A health monitoring method based on data analysis and a smart watch

Abstract

The present application relates to the technical field of health data monitoring, in particular to a health monitoring method based on data analysis and a smart watch, which obtains physiological parameters such as heart rate, blood pressure, blood oxygen and respiratory rate of a user, analyzes the interaction between the physiological parameters, judges whether the physiological parameters are abnormally high or low, and obtains the individual health state distribution result. Based on the dynamic analysis of physiological parameters and combined with multi-dimensional data measurement, the present application accurately depicts the change characteristics of the health state, avoiding the errors caused by single data point judgment. By measuring the interaction between different physiological parameters, the health state can be more comprehensively evaluated, and the judgment of individual health state is not affected by the fluctuation of single physiological index. With the help of multi-dimensional coordinate space, the distance between the health state and the steady state interval is calculated, so that the evaluation of health state change is more quantitative, and the identification ability of abnormal fluctuation is improved.

Description

Technical Field

[0001] This invention relates to the field of health data monitoring technology, and in particular to a health monitoring method and a smartwatch based on data analysis. Background Technology

[0002] The field of health data monitoring technology encompasses various monitoring methods based on physiological signal acquisition, data analysis, and real-time feedback. The core components of this technology include the acquisition of physiological parameters, data processing, feature extraction, and application analysis. Physiological parameter acquisition primarily involves non-invasive or minimally invasive measurement methods for indicators such as heart rate, blood oxygen, blood glucose, blood lipids, and uric acid, often employing technologies such as photoelectric sensing, pulse wave analysis, and electrocardiogram signal monitoring. Data processing includes signal acquisition, noise filtering, feature parameter calculation, and storage to ensure data accuracy and continuity. Feature extraction and application analysis, based on statistical models, machine learning algorithms, and physiological principles, perform trend analysis, anomaly identification, and health assessment on the monitoring data, thereby providing health status feedback, early warning information, or personalized health management plans. This technology is widely used in personal health management, medical auxiliary diagnosis, and chronic disease monitoring.

[0003] The data-driven health monitoring method refers to acquiring user health data through physiological signal acquisition devices and then processing the data using data analysis techniques to assess and manage health status. This method encompasses physiological signal acquisition, data preprocessing, feature analysis, and result output. The physiological signal acquisition section utilizes optical sensing, pulse wave detection, and electrophysiological signal monitoring to collect key health parameters such as blood oxygen saturation, blood glucose concentration, and heart rate variability. Data preprocessing involves noise reduction, signal enhancement, and feature extraction to improve data stability and effectiveness. Feature analysis employs mathematical modeling, pattern recognition, and data fitting to perform dynamic trend analysis on physiological data and identify abnormal changes. The results are presented visually, allowing users to view their health status via device terminals or mobile applications and manage their health based on the analysis results.

[0004] Relying on a single physiological signal for health status assessment is susceptible to data fluctuations and fails to accurately reflect overall health. For example, heart rate changes can be influenced by factors such as exercise and emotions; without the aid of other physiological parameters, this can lead to inaccurate health assessments. For steady-state analysis of physiological states, relying solely on a single threshold setting cannot accurately quantify the degree of change in health status, making it difficult to effectively perceive subtle changes and thus affecting early risk identification. The lack of analysis on the interrelationships between physiological parameters means that the interactions between different physiological signals are not fully considered, leading to inaccurate identification of abnormal changes and potentially missing crucial health anomalies. Health trend analysis often employs static data statistical methods, lacking the ability to predict dynamic changes, resulting in a lag in health management. Health monitoring remains primarily post-hoc, failing to provide proactive health warnings. Warning mechanisms are often crude, relying heavily on fixed thresholds to trigger alarms without incorporating trend analysis of physiological parameters, leading to a high false alarm rate and reducing the accuracy of health management and user trust. Summary of the Invention

[0005] To address the technical problems existing in the prior art, embodiments of the present invention provide a health monitoring method based on data analysis and a smartwatch. The technical solution is as follows:

[0006] A data analysis-based health monitoring method includes the following steps:

[0007] S1: Obtain the user's heart rate, blood pressure, blood oxygen, and respiratory rate physiological parameters, analyze the dynamic correlation between physiological parameters, determine whether there is an imbalance of abnormal increase or decrease in physiological parameters, and obtain the individual health status distribution results.

[0008] S2: Calculate the distance between the corresponding health status and the preset physiological steady-state interval in the individual health status distribution results in the multi-dimensional coordinate space, analyze the changes of heart rate, blood pressure, blood oxygen, and respiratory rate relative to the physiological steady-state interval, and obtain the health status stability results;

[0009] S3: Identify the inflection point of the trend in the stability results of the health status, obtain the current hemodynamic changes, oxygen delivery, and respiratory drive response, determine the phase lag or enhancement pattern of the selected physiological parameter before and after the trend inflection point compared with other parameters, filter the physiological parameters that lag or enhance, calculate the phase difference between physiological parameters, measure the linkage effect of each parameter based on the phase change, obtain the linkage degree of physiological parameters, and then obtain the physiological parameter matching result.

[0010] S4: Based on the physiological parameter matching results, extract the rate of change of similar physiological parameters within a time window, predict the continuous change of health status in the future time period, and obtain the health trend prediction results.

[0011] S5: Identify abnormal fluctuations from the health trend prediction results and establish corresponding early warning markers to obtain individual health monitoring results.

[0012] As a further aspect of the present invention, the individual health status distribution results include heart rate status, blood pressure status, blood oxygen status, and respiratory rate status; the health status stability results include heart rate stability, blood pressure stability, blood oxygen stability, and respiratory rate stability; the physiological parameter matching results include heart rate matching degree, blood pressure matching degree, blood oxygen matching degree, and respiratory rate matching degree; the health trend prediction results include heart rate change trend, blood pressure change trend, blood oxygen change trend, and respiratory rate change trend; and the individual health monitoring results include abnormal heart rate fluctuations, abnormal blood pressure fluctuations, abnormal blood oxygen fluctuations, abnormal respiratory rate fluctuations, and health warning markers.

[0013] As a further aspect of the present invention, the specific steps for obtaining the user's heart rate, blood pressure, blood oxygen, and respiratory rate physiological parameters, analyzing the interactions between these physiological parameters, determining whether there is an imbalance in the physiological parameters that is abnormally elevated or decreased, and obtaining the individual's health status distribution results are as follows:

[0014] S101: Acquire the user's heart rate, blood pressure, blood oxygen, and respiratory rate data, continuously collect multiple measurements, calculate the mean, fluctuation range, and rate of change for each parameter, filter out abnormal deviation data, determine data stability based on the rate of change and fluctuation range, and obtain stability analysis results;

[0015] S102: Based on the stability analysis results, calculate the mean and fluctuation range of heart rate, blood pressure, blood oxygen, and respiratory rate, analyze the correlation between each parameter, screen the associated parameter combinations, calculate the degree of deviation between each parameter, and determine whether there is an abnormal increase or decrease trend based on the degree of deviation and fluctuation range, and obtain the dynamic analysis results of physiological parameter deviation.

[0016] S103: Based on the dynamic analysis results of the physiological parameter deviation, calculate the balance index of individual physiological parameters, classify the health status intervals according to the index range, determine the interval to which the individual physiological parameters belong, and obtain the distribution results of individual health status.

[0017] As a further aspect of the present invention, the specific steps for calculating the distance between the corresponding health status and the preset physiological steady-state interval in the individual health status distribution results in multi-dimensional coordinate space, and analyzing the changes in heart rate, blood pressure, blood oxygen, and respiratory rate relative to the physiological steady-state interval to obtain the health status stability results are as follows:

[0018] S201: Based on the individual health status distribution results, call the heart rate, blood pressure, blood oxygen, and respiratory rate values ​​corresponding to the health status, calculate the distance between the individual health status coordinates and the physiological steady state interval coordinates, and obtain the individual health status spatial offset value.

[0019] S202: Based on the individual health status spatial offset value, calculate the offset distance of heart rate, blood pressure, blood oxygen, and respiratory rate, analyze the changing trend of the physiological steady-state interval corresponding to each parameter, filter parameters that deviate from the steady-state interval, calculate the overall offset vector of individual physiological parameters, and obtain the offset direction and amplitude of physiological parameters based on the directionality and amplitude of the offset vector.

[0020] S203: Based on the direction and magnitude of the physiological parameter offset, calculate the stability index of the individual's physiological parameters, classify the degree of health status stability, screen individuals with a stability index lower than the steady-state threshold, and obtain the health status stability result.

[0021] As a further aspect of the present invention, for calculating the stability index S of individual physiological parameters... u The formula used is:

[0022]

[0023] Among them, W i D represents the weighting coefficient of the i-th physiological parameter. i D represents the offset value of the i-th physiological parameter of an individual. ref,i The i-th baseline offset value represents the physiological homeostasis range, and n represents the total number of physiological parameters involved in the calculation. Represents the summation operation over all physiological parameter terms, |D i -D ref,i | Represents the absolute deviation between the actual offset value and the baseline offset value of the i-th physiological parameter of an individual. The square root of the sum of squares of all physiological parameter offsets is used to normalize the degree of offset. It represents the sum of the absolute values ​​of the offsets of all physiological parameters of an individual.

[0024] As a further aspect of the present invention, the following steps are taken to identify the inflection point of the trend in the stability results of the health status, obtain the current hemodynamic changes, oxygen delivery, and respiratory drive response, determine the phase lag or enhancement pattern of the selected physiological parameter before and after the trend inflection point compared with other parameters, screen the lagging or enhanced physiological parameters, calculate the phase difference between physiological parameters, measure the linkage effect of each parameter based on the phase change, obtain the linkage degree of physiological parameters, and then obtain the physiological parameter matching results:

[0025] S301: Based on the stability results of the health status, call up the values ​​of physiological parameters, detect changes in hemodynamics, oxygen delivery and respiratory drive response, identify the trend inflection point in the time series of physiological parameters, calculate the rate of change of physiological parameters before and after the inflection point, and obtain the trend inflection point offset value of physiological parameters.

[0026] S302: Based on the inflection point offset value of the physiological parameters, calculate the rate of change of hemodynamics, oxygen delivery, and respiratory drive, analyze the phase lag and enhancement mode of the selected physiological parameters before and after the inflection point, screen the physiological parameters that are lagging or enhanced, calculate the phase difference between the physiological parameters, measure the linkage effect of each parameter according to the phase change, and obtain the linkage degree of the physiological parameters.

[0027] S303: Calculate the physiological parameter matching degree based on the physiological parameter linkage degree, filter parameter combinations with matching degree lower than the matching degree threshold, and obtain the physiological parameter matching result.

[0028] As a further aspect of the present invention, for calculating the physiological parameter matching degree M v The formula used is:

[0029]

[0030] Among them, L i L represents the degree of correlation of the i-th physiological parameter of an individual. ref,i V represents the i-th reference linkage degree of the physiological homeostasis range. i Q represents the variation range of the i-th physiological parameter. j Q represents the value of the j-th environmental or individual characteristic factor for an individual. ref,j R represents the j-th reference environmental or individual characteristic factor value of the physiological homeostasis interval. j represents the influence ratio of the j-th environmental or individual characteristic factor, n represents the total number of physiological parameters involved in the calculation, and m represents the total number of additional individual or environmental parameters involved in the calculation.

[0031] As a further aspect of the present invention, based on the physiological parameter matching results, the rate of change of similar physiological parameters within a time window is extracted to predict the continuous change of health status in the future time period, and the specific steps to obtain the health trend prediction result are as follows:

[0032] S401: Based on the physiological parameter matching results, extract the rate of change of similar physiological parameters within the time window, calculate the increment of change at adjacent time points, screen parameters with stable increments of change, analyze the trend direction of physiological parameters, and obtain the physiological parameter change trend analysis results.

[0033] S402: Based on the analysis results of the physiological parameter change trends, analyze the consistency of the trend changes of each physiological parameter, analyze the change trend of health status in the future time period, screen health status parameters with abnormal trend change magnitude, and obtain health trend prediction results.

[0034] As a further aspect of the present invention, the specific steps for identifying abnormal fluctuations from the health trend prediction results and establishing corresponding early warning markers to obtain individual health monitoring results are as follows:

[0035] S501: Based on the health trend prediction results, extract the time period when the trend deviates from the stable range, set the abnormal fluctuation threshold, identify sudden or continuous abnormal fluctuations, determine the duration and intensity of abnormal fluctuations based on the degree of deviation of physiological parameters, establish abnormal marking rules, and obtain abnormal health status marking results.

[0036] S502: Based on the abnormal health status marker results, extract the physiological parameters corresponding to the abnormal status, assess the repeatability and correlation of abnormal fluctuations, classify individuals according to their risk based on the abnormal status, and obtain individual health monitoring results.

[0037] A health monitoring smartwatch based on data analysis, the smartwatch comprising:

[0038] The physiological parameter analysis module acquires the user's heart rate, blood pressure, blood oxygen, and respiratory rate, analyzes the interaction between physiological parameters, determines whether there is an imbalance of abnormally high or low physiological parameters, and obtains the individual's health status distribution results.

[0039] The steady-state distance calculation module calculates the distance between the corresponding health status and the preset physiological steady-state interval in the individual health status distribution results in a multi-dimensional coordinate space, analyzes the changes of heart rate, blood pressure, blood oxygen, and respiratory rate relative to the physiological steady-state interval, and obtains the health status stability results.

[0040] The linkage matching analysis module identifies the inflection point of the trend in the stability results of the health status, obtains the current hemodynamic changes, oxygen delivery, and respiratory drive response, determines the phase lag or enhancement pattern of the selected physiological parameter before and after the trend inflection point compared with other parameters, filters the physiological parameters that lag or enhance, calculates the phase difference between physiological parameters, measures the linkage effect of each parameter based on the phase change, obtains the linkage degree of physiological parameters, and then obtains the physiological parameter matching result.

[0041] Based on the physiological parameter matching results, the health trend prediction module extracts the rate of change of similar physiological parameters within a time window, predicts the continuous change of health status in the future time period, and obtains the health trend prediction result.

[0042] The abnormal early warning module identifies abnormal fluctuations from the health trend prediction results and establishes corresponding early warning markers to obtain individual health monitoring results.

[0043] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following:

[0044] Dynamic analysis based on physiological parameters, combined with multi-dimensional data calculations, accurately depicts the changing characteristics of health status, avoiding errors caused by judgments based on single data points. By calculating the interactions between different physiological parameters, a more comprehensive assessment of health status can be achieved, ensuring that the judgment of individual health status is not affected by fluctuations in a single physiological indicator. Calculating the distance between the health status and the steady-state interval using multi-dimensional coordinate space makes the assessment of health status changes more quantitative and improves the ability to identify abnormal fluctuations. Analysis of the changing trends of health status stability results can accurately identify inflection points of physiological parameters and deeply explore the linkages between hemodynamics, oxygen delivery, and respiratory response, making health assessment more comprehensive. By analyzing the matching of physiological parameters and measuring the degree of linkage between various physiological parameters, short-term trends in health status can be accurately identified, avoiding the impact of individual data anomalies on the overall assessment. Predicting future health trends based on the rate of change of physiological parameters within a specific time window upgrades health monitoring from static analysis to dynamic prediction, improving the foresight of health management. The identification of abnormal fluctuations combined with health trend prediction can establish a more accurate early warning mechanism, making health risk prevention and control more proactive, thereby enhancing the practicality and reliability of health management. Attached Figure Description

[0045] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0046] Figure 1 This is a flowchart of the method of the present invention;

[0047] Figure 2 This is a detailed flowchart of step S1 of the present invention;

[0048] Figure 3 This is a detailed flowchart of step S2 of the present invention;

[0049] Figure 4 This is a detailed flowchart of step S3 of the present invention;

[0050] Figure 5 This is a detailed flowchart of step S4 of the present invention;

[0051] Figure 6This is a detailed flowchart of step S5 of the present invention;

[0052] Figure 7 This is a system module diagram of the present invention. Detailed Implementation

[0053] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0054] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0055] Please see Figure 1 This invention provides a technical solution: a health monitoring method based on data analysis, comprising the following steps:

[0056] S1: Obtain the user's heart rate, blood pressure, blood oxygen, and respiratory rate physiological parameters, analyze the dynamic correlation between physiological parameters, determine whether there is an imbalance of abnormal increase or decrease in physiological parameters, and obtain the individual health status distribution results.

[0057] S2: Calculate the distance between the corresponding health status and the preset physiological steady-state interval in the individual health status distribution results in the multi-dimensional coordinate space, analyze the changes of heart rate, blood pressure, blood oxygen, and respiratory rate relative to the physiological steady-state interval, and obtain the health status stability results;

[0058] S3: Identify the inflection point of the trend in the stability results of health status, obtain the current hemodynamic changes, oxygen delivery, and respiratory drive response, determine the phase lag or enhancement pattern of the selected physiological parameter before and after the trend inflection point compared with other parameters, screen the physiological parameters that lag or enhance, calculate the phase difference between physiological parameters, measure the linkage effect of each parameter based on the phase change, obtain the linkage degree of physiological parameters, and then obtain the physiological parameter matching results.

[0059] S4: Based on the physiological parameter matching results, extract the rate of change of similar physiological parameters within a time window, predict the continuous change of health status in the future time period, and obtain the health trend prediction results.

[0060] S5: Identify abnormal fluctuations from health trend prediction results, establish corresponding early warning markers, and obtain individual health monitoring results.

[0061] Individual health status distribution results include heart rate status, blood pressure status, blood oxygen status, and respiratory rate status; health status stability results include heart rate stability, blood pressure stability, blood oxygen stability, and respiratory rate stability; physiological parameter matching results include heart rate matching degree, blood pressure matching degree, blood oxygen matching degree, and respiratory rate matching degree; health trend prediction results include heart rate change trend, blood pressure change trend, blood oxygen change trend, and respiratory rate change trend; individual health monitoring results include abnormal heart rate fluctuations, abnormal blood pressure fluctuations, abnormal blood oxygen fluctuations, abnormal respiratory rate fluctuations, and health warning markers.

[0062] Please see Figure 2 The specific steps for obtaining users' physiological parameters such as heart rate, blood pressure, blood oxygen, and respiratory rate, analyzing the interactions between these parameters, determining whether there are any imbalances in the physiological parameters, and obtaining the individual's health status distribution results are as follows:

[0063] S101: Acquire the user's heart rate, blood pressure, blood oxygen, and respiratory rate data, continuously collect multiple measurements, calculate the mean, fluctuation range, and rate of change for each parameter, filter out abnormal deviation data, determine data stability based on the rate of change and fluctuation range, and obtain stability analysis results;

[0064] To acquire user data on heart rate, blood pressure, blood oxygen saturation, and respiratory rate, sensors are used to collect these data at 5-second intervals, with 20 consecutive acquisitions forming a data sequence. For each measurement, data preprocessing is performed after acquisition, including removing outliers outside the normal physiological range and interpolating missing data. The mean of each parameter is then calculated using the following formula: in, X represents the mean of a certain physiological parameter. i N represents a single measurement value, and N represents the number of valid measurements. For example, if a user's blood oxygen saturation data in 20 measurements are 95, 96, 97, 98, 99, 95, 96, 97, 98, 99, 95, 96, 97, 98, 99, 95, 96, 97, 98, 99, then the mean is calculated as follows: Next, calculate the fluctuation range. The formula for calculating the fluctuation range is: R = X max -X min , where X max X represents the maximum value in the measured data. min The minimum value in the measured data is used as the reference. For example, if the maximum value of the blood oxygen data above is 99 and the minimum value is 95, then the fluctuation range is 99-95=4. The rate of change of the data is then calculated using the following formula: Among them, X t X is the measured value at the current time. t+1The measured value for the next moment. For example, if a user's heart rate is 80 and 85 respectively at two measurement intervals, the rate of change is calculated as When screening for abnormally deviated data, a deviation threshold T is set. Assume the mean value of ±15% is taken as the threshold range. For blood oxygen data, then T = 97 × 0.15 = 14. 55. Therefore, the judgment range is 82.45 - 111.55. If a certain measured value falls outside this range, it is determined as an abnormal value. Finally, based on the rate of change and the fluctuation range, a data stability assessment is conducted. A threshold S is set as the stability judgment criterion. If max(V) < S and R is lower than the set threshold, the data is determined to be stable; otherwise, it is determined to be unstable, and a stability analysis result is obtained.

[0065] S102: Based on the stability analysis result, calculate the mean value and the fluctuation range of heart rate, blood pressure, blood oxygen, and respiratory rate, analyze the correlation between each parameter, screen the associated parameter combinations, calculate the deviation degree between each parameter, and determine whether there is an abnormal increase or decrease trend based on the deviation degree and the fluctuation range, and obtain the dynamic analysis result of physiological parameter deviation;

[0066] Based on the stability analysis result, calculate the mean value and the fluctuation range of heart rate, blood pressure, blood oxygen, and respiratory rate. First, further refine the mean value calculation on the basis of S101, so that the mean value not only covers single physiological parameters but also includes the data mean value within a specific time window. The set time window is 1 minute to ensure that the data stability calculation is not affected by short-term fluctuations. Subsequently, for each physiological parameter, calculate their correlation. The Pearson correlation coefficient calculation formula is used: where, X i and Y i represent the measured values of two physiological parameters respectively, and represent their mean values respectively. For example, if a user's heart rate data is [70, 72, 75, 77, 80] and the blood pressure data is [120, 122, 125, 127, 130], then calculate their mean values, calculate the Pearson correlation coefficient. If the correlation coefficient is greater than 0.7, it is determined as a strong correlation. If it is between 0.3 and 0.7, it is determined as a medium correlation. Otherwise, it is a weak correlation or no correlation. When screening the associated parameter combinations, retain the parameter pairs with strong and medium correlations. For example, if the correlation between heart rate and blood pressure is strong, then retain their combination. Calculate the deviation degree between each parameter, and use the standardized deviation calculation: where, D is the standardized deviation value, and σ X is the standard deviation. If the standardized deviation value of a certain measured value is greater than 1, it is determined as an abnormal increase trend. If it is less than -1, it is determined as an abnormal decrease trend. For example, if a certain heart rate measured value is 85, Standard deviation σ X =5, then Since D>1, it was determined to be an abnormally high level, and the final result of the dynamic analysis of physiological parameter deviation was obtained.

[0067] S103: Based on the dynamic analysis results of physiological parameter deviations, calculate the balance index of individual physiological parameters, classify the health status intervals according to the index range, determine the interval to which individual physiological parameters belong, and obtain the distribution results of individual health status.

[0068] Based on the dynamic analysis results of physiological parameter deviations, the balance index of individual physiological parameters is calculated. The balance index is calculated based on the comprehensive calculation of multiple physiological parameters. After normalization, the total deviation value is calculated. The calculation formula is as follows: Where B is the balance index, w i Let |D be the weight of the i-th physiological parameter. i | Based on the standardized offset values, weights are assigned according to medical literature data. For example, the weights for heart rate, blood pressure, blood oxygen, and respiratory rate are set to 0.3, 0.4, 0.2, and 0.1, respectively. If an individual's standardized offset values ​​are 1.2, 0.8, 0.5, and 0.3, then their balance index B = 0.3 × 1.2 + 0.4 × 0.8 + 0.2 × 0.5 + 0.1 × 0.3 = 0.78. Based on the index range, the health status is categorized into intervals, and the health status intervals are set as follows: B<0.5 For health, 0.5 ≤ B < 1 is considered sub-healthy, and B ≥ 1 is considered abnormal. If the calculated B = 0.78, it is classified as a sub-healthy state. Finally, the individual's physiological parameters are determined to be within a certain range, and the individual's health status distribution is obtained.

[0069] Please see Figure 3 The specific steps for calculating the distance between the corresponding health status and the preset physiological homeostasis interval in the individual health status distribution results in multidimensional coordinate space, and analyzing the changes of heart rate, blood pressure, blood oxygen, and respiratory rate relative to the physiological homeostasis interval to obtain the health status stability results are as follows:

[0070] S201: Based on the distribution results of individual health status, call the values ​​of heart rate, blood pressure, blood oxygen, and respiratory rate corresponding to the health status, calculate the distance between the coordinates of the individual health status and the coordinates of the physiological steady state interval, and obtain the spatial offset value of the individual health status.

[0071] Based on the distribution of individual health status, the individual's health status category is first obtained, and the corresponding physiological parameters, including heart rate, blood pressure, blood oxygen saturation, and respiratory rate, are extracted. These are used as the parameter set for the individual's health status. For each physiological parameter, a reference range for the physiological steady-state interval is set. The values ​​of the steady-state interval are determined according to medical standards. For example, the steady-state range for heart rate is set to 60-100 beats / minute, blood pressure is set to systolic blood pressure 90-130 mmHg and diastolic blood pressure 60-90 mmHg, blood oxygen saturation is set to 95-100%, and respiratory rate is set to 12-20 breaths / minute. The individual's health status coordinates are calculated, and a four-dimensional coordinate system P = h,bp is defined. s ,bp d ,spo2,rr), where h is the heart rate, bp s and bp d These represent systolic and diastolic blood pressure, respectively; spo2 represents blood oxygen saturation; and rr represents respiratory rate. The coordinates of the physiological homeostasis range are also defined. in The equivalent value is the mean of the steady-state interval. For example, if the mean of the steady-state interval is set as heart rate 80, systolic blood pressure 110, diastolic blood pressure 75, blood oxygen 97.5%, and respiration 16, then P s = 80,110,75,97.5,16 ), calculate the Euclidean distance between the coordinates of an individual's health status and the coordinates of their physiological homeostasis interval:

[0072]

[0073] For example, if an individual's physiological parameters are heart rate 90, systolic blood pressure 120, diastolic blood pressure 80, blood oxygen 95%, and respiration rate 18, then their state coordinates are P = (90, 120, 80, 95, 18). Substituting these values ​​into the calculation... This value is the spatial offset of an individual's health status.

[0074] S202: Based on the spatial offset value of individual health status, calculate the offset distance of heart rate, blood pressure, blood oxygen, and respiratory rate, analyze the changing trend of the physiological steady-state interval corresponding to each parameter, screen parameters that deviate from the steady-state interval, calculate the overall offset vector of individual physiological parameters, and obtain the offset direction and amplitude of physiological parameters based on the directionality and amplitude of the offset vector.

[0075] Based on the spatial offset values ​​of individual health status, the offset distances of heart rate, blood pressure, blood oxygen, and respiratory rate are first calculated individually. The calculation method uses the absolute difference of each physiological parameter relative to the mean of the steady-state interval. For example, for heart rate, the offset distance is calculated... Blood pressure is calculated separately. and For blood oxygen calculation For respiratory rate calculation Then, the changing trends of each physiological parameter within its corresponding steady-state range are analyzed. Parameters with deviations exceeding set thresholds are filtered out. The thresholds are calculated based on the standard deviation of the steady-state range. For example, the heart rate deviation threshold is set to 10 beats / minute, the systolic blood pressure deviation threshold to 15 mmHg, the diastolic blood pressure to 10 mmHg, the blood oxygen saturation to 2%, and the respiratory rate to 3 breaths / minute. If an individual's measured data are a heart rate of 95, systolic blood pressure of 130, diastolic blood pressure of 85, blood oxygen saturation of 93, and respiratory rate of 20, then their deviation values ​​are |95-80|=15, |130-110|=20, |85-75|=10, |93-97.5|=4.5, and |20-16|=4, respectively. After comparison, it is found that all parameters exceed the thresholds. Therefore, all parameters deviating from the steady-state range are filtered out, and the overall deviation vector of the individual's physiological parameters is calculated. The deviation vector is defined. Each component value is an offset value of the corresponding parameter, i.e., V = (15, 20, 10, 4.5, 4). The directionality of this vector is then calculated. Where θ i This indicates the directionality of each parameter's offset relative to the whole; for example, for heart rate, it represents the angle of offset. The offset direction value of each parameter is calculated, and the vector magnitude is also calculated: Substitute into the calculation to obtain Finally, the direction and magnitude of the physiological parameter shift are obtained.

[0076] S203: Based on the direction and magnitude of physiological parameter deviation, calculate the stability index of individual physiological parameters, classify the degree of health status stability, screen individuals with stability index below the steady-state threshold, and obtain the health status stability result.

[0077] The stability of health status is classified using the following formula:

[0078]

[0079] Calculate the stability index, screen individuals whose stability index is below the steady-state threshold, and obtain the health status stability results;

[0080] Among them, S u W is an index representing the stability of an individual's physiological parameters. i D represents the weighting coefficient of the i-th physiological parameter. i D represents the offset value of the i-th physiological parameter of an individual. ref,i The i-th baseline offset value represents the physiological homeostasis range, and n represents the total number of physiological parameters involved in the calculation. Represents the summation operation over all physiological parameter terms, |D i -D ref,i| Represents the absolute deviation between the actual offset value and the baseline offset value of the i-th physiological parameter of an individual. The square root of the sum of squares of all physiological parameter offsets is used to normalize the degree of offset. Represents the sum of the absolute values ​​of the offsets of all physiological parameters of an individual;

[0081] Detailed explanation of the formula and the calculation derivation process:

[0082] The physiological parameters that need to be monitored include heart rate (HR), systolic blood pressure (BP_s), diastolic blood pressure (BP_d), oxygen saturation (SpO2), and respiratory rate (RR). Reference values ​​can be set through medical standards or individual historical health data. After searching for medical standards online, a reasonable health reference range can be selected.

[0083] Heart rate reference value: HR ref =70 times / minute;

[0084] Systolic blood pressure reference value: BP s,ref =120mmHg;

[0085] Diastolic blood pressure reference value: BP d,ref =80 mmHg;

[0086] Reference value for blood oxygen saturation: SpO 2,ref =98%;

[0087] Respiratory rate reference value: RR ref = 16 times / minute;

[0088] Get the current measurement value:

[0089] Obtain current physiological parameters through medical devices or wearable devices:

[0090] Heart rate: HR = 75 beats / minute;

[0091] Systolic blood pressure: BP s =125 mmHg;

[0092] Diastolic blood pressure: BP d =85 mmHg;

[0093] Blood oxygen saturation: SpO2 = 97%;

[0094] Respiratory rate: RR = 18 breaths / minute;

[0095] Calculate the offset value D i :

[0096] Calculate the offset value for each physiological parameter:

[0097] D HR =|75-70|=5;

[0098]

[0099] D RR =|18-16|=2;

[0100] Calculate weight W i :

[0101] Weights are assigned based on the importance of the parameters:

[0102] W HR =0.3;

[0103]

[0104] W RR =0.1;

[0105] Calculate the weighted absolute deviation and the denominator:

[0106] Calculate the weighted absolute deviation:

[0107]

[0108] Substitute the values:

[0109] (0.3×5)+(0.2×5)+(0.2×5)+(0.2×1)+(0.1×2)=1.5+1+1+0.2+0.2=3.9;

[0110] Calculate the sum of squares:

[0111]

[0112] Substitute the values:

[0113]

[0114] Calculate the sum of absolute offsets:

[0115]

[0116] Calculate the stability index S u :

[0117]

[0118] Results analysis:

[0119] The stability index S u A value of 0.9811 indicates that the individual's overall physiological parameters have a small deviation, and the individual is in a relatively stable state of health. The closer the index is to 1, the more stable the physiological parameters; if the index decreases, it indicates that the individual's physiological fluctuations are greater, and the health status may be more unstable.

[0120] Please see Figure 4 The specific steps for identifying inflection points in the trend of health status stability results, obtaining current hemodynamic changes, oxygen delivery, and respiratory drive response, determining the phase lag or enhancement pattern of selected physiological parameters before and after the trend inflection point compared to other parameters, screening for lagging or enhanced physiological parameters, calculating the phase difference between physiological parameters, measuring the linkage effect of each parameter based on phase changes, obtaining the linkage degree of physiological parameters, and finally obtaining the physiological parameter matching results are as follows:

[0121] S301: Based on the stability results of health status, call up the values ​​of physiological parameters, detect changes in hemodynamics, oxygen delivery and respiratory drive response, identify the trend inflection point in the time series of physiological parameters, calculate the rate of change of physiological parameters before and after the inflection point, and obtain the trend inflection point offset value of physiological parameters.

[0122] Based on the stability results of the health status, physiological parameter values ​​are retrieved to obtain time-series data such as heart rate, blood pressure, blood oxygen, and respiratory rate. The data is smoothed to remove short-term noise interference. The detection period is set to 60 seconds, divided into continuous time windows, and the mean change trend of each physiological parameter within the time window is calculated. When detecting hemodynamic changes, systolic blood pressure, diastolic blood pressure, and heart rate data are retrieved, with ΔBP = BP. s -BP d When calculating pulse pressure changes and analyzing blood oxygenation changes, blood oxygen saturation data is used to calculate the change in blood oxygen per unit time, ΔSpO2 = SpO2(t) - SpO2(t-1). When detecting respiratory drive response, respiratory rate data is used to calculate the rate of change in respiration per unit time, ΔRR = RR(t) - RR(t-1). Trend inflection points in the time series of physiological parameters are identified, and the local extremum detection method is used to determine the time point t at which the inflection point occurs. k Calculate the shift in the rate of change of physiological parameters before and after the inflection point, and define the rate of change as... Where X(t) represents the value of a certain physiological parameter at time t, the inflection point offset value is calculated. in and These represent the time points before and after the inflection point, respectively. For example, if an individual's blood pressure changes by 2 mmHg / s 10 seconds before the inflection point and by 5 mmHg / s 10 seconds after the inflection point, then the offset value ΔV = |5 - 2| = 3 mmHg / s, ultimately yielding the physiological parameter trend inflection point offset value.

[0123] S302: Based on the inflection point offset of physiological parameters, calculate the rate of change of hemodynamics, oxygen delivery, and respiratory drive, analyze the phase lag and enhancement pattern of selected physiological parameters before and after the inflection point, screen out physiological parameters with lag or enhancement, calculate the phase difference between physiological parameters, measure the linkage effect of each parameter based on the phase change, and obtain the linkage degree of physiological parameters.

[0124] Based on the inflection point offset of physiological parameters, the rates of change of hemodynamics, oxygen delivery, and respiratory drive are calculated. Systolic blood pressure, diastolic blood pressure, and heart rate data are used to calculate the rate of change of hemodynamics ΔBP. s / Δt and ΔBP d For oxygen delivery, calculate the rate of change of blood oxygen concentration per unit time ΔSpO2 / Δt; for respiratory drive, calculate the rate of change of respiratory rate per unit time ΔRR / Δt. Analyze the phase lag and enhancement patterns of selected physiological parameters before and after the inflection point, and calculate the phase lag Δφ = φ2 - φ1, where φ1 and φ2 represent the phase values ​​before and after the inflection point, respectively. For example, if the heart rate change lags behind the blood pressure change by 0.2 seconds before the inflection point and by 0.5 seconds after the inflection point, then Δφ = 0.5 - 0.2 = 0.3 seconds. Screen for physiological parameters that are lagging or enhanced, and set a lag threshold T. φ With enhancement threshold T V If the phase lag of an individual is Δφ > T φ Or the rate of change of physiological parameters ΔV>T V If the value is θ, it is determined to be a lag or enhancement parameter. The phase difference between physiological parameters is calculated using the formula θ. ij =|φ i -φ j |, where φ i ,φ j These represent the phase values ​​of two physiological parameters, for example, heart rate phase 0.8 seconds and blood oxygen phase 0.5 seconds. The linkage effect of each parameter is measured based on phase changes, and the linkage degree of physiological parameters L = Σ is calculated. i,j w ij ·θ ij , where w ij As the weight of physiological parameters, if set L is calculated, and the physiological parameter linkage degree is finally obtained.

[0125] S303: Calculate the physiological parameter matching degree based on the physiological parameter linkage degree, filter parameter combinations with matching degree lower than the matching degree threshold, and obtain the physiological parameter matching result;

[0126] The physiological parameter matching degree is calculated using the following formula:

[0127]

[0128] Calculate the physiological parameter matching degree, filter parameter combinations with matching degrees below the matching degree threshold, and obtain the physiological parameter matching results;

[0129] Among them, M v L represents the degree of matching of physiological parameters. i L represents the degree of correlation of the i-th physiological parameter of an individual. ref,i V represents the i-th reference linkage degree of the physiological homeostasis range. i Q represents the variation range of the i-th physiological parameter. j Q represents the value of the j-th environmental or individual characteristic factor for an individual. ref,j R represents the j-th reference environmental or individual characteristic factor value of the physiological homeostasis interval. j represents the influence ratio of the j-th environmental or individual characteristic factor, n represents the total number of physiological parameters involved in the calculation, and m represents the total number of additional individual or environmental parameters involved in the calculation. The summation of the products of the correlation deviation and the magnitude of change of all physiological parameters. The sum of the products of the deviations and their influence ratios representing all environmental or individual characteristic factors. This represents the sum of the reference correlation of all physiological parameters and the reference values ​​of individual environmental factors, where k represents a small positive number to avoid a denominator of zero. The square root of the sum of standardized deviations of all physiological parameters and individual environmental factors;

[0130] The formula for calculating the physiological parameter matching degree is as follows:

[0131]

[0132] Detailed explanation of the formula and the calculation derivation process:

[0133] Parameter acquisition:

[0134] Physiological parameter correlation (L) i This involves monitoring an individual's physiological indicators (such as heart rate and blood pressure) and calculating the correlation between these indicators. For example, the correlation between heart rate and blood pressure can be calculated using statistical methods.

[0135] Reference linkage (L) ref,i Based on extensive data from healthy individuals, the standard correlation between various physiological indicators is determined. For example, under normal circumstances, the reference correlation between heart rate and blood pressure can be obtained from epidemiological study data.

[0136] physiological parameter variation range (V) i ): Measures the range of fluctuations in an individual's physiological parameters. For example, the range of heart rate fluctuations can be determined by continuously monitoring the difference between the maximum and minimum values ​​over a period of time.

[0137] Environmental or individual characteristic factor value (Q) j ): Measures the influence of environmental factors (such as temperature and humidity) or individual characteristics (such as age and weight) on physiological parameters. For example, ambient temperature can be measured using a thermometer.

[0138] Reference environmental or individual characteristic factor values ​​(Q) ref,j Reference values ​​for environmental or individual characteristics are determined based on data from healthy individuals. For example, the reference temperature range for a comfortable environment is typically between 20-25°C.

[0139] Impact ratio (R) j ): Indicates the degree of influence of environmental or individual characteristics on physiological parameters. Statistical analysis determines the contribution of each factor to changes in physiological parameters. For example, studies have shown that for every 1°C increase in ambient temperature, heart rate may increase by approximately 1 beat per minute.

[0140] small positive numbers ( k ): Used to avoid the denominator being zero, usually taking a very small value, such as 0.0001.

[0141] Substitute the specific values ​​into the calculation:

[0142] Assume the monitored individual physiological parameters and environmental factors are as follows:

[0143] The correlation between heart rate and blood pressure (L1): 0.8 (calculated from monitoring data)

[0144] Reference heart rate and blood pressure correlation (L) ref,1 0.85 (based on data from healthy individuals)

[0145] Heart rate variability (V1): 10 beats / minute (obtained through continuous monitoring)

[0146] Ambient temperature (Q1): 30℃ (measured with a thermometer)

[0147] Reference ambient temperature (Q) ref,1 22℃ (Reference value for a comfortable environment)

[0148] Temperature effect ratio (R1): 1 time / minute / ℃ (derived from statistical analysis)

[0149] small positive numbers ( k ): 0.0001

[0150] Substitute the above values ​​into the formula:

[0151]

[0152] Calculation process:

[0153] Molecular part:

[0154] |0.8-0.85|·10=0.05×10=0.5;

[0155] |30-22|·1=8×1=8;

[0156] Total numerator: 0.5 + 8 = 8.5;

[0157] Denominator: 0.85 + 22 + 0.0001 = 22.8501;

[0158] Square root part:

[0159]

[0160] The sum of the square roots:

[0161] 0.00135 + 2.7826 ≈ 2.78395;

[0162]

[0163] Final calculation: M v ≈0.621;

[0164] The results indicate that the individual's current physiological parameter matching degree is 0.621, which deviates somewhat from the reference physiological state. The closer this value is to 1, the higher the matching degree between the individual's physiological state and the health reference value; the lower the value, the greater the deviation from the reference value. Based on this matching degree result, the individual's health status can be further analyzed, and it can be determined whether physiological regulation or environmental optimization is needed to bring the physiological state closer to the ideal level.

[0165] Please see Figure 5 Based on the physiological parameter matching results, the rate of change of similar physiological parameters within a time window is extracted to predict the continuous changes in health status over future periods. The specific steps to obtain the health trend prediction results are as follows:

[0166] S401: Based on the physiological parameter matching results, extract the rate of change of similar physiological parameters within the time window, calculate the increment of change at adjacent time points, screen parameters with stable increments of change, analyze the trend direction of physiological parameters, and obtain the physiological parameter change trend analysis results.

[0167] Based on the physiological parameter matching results, time series data of similar physiological parameters such as heart rate, blood pressure, blood oxygen, and respiratory rate are extracted. The rate of change of each parameter is calculated within a time window. For the time window [t], i ,t i+1 The rate of change is defined as... Where X(t) is the measured value of a certain physiological parameter at time t. Taking heart rate as an example, if an individual's heart rate data measured within 5 minutes is [78, 80, 83, 85, 87], corresponding to time points [0, 1, 2, 3, 4] minutes respectively, then the rate of change is calculated as follows: Similarly, calculate the increment of change ΔV = V at adjacent time points. i+1 -V i V i and V i+1 Let represent the rate of change at the i-th and i+1-th time points, respectively. For example, if the rate of change of the aforementioned heart rate data is [2,3,2,2], then the increment of change is calculated as 3-2=1, 2-3=-1, and 2-2=0. Parameters with stable increments of change are selected, and a stability threshold T is set. V If |ΔV| <T V Then it is determined to be stable, for example, if T is set. V =2, then the increment of change [-1,0] meets the stability condition. Analyze the trend direction of physiological parameters and calculate the trend slope θ = arctan(V). For example, if the heart rate change rate is 3 times / minute, then θ = arctan(3). If θ>0, it indicates an upward trend. If θ<0, it indicates a downward trend. Finally, the analysis results of the physiological parameter change trend are obtained.

[0168] S402: Based on the analysis results of the physiological parameter change trend, analyze the consistency of the trend change of each physiological parameter, analyze the change trend of health status in the future time period, screen the health status parameters with abnormal trend change magnitude, and obtain the health trend prediction results.

[0169] Based on the analysis of physiological parameter change trends, the consistency of the trend changes of each physiological parameter is analyzed. For different physiological parameters within the same time window, their trend change similarity is calculated, and the trend similarity is defined as... Among them W i As parameter weights, if the rates of change of heart rate, blood pressure, blood oxygen, and respiratory rate are [2, 3, -1, 0] respectively, and weights are set to [0.4, 0.3, 0.2, 0.1], then the trend similarity is calculated as follows: If a consistency threshold T is set C =1.2, then since C>T C The trend is determined to be consistent, and the changing trend of health status in the future time period is analyzed. Linear extrapolation is used to calculate the future t. f Physiological parameter value X at time f =X0+V·(t) f -t0), for example, if the current heart rate is 85 beats / minute and the rate of change is 3 beats / minute, predict the heart rate X for the next 10 minutes. f=85 + 3 × (10 - 0) = 115 times / minute, filter health status parameters with abnormal trend changes, and set the abnormal threshold T. A If |X f -X0|>T A Then it is judged as an anomaly, for example, setting T. A =20. Since |115-85|=30>20, it is judged as abnormal, and the final health trend prediction result is obtained.

[0170] Please see Figure 6 The specific steps for identifying abnormal fluctuations from health trend prediction results and establishing corresponding early warning markers to obtain individual health monitoring results are as follows:

[0171] S501: Based on the health trend prediction results, extract the time period when the trend deviates from the stable range, set the abnormal fluctuation threshold, identify sudden or continuous abnormal fluctuations, judge the duration and intensity of abnormal fluctuations according to the degree of deviation of physiological parameters, establish abnormal marking rules, and obtain the abnormal marking results of health status.

[0172] Based on health trend prediction results, the time periods during which physiological parameters such as heart rate, blood pressure, blood oxygen, and respiratory rate deviate from the stable range within a time window are extracted. A baseline value for the stable range is set, using the average of the normal physiological range as the center point. The allowable deviation range is calculated; for example, a stable heart rate range of [60, 100] beats / minute, systolic blood pressure of [90, 130] mmHg, blood oxygen of [95, 100]%, and respiratory rate of […]. 12,20 [times / minute], determine the time points when physiological parameters exceed the stable range, and record the time period [t]. s ,t e Set the abnormal fluctuation threshold T. v Calculate the rate of change of each parameter ΔX / Δt. If it exceeds T... v This is then identified as abnormal fluctuation. For example, if a blood pressure abnormal fluctuation threshold is set at 10 mmHg / minute, and an individual's blood pressure rises from 110 mmHg to 130 mmHg within 5 minutes, then the rate of change is considered abnormal. If the blood pressure does not exceed the threshold, it is not considered abnormal. However, if the blood pressure rises from 110 mmHg to 140 mmHg within 2 minutes, then... Exceeding a threshold is considered an abnormal fluctuation. This involves identifying sudden or persistent abnormal fluctuations. Sudden abnormalities are defined as rapid changes in physiological parameters within a short period, while persistent abnormalities are defined as deviations from the stable range that persist for a period of time. The duration of the abnormality, Δt = t, is calculated. e -t s Determine whether the duration exceeds the set threshold T. d For example, setting T d=10 minutes. If the abnormal fluctuation lasts for 15 minutes, it is judged as a persistent abnormality; otherwise, it is judged as a sudden abnormality. The intensity of the abnormality is calculated based on the degree of deviation of physiological parameters, and the intensity index is defined as I = |ΔX / X0|, where ΔX is the change in parameters and X0 is the baseline value. If an individual's blood oxygen saturation drops from 98% to 92%, then I = |92-98| / 98 =6.12%, and a blood oxygen abnormality threshold of 5% is set. Since 6.12% exceeds the threshold, the individual's blood oxygen abnormality is judged to be relatively strong. An abnormality marking rule is established, a classification standard is set, and the abnormality type is defined as mild, moderate and severe. The interval is divided according to the intensity index I. For example, I<3% is mild abnormality, 3%≤I<6% is moderate abnormality, and I≥6% is severe abnormality. Finally, the abnormality marking result of health status is obtained.

[0173] S502: Based on the abnormal health status marker results, extract the physiological parameters corresponding to the abnormal status, assess the repeatability and correlation of abnormal fluctuations, classify individuals according to their risk based on the abnormal status, and obtain individual health monitoring results.

[0174] Based on the abnormal health status labeling results, extract the physiological parameters corresponding to the abnormal status and count the frequency F of abnormality occurrence. s =N b / T s , where N b T represents the number of times an anomaly occurred. s To monitor the total time, for example, if an individual experiences 10 abnormal fluctuations within 24 hours, then F s =10 / 24=0.416 times / hour, assess the repeatability of abnormal fluctuations, and calculate the abnormal interval. If ΔT u Below the set threshold T v If the abnormal fluctuation is determined to have high repeatability, for example, by setting T... v =30 minutes. If the interval between abnormalities in an individual is less than 20 minutes, it is considered a highly repetitive abnormality. Assess the correlation of abnormal fluctuations, calculate the abnormal correlation between physiological parameters, and define the correlation coefficient:

[0175]

[0176] Among them, M x M y These are the outlier sequences for two physiological parameters. These are the means, if |C d A correlation coefficient > 0.7 indicates a strong association. For example, if the correlation between abnormal heart rate and blood pressure is calculated to be 0.8, then the association is considered high. Individuals are then risk-classified based on their abnormal conditions, and a risk score is assigned.

[0177] R g =Wp F s +W q I k +W r C d ;

[0178] Among them, W p W q W r I is the weighting coefficient. k For abnormal strength, such as a certain body F s =0.5, I k =6%, C d =0.8, set W p =0.3,W q =0.5,W r =0.2, then calculate R. g =0.3×0.5+0.5×6+0.2×0.8=0.15+3+0.16=3.31, set the risk classification standard, if R g <2 indicates low risk, 2≤R g <4 indicates medium risk, R g ≥4 indicates high risk, due to the individual's R g =3.31, which is between 2 and 4, so it is classified as a medium-risk individual, and the final individual health monitoring result is obtained.

[0179] Please see Figure 7 A health monitoring smartwatch based on data analysis, the smartwatch includes:

[0180] The physiological parameter analysis module acquires the user's heart rate, blood pressure, blood oxygen, and respiratory rate, analyzes the interaction between physiological parameters, determines whether there is an imbalance of abnormally high or low physiological parameters, and obtains the individual's health status distribution results.

[0181] The steady-state distance calculation module calculates the distance between the corresponding health status and the preset physiological steady-state interval in the individual health status distribution results in a multi-dimensional coordinate space, analyzes the changes of heart rate, blood pressure, blood oxygen, and respiratory rate relative to the physiological steady-state interval, and obtains the health status stability results.

[0182] The linkage matching analysis module identifies the inflection point of the trend in the stability results of health status, obtains the current hemodynamic changes, oxygen delivery, and respiratory drive response, determines the phase lag or enhancement pattern of the selected physiological parameter before and after the trend inflection point compared with other parameters, filters the physiological parameters that lag or enhance, calculates the phase difference between physiological parameters, measures the linkage effect of each parameter based on the phase change, obtains the linkage degree of physiological parameters, and then obtains the physiological parameter matching results.

[0183] The health trend prediction module extracts the rate of change of similar physiological parameters within a time window based on the physiological parameter matching results, predicts the continuous change of health status in the future time period, and obtains the health trend prediction results.

[0184] The abnormal early warning module identifies abnormal fluctuations from the health trend prediction results and establishes corresponding early warning markers to obtain individual health monitoring results.

[0185] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A health monitoring method based on data analysis, characterized in that, Includes the following steps: S1: Obtain the user's heart rate, blood pressure, blood oxygen, and respiratory rate physiological parameters, analyze the dynamic correlation between physiological parameters, determine whether there is an imbalance of abnormal increase or decrease in physiological parameters, and obtain the individual health status distribution results. S2: Calculate the distance between the corresponding health status and the preset physiological steady-state interval in the individual health status distribution results in the multi-dimensional coordinate space, analyze the changes of heart rate, blood pressure, blood oxygen, and respiratory rate relative to the physiological steady-state interval, and obtain the health status stability results; S3: Identify the inflection point of the trend in the stability results of the health status, obtain the current hemodynamic changes, oxygen delivery, and respiratory drive response, determine the phase lag or enhancement pattern of the selected physiological parameter before and after the trend inflection point compared with other parameters, filter the physiological parameters that lag or enhance, calculate the phase difference between physiological parameters, measure the linkage effect of each parameter based on the phase change, obtain the linkage degree of physiological parameters, and then obtain the physiological parameter matching result. S4: Based on the physiological parameter matching results, extract the rate of change of similar physiological parameters within a time window, predict the continuous change of health status in the future time period, and obtain the health trend prediction results. S5: Identify abnormal fluctuations from the health trend prediction results and establish corresponding early warning markers to obtain individual health monitoring results.

2. The health monitoring method based on data analysis according to claim 1, characterized in that: The individual health status distribution results include heart rate status, blood pressure status, blood oxygen status, and respiratory rate status; the health status stability results include heart rate stability, blood pressure stability, blood oxygen stability, and respiratory rate stability; the physiological parameter matching results include heart rate matching degree, blood pressure matching degree, blood oxygen matching degree, and respiratory rate matching degree; the health trend prediction results include heart rate change trend, blood pressure change trend, blood oxygen change trend, and respiratory rate change trend; and the individual health monitoring results include abnormal heart rate fluctuations, abnormal blood pressure fluctuations, abnormal blood oxygen fluctuations, abnormal respiratory rate fluctuations, and health warning markers.

3. The health monitoring method based on data analysis according to claim 1, characterized in that: The specific steps for obtaining a user's physiological parameters such as heart rate, blood pressure, blood oxygen, and respiratory rate, analyzing the interactions between these parameters, determining whether there are any imbalances in the abnormal elevation or decrease of these parameters, and obtaining the individual's health status distribution results are as follows: S101: Acquire the user's heart rate, blood pressure, blood oxygen, and respiratory rate data, continuously collect multiple measurements, calculate the mean, fluctuation range, and rate of change for each parameter, filter out abnormal deviation data, determine data stability based on the rate of change and fluctuation range, and obtain stability analysis results; S102: Based on the stability analysis results, calculate the mean and fluctuation range of heart rate, blood pressure, blood oxygen, and respiratory rate, analyze the correlation between each parameter, screen the associated parameter combinations, calculate the degree of deviation between each parameter, and determine whether there is an abnormal increase or decrease trend based on the degree of deviation and fluctuation range, and obtain the dynamic analysis results of physiological parameter deviation. S103: Based on the dynamic analysis results of the physiological parameter deviation, calculate the balance index of individual physiological parameters, classify the health status intervals according to the index range, determine the interval to which the individual physiological parameters belong, and obtain the distribution results of individual health status.

4. The health monitoring method based on data analysis according to claim 1, characterized in that: The specific steps for calculating the distance between the corresponding health status and the preset physiological homeostasis interval in the individual health status distribution results in multidimensional coordinate space, and analyzing the changes of heart rate, blood pressure, blood oxygen, and respiratory rate relative to the physiological homeostasis interval to obtain the health status stability results are as follows: S201: Based on the individual health status distribution results, call the heart rate, blood pressure, blood oxygen, and respiratory rate values ​​corresponding to the health status, calculate the distance between the individual health status coordinates and the physiological steady state interval coordinates, and obtain the individual health status spatial offset value. S202: Based on the individual health status spatial offset value, calculate the offset distance of heart rate, blood pressure, blood oxygen, and respiratory rate, analyze the changing trend of the physiological steady-state interval corresponding to each parameter, filter parameters that deviate from the steady-state interval, calculate the overall offset vector of individual physiological parameters, and obtain the offset direction and amplitude of physiological parameters based on the directionality and amplitude of the offset vector. S203: Based on the direction and magnitude of the physiological parameter offset, calculate the stability index of the individual's physiological parameters, classify the degree of health status stability, screen individuals with a stability index lower than the steady-state threshold, and obtain the health status stability result.

5. The health monitoring method based on data analysis according to claim 4, characterized in that: For calculating the stability index S of individual physiological parameters u The formula used is: Among them, W i D represents the weighting coefficient of the i-th physiological parameter. i D represents the offset value of the i-th physiological parameter of an individual. ref,i The i-th baseline offset value represents the physiological homeostasis range, and n represents the total number of physiological parameters involved in the calculation. Represents the summation operation over all physiological parameter terms, |D i -D ref,i | Represents the absolute deviation between the actual offset value and the baseline offset value of the i-th physiological parameter of an individual. The square root of the sum of squares of all physiological parameter offsets is used to normalize the degree of offset. It represents the sum of the absolute values ​​of the offsets of all physiological parameters of an individual.

6. The health monitoring method based on data analysis according to claim 1, characterized in that: The specific steps for identifying the inflection point of the trend in the stability results of the health status, obtaining the current hemodynamic changes, oxygen delivery, and respiratory drive response, determining the phase lag or enhancement pattern of the selected physiological parameter before and after the trend inflection point compared with other parameters, screening the lagging or enhanced physiological parameters, calculating the phase difference between physiological parameters, measuring the linkage effect of each parameter based on the phase change, obtaining the linkage degree of physiological parameters, and finally obtaining the physiological parameter matching results are as follows: S301: Based on the stability results of the health status, call up the values ​​of physiological parameters, detect changes in hemodynamics, oxygen delivery and respiratory drive response, identify the trend inflection point in the time series of physiological parameters, calculate the rate of change of physiological parameters before and after the inflection point, and obtain the trend inflection point offset value of physiological parameters. S302: Based on the inflection point offset value of the physiological parameters, calculate the rate of change of hemodynamics, oxygen delivery, and respiratory drive, analyze the phase lag and enhancement mode of the selected physiological parameters before and after the inflection point, screen the physiological parameters that are lagging or enhanced, calculate the phase difference between the physiological parameters, measure the linkage effect of each parameter according to the phase change, and obtain the linkage degree of the physiological parameters. S303: Calculate the physiological parameter matching degree based on the physiological parameter linkage degree, filter parameter combinations with matching degree lower than the matching degree threshold, and obtain the physiological parameter matching result.

7. The health monitoring method based on data analysis according to claim 6, characterized in that: For calculating the physiological parameter matching degree M v The formula used is: Among them, L i L represents the degree of correlation of the i-th physiological parameter of an individual. ref,i V represents the i-th reference linkage degree of the physiological homeostasis range. i Q represents the variation range of the i-th physiological parameter. j Q represents the value of the j-th environmental or individual characteristic factor for an individual. ref,j R represents the j-th reference environmental or individual characteristic factor value of the physiological homeostasis interval. j represents the influence ratio of the j-th environmental or individual characteristic factor, n represents the total number of physiological parameters involved in the calculation, and m represents the total number of additional individual or environmental parameters involved in the calculation.

8. The health monitoring method based on data analysis according to claim 1, characterized in that: Based on the physiological parameter matching results, the rate of change of similar physiological parameters within a time window is extracted to predict the continuous change of health status in the future time period. The specific steps to obtain the health trend prediction results are as follows: S401: Based on the physiological parameter matching results, extract the rate of change of similar physiological parameters within the time window, calculate the increment of change at adjacent time points, screen parameters with stable increments of change, analyze the trend direction of physiological parameters, and obtain the physiological parameter change trend analysis results. S402: Based on the analysis results of the physiological parameter change trends, analyze the consistency of the trend changes of each physiological parameter, analyze the change trend of health status in the future time period, screen health status parameters with abnormal trend change magnitude, and obtain health trend prediction results.

9. The health monitoring method based on data analysis according to claim 1, characterized in that: The specific steps for identifying abnormal fluctuations from the health trend prediction results and establishing corresponding early warning markers to obtain individual health monitoring results are as follows: S501: Based on the health trend prediction results, extract the time period when the trend deviates from the stable range, set the abnormal fluctuation threshold, identify sudden or continuous abnormal fluctuations, determine the duration and intensity of abnormal fluctuations based on the degree of deviation of physiological parameters, establish abnormal marking rules, and obtain abnormal health status marking results. S502: Based on the abnormal health status marker results, extract the physiological parameters corresponding to the abnormal status, assess the repeatability and correlation of abnormal fluctuations, classify individuals according to their risk based on the abnormal status, and obtain individual health monitoring results.

10. A health monitoring smartwatch based on data analysis, characterized in that, The health monitoring method based on data analysis according to any one of claims 1-9 is executed, wherein the smartwatch comprises: The physiological parameter analysis module acquires the user's heart rate, blood pressure, blood oxygen, and respiratory rate, analyzes the interaction between physiological parameters, determines whether there is an imbalance of abnormally high or low physiological parameters, and obtains the individual's health status distribution results. The steady-state distance calculation module calculates the distance between the corresponding health status and the preset physiological steady-state interval in the individual health status distribution results in a multi-dimensional coordinate space, analyzes the changes of heart rate, blood pressure, blood oxygen, and respiratory rate relative to the physiological steady-state interval, and obtains the health status stability results. The linkage matching analysis module identifies the inflection point of the trend in the stability results of the health status, obtains the current hemodynamic changes, oxygen delivery, and respiratory drive response, determines the phase lag or enhancement pattern of the selected physiological parameter before and after the trend inflection point compared with other parameters, filters the physiological parameters that lag or enhance, calculates the phase difference between physiological parameters, measures the linkage effect of each parameter based on the phase change, obtains the linkage degree of physiological parameters, and then obtains the physiological parameter matching result. Based on the physiological parameter matching results, the health trend prediction module extracts the rate of change of similar physiological parameters within a time window, predicts the continuous change of health status in the future time period, and obtains the health trend prediction result. The abnormal early warning module identifies abnormal fluctuations from the health trend prediction results and establishes corresponding early warning markers to obtain individual health monitoring results.

Images