A chronic disease risk management and control method based on community gridding

Abstract

This invention relates to the field of health and medical data processing technology, and particularly to a method for chronic disease risk management based on community grids. The method includes constructing an initial spatial environmental potential field based on point-of-interest weights; calculating the risk exposure integral sequence and instantaneous potential gradient by combining user trajectories; if the gradient exceeds a threshold, adjusting parameters and using a matrix contour algorithm to extract stress-related abnormal motifs from physiological monitoring data; reconstructing the potential field by statistically analyzing the frequency of group motif attacks; finally, calculating the cross-correlation between the exposure integral and the attack frequency sequence to obtain a causal index; and combining this with health records to output a chronic disease risk rating and intervention instructions. In this invention, by introducing an instantaneous potential gradient to feedforward and regulate the extraction parameters of physiological motifs, the technical problem of traditional fixed-parameter sampling easily missing environmental stress physiological responses is solved, enabling the underlying algorithm to accurately capture users' stress-related abnormal physiological motifs during sudden environmental changes.

Description

Technical Field

[0001] This invention relates to the field of health and medical data processing technology, and in particular to a method for chronic disease risk management based on community grid management. Background Technology

[0002] With the increasing demand for chronic disease management, health monitoring combining user location information with wearable device physiological data has become a routine practice. In continuous physiological data monitoring, existing data processing systems typically rely on preset and fixed sliding window lengths and distance matching thresholds to extract physiological feature sequences. However, as users move in physical space, the potential risk factors in their microenvironment dynamically change. When individuals enter or cross areas with significant environmental differences, changes in external environmental conditions often trigger short-lived stress-induced physiological fluctuations.

[0003] Because the feature extraction algorithm parameters of existing systems are static and fixed, they cannot perceive and respond to changes in external environmental risks. This makes it difficult for the system to adaptively adjust the matching sensitivity of the computing units when environmental changes occur. This computational mode, which separates environmental risk states from physiological feature extraction parameters, makes the system prone to missing abnormal physiological motifs triggered by environmental stress. This limits the ability of the underlying algorithm to capture short-term abnormal physiological responses in complex scenarios, thereby affecting the accuracy of chronic disease health risk assessment. Summary of the Invention

[0004] To overcome the above shortcomings, this invention provides a community-based grid-based method for chronic disease risk management, which aims to improve the problem that short-term stress-related abnormal physiological responses are easily missed or misjudged due to fixed feature extraction parameters.

[0005] In a first aspect, the present invention provides the following technical solution: a method for chronic disease risk management based on community grid management, comprising: Step S1: Divide the target community into multiple grid cells, extract the interest point category features within the grid cells and assign them health risk weights, and calculate and construct the initial spatial environment potential field; Step S2: Obtain the user's spatial movement trajectory data, and perform integration and difference calculations on the initial spatial environment potential energy field based on the spatial movement trajectory data to obtain the risk potential energy exposure integral sequence and instantaneous potential energy gradient; Step S3: Collect continuous physiological monitoring data. When the instantaneous potential energy gradient is greater than the preset stress trigger threshold, trigger the adjustment of feature extraction parameters and use the matrix contour algorithm to extract stress-abnormal physiological motifs from the continuous physiological monitoring data. Step S4: Calculate the cumulative frequency of the occurrence of the stress-induced abnormal physiological motifs triggered by the population passing through each grid unit within the preset observation window, update the health risk weights using the logarithmic correction function, and reconstruct the potential energy field of the space environment. Step S5: Calculate the cross-correlation between the risk potential exposure integral sequence and the frequency sequence of the stress abnormal physiological motif, extract the pathogenic causal association index, and output a comprehensive chronic disease risk rating and intervention instructions in combination with electronic health record data.

[0006] Preferably, in step S1, the steps of extracting the interest point category features within the grid cell and assigning health risk weights, and calculating the initial spatial environment potential field include: The number of interest points of different categories within each grid cell is counted. The number of interest points of each category is multiplied by their corresponding health risk weights and then summed to obtain the static discrete potential energy value of the corresponding grid cell. A Gaussian kernel function is used to perform spatial smoothing mapping on the static discrete potential energy values ​​of each grid cell to generate an initial spatial environment potential energy field. The Gaussian kernel function includes a spatial bandwidth parameter used to control the degree of diffusion and attenuation of potential energy into the surrounding physical space.

[0007] Preferably, in step S2, the step of performing integration and difference calculations on the initial spatial environment potential energy field based on the spatial movement trajectory data to obtain the risk potential energy exposure integral sequence and instantaneous potential energy gradient includes: The spatial movement trajectory data is mapped into a continuous time coordinate and spatial coordinate function; The initial spatial environment potential energy field is calculated by line integral along the spatial movement trajectory data to obtain the risk potential energy exposure integral, and then converted into the risk potential energy exposure integral sequence by time slice; The instantaneous potential gradient is calculated based on the first-order difference of the initial spatial environment potential energy field values ​​at the user's location under adjacent discrete time steps.

[0008] Preferably, in step S3, when the instantaneous potential energy gradient is greater than a preset stress trigger threshold, the step of triggering the adjustment of feature extraction parameters includes: Baseline sliding window length and baseline distance matching threshold are used as default feature extraction parameters; When the absolute value of the instantaneous potential energy gradient exceeds the stress triggering threshold, the difference between the absolute value of the instantaneous potential energy gradient and the stress triggering threshold is calculated. Based on the difference and a preset adjustment coefficient, the baseline sliding window length is shortened synchronously and the baseline distance matching threshold is reduced to obtain the adjusted feature extraction parameters.

[0009] Preferably, in step S3, the step of extracting stress-abnormal physiological motifs from the continuous physiological monitoring data using the matrix profilometry algorithm includes: The adjusted feature extraction parameters are used as input parameters for the matrix contour algorithm; Calculate the distance matrix profile of the continuous physiological monitoring data and identify abnormal subsequences in the distance matrix profile that are less than the adjusted baseline distance matching threshold; The extracted abnormal subsequences are used as the stress-related abnormal physiological motifs.

[0010] Preferably, in step S4, the step of updating the health risk weights using a logarithmic correction function and reconstructing the spatial environment potential field includes: Determine whether the cumulative frequency of attacks within a specific grid cell is greater than a preset baseline frequency threshold; When the frequency exceeds the baseline threshold, the difference between the cumulative attack frequency and the baseline frequency threshold is calculated. The logarithm of the excess difference is calculated by adding one, and then multiplied by a preset feedback damping coefficient to obtain the correction increment; The historical health risk weight of the corresponding grid cell is added to the correction increment to obtain the updated health risk weight, which is then re-input into the spatial smoothing mapping process to reconstruct the spatial environment potential field.

[0011] Preferably, in step S5, the step of calculating the cross-correlation between the risk potential exposure integral sequence and the frequency sequence of the stress abnormal physiological motif, and extracting the pathogenic causal association index, includes: Frequency data of the stress-related abnormal physiological motifs located on the same time axis are extracted and converted into the attack frequency sequence; Set a time lag constant to characterize physiological abnormalities induced by environmental exposure; The sum of the products of the risk potential exposure integral sequence and the occurrence frequency sequence with the time lag constant is calculated to obtain the cross-correlation sequence under different lag time conditions. The maximum value in the cross-correlation sequence is extracted as the pathogenic causal association index.

[0012] Preferably, in step S5, the step of outputting a comprehensive risk rating and intervention instructions for chronic diseases based on electronic health record data includes: Extract static baseline medical history data from the electronic health record data to calculate the basic static risk score; The basic static risk score and the pathogenic causal association index are assigned normalized weights and then weighted and summed to obtain a comprehensive risk score. The comprehensive risk score is mapped to a preset risk grading range, the comprehensive risk rating of chronic diseases is output, and the intervention instruction is triggered.

[0013] Preferably, the step of triggering the generation of the intervention instruction includes: Extract the time lag constant that causes the cross-correlation sequence to reach its maximum value; Based on the time lag constant backtracking, specific grid cell trajectory segments that trigger high risk and the matched stress-abnormal physiological motif features are extracted. The specific grid cell trajectory segment, together with the stress-abnormal physiological motif features, is encapsulated into a pathogenic causal evidence chain; The comprehensive risk rating of chronic diseases and the causal evidence chain of disease are packaged and sent to the management terminal to generate intervention instructions that include information on avoiding specific grid units.

[0014] Secondly, the present invention provides the following technical solution: a chronic disease risk management system based on community grid management, the system comprising: The potential energy field construction module is used to divide the target community into multiple grid cells, extract the interest point category features within the grid cells and assign health risk weights, and calculate and construct the initial spatial environment potential energy field. The data calculation module is used to acquire the user's spatial movement trajectory data, and perform integral and difference calculations on the initial spatial environment potential energy field based on the spatial movement trajectory data to obtain the risk potential energy exposure integral sequence and instantaneous potential energy gradient. The motif extraction module is used to collect continuous physiological monitoring data. When the instantaneous potential energy gradient is greater than the preset stress trigger threshold, it triggers the adjustment of feature extraction parameters and uses a matrix contour algorithm to extract stress-abnormal physiological motifs from the continuous physiological monitoring data. The potential energy field reconstruction module is used to count the cumulative frequency of the occurrence of the stress-induced abnormal physiological motif triggered by the population passing through each grid unit within a preset observation window, update the health risk weight using a logarithmic correction function, and reconstruct the potential energy field of the space environment. The risk assessment module is used to calculate the cross-correlation between the risk potential exposure integral sequence and the frequency sequence of the stress abnormal physiological motif, extract the pathogenic causal association index, and output a comprehensive chronic disease risk rating and intervention instructions in combination with electronic health record data.

[0015] The present invention has the following beneficial effects: 1. This invention solves the technical problem that traditional fixed-parameter sampling easily misses environmental stress physiological responses by introducing instantaneous potential energy gradient to feedforward control the extraction parameters of physiological motifs. When the system detects that a user is crossing a physical boundary with a large potential energy difference, it will adaptively shorten the sliding window length of the matrix contour algorithm and reduce the distance matching threshold according to the change of instantaneous potential energy gradient. This enables the underlying algorithm to accurately capture the user's abnormal physiological motifs under stress when the environment changes abruptly, effectively improving the targeting of continuous physiological data monitoring and feature extraction.

[0016] 2. This invention constructs a spatial environmental potential energy field feedback correction closed loop based on the physiological evolution characteristics of the population, overcoming the defect of static lag in physical space risk attributes in traditional grid management. The system statistically analyzes the cumulative frequency of abnormal motifs triggered by the population in each grid unit within a specific time window, uses a logarithmic correction function to update the corresponding health risk weights and reconstruct the potential energy field, and uses real physiological micro-change data of the population to objectively reflect the health impact of the physical space. It can promptly uncover hidden pathogenic factors that cannot be reflected by conventional static information, thus improving the accuracy of environmental risk assessment.

[0017] 3. This invention uses cross-correlation analysis of feature sequences to replace conventional deep learning algorithm models, which significantly improves the interpretability of comprehensive risk assessment and intervention guidance for chronic diseases. By calculating the cross-correlation between the risk potential exposure integral sequence and the frequency sequence of abnormal physiological motifs under different lag time conditions, the system can clearly extract the causal association index of physiological abnormalities induced by environmental exposure. This not only eliminates the interference of occasional physiological fluctuations, but also backtracks to generate a chain of evidence containing specific high-risk trajectory segments and abnormal physiological characteristics, providing a reliable basis for management terminals to issue clear grid avoidance and behavioral intervention instructions. Attached Figure Description

[0018] Figure 1 This is a flowchart of a chronic disease risk management method based on community grid management proposed in this invention; Figure 2 This is a system module diagram of a community-based grid-based chronic disease risk management method proposed in this invention. Detailed Implementation

[0019] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] Example 1: In the first embodiment of the present invention, the present invention provides a method for chronic disease risk management based on community grid management, such as... Figure 1 As shown, it includes the following steps: Step S1: Divide the target community into multiple grid cells, extract the interest point category features within the grid cells and assign them health risk weights, and calculate and construct the initial spatial environment potential field; In step S1, the steps of extracting the interest point category features within the grid cells and assigning health risk weights, and calculating the initial spatial environment potential field include: The number of interest points of different categories within each grid cell is counted. The number of interest points of each category is multiplied by their corresponding health risk weights and then summed to obtain the static discrete potential energy value of the corresponding grid cell. A Gaussian kernel function is used to perform spatial smoothing mapping on the static discrete potential energy values ​​of each grid cell to generate an initial spatial environment potential energy field. The Gaussian kernel function includes a spatial bandwidth parameter to control the degree of diffusion and attenuation of potential energy into the surrounding physical space.

[0021] Specifically, the geographic information system (GIS) boundary data of the target community is acquired. Based on the set spatial resolution parameters, the physical space of the target community is divided into multiple grid cells of equal area. Each grid cell is assigned spatial coordinates. ,in Indicates the number of the grid cell. ,and This represents the total number of grid cells within the target community.

[0022] Subsequently, the steps of extracting the interest point category features within the grid cells and assigning health risk weights to calculate and construct the initial spatial environment potential field specifically include the following processes: In the first step, the number of interest points of different categories in each grid cell is counted. The number of interest points of each category is multiplied by their corresponding health risk weights and then summed to obtain the static discrete potential energy value of the corresponding grid cell.

[0023] Interest points (POPs) data within each grid cell are obtained by calling the map service application programming interface (API). These POPs are then categorized according to their chronic disease risk association attributes. There are categories, and the set of categories is denoted as . The categories include fast food restaurants, convenience stores, gyms, parks, and medical facilities.

[0024] The system pre-sets a data table containing a mapping relationship between different points of interest categories and health risk weights. For each category... Assign a basic health risk weight The aforementioned health risk weights The value range is [-1, 1]. Positive values ​​indicate factors that increase the risk of chronic disease flare-ups, while negative values ​​indicate factors that decrease the risk of chronic disease flare-ups.

[0025] For the The number of grid cells is counted, and the cells belonging to the first grid cell are statistically analyzed. The number of points of interest in a category is denoted as The static discrete potential energy value of the grid cell is calculated according to the following formula. : ; In the formula, Indicates coordinates as The The static discrete potential energy value of each grid cell; M is the total number of interest point categories; For the first Within the first grid cell The number of points of interest; For the first Health risk weights corresponding to points of interest.

[0026] In the second stage, a Gaussian kernel function is used to perform spatial smoothing mapping on the static discrete potential energy values ​​of each grid cell to generate an initial spatial environment potential energy field. The Gaussian kernel function includes a spatial bandwidth parameter used to control the degree of diffusion and attenuation of potential energy into the surrounding physical space.

[0027] A two-dimensional Gaussian kernel function is used to perform spatial smoothing on the calculated static discrete potential energy matrix. The coordinates of the target smoothing evaluation point are set as follows: By combining the static discrete potential energy values ​​of all grid cells, the initial spatial environment potential energy field at the evaluation point coordinates is calculated. The specific calculation formula is as follows: ; In the formula, Indicates assessment points The initial potential energy value of the spatial environment at the location; The total number of grid cells; For the first The static discrete potential energy value of each grid cell; For the first Spatial coordinate identifiers for each grid cell; Indicates assessment points With the The square of the Euclidean distance between the spatial coordinates of each grid cell; This is the spatial bandwidth parameter, which is a constant and is used to control the effective distance at which the risk attributes of a single grid cell radiate and attenuate to the surrounding area.

[0028] Through the above calculations, a continuous spatial environment potential energy field distribution matrix is ​​output in the two-dimensional plane coordinate system of the target community, which serves as the initial spatial environment potential energy field.

[0029] Step S2: Obtain the user's spatial movement trajectory data, and perform integration and difference calculations in the initial spatial environment potential energy field based on the spatial movement trajectory data to obtain the risk potential energy exposure integral sequence and instantaneous potential energy gradient; In step S2, integration and difference calculations are performed on the initial spatial environment potential energy field based on the spatial movement trajectory data to obtain the risk potential energy exposure integral sequence and instantaneous potential energy gradient. The steps include: Map spatial movement trajectory data into a continuous time coordinate and spatial coordinate function; Line integrals are performed on the initial spatial environment potential energy field along the spatial movement trajectory data to obtain the risk potential energy exposure integral, which is then converted into a risk potential energy exposure integral sequence by time slices. The instantaneous potential gradient is calculated based on the first-order difference of the initial spatial environment potential field values ​​at the user's location in adjacent discrete time steps.

[0030] Specifically, spline interpolation algorithm is used to numerically fit the discrete set of trajectory points to generate time variables. Two-dimensional coordinates in space A continuous mapping relationship between them. The user's monitoring time domain is set as... Within this time domain, construct time-continuous coordinate functions, denoted as x-coordinate functions. and ordinate function Therefore, the user's spatial movement trajectory is transformed into a system of time-parametric equations. .

[0031] In the second stage, the initial spatial environment potential energy field is calculated by line integral along the spatial movement trajectory data to obtain the risk potential energy exposure integral, which is then converted into a risk potential energy exposure integral sequence by time slice.

[0032] The initial spatial environment potential field matrix data constructed in step S1 is denoted as... Set a fixed time slice length. Total monitoring time domain Divided into The nth consecutive non-overlapping time slices The time intervals corresponding to each time slice are denoted as . ,in .

[0033] In the Within a time slice, along the continuous movement trajectory For the initial space environment potential field Perform the first type of line integral calculation and output the first type of line integral. Risk potential exposure integral for each time slice The specific calculation formula is as follows: ; In the formula, Indicates the first Risk potential exposure integral over a time interval; Indicates a point in time The initial spatial environment potential energy value at the time trajectory coordinates; Indicates a point in time At the time user The component of the velocity along the axial direction; Indicates a point in time At the time user The component of the velocity along the axial direction.

[0034] Based on chronological order, combine the above divisions. The risk potential energy exposure integral calculation results for each time slice are used to generate a structured one-dimensional array, which serves as the risk potential energy exposure integral sequence, denoted as... .

[0035] In the third stage, the instantaneous potential gradient is calculated based on the first-order difference of the initial spatial environment potential energy field values ​​at the user's location under adjacent discrete time steps.

[0036] Set the discrete sampling time step parameter as follows Get the current evaluation time point. and its previous sampling time point The corresponding trajectory coordinates are respectively and .

[0037] Based on the coordinates, the initial spatial environment potential energy field matrix is ​​queried, and the potential energy values ​​at the corresponding locations are extracted, respectively. and Perform a first-order difference operation on the potential energy value with respect to the sampling time step, and output the current evaluation time point. instantaneous potential gradient The specific calculation formula is as follows: ; In the formula, Indicates a point in time The instantaneous potential energy gradient under changes in physical environmental risk; Indicates a point in time The initial spatial potential energy value of the user's current location; Indicates the previous sampling time point The initial spatial environmental potential energy value of the user's historical location; This represents the time step constant between adjacent sampling points. The system continuously calculates the instantaneous potential energy gradient at each time point and outputs a set of gradient values ​​characterizing the degree of drastic change in the user's environmental risk, which is used to trigger parameter adjustments in the subsequent sequence extraction module.

[0038] Step S3: Collect continuous physiological monitoring data. When the instantaneous potential energy gradient is greater than the preset stress trigger threshold, trigger the adjustment of feature extraction parameters and use the matrix contour algorithm to extract stress-abnormal physiological motifs from the continuous physiological monitoring data. In step S3, when the instantaneous potential energy gradient is greater than the preset stress trigger threshold, the step of triggering the adjustment of feature extraction parameters includes: Baseline sliding window length and baseline distance matching threshold are used as default feature extraction parameters; When the absolute value of the instantaneous potential energy gradient exceeds the stress triggering threshold, calculate the difference between the absolute value of the instantaneous potential energy gradient and the stress triggering threshold. Based on the difference and the preset adjustment coefficient, the baseline sliding window length is shortened simultaneously and the baseline distance matching threshold is reduced to obtain the adjusted feature extraction parameters. Step S3, which involves extracting stress-related abnormal physiological motifs from continuous physiological monitoring data using a matrix profilometry algorithm, includes: The adjusted feature extraction parameters are used as input parameters for the matrix contour algorithm; Calculate the distance matrix profile of continuous physiological monitoring data and identify anomalous subsequences in the distance matrix profile that are less than the adjusted baseline distance matching threshold; The extracted abnormal subsequences were used as stress-induced abnormal physiological motifs.

[0039] Specifically, through a physiological data acquisition device connected to the system, continuous physiological monitoring data of the user is acquired at a set sampling frequency. The continuous physiological monitoring data is arranged in chronological order, forming a one-dimensional time series data set.

[0040] In the first stage of the process, when the instantaneous potential energy gradient is greater than the preset stress trigger threshold, the feature extraction parameters are adjusted.

[0041] The system memory pre-stores default feature extraction parameters. These default parameters include the baseline sliding window length and the baseline distance matching threshold. When the system determines that the environmental risk is stable, the baseline sliding window length and the baseline distance matching threshold are used as static input parameters for the underlying feature extraction algorithm.

[0042] The system receives the instantaneous potential energy gradient data sequence calculated in step S2 and extracts the preset stress triggering threshold constant. Get the current evaluation time point. instantaneous potential gradient Performing the absolute value operation yields and judge Is it greater than the stress trigger threshold? .

[0043] When the condition is determined Upon activation, the system triggers a dynamic adjustment mechanism for feature extraction parameters. First, it calculates the difference between the absolute value of the instantaneous potential energy gradient and the stress trigger threshold. The specific calculation formula is as follows: ; In the formula, This represents the potential energy gradient difference exceeding the threshold. Indicates the assessment time point The absolute value of the instantaneous potential energy gradient; This indicates the preset stress trigger threshold.

[0044] Subsequently, based on the calculated difference In conjunction with pre-configured adjustment coefficients, the baseline sliding window length is shortened and the baseline distance matching threshold is lowered, outputting the adjusted feature extraction parameters. The specific update calculation formula is as follows: ; ; In the formula, Indicates the adjusted length of the sliding window; Indicates the default baseline sliding window length; This represents the preset constant coefficient used for window length adjustment; This indicates a floor function, ensuring the window length is an integer. This represents the minimum lower limit of the sliding window length set by the system. This indicates the adjusted distance matching threshold; This indicates the default baseline distance matching threshold; This represents the preset constant coefficient used for adjusting the distance matching threshold; This represents the minimum lower limit of the distance matching threshold set by the system. Through the above calculations, the feature extraction parameters are updated in the corresponding triggered state.

[0045] In the second stage, the matrix contour algorithm is used to extract stress-related abnormal physiological motifs from continuous physiological monitoring data.

[0046] Adjust the sliding window length of the above calculation output. And the adjusted distance matching threshold It is passed to the algorithm module as an execution parameter for the matrix contour algorithm.

[0047] The sequence of continuously collected physiological monitoring data is set as follows: ,in This represents the total number of data points contained in the sequence. Using a length of... Sliding window in time series The algorithm iterates through the data at fixed steps, generating a series of continuous subsequences. It calculates the Euclidean distance between each subsequence and records the minimum distance between each subsequence and its nearest neighbor, combining these values ​​to form a distance matrix contour vector. The vector The first in element Corresponding to time series The Middle The shortest matching distance of a continuous subsequence starting from a data point.

[0048] System traversal distance matrix contour vector All elements in the set, with each element's value... Adjusted distance matching threshold with the input algorithm module Numerical comparison is performed to identify anomalous subsequences in the distance matrix contour that are less than the adjusted baseline distance matching threshold. The comparison logic is as follows: ; When the above judgment logic is true, the system extracts the continuous physiological monitoring data sequence. From the middle Starting from a time point, with a length of [number]... Data segmentation. All data segments that meet this condition are extracted and marked as anomalous subsequences. The extracted anomalous subsequences are then used as stress-related abnormal physiological motifs for the assessment period and stored in the system database for subsequent correlation index calculation and risk assessment.

[0049] Step S4: Calculate the cumulative frequency of occurrence of stress-induced abnormal physiological motifs triggered by the population passing through each grid unit within the preset observation window, update the health risk weights using the logarithmic correction function, and reconstruct the potential energy field of the space environment. In step S4, the steps of updating the health risk weights using the logarithmic correction function and reconstructing the potential energy field of the space environment include: Determine whether the cumulative frequency of attacks within a specific grid cell is greater than a preset baseline frequency threshold; When the frequency exceeds the baseline threshold, the difference between the cumulative frequency of attacks and the baseline frequency threshold is calculated. The logarithm of the deviation exceeding the standard is calculated by adding one, and then multiplied by the preset feedback damping coefficient to obtain the correction increment; The historical health risk weights of the corresponding grid cells are added to the correction increments to obtain the updated health risk weights, which are then re-input into the spatial smoothing mapping process to reconstruct the potential energy field of the space environment.

[0050] Specifically, the system sets a preset observation window time parameter. It extracts the group identifiers for all spatial movement trajectory data generated within the preset observation window time range, where the trajectory coordinates intersect with the spatial coordinate range of each grid cell. Based on the stress-related abnormal physiological motif data extracted and stored in step S3, it statistically analyzes specific grid cells (set as the [number]th [cell]). The cumulative frequency of occurrence of the stress-induced abnormal physiological motif triggered by this population within a grid cell is denoted as . .

[0051] Subsequently, the steps of updating the health risk weights using a logarithmic correction function and reconstructing the potential energy field of the space environment are performed, including: In the first step, it is determined whether the cumulative frequency of attacks within a specific grid cell is greater than the preset baseline frequency threshold.

[0052] The system reads the pre-configured baseline frequency threshold parameter, denoted as... The statistical results obtained are the first... Cumulative frequency of occurrence within each grid cell With the baseline frequency threshold Perform numerical comparisons and execute conditional judgment logic. .

[0053] In the second stage of the process, when the frequency exceeds the baseline threshold, the difference between the cumulative frequency of attacks and the baseline threshold is calculated.

[0054] When the above judgment logic When the result is true, the weight value update calculation process for the target grid cell is triggered. The system performs a subtraction operation to calculate the cumulative attack frequency. Baseline frequency threshold The difference is output as the excess difference. The specific calculation formula is as follows: ; In the formula, Indicates the first The deviation of the group outbreak frequency in each grid cell; Indicates the first The cumulative frequency of triggering abnormal physiological motifs of stress within each grid cell; This represents the preset baseline frequency threshold constant.

[0055] In the third stage, the logarithm of the excess difference is calculated by adding one, and then multiplied by the preset feedback damping coefficient to obtain the correction increment.

[0056] The above calculated output of the deviation value Substitute the values ​​into the system's preset logarithmic correction function module. For the out-of-scalar difference... Perform an increment operation, then calculate the natural logarithm of the sum. Multiply the logarithm result by the system's preset feedback damping coefficient, and output the health risk weight correction increment for that specific grid cell. The specific calculation formula is as follows: ; In the formula, Indicates the first Incremental adjustment of health risk weights for each grid cell; This represents the preset feedback damping coefficient constant, which controls the upper limit of the numerical change range in a single weight update calculation; Represented by constants The natural logarithm operation with base 0; This represents the excess value of the cumulative seizure frequency. The calculation includes an increment operation to maintain the mathematical validity of the logarithmic function and the smoothness of the output when the input value is zero or in the low numerical range.

[0057] In the fourth stage, the historical health risk weights of the corresponding grid cells are added to the correction increments to obtain the updated health risk weights, which are then re-input into the spatial smoothing mapping process to reconstruct the spatial environment potential energy field.

[0058] Extract the historical health risk weight data of the i-th grid cell before the current iteration cycle, denoted as . The historical health risk weights The correction increment calculated from the above steps Perform scalar addition and output the updated health risk weights. The specific calculation formula is as follows: ; In the formula, This indicates the updated calculated health risk weight value; This represents the historical health risk weight value extracted from the corresponding grid cell; This represents the correction increment value obtained after calculation using the logarithmic correction function.

[0059] After calculating the weights of all grid cells within the target community that triggered the update conditions, the system uses the grid cell data matrix containing the updated health risk weights as input parameters and re-substitutes it into the space environment potential energy field calculation module established in step S1. Using the updated health risk weights, the system calculates the updated static discrete potential energy value of each grid cell and calls the Gaussian kernel function to perform a spatial smoothing mapping operation on the static discrete potential energy value data. After the operation is completed, the system outputs the reconstructed space environment potential energy field distribution matrix and writes this matrix into memory, completing the closed-loop numerical update of the space environment risk assessment data.

[0060] Step S5: Calculate the cross-correlation between the risk potential exposure integral sequence and the frequency sequence of the abnormal physiological motifs of stress, extract the pathogenic causal association index, and output a comprehensive chronic disease risk rating and intervention instructions in combination with electronic health record data. In step S5, the steps of calculating the cross-correlation between the risk potential exposure integral sequence and the frequency sequence of abnormal physiological motifs under stress, and extracting the pathogenic causal association index, include: Frequency data of abnormal physiological motifs of stress located on the same time axis are extracted and converted into an attack frequency sequence; Set a time lag constant to characterize physiological abnormalities induced by environmental exposure; The sum of the products of the risk potential exposure integral sequence and the occurrence frequency sequence with a time lag constant is calculated to obtain the cross-correlation sequence under different lag time conditions. The maximum value in the cross-correlation sequence is extracted as the pathogenic causal association index; Step S5, which combines electronic health record data to output a comprehensive risk assessment and intervention instructions for chronic diseases, includes the following steps: Static baseline medical history data is extracted from electronic health records to calculate the baseline static risk score. The basic static risk score and the pathogenic causal association index were assigned normalized weights and then weighted and summed to obtain the comprehensive risk score. The comprehensive risk score is mapped to a preset risk level range, outputting a comprehensive chronic disease risk rating and triggering the generation of intervention instructions; The steps to trigger the generation of intervention instructions include: Extract the time lag constant that causes the cross-correlation sequence to reach its maximum value; Based on the time lag constant backtracking, specific grid cell trajectory segments that trigger high risk and the matched stress abnormal physiological motif features are extracted. Encapsulate specific grid cell trajectory segments along with abnormal physiological motifs of stress into a chain of causal evidence for pathogenesis. The comprehensive risk assessment of chronic diseases and the evidence chain of causal causes are packaged and sent to the management terminal to generate intervention instructions that include information on avoiding specific grid units.

[0061] Specifically, the system obtains the risk potential energy exposure integral sequence generated in step S2. And the stress-related abnormal physiological motifs extracted in step S3.

[0062] Subsequently, the steps of calculating the cross-correlation between the risk potential exposure integral sequence and the frequency sequence of stress-related abnormal physiological motifs, and extracting the pathogenic causal association index, are performed, specifically including: In the first step, frequency data of abnormal physiological motifs of stress located on the same time axis are extracted and converted into an attack frequency sequence.

[0063] The system divides the monitoring time domain into integral sequences related to risk potential exposure. Consistent Each time slice is analyzed. The number of abnormal physiological motifs triggered within each time slice is counted to construct a one-dimensional attack frequency sequence, denoted as [missing information]. .in, Indicates the first The number of times stress-related abnormal physiological motifs were detected within a time slice.

[0064] In the second stage, a time lag constant is set to characterize the physiological abnormalities induced by environmental exposure.

[0065] Considering the delayed nature of the impact of environmental exposure on physiological state, the system pre-sets a set of discrete time-lag constants. .in, This represents the offset of the frequency sequence relative to the risk potential exposure integral sequence on the time axis, expressed in the number of time slices.

[0066] In the third step, the sum of the products of the risk potential exposure integral sequence and the occurrence frequency sequence with a time lag constant is calculated to obtain the cross-correlation sequence under different lag time conditions.

[0067] For sets For each time lag constant in the calculation, the system performs a cross-correlation operation. The calculation formula is as follows: ; In the formula, This indicates that the time lag is Cross-correlation value at time; The first element in the risk potential exposure integral sequence represents the... The values ​​of each time slice; Indicating the number of seizures in the sequence of seizure frequencies The values ​​of each time slice; This is the total length of the sequence. (This is achieved by analyzing the set...) The cross-correlation sequence vector is output by iterating through all offsets within the range. .

[0068] In the fourth stage, the maximum value in the cross-correlation sequence is extracted as the pathogenic causal association index.

[0069] The system performs a maximum value retrieval operation from the cross-correlation sequence vector. Extract the maximum element value from the list and denote it as the pathogenic causal association index. The specific formula is as follows: ; In the formula, This represents a numerical value that characterizes the strength of the causal association between environmental risk exposure and the onset of physiological abnormalities.

[0070] Subsequently, the steps of combining electronic health record data to output a comprehensive risk assessment and intervention instructions for chronic diseases are implemented, specifically including: In the first step, static baseline medical history data is extracted from the electronic health record data to calculate the basic static risk score.

[0071] The system accesses electronic health records stored in a medical database. It extracts static baseline medical history data, including age, past medical history, family history, and baseline body mass index. Based on a pre-defined scoring matrix, it quantifies and assigns values ​​to each indicator, performs a weighted summation, and outputs a baseline static risk score. .

[0072] In the second stage, normalized weights are assigned to the basic static risk score and the pathogenic causal association index, and a weighted sum is performed to obtain the comprehensive risk score.

[0073] The system sets two normalized weight parameters. and And satisfy Perform a weighted summation operation to calculate the overall risk score. The calculation formula is as follows: ; In the formula, This represents an individual's overall risk score for chronic diseases. This represents the normalized value of the pathogenic causal correlation index.

[0074] In the third stage, the comprehensive risk score is mapped to a preset risk grading range, outputting a comprehensive chronic disease risk rating and triggering the generation of intervention instructions.

[0075] The system has a built-in risk level mapping table, which will... The corresponding numerical ranges are mapped to low-risk, medium-risk, and high-risk levels. Simultaneously, an operation is executed to trigger the generation of intervention commands.

[0076] The steps to trigger the generation of intervention instructions include: First, extract the time lag constant that maximizes the cross-correlation sequence, denoted as the target lag. .

[0077] Secondly, based on the target lag Perform time regression. Locate the time points in the seizure frequency sequence where the values ​​are greater than zero. Tracing back to the time point when the corresponding environmental exposure occurred. Extract from trajectory database The spatial coordinates corresponding to a given moment are used to pinpoint the trajectory segment of a specific grid cell that triggers a high risk, and the physiological motif characteristic parameters of the stress-induced abnormality corresponding to that time period are simultaneously matched.

[0078] Then, the specific grid cell trajectory segments, environmental potential energy attribute data, and stress-related abnormal physiological motif features are encapsulated into a structured chain of pathogenic causal evidence.

[0079] Finally, the comprehensive risk assessment of chronic diseases is packaged and encapsulated with the aforementioned causal evidence chain. This is then sent to the management terminal via a communication interface, generating intervention instructions that include suggestions for avoiding specific grid cell locations, adjusting travel routes, and early warning information about physiological abnormalities.

[0080] Example 2: When a hypertensive patient moves from a quiet park into a noisy, crowded commercial area with numerous high-fat restaurants, the environmental risk they face changes drastically. Existing systems, due to their fixed physiological sampling windows, often miss the brief, stress-induced abnormal fluctuations in blood pressure that occur at the moment of this environmental change. Furthermore, existing systems cannot collect abnormal physiological responses from the population in that area to assess and update the actual health threats in that physical space, resulting in significant delays and limitations in risk warnings. To address these issues, this invention provides a community-based grid-based chronic disease risk management system, the structure of which is as follows: Figure 2 As shown. The specific implementation process of this system is as follows: The potential energy field construction module is used to divide the target community into multiple grid cells, extract the interest point category features within the grid cells and assign health risk weights, and calculate and construct the initial spatial environment potential energy field. The data calculation module is used to acquire the user's spatial movement trajectory data, and perform integral and difference calculations on the initial spatial environment potential energy field based on the spatial movement trajectory data to obtain the risk potential energy exposure integral sequence and instantaneous potential energy gradient. The motif extraction module is used to collect continuous physiological monitoring data. When the instantaneous potential energy gradient is greater than the preset stress trigger threshold, it triggers the adjustment of feature extraction parameters and uses the matrix contour algorithm to extract stress-abnormal physiological motifs from the continuous physiological monitoring data. The potential energy field reconstruction module is used to count the cumulative frequency of the occurrence of stress-induced abnormal physiological motifs triggered by the population passing through each grid unit within a preset observation window, and to reconstruct the potential energy field of the space environment by updating the health risk weights using a logarithmic correction function. The risk assessment module is used to calculate the cross-correlation between the risk potential exposure integral sequence and the frequency sequence of the abnormal physiological motifs of stress, extract the pathogenic causal association index, and output a comprehensive risk rating and intervention instructions for chronic diseases by combining electronic health record data.

[0081] Specifically, the potential energy field construction module is configured to divide the target community into multiple grid cells, extract the category features of points of interest within each grid cell and assign health risk weights, and calculate and construct the initial spatial environment potential energy field. This module receives geographic information system boundary data and map point of interest data through a communication interface. Extracting the first... Within the first grid cell Number of points of interest The corresponding health risk weights preset in the database The built-in processor calculates the first... Static discrete potential energy value of each grid cell : ; In the formula, This represents the total number of interest point categories within the target community. Subsequently, this module uses a spatial bandwidth parameter of... The Gaussian kernel function performs a spatial smoothing mapping operation on the data matrix containing static discrete potential energy values, and calculates the coordinates of the evaluation points in the two-dimensional plane according to the following formula. Initial spatial potential energy value : ; In the formula, This represents the total number of grid cells in the division. Indicates the first The center coordinates of each grid cell are identified. The generated initial spatial environment potential energy field distribution matrix is ​​output and stored in the system memory, serving as the input source of the basic physical environment data for subsequent calculations.

[0082] The data calculation module, communicatively connected to the potential energy field construction module, is configured to acquire the user's spatial movement trajectory data. Based on this data, it performs integration and difference calculations within the initial spatial environment potential energy field to obtain the risk potential energy exposure integral sequence and instantaneous potential energy gradient. This module receives a set of discrete trajectory point coordinates uploaded by the mobile terminal positioning module and generates a continuous-time coordinate function using a spline interpolation algorithm. Based on the read initial spatial environment potential energy field distribution matrix, this module calculates the first... along the spatial trajectory using a numerical integration logic unit. Time slice interval Risk potential energy exposure integral within : ; The risk potential exposure integral sequence is output by combining the integral results of each slice in chronological order. Simultaneously, a first-order differential logic unit is used to combine the time step constant between adjacent discrete sampling points. Calculate the current sampling time point instantaneous potential gradient : ; The serialized result is then output to the motif extraction module.

[0083] The motif extraction module, communicatively connected to the data calculation module and the peripheral physiological data acquisition device, is configured to acquire continuous physiological monitoring data. When the instantaneous potential energy gradient exceeds a preset stress trigger threshold, it triggers the adjustment of feature extraction parameters and uses a matrix contour algorithm to extract stress-induced abnormal physiological motifs from the continuous physiological monitoring data. This module includes a threshold determination logic unit, a parameter adjustment logic unit, and a matrix contour calculation unit. The threshold determination logic unit determines the absolute value of the instantaneous potential energy gradient in real time. Does it exceed the preset stress trigger threshold constant? When the determination result is true, the parameter adjustment logic unit calculates the absolute difference of the potential energy gradient exceeding the threshold. : ; Based on the difference, the baseline sliding window length is shortened synchronously and the baseline distance matching threshold is reduced, outputting the adjusted sliding window length. And the adjusted distance matching threshold : ; ; In the formula, and These represent the default baseline sliding window length and the distance matching threshold, respectively. and The preset adjustment coefficient; and This is the set minimum lower limit value. The matrix contour calculation unit uses the above-adjusted extraction parameters as algorithm input to calculate the distance matrix contour vector of continuous physiological monitoring data. When the first value in the distance matrix contour vector... The value of each element Satisfy the judgment logic At that time, the data segment is searched and marked, and the extracted stress-related abnormal physiological motifs are output to the potential energy field reconstruction module and the risk assessment module.

[0084] The potential energy field reconstruction module, communicatively connected to the sequence extraction module and the potential energy field construction module, is configured to statistically analyze the cumulative frequency of occurrences of the stress-induced abnormal physiological sequence triggered by the population passing through each grid cell within a preset observation window, update the health risk weights using a logarithmic correction function, and reconstruct the potential energy field of the space environment. This module receives anomalous sequence markers for each time period; when the... Cumulative frequency of occurrence within each grid cell Greater than the system's preset baseline frequency threshold constant At that time, calculate the frequency of exceeding the standard difference. : ; Utilizing the built-in logarithmic correction function and the feedback damping coefficient constant Calculate the health risk weight correction increment for the corresponding grid cell. : ; The historical health risk weight of the grid cell Perform a scalar addition operation with the correction increment to obtain the updated calculated health risk weights. : ; This module sends data instructions containing updated health risk weights to the control terminal of the potential energy field construction module, driving it to re-execute potential energy value calculation and spatial smoothing mapping, thereby completing the closed-loop data reconstruction of the underlying spatial environment risk status.

[0085] The risk assessment module, communicatively connected to the data calculation module, the motif extraction module, and the electronic health record server, is configured to calculate the cross-correlation between the risk potential exposure integral sequence and the frequency sequence of the stress-related abnormal physiological motifs, extract the pathogenic causal association index, and output a comprehensive chronic disease risk rating and intervention instructions based on the electronic health record data. This module iterates through a discrete set of time lag constants. Calculate the risk potential exposure integral sequence of the th slice values The first in the sequence of attack frequency slice values In the total length of the sequence The sum of the products within the product, the output time lag is Cross-correlation calculation value at time : ; Extract the maximum value from the cross-correlation sequences as the pathogenic causal association index. : ; In parallel, this module parses static baseline medical history data from electronic health record data to calculate a basic static risk score. The basic static risk score and the normalized pathogenic causality index were compared. Assign normalized weights as parameters Then, a weighted summation is performed to generate a comprehensive risk score for chronic diseases. : ; This module maps the comprehensive risk score to a preset risk grading range. Finally, based on the time lag constant that maximizes the cross-correlation, the module backtracks to extract specific grid cell trajectory segments and stress-related abnormal physiological motif features, encapsulates them into a pathogenic causal evidence chain, packages it together with the comprehensive chronic disease risk rating, and sends it to the management terminal via the communication module to generate an intervention instruction that includes information on avoiding specific grid cells and route adjustment suggestions.

[0086] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for chronic disease risk management based on community grid management, characterized in that, include: Step S1: Divide the target community into multiple grid cells, extract the interest point category features within the grid cells and assign them health risk weights, and calculate and construct the initial spatial environment potential field; Step S2: Obtain the user's spatial movement trajectory data, and perform integration and difference calculations on the initial spatial environment potential energy field based on the spatial movement trajectory data to obtain the risk potential energy exposure integral sequence and instantaneous potential energy gradient; Step S3: Collect continuous physiological monitoring data. When the instantaneous potential energy gradient is greater than the preset stress trigger threshold, trigger the adjustment of feature extraction parameters and use the matrix contour algorithm to extract stress-abnormal physiological motifs from the continuous physiological monitoring data. Step S4: Calculate the cumulative frequency of the occurrence of the stress-induced abnormal physiological motifs triggered by the population passing through each grid unit within the preset observation window, update the health risk weights using the logarithmic correction function, and reconstruct the potential energy field of the space environment. Step S5: Calculate the cross-correlation between the risk potential exposure integral sequence and the frequency sequence of the stress abnormal physiological motif, extract the pathogenic causal association index, and output a comprehensive chronic disease risk rating and intervention instructions in combination with electronic health record data.

2. The chronic disease risk management method based on community grid management according to claim 1, characterized in that, In step S1, the steps of extracting the interest point category features within the grid cell and assigning health risk weights, and calculating the initial spatial environment potential field include: The number of interest points of different categories within each grid cell is counted. The number of interest points of each category is multiplied by their corresponding health risk weights and then summed to obtain the static discrete potential energy value of the corresponding grid cell. A Gaussian kernel function is used to perform spatial smoothing mapping on the static discrete potential energy values ​​of each grid cell to generate an initial spatial environment potential energy field. The Gaussian kernel function includes a spatial bandwidth parameter used to control the degree of diffusion and attenuation of potential energy into the surrounding physical space.

3. The chronic disease risk management method based on community grid management according to claim 1, characterized in that, In step S2, the process of integrating and differencing the initial spatial environment potential energy field based on the spatial movement trajectory data to obtain the risk potential energy exposure integral sequence and instantaneous potential energy gradient includes: The spatial movement trajectory data is mapped into a continuous time coordinate and spatial coordinate function; The initial spatial environment potential energy field is calculated by line integral along the spatial movement trajectory data to obtain the risk potential energy exposure integral, and then converted into the risk potential energy exposure integral sequence by time slice; The instantaneous potential gradient is calculated based on the first-order difference of the initial spatial environment potential energy field values ​​at the user's location under adjacent discrete time steps.

4. The chronic disease risk management method based on community grid management according to claim 1, characterized in that, In step S3, when the instantaneous potential energy gradient is greater than the preset stress trigger threshold, the step of triggering the adjustment of feature extraction parameters includes: Baseline sliding window length and baseline distance matching threshold are used as default feature extraction parameters; When the absolute value of the instantaneous potential energy gradient exceeds the stress triggering threshold, the difference between the absolute value of the instantaneous potential energy gradient and the stress triggering threshold is calculated. Based on the difference and a preset adjustment coefficient, the baseline sliding window length is shortened synchronously and the baseline distance matching threshold is reduced to obtain the adjusted feature extraction parameters.

5. A method for chronic disease risk management based on community grid management according to claim 1, characterized in that, In step S3, the step of extracting stress-related abnormal physiological motifs from the continuous physiological monitoring data using the matrix contour algorithm includes: The adjusted feature extraction parameters are used as input parameters for the matrix contour algorithm; Calculate the distance matrix profile of the continuous physiological monitoring data and identify abnormal subsequences in the distance matrix profile that are less than the adjusted baseline distance matching threshold; The extracted abnormal subsequences are used as the stress-related abnormal physiological motifs.

6. The chronic disease risk management method based on community grid management according to claim 1, characterized in that, In step S4, updating the health risk weights using a logarithmic correction function and reconstructing the spatial environment potential field includes: Determine whether the cumulative frequency of attacks within a specific grid cell is greater than a preset baseline frequency threshold; When the frequency exceeds the baseline threshold, the difference between the cumulative attack frequency and the baseline frequency threshold is calculated. The logarithm of the excess difference is calculated by adding one, and then multiplied by a preset feedback damping coefficient to obtain the correction increment; The historical health risk weight of the corresponding grid cell is added to the correction increment to obtain the updated health risk weight, which is then re-input into the spatial smoothing mapping process to reconstruct the spatial environment potential field.

7. A method for chronic disease risk management based on community grid management according to claim 1, characterized in that, In step S5, the step of calculating the cross-correlation between the risk potential exposure integral sequence and the frequency sequence of the stress abnormal physiological motif, and extracting the pathogenic causal association index, includes: Frequency data of the stress-related abnormal physiological motifs located on the same time axis are extracted and converted into the attack frequency sequence; Set a time lag constant to characterize physiological abnormalities induced by environmental exposure; The sum of the products of the risk potential exposure integral sequence and the occurrence frequency sequence with the time lag constant is calculated to obtain the cross-correlation sequence under different lag time conditions. The maximum value in the cross-correlation sequence is extracted as the pathogenic causal association index.

8. A method for chronic disease risk management based on community grid management according to claim 1, characterized in that, In step S5, the steps of outputting a comprehensive risk rating and intervention instructions for chronic diseases based on electronic health record data include: Extract static baseline medical history data from the electronic health record data to calculate the basic static risk score; The basic static risk score and the pathogenic causal association index are assigned normalized weights and then weighted and summed to obtain a comprehensive risk score. The comprehensive risk score is mapped to a preset risk grading range, the comprehensive risk rating of chronic diseases is output, and the intervention instruction is triggered.

9. A method for chronic disease risk management based on community grid management according to claim 8, characterized in that, The step of triggering the generation of the intervention instruction includes: Extract the time lag constant that causes the cross-correlation sequence to reach its maximum value; Based on the time lag constant backtracking, specific grid cell trajectory segments that trigger high risk and the matched stress-abnormal physiological motif features are extracted. The specific grid cell trajectory segment, together with the stress-abnormal physiological motif features, is encapsulated into a pathogenic causal evidence chain; The comprehensive risk rating of chronic diseases and the causal evidence chain of disease are packaged and sent to the management terminal to generate intervention instructions that include information on avoiding specific grid units.

10. A chronic disease risk management system based on community grid management, characterized in that, A method for chronic disease risk management based on community grid management as described in any one of claims 1-9, the system comprising: The potential energy field construction module is used to divide the target community into multiple grid cells, extract the interest point category features within the grid cells and assign health risk weights, and calculate and construct the initial spatial environment potential energy field. The data calculation module is used to acquire the user's spatial movement trajectory data, and perform integral and difference calculations on the initial spatial environment potential energy field based on the spatial movement trajectory data to obtain the risk potential energy exposure integral sequence and instantaneous potential energy gradient. The motif extraction module is used to collect continuous physiological monitoring data. When the instantaneous potential energy gradient is greater than the preset stress trigger threshold, it triggers the adjustment of feature extraction parameters and uses a matrix contour algorithm to extract stress-abnormal physiological motifs from the continuous physiological monitoring data. The potential energy field reconstruction module is used to count the cumulative frequency of the occurrence of the stress-induced abnormal physiological motif triggered by the population passing through each grid unit within a preset observation window, update the health risk weight using a logarithmic correction function, and reconstruct the potential energy field of the space environment. The risk assessment module is used to calculate the cross-correlation between the risk potential exposure integral sequence and the frequency sequence of the stress abnormal physiological motif, extract the pathogenic causal association index, and output a comprehensive chronic disease risk rating and intervention instructions in combination with electronic health record data.

Images