A self-adaptive hierarchical management method and system based on crowd health portrait

Abstract

The application discloses a kind of self-adapting hierarchical management method and system based on crowd health portrait, it is related to medical health management technical field.The method includes: collecting target crowd living location, work characteristic parameter, obtains commuting parameter by commuting analysis;Combining living, work location and commuting parameter, personnel environmental parameter sequence is generated from environmental parameter distribution by fitting;First, second health parameters related to environment and work are obtained respectively by environmental health analyzer and health influence classifier and are fused;The health dispersion of fused health parameter, the characteristic dispersion of living and work characteristics are analyzed, the health portrait accuracy is verified in combination with crowd baseline dispersion, and adaptive grading is realized by dynamically configuring grading standard.The application solves the problem of single data and fixed grading in traditional health management by multi-source data fusion and dynamic grading mechanism, and improves the pertinence and effectiveness of crowd health management.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of medical health management, and in particular to a self-adaptive hierarchical management method and system based on population health portraits. BACKGROUND

[0002] In today's society, with the improvement of people's living standards and the increase of health concerns, effective population health management is increasingly important. Health management not only relates to the quality of life of individuals, but also has key significance for the optimization of regional public health resources, the improvement of disease prevention and treatment efficiency, and the improvement of overall health level.

[0003] However, the traditional health management mode has significant limitations, and is mostly carried out in a one-size-fits-all manner, which is difficult to accurately meet the health needs of different individuals or groups. When facing complex and diverse populations, the individual health characteristics are not fully explored, and it is not possible to provide highly personalized management solutions. SUMMARY

[0004] The present application provides a self-adaptive hierarchical management method and system based on population health portraits to address the technical problems of lack of multi-dimensional data fusion analysis in existing health management, insufficient health portrait accuracy, fixed grading standards and difficulty in adapting to population characteristic differences, resulting in poor management relevance and low efficiency.

[0005] The technical solution of the present application to solve the above technical problems is as follows:

[0006] In a first aspect, the present application provides a self-adaptive hierarchical management method based on population health portraits, comprising:

[0007] Collecting the residence locations and work characteristic parameters of multiple individuals in a target population in a target area, performing a commute analysis based on the residence locations and work characteristic parameters of each individual, and obtaining multiple commute parameters, wherein the work characteristic parameters include work location and work time;

[0008] Obtaining the distribution of environmental parameters in the target environment, fitting and processing the multiple residence locations, multiple work locations and multiple commute parameters to obtain multiple individual environmental parameter sequences;

[0009] Performing health impact analysis based on the multiple individual environmental parameter sequences to obtain multiple first health parameters, performing health impact analysis based on the multiple commute parameters and multiple work characteristic parameters to obtain multiple second health parameters, and processing to obtain multiple fused health parameters;

[0010] The health discreteness of the plurality of fusion health parameters is analyzed, the feature discreteness of the plurality of living positions and the plurality of work feature parameters is analyzed, the health discreteness is verified in combination with a population baseline discreteness, health portrait precision is obtained, a grading standard is configured, adaptive grading is performed on the plurality of personnel according to the plurality of fusion health parameters, and a grading result is obtained.

[0011] In a second aspect, the present application provides an adaptive grading management system based on a population health portrait, comprising:

[0012] The commuting parameter acquisition module is configured to collect the living positions and work feature parameters of the plurality of personnel in the target population in the target region, perform commuting analysis according to the living positions and work feature parameters of each personnel, and obtain a plurality of commuting parameters, wherein the work feature parameters include a work position and a work time.

[0013] The environment parameter sequence construction module is configured to obtain an environment parameter distribution in a target environment, fit and process the plurality of living positions, the plurality of work positions and the plurality of commuting parameters to obtain a plurality of personnel environment parameter sequences.

[0014] The fusion health parameter calculation module is configured to perform health impact analysis according to the plurality of personnel environment parameter sequences to obtain a plurality of first health parameters, perform health impact analysis according to the plurality of commuting parameters and the plurality of work feature parameters to obtain a plurality of second health parameters, and process the plurality of first health parameters and the plurality of second health parameters to obtain a plurality of fusion health parameters.

[0015] The adaptive grading execution module is configured to analyze the health discreteness of the plurality of fusion health parameters, analyze the feature discreteness of the plurality of living positions and the plurality of work feature parameters, verify the health discreteness in combination with a population baseline discreteness, obtain health portrait precision, configure a grading standard, perform adaptive grading on the plurality of personnel according to the plurality of fusion health parameters, and obtain a grading result.

[0016] The present application has the following beneficial effects:

[0017] Compared with the prior art, the present application firstly collects the residential location and work characteristic parameters through the data acquisition module and obtains the commuting parameters through the commuting analysis module, realizes the accurate integration of multi-dimensional basic data, and provides data for health assessment; secondly, the environment parameter sequence construction module generates the personnel environment parameter sequence from the environment parameter distribution by combining the residential location, the work location and the commuting parameters, and quantifies the spatial environment factors into the health analysis; thirdly, the fusion health parameter calculation module obtains the first and second health parameters through the environment health analyzer and the health impact classifier respectively and fuses and processes them, and comprehensively considers the influence of the environment and the work factors on the health; finally, the adaptive grading execution module analyzes the health dispersion and the characteristic dispersion, verifies the health portrait accuracy in combination with the population baseline dispersion, dynamically configures the grading standard, and realizes the closed loop of “collection, analysis, fusion and grading” to comprehensively consider the multi-source data fusion, the dynamic accuracy verification and the adaptive grading, supports the differentiated management with the accurate health portrait, effectively avoids the defects of single data and fixed grading in the traditional health management, and improves the pertinence and effectiveness of the population health management. BRIEF DESCRIPTION OF DRAWINGS

[0018] Figure 1 A flowchart of an adaptive grading management method based on a population health portrait provided by the present application is shown.

[0019] Figure 2 A structure diagram of an adaptive grading management system based on a population health portrait provided by the present application is shown.

[0020] In the drawings, the components represented by the respective reference numerals are as follows:

[0021] The commuting parameter acquisition module 11, the environment parameter sequence construction module 12, the fusion health parameter calculation module 13 and the adaptive grading execution module 14. DETAILED DESCRIPTION

[0022] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all the other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.

[0023] In the description of the present application, the terms “first” and “second” are only used for description purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined with “first” and “second” can explicitly or implicitly include one or more of the features. In the description of the present application, the meaning of “multiple” is two or more, unless otherwise specifically limited.

[0024] In the description of this invention, the term "for example" is used to mean "used as an example, illustration, or description." Any embodiment described as "for example" in this invention is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is provided to enable any person skilled in the art to make and use the invention. Details are set forth in the following description for purposes of explanation. It should be understood that those skilled in the art will recognize that the invention can be made without using these specific details. In other instances, well-known structures and processes will not be described in detail to avoid obscuring the description of the invention with unnecessary detail. Therefore, the invention is not intended to be limited to the embodiments shown, but is consistent with the broadest scope of the principles and features disclosed herein.

[0025] Example 1, as Figure 1 As shown, this embodiment of the invention provides an adaptive hierarchical management method based on population health profiles, including:

[0026] S10: Collect residential location and work characteristic parameters of multiple individuals within the target population in the target area, and perform commuting analysis based on each individual's residential location and work characteristic parameters to obtain multiple commuting parameters, including work characteristic parameters such as work location and work time;

[0027] The target area refers to a management unit with clear geographical boundaries, such as a county, urban area, or specific community. Health management needs to focus on a specific geographical area to ensure the targeted allocation of resources and intervention measures, and avoid data clutter or intervention failure due to an overly large area. The target population refers to a specific group within the target area that needs health management, such as residents within the target area.

[0028] Collect residential location and work characteristic parameters of multiple individuals within the target population area. Perform commuting analysis based on each individual's residential location and work characteristic parameters to obtain multiple commuting parameters, including:

[0029] Collect residential location and work characteristic parameters of multiple individuals within the target population in the target area. Among them, the work characteristic parameters include work location and work time.

[0030] Commuting analysis is performed based on each person's residential location and work characteristics to obtain multiple commuting parameters, each of which includes a commuting location sequence and commuting time.

[0031] Residential location refers to the specific place where an individual lives on a daily basis. Work characteristic parameters include work location and work time. Since work is an individual's main activity scenario besides residence, work characteristics and residential characteristics together constitute an individual's main spatiotemporal activity trajectory. Based on residential location, work location, and work time, commuting routes and commuting times can be obtained as multiple commuting parameters.

[0032] Commuting analysis is performed based on each individual's residential location and work characteristics to obtain multiple commuting parameters, including:

[0033] Based on the working time within each job feature parameter, multiple commuting times are extracted;

[0034] Multiple commuting location sequences are generated based on multiple residential locations, multiple commuting times, and multiple work locations. Combined with multiple commuting times, multiple commuting parameters are obtained.

[0035] Among them, commuting parameters are parameters that reflect an individual's round-trip journey from their residence to their workplace. They include "commuting location sequence" and "commuting time". The commuting location sequence refers to the set of key location points recorded in chronological order during an individual's commute. The commuting time refers to the total duration of an individual's single or daily commute.

[0036] For example, if we read the working hours (e.g., 09:00-18:00) from the work characteristic parameters, and assume that the regional commuting habit is to leave 45 minutes in advance and return 30 minutes in advance, then the early commute time = 09:00-08:15 = 45 minutes, the late commute time = 18:30-18:00 = 30 minutes, and the total commute time = 45 + 30 = 75 minutes;

[0037] Given the residential location (L1: N30.2°, E120.3°), work location (L2: N30.3°, E120.5°), and early commute time (45 min), the commute route is obtained through map route planning. Coordinates are extracted at 15-minute intervals: 08:15 (departure): L1 (N30.2°, E120.3°), 08:30 (15 min later): L3 (N30.23°, E120.35°), 08:45 (30 min later): L4 (N30.27°, E120.42°), 09:00 (arrival): L2 (N30.3°, E120.5°), forming the early commute location sequence: [L1, L3, L4, L2].

[0038] Commuting is an intermediate scenario connecting residence and work, and is used to analyze the environment of an individual's daily activity trajectory. Only by clearly knowing when and where an individual is can we accurately quantify their environmental exposure and work intensity, providing spatiotemporal and health-related support for the construction of health profiles.

[0039] S20: Obtain the distribution of environmental parameters within the target environment, and obtain multiple personnel environmental parameter sequences by fitting and processing based on multiple residential locations, multiple work locations, and multiple commuting parameters;

[0040] This involves acquiring the distribution of environmental parameters within the target environment, and, based on multiple residential locations, multiple work locations, and multiple commuting parameters, fitting and processing to obtain multiple sequences of environmental parameters for individuals, including:

[0041] The mean environmental parameters of all monitored coordinates within the target environment within the most recent preset time range are obtained, and the distribution of environmental parameters within the target environment is obtained by interpolation.

[0042] Based on multiple residential locations, multiple work locations, and multiple commuting parameters, the environmental parameters of each person at each time are obtained by indexing and extracting within the environmental parameter distribution, resulting in multiple environmental parameter sequences for each person.

[0043] Environmental parameter distribution refers to continuous environmental indicator data covering the entire target environment. It is generated by processing discrete monitoring point data and can reflect the environmental parameter values ​​at any location within the target environment, such as the air pollution index. The air pollution index comprehensively covers the concentrations of pollutants such as PM2.5, PM10, sulfur dioxide, and nitrogen dioxide that have a direct impact on human health. It can comprehensively reflect the potential risks of environmental exposure to human health. At the same time, the air pollution index is a standardized quantitative indicator in environmental monitoring, and the data is easy to obtain and facilitates subsequent calculations. The set of environmental parameters that an individual is exposed to at different times and locations (at home, at work, or during commuting) arranged in chronological order, which completely records the individual's environmental exposure trajectory, can yield the personnel environmental parameter sequence.

[0044] First, the coordinates and corresponding environmental parameter data of all environmental monitoring points within the target environment over the most recent preset time range, such as the last 7 days, are collected. The mean parameter value of each monitoring point within this time range is calculated, resulting in a discrete monitoring coordinate-mean parameter dataset. Then, an interpolation algorithm is used to process the discrete data, expanding it into a continuous environmental parameter distribution covering the entire target environment, thus fully presenting the spatial variation trend of environmental parameters.

[0045] In the process of obtaining the distribution of environmental parameters within the target environment, interpolation algorithms are used to process the average environmental parameters of all monitored coordinates within the most recent preset time range in the target environment to obtain a continuous distribution of environmental parameters covering the entire target environment, such as inverse distance weighted interpolation, Kriging interpolation, etc.

[0046] For example, the Kriging interpolation algorithm is selected: first, the mean environmental parameters of all monitored coordinates within the most recent preset time range in the target environment are calculated to obtain discrete monitoring data; then, the spatial variation characteristics between monitoring points are fitted by a semi-variogram function to determine the weight coefficients of different locations; finally, the discrete mean is weighted using the weight coefficients to generate a continuous environmental parameter distribution covering the target environment.

[0047] Secondly, based on the generated environmental parameter distribution, combined with multiple residential locations, work locations, and commuting parameters obtained in the early stages, the specific location of each person at different times is determined. Thirdly, the environmental parameter values ​​corresponding to each location at each time moment are extracted from the environmental parameter distribution, and the parameter values ​​are organized in chronological order to form an environmental parameter sequence for each person, thus comprehensively reflecting the individual's environmental exposure in residential, work, and commuting scenarios.

[0048] For example, given the PM2.5 monitoring data for the past 7 days in the target environment, the mean values ​​for three monitoring points were calculated: 45 μg / m³ for monitoring point A (N 30.18°, E 120.28°), 55 μg / m³ for monitoring point B (N 30.25°, E 120.38°), and 40 μg / m³ for monitoring point C (N 30.32°, E 120.52°). The discrete mean values ​​were then processed using the Kriging interpolation algorithm to generate a PM2.5 parameter distribution covering the target environment.

[0049] Based on the early morning commute location sequence [L1 (N 30.2°, E 120.3°), L3 (N 30.23°, E 120.35°), L4 (N 30.27°, E 120.42°), L2 (N 30.3°, E 120.5°)], the location parameters are indexed in the interpolated PM2.5 distribution: L1 corresponds to 48 μg / m³, L3 corresponds to 52 μg / m³, L4 corresponds to 49 μg / m³, and L2 corresponds to 42 μg / m³. Arranged according to the early morning commute time, the early morning commute environmental parameter sequence for this person is formed as: [48, 52, 49, 42].

[0050] S30: Perform health impact analysis based on multiple personnel environmental parameter sequences to obtain multiple first health parameters; perform health impact analysis based on the multiple commuting parameters and multiple work characteristic parameters to obtain multiple second health parameters; and process to obtain multiple fused health parameters.

[0051] First, a health impact analysis was conducted based on multiple personnel environmental parameter sequences to obtain several primary health parameters, including:

[0052] The environmental health analyzer is invoked, wherein the environmental health analyzer is trained based on the set of environmental parameter sequences of sample personnel and the labeled first health parameter set of the sample;

[0053] Each person's environmental parameter sequence is input into the environmental health analyzer, which outputs multiple first health parameters.

[0054] The Environmental Health Analyzer is a modeling tool used to quantify the health impacts of environmental exposures. Its core function is to output corresponding health impact assessment results, i.e., the primary health parameters, based on the input sequence of environmental parameters of the individuals. The set of environmental parameter sequences of the sample individuals is essentially a collection of historical PM2.5 exposure data, used to record the PM2.5 exposure of individuals at different times and in different scenarios, obtained from the air quality automatic monitoring stations of the official ecological and environmental departments. The labeled set of primary health parameters of the sample is the health data of the same period and the same population, such as the incidence rate of respiratory diseases, which is quantified into a score through statistical analysis.

[0055] For example, a health score of 1. If the incidence of respiratory diseases in a certain population during a historical period was 5%, the health score would be 1 − 5% = 95%.

[0056] Based on the set of environmental parameter sequences of sample personnel and the set of labeled first health parameters of the sample, a model is constructed and trained, which can be a neural network, regression model, etc.

[0057] For example, due to the complex nonlinear correlation between personnel environmental parameter sequences and health parameters, and the significant interaction between environmental parameters, the data relationships are difficult to accurately characterize using traditional linear models. Therefore, a neural network model is chosen as the core algorithm of the environmental health analyzer. A neural network is a computational model that mimics the structure and function of a biological nervous system. It consists of a large number of artificial neurons connected together to form a network, capable of learning patterns from data and making predictions or classifications. Its basic structure includes an input layer, hidden layers, and an output layer. The input layer receives raw data; the hidden layer, located between the input and output layers, processes information through the connection weights between neurons; and the output layer outputs the final result. The core principle of a neural network is to input data into the input layer, perform weighted calculations by the hidden layer neurons, and finally obtain the prediction result from the output layer. Simultaneously, by comparing the error between the predicted result and the actual result, the connection weights of neurons in each layer are adjusted in reverse from the output layer to gradually reduce the error.

[0058] During the training process, the collected sets of environmental parameter sequences for sample personnel and the corresponding sets of first health parameters for labeled samples are first preprocessed to remove outliers and eliminate interference caused by differences in magnitude. The preprocessed data is then divided in a 7:3 ratio: 70% is used as the training set for network parameter learning and weight updates; 30% is used as the validation set to assess generalization ability during health training and avoid overfitting the training data.

[0059] During training, the input layer receives a normalized set of environmental parameter sequences for sample personnel. After weighted calculation by neurons in the hidden layer, the output layer yields the predicted set of first health parameters. By comparing the error between the predicted values ​​and the first health parameter sets of the labeled samples, the backpropagation algorithm is used to calculate the weight gradient layer by layer. Then, the optimizer dynamically adjusts the connection weights of each layer, iterating repeatedly until the validation set error stabilizes and reaches 95% accuracy, ultimately forming a model that can accurately predict the first health parameter set.

[0060] Finally, the environmental parameter sequence for each person is input into the trained environmental health analyzer, which outputs multiple primary health parameters.

[0061] Furthermore, based on the multiple commuting parameters and multiple job characteristic parameters, a health impact analysis is performed to obtain multiple second health parameters, which are then processed to obtain multiple fused health parameters, including:

[0062] Based on the multiple commuting parameters and multiple work characteristic parameters, multiple total working times are calculated;

[0063] Multiple total working times are input into a health impact classifier to obtain multiple working impact health parameters, and multiple second health parameters are calculated. The health impact classifier is constructed based on the mapping relationship between the sample total working time set and the sample working impact health parameter set.

[0064] Multiple fusion health parameters are calculated based on multiple first health parameters and multiple second health parameters.

[0065] Among them, the total working time set of the sample refers to the different total working time data in the historical records, and the health parameter set of the sample work is the health risk quantification value corresponding to the total working time. The annotation logic is the same as that of the first health parameter of the sample above. Multiple total working times are input into the health impact classifier to obtain multiple health parameters of work impact.

[0066] First, based on the multiple commuting parameters and multiple work characteristic parameters, multiple total working times are calculated. Commuting time is extracted from the commuting parameters, and working time is extracted from the work parameters. Total working time = commuting time + working time. For example, if the commuting time is 75 minutes and the working time is 7 hours, the total working time = 8.25 hours.

[0067] For example, total working time and the magnitude of the work-related health parameter are positively correlated. The work-related health parameter = total working time / maximum total working time of all personnel × reduction factor of 10%. For example, if the maximum total working time of all personnel in the area is 10 hours and personnel A's working time is 8.25 hours, then the work-related health parameter = 8.25 / 10 × 10% = 8.25%. The second health parameter is 1 - work-related health parameter, which is 91.75%.

[0068] Secondly, a health impact classifier can be constructed and trained based on the total working time set of the samples and the set of health parameters affected by the working time of the samples. This can also be done using a neural network algorithm, following the same approach as the first health parameter mentioned above.

[0069] Based on the multiple commuting parameters and multiple work characteristic parameters, multiple total working times are calculated. The calculated total working times are then input into a trained health impact classifier, which outputs a work impact health parameter, which is the second health parameter.

[0070] Based on multiple primary health parameters and multiple secondary health parameters, multiple integrated health parameters are calculated. Since the primary health parameters are mainly affected by the environment and the secondary health parameters are mainly affected by the total working time, a weighted sum is needed to reflect the contribution weights of the two. Integrated health parameter = primary health parameter × weight 1 + secondary health parameter × weight 2. The weights are set based on the degree of health impact. For example, if the environment accounts for 40% and the working time accounts for 60%, the integrated health parameter = 0.4 × 95% + 0.6 × 91.75% = 93.05%, or 93.05 points.

[0071] S40: Analyze the health dispersion of the multiple fused health parameters, analyze the feature dispersion of multiple residential locations and multiple work characteristic parameters, verify the health dispersion by combining the population baseline dispersion, obtain the health profile accuracy, configure the grading standard, and adaptively grade multiple individuals according to the multiple fused health parameters to obtain the grading result.

[0072] Specifically, the health dispersion of the multiple fused health parameters is analyzed, as are the feature dispersions of multiple residential locations and multiple work characteristic parameters. The health dispersion is then validated using the population baseline dispersion to obtain the accuracy of the health profile. A grading standard is configured, and adaptive grading is performed on multiple individuals based on the multiple fused health parameters to obtain grading results, including:

[0073] Calculate and obtain the health dispersion of the multiple fused health parameters;

[0074] Calculate the dispersion of multiple residential locations and multiple work characteristic parameters and calculate the mean to obtain the feature dispersion;

[0075] Obtain the baseline dispersion of the population;

[0076] The ratio of the health dispersion to the sum of the feature dispersion and the population baseline dispersion is calculated as the accuracy of the health profile.

[0077] Based on the accuracy of the health profile, a grading standard is configured, and multiple individuals are adaptively graded using multiple fused health parameters to obtain the grading results.

[0078] Among them, health dispersion refers to the degree of difference between multiple integrated health parameters, reflecting the level of differentiation in the health status of the population. The greater the difference, the higher the dispersion, which is calculated through variance. Feature dispersion refers to the degree of comprehensive difference between residential location and work feature parameters, reflecting the inherent differences of the population in their living and working scenarios. Population baseline dispersion refers to the baseline value of inherent differences of the target population in dimensions such as health, residence, and work, such as the average health difference level of the regional population.

[0079] First, calculate the variance of the multiple fused health parameters, which is the health dispersion = (standard deviation of fused health parameters / mean of fused health parameters) × 100%;

[0080] For example, the integrated health parameters of 5 individuals are 65, 72, 58, 80, and 60: Mean = (65+72+58+80+60) / 5 = 67; Standard deviation = √[((65-67)²+(72-67)²+...+(60-67)²) / 5] = 8.7; Health dispersion = (8.7 / 67) × 100% = 13%.

[0081] Secondly, calculate the residential location dispersion = the sum of the straight-line distances between the residential locations of all individuals / the number of pairs of individuals; calculate the job characteristic dispersion = (standard deviation of total working time / mean of total working time) × 100%; calculate the characteristic dispersion as (standardized value of residential location dispersion + job characteristic dispersion) / 2.

[0082] For example, three people reside at locations L1 (N 30.2°, E 120.3°), L2 (N 30.3°, E 120.5°), and L3 (N 30.1°, E 120.4°): the pairwise distances are L1-L2≈22 km, L1-L3≈15 km, and L2-L3≈25 km; the location dispersion = (22+15+25) / 3≈20.7 km. Total working hours: 10.25 hours, 8.5 hours, and 11 hours: mean = 9.92 hours; standard deviation ≈ 1.05 hours. Work characteristic dispersion = 1.05 / 9.92 = 10.6%;

[0083] Standardizing the residential location dispersion to a percentage ensures consistency of dimensions. Mapping it to 15%, which represents the relative level of residential differences within the region, the characteristic dispersion = (15% + 10.6%) / 2 ≈ 12.8%.

[0084] Furthermore, the baseline dispersion of the population is the inherent difference benchmark of the population in the target area, which can be obtained through historical data statistics. The accuracy of the health profile is measured by the ratio of "health dispersion" to "feature dispersion + population baseline dispersion". Assuming that the population baseline dispersion of the area is 8%, the health dispersion is 13%, and the feature dispersion is 12.8%, then the accuracy of the health profile is 13% / (12.8%+8%)×100%≈62.5%, which means that about 62.5% of the health differences can be explained by residential / work characteristics and population baseline.

[0085] Finally, based on the accuracy of the health profile, a grading standard is configured, and multiple individuals are adaptively graded using multiple fused health parameters to obtain grading results, including:

[0086] Get the preset number of levels;

[0087] The number of configured levels is calculated based on the accuracy of the health profile and the preset number of levels, and is used as the leveling standard.

[0088] According to the grading criteria, the multiple fused health parameters are equidistantly graded to obtain multiple health parameter grades, which are used as the grading results.

[0089] The preset number of grades refers to the default optional grade scheme, the configured number of grades refers to the final number of grades selected from the preset number based on the accuracy of the health profile, and the equidistant grade divides the range of values ​​of the integrated health parameters into equal intervals according to the configured number of grades, with each interval corresponding to a health level.

[0090] The preset number of levels is usually set based on the needs of the health management scenario. For example, the preset number of levels is 2, 3, or 5.

[0091] Level 2: Suitable for coarse-grained scenarios that quickly distinguish between "low risk" and "high risk";

[0092] Level 3: Applicable to "low-risk / medium-risk / high-risk" scenarios for routine health management;

[0093] Level 5: Applicable to "extremely low risk / low risk / medium risk / high risk / extremely high risk" scenarios for refined health intervention.

[0094] Furthermore, the product of "health profile accuracy × preset number of levels" is rounded down to obtain the number of levels to be configured. The higher the accuracy, the larger the product, the more levels to be configured, and the finer the levels; the lower the accuracy, the smaller the product, the fewer levels to be configured, and the coarser the levels, thus achieving an adaptive match between the grading standard and the data reliability.

[0095] For example, if the accuracy of the health profile is calculated to be 62.5%, and the preset number of levels is 5, the product = 0.625 × 5 ≈ 3.125, which is rounded to 3 levels, that is, the number of levels is configured to be 3.

[0096] According to the grading standard, the multiple fused health parameters are equidistantly graded to obtain multiple health parameter grades as grading results. This ensures that the grading granularity is dynamically adjusted with data reliability, avoiding excessive subdivision under low precision while achieving refined risk differentiation under high precision.

[0097] In summary, the embodiments of this application have at least the following technical effects:

[0098] Compared to existing technologies, this application first collects residential location and work characteristic parameters through a commuting parameter acquisition module. After analysis, commuting parameters containing commuting location sequences and times are obtained, linking individual spatial activity data with health management to achieve accurate collection of basic data. Second, an environmental parameter sequence construction module obtains the distribution of environmental parameters and combines residential, work location, and commuting parameters to generate a sequence of environmental parameters for individuals, quantifying the impact of environmental factors on health into analyzable data. Third, a health parameter calculation module calls an environmental health analyzer and a health impact classifier to obtain and fuse health parameters related to the environment and work, comprehensively assessing health status from multiple dimensions. Finally, an adaptive grading execution module analyzes the dispersion of health and characteristics, verifies the accuracy of health profiles by combining the dispersion of the population baseline, dynamically configures grading standards to achieve adaptive grading, and forms a closed loop through the entire data flow process.

[0099] Through the above technical solution, this application deeply integrates multi-source data fusion with a dynamic grading mechanism, taking into account factors such as data comprehensiveness, assessment accuracy, and grading adaptability. By using dynamically optimized health profiles and grading standards, it effectively avoids the shortcomings of traditional health management data being singular and grading fixed, and significantly improves the pertinence and effectiveness of population health management.

[0100] Example 2, as Figure 2 As shown, based on the same inventive concept as the adaptive hierarchical management method based on population health profiles provided in Embodiment 1, this embodiment of the invention also provides an adaptive hierarchical management system based on population health profiles, including:

[0101] The commuting parameter acquisition module 11 is used to collect the residential location and work characteristic parameters of multiple people in the target population within the target area, and to perform commuting analysis based on the residential location and work characteristic parameters of each person to obtain multiple commuting parameters, among which the work characteristic parameters include work location and work time.

[0102] The environmental parameter sequence construction module 12 is used to obtain the distribution of environmental parameters within the target environment. Based on multiple residential locations, multiple work locations, and multiple commuting parameters, it is fitted and processed to obtain multiple environmental parameter sequences for personnel.

[0103] The integrated health parameter calculation module 13 is used to perform health impact analysis based on multiple personnel environmental parameter sequences to obtain multiple first health parameters, perform health impact analysis based on the multiple commuting parameters and multiple work characteristic parameters to obtain multiple second health parameters, and process to obtain multiple integrated health parameters;

[0104] The adaptive grading execution module 14 is used to analyze the health dispersion of the multiple fused health parameters, analyze the feature dispersion of multiple residential locations and multiple work characteristic parameters, verify the health dispersion by combining the population baseline dispersion, obtain the accuracy of the health profile, configure the grading standard, and adaptively grade multiple people according to the multiple fused health parameters to obtain the grading result.

[0105] The commuting parameter acquisition module 11 is specifically used for:

[0106] Collect residential location and work characteristic parameters of multiple individuals within the target population area. Perform commuting analysis based on each individual's residential location and work characteristic parameters to obtain multiple commuting parameters, including:

[0107] Collect residential location and work characteristic parameters of multiple individuals within the target population in the target area. Among them, the work characteristic parameters include work location and work time.

[0108] At the same time, commuting analysis is performed based on each person's residential location and work characteristics to obtain multiple commuting parameters, each of which includes a commuting location sequence and commuting time.

[0109] Commuting analysis is performed based on each individual's residential location and work characteristics to obtain multiple commuting parameters, including:

[0110] Based on the working time within each job feature parameter, multiple commuting times are extracted;

[0111] Multiple commuting location sequences are generated based on multiple residential locations, multiple commuting times, and multiple work locations. Combined with multiple commuting times, multiple commuting parameters are obtained.

[0112] The environmental parameter sequence construction module 12 is specifically used for:

[0113] The distribution of environmental parameters within the target environment is obtained. Based on multiple residential locations, multiple work locations, and multiple commuting parameters, multiple environmental parameter sequences for individuals are obtained through fitting and processing, including:

[0114] The mean environmental parameters of all monitored coordinates within the target environment within the most recent preset time range are obtained, and the distribution of environmental parameters within the target environment is obtained by interpolation.

[0115] Based on multiple residential locations, multiple work locations, and multiple commuting parameters, the environmental parameters of each person at each time are obtained by indexing and extracting within the environmental parameter distribution, resulting in multiple environmental parameter sequences for each person.

[0116] The integrated health parameter calculation module 13 is specifically used for:

[0117] Health impact analysis was conducted based on multiple personnel and environmental parameter sequences to obtain several primary health parameters, including:

[0118] The environmental health analyzer is invoked, wherein the environmental health analyzer is trained based on the set of environmental parameter sequences of sample personnel and the labeled first health parameter set of the sample;

[0119] Each person's environmental parameter sequence is input into the environmental health analyzer, which outputs multiple first health parameters.

[0120] Furthermore, based on the multiple commuting parameters and multiple job characteristic parameters, a health impact analysis is performed to obtain multiple second health parameters, which are then processed to obtain multiple fused health parameters, including:

[0121] Based on the multiple commuting parameters and multiple work characteristic parameters, multiple total working times are calculated;

[0122] Multiple total working times are input into a health impact classifier to obtain multiple working impact health parameters, and multiple second health parameters are calculated. The health impact classifier is constructed based on the mapping relationship between the sample total working time set and the sample working impact health parameter set.

[0123] Multiple fusion health parameters are calculated based on multiple first health parameters and multiple second health parameters.

[0124] The adaptive hierarchical execution module 14 is specifically used for:

[0125] The health dispersion of the multiple fused health parameters is analyzed, as are the feature dispersions of multiple residential locations and multiple work characteristic parameters. The health dispersion is then validated against the population baseline dispersion to obtain the accuracy of the health profile. A grading standard is configured, and adaptive grading is performed on multiple individuals based on the multiple fused health parameters to obtain grading results, including:

[0126] Calculate and obtain the health dispersion of the multiple fused health parameters;

[0127] Calculate the dispersion of multiple residential locations and multiple work characteristic parameters and calculate the mean to obtain the feature dispersion;

[0128] Obtain the baseline dispersion of the population;

[0129] The ratio of the health dispersion to the sum of the feature dispersion and the population baseline dispersion is calculated as the accuracy of the health profile.

[0130] Based on the accuracy of the health profile, a grading standard is configured, and multiple individuals are adaptively graded using multiple fused health parameters to obtain the grading results.

[0131] Furthermore, based on the accuracy of the health profile, a grading standard is configured, and multiple individuals are adaptively graded using multiple fused health parameters to obtain grading results, including:

[0132] Get the preset number of levels;

[0133] The number of configured levels is calculated based on the accuracy of the health profile and the preset number of levels, and is used as the leveling standard.

[0134] According to the grading criteria, the multiple fused health parameters are equidistantly graded to obtain multiple health parameter grades, which are used as the grading results.

[0135] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0136] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0137] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and modifications fall within the scope of this application and its equivalents, this application intends to include such modifications and modifications.

Claims

1. A self-adaptive hierarchical management method based on population health portrait, characterized in that, The method comprises: Collecting the living positions and work characteristic parameters of a plurality of persons in a target population in a target area, performing commuting analysis on the basis of the living positions and work characteristic parameters of each person to obtain a plurality of commuting parameters, wherein the work characteristic parameters comprise a work position and a work time; Obtaining an environmental parameter distribution in a target environment, fitting and processing the plurality of living positions, work positions and commuting parameters to obtain a plurality of personnel environmental parameter sequences; Performing health influence analysis on the basis of the plurality of personnel environmental parameter sequences to obtain a plurality of first health parameters, performing health influence analysis on the basis of the plurality of commuting parameters and work characteristic parameters to obtain a plurality of second health parameters, and processing to obtain a plurality of fused health parameters, comprising: Calling an environmental health analyzer, wherein the environmental health analyzer is trained on the basis of a sample personnel environmental parameter sequence set and a labeled sample first health parameter set; Inputting each personnel environmental parameter sequence into the environmental health analyzer to output a plurality of first health parameters; Calculating a plurality of total work times on the basis of the plurality of commuting parameters and work characteristic parameters; Inputting the plurality of total work times into a health influence classifier respectively to obtain a plurality of work influence health parameters, and calculating a plurality of second health parameters, wherein the health influence classifier is constructed on the basis of a mapping relationship between a sample total work time set and a sample work influence health parameter set; Calculating a plurality of fused health parameters on the basis of the plurality of first health parameters and second health parameters; Analyzing a health dispersion degree of the plurality of fused health parameters, analyzing a feature dispersion degree of the plurality of living positions and work characteristic parameters, verifying the health dispersion degree in combination with a population baseline dispersion degree to obtain a health portrait accuracy, configuring a grading standard, and adaptively grading a plurality of persons on the basis of the plurality of fused health parameters to obtain a grading result, comprising: Calculating a health dispersion degree of the plurality of fused health parameters, wherein the health dispersion degree is a difference degree between the plurality of fused health parameters; Calculating a dispersion degree of the plurality of living positions and work characteristic parameters and calculating a mean value to obtain a feature dispersion degree, wherein the feature dispersion degree is a mean value of a living position dispersion degree standardized value and a work characteristic dispersion degree, and the living position dispersion degree and the work characteristic dispersion degree are difference degrees between the plurality of living positions and work characteristic parameters respectively; Obtaining a population baseline dispersion degree, wherein the population baseline dispersion degree is an average health difference level of a population in the target area; Calculating a ratio of the health dispersion degree to a sum of the feature dispersion degree and the population baseline dispersion degree as a health portrait accuracy; Configuring a grading standard on the basis of the health portrait accuracy, and adaptively grading a plurality of persons on the basis of the plurality of fused health parameters to obtain a grading result, comprising: Obtaining a preset grading number; Calculating a configured grading number on the basis of the health portrait accuracy and the preset grading number as a grading standard; According to the grading standard, equally grading the plurality of fused health parameters to obtain a plurality of health parameter grades as a grading result.

2. The adaptive stratification management method based on the crowd health portrait according to claim 1, characterized in that, The living positions and work characteristic parameters of a plurality of persons in a target population in a target area are collected, commuting analysis is performed according to the living positions and work characteristic parameters of each person, and a plurality of commuting parameters are obtained, including: The living positions and work characteristic parameters of a plurality of persons in a target population in a target area are collected, wherein the work characteristic parameters include work positions and work times; Commuting analysis is performed according to the living positions and work characteristic parameters of each person, and a plurality of commuting parameters are obtained, wherein each commuting parameter includes a commuting position sequence and a commuting time. 3.The adaptive hierarchical management based on crowd health portrait according to claim 2, characterized in that, The living positions and work characteristic parameters of a plurality of persons in a target population in a target area are collected, commuting analysis is performed according to the living positions and work characteristic parameters of each person, and a plurality of commuting parameters are obtained, including: According to the work times in each work characteristic parameter, a plurality of commuting times are extracted; According to the plurality of living positions, the plurality of commuting times and the plurality of work positions, a plurality of commuting position sequences are generated, and a plurality of commuting parameters are obtained in combination with the plurality of commuting times. 4.The crowd health portrait-based adaptive hierarchical management method according to claim 1, characterized in that, An environment parameter distribution in a target environment is obtained, and a plurality of personnel environment parameter sequences are obtained by fitting processing according to the plurality of living positions, the plurality of work positions and the plurality of commuting parameters, including: An environment parameter distribution in a target environment is obtained, and a plurality of personnel environment parameter sequences are obtained by fitting processing according to the plurality of living positions, the plurality of work positions and the plurality of commuting parameters, including: An environment parameter distribution in a target environment is obtained, and a plurality of personnel environment parameter sequences are obtained by fitting processing according to the plurality of living positions, the plurality of work positions and the plurality of commuting parameters, including:

5. A self-adapting hierarchical management system based on population health profiling, characterized in that, For performing the method of any one of claims 1-4, including: The commuting parameter acquisition module is configured to collect the living positions and work characteristic parameters of a plurality of persons in a target population in a target area, perform commuting analysis according to the living positions and work characteristic parameters of each person, and obtain a plurality of commuting parameters, wherein the work characteristic parameters include work positions and work times; The environment parameter sequence construction module is configured to obtain an environment parameter distribution in a target environment, and obtain a plurality of personnel environment parameter sequences by fitting processing according to the plurality of living positions, the plurality of work positions and the plurality of commuting parameters; The fusion health parameter calculation module is configured to perform health influence analysis according to the plurality of personnel environment parameter sequences, obtain a plurality of first health parameters, perform health influence analysis according to the plurality of commuting parameters and the plurality of work characteristic parameters, obtain a plurality of second health parameters, and process to obtain a plurality of fusion health parameters, including: The environment health analyzer is trained based on a sample personnel environment parameter sequence set and a labeled sample first health parameter set; Each personnel environment parameter sequence is input into the environment health analyzer, and a plurality of first health parameters are output; A plurality of total work times are calculated according to the plurality of commuting parameters and the plurality of work characteristic parameters; The plurality of total work times are input into the health influence classifier respectively, a plurality of work influence health parameters are obtained, and a plurality of second health parameters are calculated, wherein the health influence classifier is constructed based on a mapping relationship between a sample total work time set and a sample work influence health parameter set; The plurality of first health parameters and the plurality of second health parameters are used to calculate a plurality of fusion health parameters. An adaptive grading execution module is configured to analyze health dispersion of the plurality of fusion health parameters, analyze feature dispersion of a plurality of residence locations and a plurality of work feature parameters, verify the health dispersion in combination with population baseline dispersion, obtain health portrait accuracy, configure grading standards, and perform adaptive grading on a plurality of persons according to the plurality of fusion health parameters to obtain grading results, including: calculating health dispersion of the plurality of fusion health parameters, wherein the health dispersion is a difference degree between the plurality of fusion health parameters; calculating dispersion of the plurality of residence locations and the plurality of work feature parameters and calculating a mean value to obtain feature dispersion, wherein the feature dispersion is a mean value of residence location dispersion standardized value and work feature dispersion, and the residence location dispersion and the work feature dispersion are difference degrees between the plurality of residence locations and the plurality of work feature parameters, respectively; obtaining population baseline dispersion, wherein the population baseline dispersion is an average health difference level of a target area population; calculating a ratio of the health dispersion to a sum of the feature dispersion and the population baseline dispersion as health portrait accuracy; configuring grading standards according to the health portrait accuracy, performing adaptive grading on a plurality of persons according to the plurality of fusion health parameters to obtain grading results, including: obtaining a preset grading number; calculating a configured grading number as the grading standards according to the health portrait accuracy and the preset grading number; performing equidistant grading on the plurality of fusion health parameters according to the grading standards to obtain a plurality of health parameter gradings as the grading results.

Images