds, it clarifies the degree of health deviation, providing a clear grouping basis for subsequent overall health status assessments. Finally, it avoids the blind spots inherent in health combination analysis, improving the practicality and reliability of the health combination analysis data.

[0026] In one embodiment of the present invention, the step of analyzing user health status based on health portfolio analysis data to obtain health status assessment analysis data includes: Obtain the preset health portfolio weights corresponding to the health portfolio analysis data; Calculate the product of the health portfolio analysis data and the preset health portfolio weights to obtain the weighted portfolio analysis data; Calculate the average weighted combination analysis data of all health combination analysis data to obtain the user's health status coefficient; The user's health status coefficient is compared with a preset health status threshold to obtain health status assessment and analysis data.

[0027] The working principle and technical effect of the above technical solution are as follows: This method achieves accurate quantitative assessment of the user's overall health status through weighted calculation and threshold comparison. Based on different types of health combination analysis data (i.e., characteristic combinations of different health systems), preset health combination weights are obtained. These weights are set according to the importance of each health system (e.g., the cardiovascular and metabolic systems have higher weights than other systems), reflecting the differences in the impact of different health systems on the overall health status. Each set of health combination analysis data is multiplied by its corresponding preset health combination weight to obtain weighted combination analysis data, which reflects the contribution and influence of different health systems on overall health. The average value of the weighted combination analysis data corresponding to all health combination analysis data is calculated to obtain the user's health status coefficient. This coefficient is a quantitative indicator of the user's overall health status, comprehensively reflecting the status of each health system. The user's health status coefficient is compared with a preset health status threshold. Based on the threshold range in which the coefficient falls, the user's current overall health status level (e.g., healthy, sub-healthy, high-risk, disease-related status) is determined. Simultaneously, the deviation of the health status coefficient from the threshold is identified, obtaining health status assessment analysis data that includes the health status level, health status coefficient, and deviation.

[0028] This method addresses the technical problems of traditional health status assessments, such as failing to consider the importance of different health systems and using equal weights leading to inaccurate results. It also overcomes the deficiency of health status assessments lacking quantitative indicators, thus failing to accurately reflect the user's overall health level. Furthermore, it resolves the disconnect between overall health status assessment and grouped health analysis, resulting in insufficient assessment data. It achieves a quantitative assessment of the user's overall health status, intuitively reflecting the user's health level through a health status coefficient, thereby improving the accuracy and objectivity of health assessments. By pre-setting health combination weights, it reflects the importance of different health systems, making the assessment results more closely reflect the actual state of human health and avoiding assessment bias caused by equal weights. By comparing with preset health status thresholds, it clarifies the user's health status level and deviations. This enhances the scientific rigor and rationality of health status assessments, enabling a more accurate reflection of the user's actual health condition.

[0029] In one embodiment of the present invention, S3 includes: Obtain users' personalized basic information, combine and analyze the user's personalized basic information with health status assessment and analysis data to obtain personalized health combined analysis data; Based on personalized health data analysis, user health baseline values ​​and user health status values ​​are determined. Compare the user's health baseline value with the user's health status value to obtain health comparison data; Based on health comparison data, corresponding intensity warnings are issued to obtain assessment and warning information.

[0030] The working principle and technical effect of the above technical solution are as follows: This method addresses the problems of traditional early warning methods, such as lack of personalization, single early warning method, and insufficient early warning targeting. By combining users' personalized basic information, it achieves accurate and differentiated risk early warning. The system acquires personalized basic information about users, including age, gender, past medical history, family medical history, and lifestyle habits—personalized data that can influence health status assessment and early warning thresholds. This personalized basic information is then combined with health status assessment data for analysis, linking personalized information to health status coefficients and levels to correct the health status assessment results and obtain personalized health-integrated analysis data, ensuring the assessment results align with the individual user's situation. Based on this personalized health-integrated analysis data and the user's personalized basic information, a personalized health baseline value (i.e., the user's own normal health status quantification value) and the current user health status value (i.e., the corrected health status coefficient) are determined. The user's health baseline value and the user health status value are compared, and the difference is calculated to obtain health comparison data, which quantifies the degree of deviation between the user's current health status and their normal state. Based on the health comparison data and a preset early warning intensity level standard, the user's current health risk level is determined, and differentiated early warning feedback is implemented for different risk levels (e.g., low risk only pushes a reminder, high risk triggers an audible and visual alarm and contacts emergency contacts), obtaining assessment and early warning information including early warning level, warning method, and warning content, ensuring the targetedness and effectiveness of the early warning information.

[0031] This method addresses the technical problems of traditional early warning methods, which use a uniform standard and fail to consider individual user differences, leading to inaccurate warnings and high false alarm / missed alarm rates. It also overcomes the shortcomings of traditional warning methods, which are too simplistic and unable to provide differentiated feedback based on risk levels, resulting in poor warning effectiveness. Furthermore, it addresses the disconnect between warning feedback and individual user needs, failing to meet the diverse warning requirements of different users. This method achieves personalized risk warnings by combining users' individual basic information to adjust health baseline values ​​and health status assessment results, avoiding a one-size-fits-all warning approach and improving accuracy. By comparing health baseline values ​​and health status values, it quantifies the degree of deviation from the user's health status. Differentiated warning feedback employs different warning methods for different risk levels, improving the targeting and effectiveness of warnings. Low-risk warnings prevent user panic, while high-risk warnings ensure users receive timely alerts and assistance. This reduces false alarm and missed alarm rates, increases user acceptance of warning information, and achieves personalized warning feedback to meet the health monitoring needs of different users.

[0032] In one embodiment of the present invention, the step of obtaining user personalized basic information and combining and analyzing the user personalized basic information with health status assessment and analysis data to obtain personalized health combined analysis data includes: Obtain basic personalized information from users, preprocess the basic personalized information to obtain personalized processing information; The health trait association database will associate and match personalized processing information with health status assessment and analysis data to obtain personalized association matching information. The preset weights of health combinations are adjusted based on personalized correlation and matching information to obtain the adjustment weights; The health status assessment and analysis data are updated based on the adjustment weights to obtain personalized health combined analysis data.

[0033] The working principle and technical effect of the above technical solution are as follows: This method addresses the problems of insufficient utilization of personalized basic information and lack of personalized correction of health status assessment results, and realizes the deep integration of personalized information and health assessment data to improve the personalization of assessment results. The system acquires personalized basic information about users, including age, gender, past medical history, family medical history, lifestyle habits, and medication use. This data is preprocessed (including classification, removal of invalid information, and standardization) to remove redundant and invalid data, standardize data formats, and obtain personalized processing information, ensuring effective utilization of personalized data. A health feature association database is accessed to correlate and match the personalized processing information with health status assessment data, uncovering the correlation between personalized information and health status assessment results (e.g., the correlation between a history of hypertension and cardiovascular health assessment results), obtaining personalized correlation matching information. Based on this personalized correlation matching information, the preset health combination weights are dynamically adjusted. Adjustment weights are obtained for individual user circumstances (e.g., increasing the weight of metabolic system feature combinations for users with diabetes), making the weight settings more suitable for individual user situations. Based on the adjusted weights, the health status assessment data is updated and corrected, and the user's health status coefficient and health status level are recalculated, resulting in personalized health combination analysis data that includes personalized corrected health status information, adjustment weights, and personalized correlation information.

[0034] This method addresses the technical problems of traditional health assessments, which fail to fully utilize users' personalized basic information, resulting in a lack of personalized assessment results and an inability to match individual user circumstances. It also resolves the deficiency of fixed health combination weights, which cannot be dynamically adjusted according to user individualization, leading to inaccurate assessment results. Furthermore, it addresses the issue of the disconnect between personalized information and health assessment data, resulting in low efficiency in the utilization of personalized information. It achieves deep integration of personalized information and health assessment data, improving the personalization and accuracy of health status assessment results, making the assessment results more closely reflect the user's actual health condition. By dynamically adjusting the health combination weights, it avoids the assessment bias caused by fixed weights, further improving the accuracy of health assessments. It fully utilizes users' personalized basic information, exploring the impact of personalized information on health status and improving the efficiency of personalized information utilization. Finally, it ensures that subsequent alerts truly meet individual user needs, reducing the false alarm rate and false negative rate of alerts.

[0035] According to one embodiment of the present invention, the system includes: The data acquisition and processing module is used to acquire and process physiological information data through biosensors to obtain physiological data acquisition and processing information. The feature analysis module is used to perform health feature processing and analysis based on physiological data collection and processing information to obtain health feature processing and analysis data, and to perform health status assessment and analysis based on the health feature processing and analysis data to obtain health status assessment and analysis data. The status assessment module is used to perform risk warning analysis based on health status assessment data, obtain risk warning analysis data, provide differentiated warning feedback based on the risk warning analysis data, and obtain assessment warning information.

[0036] The working principle and technical effects of the above technical solution are as follows: This method acquires multi-dimensional physiological information of the human body through biosensors to complete the initial data collection and preprocessing; then, based on the preprocessed physiological data, it extracts, analyzes and combines health characteristics to further complete the accurate assessment of the user's health status, clarifying the user's current health level and potential risks; finally, combined with the health status assessment results, it conducts risk warning analysis and implements differentiated warning feedback according to the risk level to ensure the pertinence and effectiveness of the warning information, while forming a complete closed-loop process to ensure smooth connection of each link and realize dynamic monitoring and timely warning of health status.

[0037] This method addresses the technical challenges of fragmented processes and disconnected stages in traditional health monitoring methods, which prevent the achievement of closed-loop monitoring throughout the entire process. It also overcomes the limitation of traditional methods, which can only monitor single physiological indicators and cannot achieve integrated processing from data collection to early warning feedback. This method automates and integrates the entire process of health status assessment and early warning, completing the entire operation from physiological data collection to early warning information delivery without human intervention. It improves the continuity and completeness of health monitoring, ensuring efficient use of physiological data and timely conversion of health assessment results into early warning feedback. Furthermore, it reduces manual operation costs and human error, avoiding problems such as untimely monitoring and delayed early warnings caused by process disconnects.

[0038] 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 status assessment and early warning method based on biosensors, characterized in that, The method includes: S1. Physiological information data is collected and processed through biosensors to obtain physiological data collection and processing information; S2. Based on the physiological data collection and processing information, perform health characteristic processing and analysis to obtain health characteristic processing and analysis data. Based on the health characteristic processing and analysis data, perform health status assessment and analysis to obtain health status assessment and analysis data. S3. Conduct risk warning analysis based on health status assessment data to obtain risk warning analysis data, and provide differentiated warning feedback based on the risk warning analysis data to obtain assessment warning information.

2. The health status assessment and early warning method based on biosensors according to claim 1, characterized in that, S1 includes: A biosensor is obtained by combining multiple preset acquisition sensors; Physiological data is acquired through biosensors. Physiological data collection is divided into time series to obtain time series physiological data. Based on the temporal physiological segmentation data, temporal physiological abnormality analysis was performed to obtain temporal physiological abnormality analysis data; Based on the analysis data of time-series physiological abnormalities, the frequency of physiological sensor acquisition is adjusted to obtain frequency adjustment information; Physiological data is updated, collected, and preprocessed based on frequency adjustment information to obtain physiological data acquisition and processing information.

3. The health status assessment and early warning method based on biosensors according to claim 2, characterized in that, The step of performing temporal physiological abnormality analysis based on temporal physiological segmentation data to obtain temporal physiological abnormality analysis data includes: Acquire preset time-series physiological standard data, and determine the preset time-series physiological standard curve based on the preset time-series physiological standard data; Determine the actual time-series physiological curves based on time-series physiological segmentation data; The time-series physiological standard curve is compared with the actual time-series physiological curve to obtain the time-series physiological comparison curve; The maximum and minimum values ​​of time-series physiological differences were obtained from the time-series physiological comparison curves. The difference between the maximum and minimum values ​​of temporal physiological differences is calculated to obtain temporal physiological fluctuation data; Compare the time-series physiological fluctuation data with the preset time-series physiological fluctuation threshold to obtain fluctuation comparison data; The time-series physiological fluctuation data is compared with a preset time-series time fluctuation threshold to obtain time comparison data; By combining fluctuation comparison data with time comparison data, time-series physiological abnormalities are determined, and time-series physiological abnormality analysis data are obtained.

4. The health status assessment and early warning method based on biosensors according to claim 1, characterized in that, S2 includes: Information feature extraction and temporal feature extraction are performed on physiological data acquisition and processing information to obtain information feature extracted data and temporal feature extracted data. The information feature extraction data and the time series feature extraction data are combined accordingly to obtain feature combination data; Perform feature-health correlation analysis based on feature combination data to obtain feature-health correlation analysis data; Based on the feature-based health association analysis data, health combination analysis is performed on the feature combination data to obtain health combination analysis data; Based on the health portfolio analysis data, user health status analysis is performed to obtain health status assessment analysis data.

5. The health status assessment and early warning method based on biosensors according to claim 4, characterized in that, The step of performing feature health association analysis based on feature combination data to obtain feature health association analysis data includes: Construct a health feature association database, and determine the association relationships of health features based on the health feature association database; Calculate the correlation degree of each feature relationship to obtain feature correlation degree data; The feature correlation data is compared with a preset feature correlation threshold to obtain the feature correlation comparison results; Based on the feature association comparison results, the features corresponding to the feature association relationships are combined to obtain key feature combination data; The association relationships of feature combinations are determined by combining key feature combination data with a health feature association database; Calculate the degree of association for each feature combination to obtain the combination association degree data; The combined correlation data is compared with the preset combined correlation threshold to obtain the combined correlation comparison results; The results of feature association comparison and combined association comparison are integrated to obtain feature health association analysis data.

6. The health status assessment and early warning method based on biosensors according to claim 4, characterized in that, The step of performing health combination analysis on feature combination data based on feature health association analysis data to obtain health combination analysis data includes: Based on the physiological data acquisition and processing information, the ratio of time-series physiological abnormality analysis data to time-series physiological normality analysis data is obtained from the characteristic health association analysis data to obtain the characteristic association health assessment coefficient. Based on the characteristic health association analysis data, health associations are grouped to obtain health association analysis data; The feature-related health assessment coefficient is compared with the preset feature-related health assessment threshold to obtain the associated health assessment comparison data. Based on the comparative data of associated health assessments, the health status of the feature combination data is determined, and health combination analysis data is obtained.

7. The health status assessment and early warning method based on biosensors according to claim 4, characterized in that, The step of analyzing user health status based on health portfolio analysis data to obtain health status assessment analysis data includes: Obtain the preset health portfolio weights corresponding to the health portfolio analysis data; Calculate the product of the health combination analysis data and the preset health combination weights to obtain the weighted combination analysis data; Calculate the average weighted combination analysis data of all health combination analysis data to obtain the user's health status coefficient; The user's health status coefficient is compared with a preset health status threshold to obtain health status assessment and analysis data.

8. The health status assessment and early warning method based on biosensors according to claim 1, characterized in that, S3 includes: Obtain users' personalized basic information, combine and analyze the user's personalized basic information with health status assessment and analysis data to obtain personalized health combined analysis data; Based on personalized health data analysis, user health baseline values ​​and user health status values ​​are determined. Compare the user's health baseline value with the user's health status value to obtain health comparison data; Based on health comparison data, corresponding intensity warnings are issued to obtain assessment and warning information.

9. The health status assessment and early warning method based on biosensors according to claim 8, characterized in that, The process of obtaining personalized basic user information and combining it with health status assessment and analysis data to obtain personalized health combined analysis data includes: Obtain basic personalized information from users, preprocess the basic personalized information to obtain personalized processing information; The health trait association database will associate and match personalized processing information with health status assessment and analysis data to obtain personalized association matching information. The preset weights of health combinations are adjusted based on personalized correlation and matching information to obtain the adjustment weights; The health status assessment and analysis data are updated based on the adjustment weights to obtain personalized health combined analysis data.

10. A health status assessment and early warning system based on biosensors, characterized in that, The system includes: The data acquisition and processing module is used to acquire and process physiological information data through biosensors to obtain physiological data acquisition and processing information. The feature analysis module is used to perform health feature processing and analysis based on physiological data collection and processing information to obtain health feature processing and analysis data, and to perform health status assessment and analysis based on the health feature processing and analysis data to obtain health status assessment and analysis data. The status assessment module is used to perform risk warning analysis based on health status assessment data, obtain risk warning analysis data, provide differentiated warning feedback based on the risk warning analysis data, and obtain assessment warning information.

Images