A soil heavy metal health risk prediction method for human risk differentiation

Abstract

The embodiment of the application discloses a soil heavy metal health risk prediction method for a crowd risk differentiation; the embodiment of the application can grid the region to be evaluated, obtain the region type and the soil heavy metal content of each subregion; in combination with the individual type and the space-time activity data of the individual in each subregion, the activity subregion set of each individual is determined, and multiple homogeneous groups are formed based on the region type, the individual type and the activity subregion set clustering; for each group, the heavy metal content, the differentiated exposure parameter and the residence time length distribution characteristics of the activity subregion are fused, the daily average intake of heavy metals in multiple ways is calculated, and then the health risk score is obtained; finally, the group risk score is mapped and accumulated to the spatial subregion according to the residence time length distribution, the crowd health risk load value is generated, and the subregion to be treated is identified accordingly. Thus, the embodiment of the application is fair and targeted in the treatment of soil heavy metal pollution.

Description

Technical Field

[0001] This application relates to the field of environmental health risk assessment technology, specifically to a method for predicting soil heavy metal health risks based on population risk differentiation. Background Technology

[0002] Heavy metal pollution in soil is characterized by its insidious, cumulative, and non-degradable nature. It can enter the human body through exposure routes such as ingestion, skin contact, and inhalation, accumulating in organs such as the liver, kidneys, and bones, causing serious health hazards such as damage to the nervous system, immune dysfunction, and even cancer and birth defects. Especially for certain sensitive populations, even low-dose long-term exposure can cause irreversible physiological effects. With the rapid advancement of industrialization and urbanization, the superposition of multiple pollution sources, including industrial and mining enterprises, major transportation routes, and agricultural inputs, has led to prominent problems of excessive heavy metal levels in soil in some areas, becoming a significant environmental risk source threatening public health and ecological security.

[0003] However, current methods for predicting the health risks of heavy metals in soil are generally based on the assumption of "population homogeneity," typically treating regional residents as a single entity and using uniform exposure parameters (such as fixed soil ingestion rates and annual average exposure frequencies) and pre-defined land use types for risk assessment. This makes it difficult to reflect the actual differences in exposure intensity among different individuals. Such methods tend to overlook the serious health risks faced by a few highly exposed individuals (such as those frequently exposed to polluted farmland), causing their individual risks to be masked by regional averages. This makes it impossible to identify and intervene in a timely manner in pollution sources or high-risk areas that pose a significant health threat to specific populations. Summary of the Invention

[0004] This application provides a method for predicting the health risks of heavy metals in soil based on population risk differentiation. It can accurately identify areas that pose significant health risks to specific populations and effectively avoid ignoring high-risk individuals due to regional average effects, thereby improving the fairness and targeting of soil heavy metal pollution control.

[0005] This application provides a method for predicting the health risk of heavy metals in soil based on population risk differentiation. The method includes:

[0006] The area to be evaluated is divided into grids to obtain multiple sub-regions, and the region type and soil heavy metal content of each sub-region are obtained.

[0007] Obtain the individual types corresponding to multiple individuals in the region to be evaluated, as well as the spatiotemporal activity data of each individual in each sub-region. The spatiotemporal activity data shall include at least the duration of stay of the individual in each sub-region.

[0008] Based on spatiotemporal activity data, determine the set of activity sub-regions associated with each individual;

[0009] Based on the region type of each sub-region, the individual type corresponding to the individual, and the set of active sub-regions, multiple individuals are clustered to obtain multiple groups. Individuals in the same group have the same individual type and the same set of active sub-regions.

[0010] For each group, based on the soil heavy metal content in each activity sub-region of the group's activity sub-region set, the differential exposure parameters corresponding to the group, and the distribution characteristics of the group's residence time in each activity sub-region, the average daily intake of heavy metals of the group under multiple preset exposure pathways is calculated, and the health risk score of the group is obtained based on the average daily intake of heavy metals.

[0011] The health risk scores of each group are spatially mapped and accumulated based on their corresponding length of stay distribution characteristics to obtain the population health risk load value of each sub-region. Based on the population health risk load value of each sub-region, the sub-regions to be treated are determined from multiple sub-regions.

[0012] This application also provides a soil heavy metal health risk prediction device based on population risk differentiation, the device comprising:

[0013] The regional division unit is used to divide the area to be evaluated into grids, resulting in multiple sub-regions, and to obtain the regional type and soil heavy metal content of each sub-region;

[0014] The data acquisition unit is used to acquire the individual types corresponding to multiple individuals in the area to be evaluated, as well as the spatiotemporal activity data of each individual in each sub-area. The spatiotemporal activity data includes at least the duration of the individual's stay in each sub-area.

[0015] Regional association unit, used to determine the set of activity sub-regions associated with each individual based on spatiotemporal activity data;

[0016] Clustering unit is used to cluster multiple individuals based on the region type of each sub-region, the individual type corresponding to the individual, and the set of active sub-regions to obtain multiple groups. Individuals in the same group have the same individual type and the same set of active sub-regions.

[0017] The intake calculation unit is used to calculate the average daily intake of heavy metals for each group under multiple preset exposure pathways, based on the soil heavy metal content of each activity sub-region in the group's activity sub-region set, the differential exposure parameters corresponding to the group, and the distribution characteristics of the group's residence time in each activity sub-region. The health risk score of the group is obtained based on the average daily intake of heavy metals.

[0018] The risk determination unit is used to spatially map and accumulate the health risk scores of each group according to their corresponding length of stay distribution characteristics, to obtain the population health risk load value of each sub-region, and to determine the sub-region to be treated from multiple sub-regions based on the population health risk load value of each sub-region.

[0019] This application also provides an electronic device, including a processor and a memory, wherein the memory stores multiple instructions; the processor loads instructions from the memory to execute the steps in any of the soil heavy metal health risk prediction methods for population risk differentiation provided in this application.

[0020] This application also provides a computer-readable storage medium storing multiple instructions adapted for loading by a processor to execute steps in any of the soil heavy metal health risk prediction methods for population risk differentiation provided in this application.

[0021] This application also provides a computer program product, including a computer program / instructions, which, when executed by a processor, implement the steps in any of the soil heavy metal health risk prediction methods for population risk differentiation provided in this application.

[0022] In this application, based on obtaining the regional type and soil heavy metal content of each sub-region in the area to be assessed, the individual types of multiple individuals and their spatiotemporal activity data in each sub-region are further obtained. Based on this spatiotemporal activity data, the set of activity sub-regions associated with each individual is determined, and the individuals are clustered by combining the regional type of the sub-region, the individual type, and the set of activity sub-regions to obtain multiple groups. Since individuals in the same group have the same individual type and consistent activity patterns, a structured characterization of population risk heterogeneity is achieved. For each group, the soil heavy metal content of its activity sub-regions, the matched differential exposure parameters, and the distribution characteristics of residence time are integrated to calculate its average daily intake of heavy metals under multiple exposure pathways. This method abandons the assumption of traditional uniform exposure parameters, making dose estimation closer to real-world behavioral scenarios and significantly improving the physiological rationality and accuracy of individualized risk assessment. Furthermore, based on the average daily intake, the health risk score of each group is calculated, realizing a paradigm shift from "regional average risk" to "population-specific risk," effectively revealing highly sensitive populations masked by overall low risk. Finally, the health risk scores of each group are allocated to the relevant activity sub-regions based on their residence duration distribution characteristics and accumulated to generate a population health risk load value for each sub-region, thereby identifying sub-regions to be treated. This mechanism ensures that treatment decisions not only reflect pollution intensity but also the actual health impact on vulnerable groups. Therefore, it can accurately identify areas posing significant health risks to specific populations, effectively preventing high-risk individuals from being overlooked due to regional average effects, and improving the fairness and targeting of soil heavy metal pollution treatment. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 This is a flowchart illustrating a method for predicting soil heavy metal health risks based on population risk differentiation, as provided in an embodiment of this application.

[0025] Figure 2 This is a schematic diagram of a soil heavy metal health risk prediction device for population risk differentiation provided in an embodiment of this application. Detailed Implementation

[0026] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0027] This application provides a method for predicting the health risks of heavy metals in soil based on population risk differentiation.

[0028] One such device, a soil heavy metal health risk prediction device based on population risk differentiation, can be integrated into an electronic device, such as a terminal or server. The terminal can be a mobile phone, tablet, smart Bluetooth device, laptop, or personal computer (PC); the server can be a single server or a server cluster consisting of multiple servers.

[0029] In some embodiments, the soil heavy metal health risk prediction device based on population risk differentiation can also be integrated into multiple electronic devices. For example, the soil heavy metal health risk prediction device based on population risk differentiation can be integrated into multiple servers, and the soil heavy metal health risk prediction method based on population risk differentiation of this application can be implemented by multiple servers.

[0030] In some embodiments, the server may also be implemented as a terminal.

[0031] The following sections provide detailed descriptions of each example. It should be noted that the sequence numbers of the following embodiments are not intended to limit the preferred order of the embodiments.

[0032] In this embodiment, a method for predicting the health risks of heavy metals in soil based on population risk differentiation is provided, such as... Figure 1 The specific process of this method for predicting soil heavy metal health risks based on population risk differentiation can be as follows:

[0033] 101. The area to be evaluated is divided into grids to obtain multiple sub-regions, and the region type and soil heavy metal content of each sub-region are obtained.

[0034] The area to be assessed refers to the geographical area where soil heavy metal health risk prediction needs to be carried out. This area usually has clear administrative boundaries or environmental management units, such as a county, township, industrial park, urban built-up area, or specific watershed.

[0035] A subregion refers to a basic computational unit obtained by dividing the area to be evaluated into regular or irregular spatial divisions. Subregions are used to carry multi-source data such as environmental parameters, heavy metal concentrations, and human activities, and are the basic granularity for risk modeling and spatial mapping.

[0036] In some embodiments, the area to be evaluated is divided into regular grids according to a preset grid size (e.g., 100m × 100m) to obtain multiple square sub-regions.

[0037] In some embodiments, the area to be assessed is irregularly divided according to preset functions (such as natural villages, communities, or land use patches) to obtain multiple sub-areas with consistent functions.

[0038] Region type refers to the dominant land use or functional attribute category corresponding to each sub-region, used to characterize the environmental exposure background characteristics of that sub-region. Region type can be determined based on remote sensing image interpretation results, through cluster analysis of the spatial distribution of points of interest (POIs) in electronic maps, etc.

[0039] In some embodiments, the area type includes, but is not limited to: farmland, industrial land, residential area, woodland, water body, transportation land, or urban-rural fringe. For example, if more than 90% of the area of ​​a sub-area is arable land, its area type is marked as "farmland".

[0040] Soil heavy metal content refers to the mass concentration of a target heavy metal element per unit mass of soil, usually expressed in milligrams per kilogram (mg / kg). Heavy metals include one or more of cadmium (Cd), lead (Pb), arsenic (As), chromium (Cr), mercury (Hg), copper (Cu), and zinc (Zn). The values ​​can be obtained through on-site sampling and laboratory chemical analysis, or estimated through predictive models.

[0041] It should be noted that the remote sensing images, land survey data, environmental monitoring data, etc., related to the regional type and soil heavy metal content of each sub-region in the area to be evaluated were all obtained and used with the authorization or permission of the relevant competent authorities or data holders in accordance with the law.

[0042] In some embodiments, the fusion of multiple prediction models trained based on combinations of different land environmental characteristics effectively improves the robustness and generalization ability of soil heavy metal content estimation under small sample conditions, making it particularly suitable for underdeveloped areas with high sampling costs and sparse monitoring points. Specifically, before obtaining the regional type and soil heavy metal content of each sub-region, the method further includes:

[0043] Obtain the land environmental characteristic parameters of each sub-region in the area to be evaluated, as well as the measured soil heavy metal content of some sub-regions;

[0044] Based on the land environmental characteristic parameters of some sub-regions and their corresponding measured soil heavy metal content, multiple heavy metal prediction models are constructed. Different heavy metal prediction models are constructed based on different combinations of land environmental characteristics.

[0045] Using multiple heavy metal prediction models, the predicted soil heavy metal content of the remaining sub-regions is predicted based on the land environmental characteristic parameters of each sub-region. The remaining sub-regions are the sub-regions in the area to be evaluated, excluding some sub-regions.

[0046] The predicted soil heavy metal contents of the remaining sub-regions are fused together to obtain the soil heavy metal contents of the remaining sub-regions.

[0047] Among them, land environmental characteristic parameters refer to quantitative indicators used to characterize the natural and human environmental attributes of a sub-region, covering multiple dimensions such as topography, climate, distance, land use, socio-economic factors, and vegetation.

[0048] In some embodiments, land environmental characteristic parameters include: altitude, slope, average annual precipitation, straight-line distance to the nearest factory, density of industrial points of interest, proportion of farmland, nighttime light index, normalized vegetation index, etc., totaling 14 dimensions.

[0049] A sub-region refers to a sub-region within the area to be evaluated where soil sampling points have been set up and the measured heavy metal content has been obtained, serving as a supervised sample for model training.

[0050] For example, 185 sampling points were evenly distributed across the 1,200 km² county to be evaluated, corresponding to 185 sub-regions, accounting for 0.15% of the total number of sub-regions (approximately 120,000).

[0051] The measured soil heavy metal content refers to the actual heavy metal concentration value obtained by collecting soil samples from a portion of the sub-region and conducting laboratory chemical analysis.

[0052] In some embodiments, for a certain sub-region, multiple soil sampling points are set up within it to collect surface soil samples and mix them thoroughly to form a mixed soil sample for that sub-region; then the mixed soil sample is tested for heavy metal content, and the result is used as the measured soil heavy metal content for that sub-region.

[0053] Heavy metal prediction models refer to prediction models that take land environmental characteristic parameters as input and soil heavy metal content as output, and are used to predict the heavy metal content in unsampled sub-regions. In some embodiments, heavy metal prediction models include, but are not limited to, random forest models, gradient boosting tree models, support vector regression models, or neural network models.

[0054] Multiple heavy metal prediction models refer to a set of heavy metal prediction models with the same structure but different feature subsets used for training. Each set of models focuses on a specific pollution driving mechanism (such as industrial sources or agricultural sources).

[0055] For example, three heavy metal prediction models were constructed: the first heavy metal prediction model uses the distance from the factory, the density of industrial points of interest, and the density of the transportation network as input features to predict the cadmium (Cd) content in industrial-dominated areas; the second heavy metal prediction model uses the proportion of farmland, the amount of fertilizer applied, and the density of irrigation canals as input features to predict the cadmium (Cd) content in agricultural-dominated areas; and the third heavy metal prediction model uses all 14-dimensional land environmental feature parameters as input to predict the global cadmium content of the entire area to be evaluated.

[0056] A land environmental feature combination refers to a subset selected from all land environmental feature parameters for training a specific heavy metal prediction model.

[0057] In some embodiments, the “land environment feature combination” is determined by expert knowledge or feature importance ranking. For example, for areas surrounding a mining area, combinations such as “distance from mining company, tailings dam density, and wind direction frequency” are preferred.

[0058] The remaining sub-regions refer to all unsampled sub-regions in the region to be evaluated, excluding some sub-regions. The soil heavy metal content of these sub-regions needs to be obtained through model prediction.

[0059] For example, if the area to be evaluated is divided into 120,000 sub-areas, of which 185 are partial sub-areas, then the remaining 119,815 are the remaining sub-areas.

[0060] Predicting soil heavy metal content refers to estimating the heavy metal concentration of any remaining sub-region using a single heavy metal prediction model.

[0061] For example, Model A predicts the Cd content in a certain farmland sub-region to be 0.65 mg / kg, while Model B predicts it to be 0.72 mg / kg.

[0062] Multiple predicted soil heavy metal contents refer to the set of multiple predicted values ​​output by multiple heavy metal prediction models for the same remaining sub-region.

[0063] In some embodiments, for the same sub-region, the three models output predicted Cd values ​​of 0.65 mg / kg, 0.72 mg / kg, and 0.68 mg / kg, respectively, constituting multiple predicted soil heavy metal contents for that sub-region.

[0064] In some embodiments, a weighted average fusion is performed on the multiple predicted soil heavy metal contents, with the weights dynamically allocated based on the R² scores of each model in cross-validation, ultimately yielding the fused soil heavy metal contents for the remaining sub-region. This strategy significantly reduces the prediction error caused by feature bias in a single model and improves spatial interpolation accuracy (MAE reduced by 18.7%).

[0065] In some embodiments, the problem of weak model generalization ability caused by insufficient training data in small sample areas is effectively solved, significantly improving the transfer adaptability and accuracy stability of heavy metal spatial prediction. Specifically, based on the land environmental characteristic parameters of some sub-regions and their corresponding measured soil heavy metal content, multiple heavy metal prediction models are constructed, including:

[0066] Acquire multiple pre-set reference heavy metal prediction models, as well as target feature information characterizing the regional characteristics of the area to be evaluated;

[0067] From the land environmental characteristic parameters, select the target land environmental characteristic parameters that match the target characteristic information;

[0068] From multiple reference heavy metal prediction models, candidate reference heavy metal prediction models whose input variables include target land environmental characteristic parameters are selected.

[0069] By using the target land environmental characteristic parameters of some sub-regions and their corresponding measured soil heavy metal content, the model parameters of the candidate reference heavy metal prediction model are calibrated to obtain the calibrated heavy metal prediction model.

[0070] The calibrated heavy metal prediction model, along with the remaining reference heavy metal prediction models (excluding the candidate reference heavy metal prediction model) from among the multiple reference heavy metal prediction models, are identified as multiple heavy metal prediction models.

[0071] Among them, the reference heavy metal prediction model refers to a heavy metal prediction model that has been trained and validated in other geographical areas or historical projects. Its input is land environmental characteristic parameters and its output is soil heavy metal content. It can be used as an initial model library for modeling new areas.

[0072] In some embodiments, the reference heavy metal prediction model library may include five trained random forest models, two of which are derived from the Yangtze River Delta industrial zone (based on distance from factories, PM2.5, and PM2.5). 2.5 Two models are based on sedimentation (with sedimentation as the main feature), two are from the Northeast Black Soil Agricultural Area (mainly based on organic matter and fertilizer application), and one is a national-scale general model (containing 14-dimensional full features).

[0073] In some embodiments, multiple models in the reference heavy metal prediction model library may all originate from the same geographical region, but be constructed based on different combinations of land environmental features or sampling periods to enhance model diversity.

[0074] Regional characteristics refer to the typical attributes of the area to be evaluated that distinguish it from other areas in terms of natural or human environment, and are used to guide model selection and feature screening.

[0075] In some embodiments, regional characteristics may include climate type (e.g., arid / humid), dominant industry (e.g., mining / agriculture), topography (e.g., plains / mountains) or pollution source structure (e.g., dominated by industrial emissions or dominated by agricultural non-point sources).

[0076] Target feature information refers to the quantitative or labeled description of regional characteristics, which is used to match with land environmental feature parameters.

[0077] For example, if the area to be evaluated is a county in a drought-stricken mining area in Northwest China, its target characteristic information can be represented as {"average annual precipitation < 300 mm", "dominant industry = mining", "topography = hills"}.

[0078] The target land environmental characteristic parameters refer to the subset of parameters that are highly correlated with the target characteristic information selected from all land environmental characteristic parameters, and are used for localized modeling.

[0079] For example, based on the target characteristic information of "dominant industry = mining", "distance from mining enterprises", "tailings dam density" and "wind direction frequency (dominant wind direction)" are selected as target land environmental characteristic parameters.

[0080] Candidate reference heavy metal prediction models refer to models whose original input variables include all or most of the target land environmental characteristic parameters among multiple reference heavy metal prediction models, and have local adaptation potential.

[0081] The calibrated heavy metal prediction model refers to the local optimized model obtained by fine-tuning the internal parameters (such as decision tree splitting threshold, weights, etc.) of the candidate reference model using the measured soil heavy metal content of some sub-regions.

[0082] For example, using 185 local sampling points to perform transfer learning calibration on the candidate reference heavy metal prediction model improved its R² on the test set from 0.52 to 0.76.

[0083] The remaining reference heavy metal prediction models refer to the reference models that were not selected as candidates. Their original parameters are retained as a supplement to the diversity in multi-model fusion.

[0084] For example, the calibrated model can be combined with the three uncalibrated reference models to form a final set of multiple heavy metal prediction models, so as to balance local adaptability and global robustness.

[0085] It should be noted that this method implements a three-stage modeling process of "prior knowledge transfer + local data calibration + multi-model integration". Compared with training a single model from scratch, this method reduces the mean absolute error (MAE) by 22.3% under sparse sampling conditions (<0.2% coverage), and performs better in highly heterogeneous regions.

[0086] 102. Obtain the individual types corresponding to multiple individuals in the region to be evaluated, as well as the spatiotemporal activity data of each individual in each sub-region. The spatiotemporal activity data shall include at least the duration of stay of the individual in each sub-region.

[0087] Among them, an individual refers to a natural person within the area to be assessed who has an independent or proxy spatiotemporal behavioral trajectory, and is the basic unit of health risk assessment.

[0088] In some embodiments, individuals include permanent residents, commuters, students, outdoor workers, etc.; for vulnerable individuals in the developmental stage or the elderly who lack independent mobility, their activity trajectories can be represented by guardians or fixed locations.

[0089] Individual type refers to a category label based on an individual's basic attributes (such as age, occupation, behavioral patterns, etc.), used to distinguish the exposure characteristics and sensitivities of different groups of people.

[0090] In some embodiments, individual types include: susceptible individuals during development, rural farmers, urban office workers, retired seniors, and industrial workers. For example, a 50-year-old fruit farmer is classified as a "rural farmer," whose typical activity area is their family home and contracted orchard.

[0091] Spatiotemporal activity data refers to a set of location information describing an individual's location in geographic space over time, used to characterize their actual exposure environment.

[0092] In some embodiments, spatiotemporal activity data is generated from satellite positioning trajectories, cellular base station positioning logs, or wireless local area network (Wi-Fi) access logs recorded by mobile terminals (such as smartphones, smart wearable devices, etc.) and mapped to a gridded sub-regional system of the area to be evaluated.

[0093] The duration of an individual's stay in each sub-region refers to the cumulative time an individual spends in a certain sub-region within a preset time period (such as one day or one week), which is used to quantify the exposure weight of that individual in that region.

[0094] In some embodiments, an individual working on a workday spends 8 hours in area A (residential area), 10 hours in sub-area B (office area), and 2 hours in sub-area C (commuting route). The remaining time is considered a brief pass-through and is not counted as a valid stay. The stay duration is extracted using trajectory point clustering and time series analysis algorithms.

[0095] It should be noted that the collection and processing of spatiotemporal activity data comply with the Personal Information Protection Law and ethical review requirements. The original trajectories are anonymized and aggregated, and only the statistical characteristics at the group level are retained for risk modeling to ensure personal privacy and security.

[0096] In some embodiments, this method effectively solves the challenge of exposure behavior modeling for vulnerable populations who do not travel independently and do not carry mobile devices, significantly improving the completeness and accuracy of health risk assessment coverage for vulnerable groups. Specifically, before obtaining the individual types corresponding to multiple individuals in the area to be assessed, and the spatiotemporal activity data of each individual in each sub-area, the method further includes:

[0097] Obtain basic attribute information of multiple individuals in the area to be evaluated;

[0098] Based on basic attribute information, multiple individuals are divided into independent individuals and non-independent individuals;

[0099] For an individual, the spatiotemporal activity data of the individual is obtained by mapping the original trajectory data generated by the mobile terminal held by the individual within a preset time period to the sub-regions of the area to be evaluated.

[0100] For non-independent individuals, based on the guardianship relationship identifier in the basic attribute information, the associated independent individuals bound to the non-independent individuals are identified, and the original trajectory data of the associated independent individuals within a preset time period, as well as the independent activity time periods and corresponding fixed activity locations of the non-independent individuals are obtained.

[0101] When a non-independent individual is in a non-independent activity period, the original trajectory data of the associated independent individual is used as the proxy trajectory data of the non-independent individual.

[0102] When a non-independent individual is in an independent activity period, fixed trajectory data of the non-independent individual is generated based on the geographical location of the fixed activity location and the preset stay duration.

[0103] By temporally fusing proxy trajectory data with fixed trajectory data, complete trajectory data of non-independent individuals is obtained. The complete trajectory data is then mapped to each sub-region of the area to be evaluated to obtain spatiotemporal activity data of non-independent individuals.

[0104] Basic attribute information refers to data used to describe an individual's basic sociodemographic characteristics, including but not limited to age, gender, occupation, household registration type, family member relationships, and guardianship status.

[0105] In some embodiments, the basic attribute information is derived from the community registration system or desensitized statistical data. For example, for individuals under care without independent signaling, the structured expression of their basic attribute information includes: physiological characteristic label = younger age group (group A), agent tracking node (such as guardian ID) = P1023, and high-frequency fixed activity location = XX centralized care area.

[0106] An independent individual refers to an individual with autonomous mobility and a mobile terminal that can record location information, whose activity trajectory can be directly obtained.

[0107] In some embodiments, the individual includes office workers, students (those carrying mobile phones), food delivery riders, etc.

[0108] Non-independent individuals refer to individuals who, due to age, physical condition, or equipment limitations, are unable to continuously generate autonomous trajectory data and need to have their activity patterns reconstructed through proxy methods.

[0109] In some embodiments, non-independent individuals mainly include individuals susceptible during development, elderly people living alone, and people with disabilities.

[0110] Mobile terminals refer to portable electronic devices with positioning and communication functions, used to collect individual location information.

[0111] In some embodiments, the mobile terminal includes smartphones, smart wearable devices (such as smartwatches, smart bracelets, smart rings, smart glasses, etc.).

[0112] The preset time period refers to the time window used to analyze individual behavioral patterns, usually a day, a week, or a month.

[0113] In some embodiments, the preset time period is a combination of seven consecutive weekdays and weekends to cover a typical activity cycle.

[0114] Raw trajectory data refers to the sequence of location points recorded by the mobile terminal in chronological order within a preset time period, including information such as latitude and longitude and timestamps.

[0115] In some embodiments, the raw trajectory data of an individual includes a location navigation point (such as BeiDou navigation coordinates, global positioning coordinates, etc.) every 5 minutes, for a total of 2,016 points per week.

[0116] Spatiotemporal activity data of an independent individual refers to the distribution of dwell time in each sub-region obtained by mapping the original trajectory data to gridded sub-regions.

[0117] In some embodiments, an office worker spends 8 hours / day in the residential sub-area, 9.5 hours / day in the office sub-area, and the rest is for commuting or short-term passing through.

[0118] Guardianship identifier refers to a unique identifier or relationship field used to associate a non-independent individual with their legal or de facto guardian.

[0119] In some embodiments, the guardianship relationship identifier is the guardian's identity identifier, which is stored in the family member relationship table.

[0120] An associated independent individual refers to an independent individual who has a guardianship relationship with a non-independent individual and whose trajectory can be used to act as an agent for the non-independent individual's activities.

[0121] For example, the independent related individual of a developmentally susceptible individual is his father (P1023), who picks up and drops off the developmentally susceptible individual from school every day.

[0122] Independent activity periods refer to the time periods during which a non-independent individual is separated from their guardian and engages in independent activities in a fixed location.

[0123] For example, a susceptible individual in the developmental stage is in an independent activity period from 8:00 to 16:30 on weekdays, and the location is school.

[0124] Fixed activity locations refer to definite locations where non-independent individuals stay for extended periods during independent activity periods.

[0125] For example, fixed activity venues can include schools, nursing homes, rehabilitation centers, and family residences.

[0126] Non-independent activity periods refer to the time periods during which non-independent individuals and associated independent individuals engage in activities together, and their spatial location is determined by the latter.

[0127] For example, susceptible individuals during developmental stages are not allowed to be independent during the period from 5:00 PM to 7:59 AM the next day, when they stay at home or go out with their parents.

[0128] Proxy trajectory data refers to the use of the original trajectory data of related independent individuals as a location proxy for the non-independent individuals during non-independent activity periods.

[0129] For example, the location of a susceptible individual at home at night during development can be represented by the location of their father's mobile phone.

[0130] Independent activity periods refer to the geographical coordinates or sub-regions corresponding to a fixed activity location.

[0131] The geographical location of a fixed activity venue refers to the geographical coordinates or sub-region to which the fixed activity venue belongs.

[0132] For example, the geographical location of XX school is XX° east longitude and XX° north latitude, corresponding to sub-region G-452.

[0133] The preset stay duration refers to the default daily stay time within a fixed activity venue, which is set based on the venue type or historical experience.

[0134] For example, the preset stay duration for primary schools is 8.5 hours / day, and for nursing homes it is 22 hours / day.

[0135] Fixed trajectory data refers to a virtual trajectory generated within an independent activity period, centered on a fixed activity location and based on a preset dwell time.

[0136] For example, the system locks the location of susceptible individuals in the developmental stage in sub-region G-452 between 8:00 and 16:30, forming fixed trajectory data.

[0137] Complete trajectory data refers to the full-time trajectory formed by splicing and fusing proxy trajectory data and fixed trajectory data in chronological order, and is used to characterize the overall activity pattern of non-independent individuals.

[0138] For example, the complete trajectory data of susceptible individuals during development includes: 7:00–8:00 (commuting with father), 8:00–16:30 (fixed time at school), and 16:30–22:00 (staying at home with father), achieving full 24-hour coverage.

[0139] It should be noted that all individual data has been anonymized, with basic attribute information and trajectory data stored separately. Only group statistical characteristics are retained for risk modeling, complying with the Personal Information Protection Law and ethical review requirements. This application employs "de-identification" when processing data on guardians and non-independent individuals. The system only extracts spatial statistical features of the trajectory (such as the duration of stay in a grid cell) and does not store the original latitude and longitude coordinates; furthermore, guardianship relationship matching is performed only in a local encrypted environment to ensure personal privacy and security.

[0140] 103. Based on spatiotemporal activity data, determine the set of activity sub-regions associated with each individual.

[0141] The activity sub-region set refers to the set of all sub-regions in which an individual actually stays (e.g., the cumulative stay time exceeds a threshold) within a preset time period, based on the individual's spatiotemporal activity data. It is used to characterize the spatial range of the individual's daily exposure.

[0142] For example, if a student spends significant time in only three sub-regions within a week—sub-region A where their home is located, sub-region B where their school is located, and sub-region C where they frequently visit a park on weekends (spending an average of ≥1 hour per day in each sub-region), then their activity sub-region set is {A,B,C}.

[0143] In other embodiments, the set of activity sub-areas for commuters typically includes residential sub-areas, office sub-areas, and several transportation transfer node sub-areas, while sub-areas that are passed through briefly without a significant stop are not included in the set.

[0144] 104. Based on the region type of each sub-region, the individual type corresponding to each individual, and the set of active sub-regions, cluster multiple individuals to obtain multiple groups. Individuals in the same group have the same individual type and the same set of active sub-regions.

[0145] Here, a group refers to a homogeneous subset obtained by clustering multiple individuals, where each individual has the same individual type and their activity sub-regions are completely identical. The group represents a population highly similar in demographic attributes and daily spatial behavior patterns, and serves as the basic unit for subsequent differentiated exposure assessments and health risk calculations.

[0146] For example, all "developmentally susceptible individuals" can be grouped according to their activity sub-region sets: if 12 developmentally susceptible individuals only operate in the two sub-regions {residential sub-region R-087, school sub-region S-215}, they will be classified into the same group; while another developmentally susceptible individual, due to frequent visits to the town health center (sub-region H-033), has an activity sub-region set of {R-087, S-215, H-033}, and therefore belongs to a different group.

[0147] Among urban office workers, individuals whose activity range is limited to {residential area A, office building B} constitute one group, while individuals who frequently travel for business and whose activity sub-area includes airport C form another independent group, even if the two groups are of the same type (both are "office workers").

[0148] 105. For each group, based on the soil heavy metal content of each activity sub-region in the group's activity sub-region set, the differential exposure parameters corresponding to the group, and the distribution characteristics of the group's residence time in each activity sub-region, calculate the group's average daily intake of heavy metals under multiple preset exposure pathways, and obtain the group's health risk score based on the average daily intake of heavy metals.

[0149] Among them, differentiated exposure parameters refer to personalized exposure parameters used for different individual types in specific regional types, corresponding to different exposure pathways, to more realistically reflect actual exposure behavior. These parameters differ from the default values ​​uniformly used in traditional risk assessments.

[0150] For example, for "developmentally susceptible individuals" in "farmland" areas, the rate of soil ingestion via oral administration was set at 200 mg / day (due to frequent hand-mouth interactions); while for "urban office workers" in "residential areas," this parameter was set at 50 mg / day. Similarly, skin contact area, respiratory inhalation rate, etc., were also dynamically adjusted according to population-region combinations.

[0151] The distribution characteristics of the duration of stay of a group in each activity sub-region refer to the proportion of time allocated to each sub-region in the set of activity sub-regions of the group within a preset time period, which is used to characterize the spatial exposure weight.

[0152] For example, if a group of susceptible individuals in the developmental stage spends an average of 6 hours and 4 hours per day in the residential sub-region R-12 and the farmland sub-region F-34 respectively, then the distribution characteristics of their stay duration are: 60% in R-12 and 40% in F-34.

[0153] Preset exposure pathways refer to the main routes through which heavy metals enter the human body, and are pre-defined and used for systematic risk assessment.

[0154] In some embodiments, the preset exposure routes include three main routes: oral ingestion of soil, skin contact with soil, and inhalation of dust particles.

[0155] Pre-defined exposure pathways for soil ingestion via oral route: ;

[0156] Pre-defined exposure routes for skin contact with soil: ;

[0157] Pre-defined exposure pathways for inhaling particulate matter: ;

[0158] Subscript It indicates a specific type of individual (such as the aforementioned developmentally susceptible individuals, adults, and other different groups). The average daily dose ingested orally, Soil ingestion rate refers to the daily soil mass ingested by an individual (unit: mg / day), which is dynamically adjusted according to individual type and regional type (e.g., 200 mg / day for susceptible individuals in the developmental stage, and 50 mg / day for adults). For the first Exposure frequency for an individual refers to the number of days of exposure per year (unit: d / year), usually taken as 365 d / year (exposure throughout the year) or adjusted for seasonality. Exposure Duration refers to the number of years an individual is exposed to a particular area (unit: year). For example, for susceptible individuals during developmental stages, the exposure duration is typically defined as the number of years during a specific developmental stage (e.g., 6 years). The conversion factor is used to convert soil concentration from mg / kg to mg / g, with a value of 1×10⁻⁶. -3 , This refers to individual body weight, measured in kg, and is set according to different individual types (e.g., 18 kg for susceptible individuals during development, and 70 kg for adults). The average exposure time is measured in days (d), typically equal to the exposure duration multiplied by 365. , Average daily dose for skin contact, Skin area refers to the body surface area in daily contact with soil (unit: cm²), and is set according to individual type and activity scenario (e.g., 2,000 cm² for susceptible individuals during development). Soil adhesion factor (SAF) refers to the mass of soil adhering to a unit area of ​​skin (unit: mg / cm²), with a typical value of 10 mg / cm². ABS stands for Absorption Fraction, which refers to the proportion of heavy metals absorbed through the skin, usually taken as 0.1 (10%). The average daily inhaled dose, Inhalation rate refers to the daily volume of air inhaled (unit: m³ / day). For adults, it is taken as 20 m³ / day, and for susceptible individuals during development, it is taken as 10 m³ / day. The Particulate Exposure Factor (PEF) is the conversion factor between the concentration of particulate matter in the air and the concentration of heavy metals in the soil. The unit is m³ / kg, and a typical value is 1×10⁻⁶. 6 m 3 / kg.

[0159] It is understandable that differential exposure parameters include, but are not limited to: soil uptake rate ( ), skin exposed area ( ), respiratory intake rate ( and exposure frequency () These parameters are dynamically matched based on individual type (e.g., developmentally susceptible individuals, adults, workers) and the type of their sub-region (e.g., farmland, industrial area, residential area) to reflect the actual exposure scenario. Non-differentiated parameters: , , , , (Mostly constants or standard values).

[0160] Daily heavy metal intake refers to the average daily intake of heavy metals by a group through a pre-defined exposure pathway (usually expressed in mg / kg·bw / day), and is a core intermediate variable in health risk calculation.

[0161] For example, the average daily intake of Cd orally in a certain group is 1.2 × 10⁻⁶. -6 mg / kg·bw / day, skin contact: 3.5×10 -7 mg / kg·bw / day, inhalation dose 8.0×10 -8 mg / kg·bw / day.

[0162] Health risk score is a quantitative health risk indicator calculated based on the average daily intake of heavy metals by a group under various preset exposure pathways. It is used to characterize the carcinogenic or non-carcinogenic risk level faced by the group due to soil heavy metal exposure.

[0163] In some embodiments, individual behavioral differences and group commonalities are effectively integrated, avoiding the bias of substituting a single individual's trajectory for group exposure, and significantly improving the statistical robustness and population representativeness of dose estimation. Specifically, before calculating the average daily heavy metal intake of the group under multiple preset exposure pathways based on the soil heavy metal content of each activity sub-region in the group's activity sub-region set, the differentiated exposure parameters corresponding to the group, and the distribution characteristics of the group's residence time in each activity sub-region, the method further includes:

[0164] For any activity sub-region in the set of activity sub-regions of the group, calculate the average stay time of all individuals in the group in any activity sub-region.

[0165] Calculate the total average dwell time of all individuals in the group across all active sub-regions in the set of active sub-regions;

[0166] The ratio of the average stay time in any activity sub-region to the total average stay time is determined as the proportion of time the group spends in any activity sub-region.

[0167] The distribution characteristics of the duration of a group's stay in each activity sub-region are determined by the proportion of time the group spends in each activity sub-region.

[0168] In this context, any active subregion refers to any subregion within the set of active subregions of a group.

[0169] For example, if the set of active sub-regions of a certain group is {A,B,C}, then “A”, “B”, or “C” can all be treated as “any active sub-region” and processed separately.

[0170] Average stay duration refers to the arithmetic mean of the daily stay duration of all individuals in the group in a specific activity sub-area within a preset time period.

[0171] For example, a developmentally susceptible group consisting of 10 individuals spent an average of 3.2 hours, 2.8 hours, 3.5 hours, etc. in the farmland sub-area F-34, with an average stay of 3.1 hours per day.

[0172] Total average stay duration refers to the average of the sum of the daily average stay durations of all individuals in the group across all sub-regions in their entire set of activity sub-regions, reflecting the total daily active exposure duration of the group.

[0173] For example, the average total daily stay of the 10 susceptible individuals in the developmental stage in the three sub-areas {residential R-12, farmland F-34, village school S-09} was 14.5h, 13.8h, 15.2h, ... respectively, so the total average stay was 14.6 hours / day.

[0174] Time percentage refers to the proportion of a group's average time spent in a certain activity sub-region to its total average time spent in the activity sub-region. It is used to characterize the relative importance of the sub-region in the group's daily activities and serves as an exposure weight.

[0175] For example, if the average stay duration of this group in farmland F-34 is 3.1 hours and the total average stay duration is 14.6 hours, then the proportion of their time spent in F-34 is approximately 3.1 ÷ 14.6 ≈ 21.2%. This value is the characteristic component of their stay duration distribution in F-34.

[0176] Understandably, the sum of time percentages is always 1 (or 100%), forming a probability distribution vector, which is naturally suitable as a weighting coefficient for the contribution of multi-source exposure (see weight 9). Compared to directly using absolute dwell time, this normalization method eliminates the influence of differences in total individual activity, making the exposure patterns between different groups more comparable.

[0177] In some embodiments, a scientifically normalized assessment of population health risk is achieved. This method overcomes the limitations of traditional approaches that focus solely on pollutant concentration or a single exposure pathway, and can accurately reflect the cumulative health threats under combined exposure, providing an objective basis for identifying high-risk populations and prioritizing interventions. Specifically, a population health risk score is obtained based on the average daily intake of heavy metals, including:

[0178] The carcinogenic slope factor for each of the various preset exposure pathways was obtained. The carcinogenic slope factor represents the probability of carcinogenesis caused by a unit of daily intake of heavy metals.

[0179] For the average daily intake of heavy metals in a group under each preset exposure route, the intake is multiplied by the carcinogenic slope factor corresponding to the preset exposure route to obtain the individual carcinogenic risk value of the group through the preset exposure route.

[0180] The total carcinogenic risk value of the population is obtained by summing up the individual carcinogenic risk values ​​of the population through all preset exposure pathways.

[0181] A health risk score for the population is determined based on the total carcinogenic risk value.

[0182] The carcinogenic slope factor, also known as the lifetime excess risk of cancer corresponding to a unit of daily heavy metal intake, is an internationally recognized toxicity parameter used to convert exposure dose into a health risk value. Its unit is typically (mg / kg·bw / day). -1 .

[0183] For example, the carcinogenic slope factor of cadmium (Cd) through oral ingestion is 6.1 (mg / kg·bw / day). -1 The inhalation slope factor for arsenic (As) is 15.0 (mg / kg·bw / day). -1 .

[0184] The daily intake of heavy metals per unit refers to the exposure level corresponding to the intake of 1 mg of heavy metal per kilogram of body weight per day (i.e., 1 mg / kg·bw / day), which is used as the benchmark dose unit for risk calculation.

[0185] In some embodiments, "daily intake per unit of heavy metal" is not the actual intake, but a reference dose unit used for standardized risk calculation. Actual risk = actual intake × carcinogenic slope factor.

[0186] The probability of carcinogenicity refers to the likelihood that an individual will develop cancer due to lifetime exposure to a certain dose of heavy metal, and is usually expressed as a dimensionless value (e.g., 1 × 10⁻⁶). -4 (This represents one in ten thousand).

[0187] For example, if the total cancer risk value for a certain group is 3×10 -5 This means that approximately 3 out of every 100,000 people in this group may develop cancer due to exposure to heavy metals in the soil.

[0188] The single carcinogenic risk value refers to the contribution of a group to the carcinogenic risk through a specific exposure route (such as oral intake), which is obtained by multiplying the average daily intake of heavy metals through that route by the corresponding carcinogenic slope factor.

[0189] For example, the average daily oral intake of Cd for an individual in a sensitive developmental period is 2.0 × 10⁻⁶. -6 Multiplying mg / kg bw / day by SF=6.1 yields a single carcinogenic risk value of 1.22×10⁻⁶. -5 .

[0190] The total carcinogenic risk value refers to the sum of individual carcinogenic risk values ​​generated by a group through all pre-specified exposure routes (oral, skin, inhalation), representing its overall carcinogenic risk level.

[0191] For example, the oral risk for this group is 1.22 × 10⁻⁶. -5 The risk of skin contact is 0.18 × 10⁻⁶. -5 The inhalation risk is 0.05 × 10⁻⁶.-5 The overall carcinogenic risk value is 1.45 × 10⁻⁶. -5 .

[0192] In some embodiments, the total carcinogenic risk value is directly used as the health risk score of the population; when this value is ≥ a preset health risk score threshold (e.g., 1 × 10⁻⁶), the risk score is considered to be positive. -4 When a group is identified as being at high risk, it should be prioritized for inclusion in the governance scope.

[0193] In other embodiments, the total carcinogenic risk value is logarithmically transformed or normalized to generate a health risk score of 0–100, which facilitates comparisons across regions and populations.

[0194] Total carcinogenic risk value It is the sum of the average daily dose (ADD) of each exposure route multiplied by the corresponding carcinogenic slope factor (SF).

[0195] In some embodiments, it also includes:

[0196] Based on the average daily intake of heavy metals by the population under each pre-specified exposure route, the total non-carcinogenic risk value was calculated, including:

[0197] For each preset exposure pathway, the average daily intake of heavy metals under the preset exposure pathway was calculated. Divide by the corresponding reference dose This yields a single non-carcinogenic risk value.

[0198] The total non-carcinogenic risk value of the population is obtained by summing up all the individual non-carcinogenic risk values.

[0199] A health risk score for the population is determined based on the total non-carcinogenic risk value and the total carcinogenic risk value.

[0200] Total non-carcinogenic risk value The hazard quotient (RfD) is calculated by dividing the daily average dose from each exposure route by the reference dose (RfD) and is used to determine non-carcinogenic effects.

[0201] when When the carcinogenic risk is negligible, it indicates that the risk is negligible.

[0202] when When the time is right, it indicates a significant carcinogenic risk;

[0203] when When this occurs, it indicates a serious carcinogenic risk.

[0204] In some embodiments, the assumption of using default parameters uniformly for all populations in traditional risk assessments is abandoned, significantly improving the behavioral realism and physiological rationality of exposure dose estimation, especially applicable to complex areas with large urban-rural differences and strong population heterogeneity. Specifically, before calculating the average daily heavy metal intake of the population under multiple preset exposure pathways based on the soil heavy metal content of each activity sub-region in the population's activity sub-region set, the differentiated exposure parameters corresponding to the population, and the distribution characteristics of the population's residence time in each activity sub-region, the following steps are included:

[0205] Obtain a pre-set exposure parameter library, which stores baseline exposure parameters for different individual types and different regional types corresponding to various preset exposure pathways;

[0206] Based on the individual type corresponding to the group and the region type of each activity sub-region in its activity sub-region set, the corresponding baseline exposure parameters are matched and extracted from the exposure parameter library and determined as the differentiated exposure parameters corresponding to the group in each activity sub-region.

[0207] The exposure parameter library refers to a pre-built structured database used to store baseline exposure parameters for different individual types in different regions corresponding to various preset exposure pathways, and supports rapid retrieval and matching by "population-environment" combination.

[0208] The exposure parameter library is pre-generated based on large-scale population behavioral survey data, localized questionnaire survey results, etc., after statistical analysis.

[0209] For example, the exposure parameter library is stored on a server or local storage medium and organized in the form of a three-dimensional mapping table: Dimension 1 is the individual type (such as "susceptible individuals in the developmental stage" or "rural farmers"), Dimension 2 is the area type (such as "farmland", "industrial area", or "residential area"), and Dimension 3 is the exposure route (such as "oral ingestion", "skin contact", or "inhalation"). Each cell contains the corresponding baseline exposure parameter value.

[0210] The baseline exposure parameter refers to the standardized parameter value that characterizes typical exposure behavior under a specific combination of "individual type – region type – exposure pathway", and serves as the basic input for differentiated exposure modeling.

[0211] For example, regarding the oral intake pathway in farmland areas for "susceptible groups during developmental years," baseline exposure parameters include: soil intake rate. ,weight (This reflects the physiological vulnerability of this group); while susceptible individuals at the same developmental stage in "residential areas" This is 100mg / day, reflecting the influence of the environment on behavior.

[0212] For "inhalation" routes of exposure in "industrial workers" within "industrial land," the baseline exposure parameters include: respiratory rate. Exposure frequency The rate is significantly higher than that of ordinary residents.

[0213] In some embodiments, a multi-level coupling calculation of soil heavy metal content, differentiated exposure parameters, and residence time weights is used to construct an integrated "space-behavior-dose" daily intake model. This method not only considers the spatial heterogeneity of pollutant concentrations but also integrates individual differences in population behavior patterns and exposure sensitivities, significantly improving the spatiotemporal accuracy and population representativeness of heavy metal exposure dose estimation. Specifically, based on the soil heavy metal content of each activity sub-region in the population's activity sub-region set, the differentiated exposure parameters corresponding to the population, and the distribution characteristics of the population's residence time in each activity sub-region, the daily heavy metal intake of the population under various preset exposure pathways is calculated, including:

[0214] For each of the multiple preset exposure pathways, obtain the differential exposure parameters of the population in any of its activity sub-regions corresponding to the preset exposure pathway. The differential exposure parameters are determined based on the individual type of the population and the region type of any activity sub-region.

[0215] The soil heavy metal content of any active sub-region is multiplied with its corresponding differential exposure parameter to obtain the unit exposure intensity of the population in any active sub-region via a preset exposure pathway.

[0216] Based on the time proportion of any activity sub-region in the distribution characteristics of the group's stay time in each activity sub-region, the unit exposure intensity is multiplied by the time proportion to obtain the weighted exposure contribution value of the group in any activity sub-region via a preset exposure path.

[0217] The weighted exposure contribution values ​​corresponding to all activity sub-regions in the activity sub-region set are summed to obtain the average daily intake of heavy metals of the group through the preset exposure pathway.

[0218] Unit exposure intensity refers to the exposure level per unit time of a population in a specific activity sub-region via a predetermined exposure pathway. It is obtained by multiplying the soil heavy metal content of the sub-region with its corresponding differential exposure parameter, reflecting the potential exposure capacity under the "location-pathway" combination.

[0219] The weighted exposure contribution value refers to the actual contribution of a group to the total daily intake through a certain exposure pathway in a certain activity sub-region. It is obtained by multiplying the unit exposure intensity of the sub-region by its time proportion in the distribution characteristics of the group's stay time, reflecting the reality that "if the stay time in a high-exposure area is short, the contribution is limited".

[0220] 106. Spatially map and accumulate the health risk scores of each group according to their corresponding length of stay distribution characteristics to obtain the population health risk load value of each sub-region, and determine the sub-region to be treated from multiple sub-regions based on the population health risk load value of each sub-region.

[0221] The population health risk load value refers to the comprehensive risk index obtained by cumulatively assigning time-weighted health risk scores to all groups passing through or staying in a sub-region. It is used to quantify the overall threat level to population health in that sub-region. The higher the value, the heavier the cumulative burden on public health in that region.

[0222] Sub-regions to be addressed refer to sub-regions where the population health risk load exceeds a preset risk threshold, and are identified as key areas for priority soil remediation, pollution source control, or health intervention.

[0223] For example, setting the risk load threshold as If the load value of sub-region F-34 is Areas with loads exceeding the threshold are marked as sub-areas requiring remediation; while the load value for residential areas R-12 is... If the value is below the threshold, it will not be included in the scope of governance for the time being.

[0224] In other embodiments, all sub-regions are sorted in descending order of population health risk load value, and the top 10% are selected as sub-regions to be treated, so as to achieve precise allocation of resources.

[0225] Understandably, traditional methods, which delineate remediation areas solely based on soil heavy metal concentrations or single-population exposure assessments, tend to overlook scenarios characterized by "low pollution but high population density" or "high-risk population concentrations." This application, through a risk-spatiotemporal coupling mapping mechanism, achieves a paradigm shift from "pollution-oriented" to "health impact-oriented" remediation, significantly enhancing the fairness and effectiveness of environmental management.

[0226] In some embodiments, this method overcomes the limitations of the traditional approach of "prioritizing governance based on pollution concentration," enabling the quantification of regional governance urgency from the perspective of "population health impact." In particular, it can identify key areas where high-risk populations are densely concentrated or spend long periods of time in areas with moderate pollution levels. Specifically, the health risk scores of each group are spatially mapped and accumulated based on their corresponding duration of stay distribution characteristics to obtain the population health risk load value for each sub-region, including:

[0227] For each group, the risk contribution value of the group to each activity sub-area is calculated based on the group's health risk score, the number of individuals in the group, and the distribution characteristics of their corresponding stay duration.

[0228] The risk contribution values ​​of all groups mapped to the same sub-region are summed to obtain the population health risk load value of the sub-region.

[0229] Among them, the risk contribution value refers to the share of health risk allocated to each activity sub-region by a specific group. It is determined by the group's health risk score, the number of people in the group, and the proportion of time spent in the sub-region. It is used to characterize the actual contribution of the group to the health burden of the sub-region.

[0230] Understandably, the calculation of risk contribution value reflects the triple fairness principle of "who is exposed, how long is exposed, and how many people are exposed": high-risk groups, areas with long-term stays, and large-scale gatherings will receive higher load values, thus gaining priority in the allocation of governance resources, and avoiding the crowding out of governance needs that truly affect public health by areas with "high pollution but no inhabitants".

[0231] In some embodiments, by introducing time-based weighting, the health risk score of a group is spatially decomposed according to its actual residence proportion in each activity sub-area, and then the risk contribution value is calculated in combination with the group size. This method realizes a scientific mapping of health risk from the "population dimension" to the "spatial unit," effectively solving the problem of "risk and space disconnect" in traditional governance, and providing a quantitative basis for accurately identifying high-risk areas. Specifically, based on the group's health risk score, the number of individuals in the group, and their corresponding residence duration distribution characteristics, the risk contribution value of the group to each activity sub-area is calculated, including:

[0232] For any activity sub-region in the set of activity sub-regions of a group, the time proportion of any activity sub-region in the dwell time distribution characteristics is determined as the time allocation weight;

[0233] The risk contribution of the group to any activity sub-region is obtained by calculating the health risk score of the group, the product of the number of individuals in the group and the time allocation weight.

[0234] The time allocation weight refers to the proportion of the average daily stay in a specific sub-region of a group's activity sub-region set to the total active stay time. It is used to characterize the time share of that sub-region in the group's daily activities and serves as a spatial allocation coefficient for risk sharing. Its value ranges from (0,1], and the sum of the time allocation weights of all activity sub-regions is 1.

[0235] Understandably, time-based weighting ensures the authenticity of risk allocation behavior: even if a sub-area has a high pollution level, if a group only passes through briefly (weighting close to 0), its risk contribution is minimal; conversely, areas with low pollution but long-term stays (such as residences and schools) may become the main risk-bearing units. This mechanism is more in line with the basic environmental health principle of "exposure = concentration × time × number of people". In this application, taking District A of a city as an example, the soil lead content of the entire area is first predicted using a model that integrates measured points and environmental characteristics; for vulnerable individuals in their developmental stage without mobile phone signaling, the complete activity trajectory is reconstructed using the time-series splicing technology of "guardian trajectory proxy" and "fixed location and fixed rest"; then, the population is clustered into groups such as "vulnerable population in developmental stage" and "adult commuters" and matched with differentiated exposure parameters (such as specific high-frequency surface contact rate); finally, the risk load is calculated by "risk score × number of people × length of stay", accurately identifying concentrated residence areas (such as sub-areas) with extremely high risk loads due to high-density and long-term stays of vulnerable populations in developmental stage, even though the soil concentration is moderate. This makes it superior to abandoned factory areas with high concentrations but low exposure (such as sub-regions). It was identified as the primary target for treatment, realizing a shift in decision-making from "single concentration assessment" to "precise prevention and control".

[0236] In summary, this application can accurately identify areas that pose significant health risks to specific populations and effectively prevent high-risk individuals from being overlooked due to regional average effects, thereby improving the fairness and targeting of soil heavy metal pollution remediation.

[0237] To better implement the above methods, this application also provides a soil heavy metal health risk prediction device based on population risk differentiation. This device can be integrated into an electronic device, such as a terminal or server. The terminal can be a mobile phone, tablet, smart Bluetooth device, laptop, or personal computer; the server can be a single server or a server cluster consisting of multiple servers.

[0238] For example, in this embodiment, a soil heavy metal health risk prediction device for population risk differentiation will be specifically integrated into an electronic device to illustrate the method of this application embodiment in detail.

[0239] For example, such as Figure 2 As shown, the soil heavy metal health risk prediction device for population risk differentiation may include a region division unit 201, a data acquisition unit 202, a region association unit 203, a clustering unit 204, an intake calculation unit 205, and a risk determination unit 206, as follows:

[0240] (I) Regional division unit 201.

[0241] The region division unit 201 is used to perform grid division processing on the region to be evaluated, obtain multiple sub-regions, and obtain the region type and soil heavy metal content of each sub-region.

[0242] In some embodiments, before obtaining the region type and soil heavy metal content of each sub-region, the method further includes:

[0243] Obtain the land environmental characteristic parameters of each sub-region in the area to be evaluated, as well as the measured soil heavy metal content of some sub-regions;

[0244] Based on the land environmental characteristic parameters of some sub-regions and their corresponding measured soil heavy metal content, multiple heavy metal prediction models are constructed. Different heavy metal prediction models are constructed based on different combinations of land environmental characteristics.

[0245] Using multiple heavy metal prediction models, the predicted soil heavy metal content of the remaining sub-regions is predicted based on the land environmental characteristic parameters of each sub-region. The remaining sub-regions are the sub-regions in the area to be evaluated, excluding some sub-regions.

[0246] The predicted soil heavy metal contents of the remaining sub-regions are fused together to obtain the soil heavy metal contents of the remaining sub-regions.

[0247] In some embodiments, based on the land environmental characteristic parameters of certain sub-regions and their corresponding measured soil heavy metal contents, multiple heavy metal prediction models are constructed, including:

[0248] Acquire multiple pre-set reference heavy metal prediction models, as well as target feature information characterizing the regional characteristics of the area to be evaluated;

[0249] From the land environmental characteristic parameters, select the target land environmental characteristic parameters that match the target characteristic information;

[0250] From multiple reference heavy metal prediction models, candidate reference heavy metal prediction models whose input variables include target land environmental characteristic parameters are selected.

[0251] By using the target land environmental characteristic parameters of some sub-regions and their corresponding measured soil heavy metal content, the model parameters of the candidate reference heavy metal prediction model are calibrated to obtain the calibrated heavy metal prediction model.

[0252] The calibrated heavy metal prediction model, along with the remaining reference heavy metal prediction models (excluding the candidate reference heavy metal prediction model) from among the multiple reference heavy metal prediction models, are identified as multiple heavy metal prediction models.

[0253] (ii) Data acquisition unit 202.

[0254] The data acquisition unit 202 is used to acquire the individual types corresponding to multiple individuals in the area to be evaluated, as well as the spatiotemporal activity data of each individual in each sub-area. The spatiotemporal activity data includes at least the duration of stay of the individual in each sub-area.

[0255] In some embodiments, before obtaining the individual types corresponding to multiple individuals in the region to be evaluated, and the spatiotemporal activity data of each individual in each sub-region, the method further includes:

[0256] Obtain basic attribute information of multiple individuals in the area to be evaluated;

[0257] Based on basic attribute information, multiple individuals are divided into independent individuals and non-independent individuals;

[0258] For an individual, the spatiotemporal activity data of the individual is obtained by mapping the original trajectory data generated by the mobile terminal held by the individual within a preset time period to the sub-regions of the area to be evaluated.

[0259] For non-independent individuals, based on the guardianship relationship identifier in the basic attribute information, the associated independent individuals bound to the non-independent individuals are identified, and the original trajectory data of the associated independent individuals within a preset time period, as well as the independent activity time periods and corresponding fixed activity locations of the non-independent individuals are obtained.

[0260] When a non-independent individual is in a non-independent activity period, the original trajectory data of the associated independent individual is used as the proxy trajectory data of the non-independent individual.

[0261] When a non-independent individual is in an independent activity period, fixed trajectory data of the non-independent individual is generated based on the geographical location of the fixed activity location and the preset stay duration.

[0262] By temporally fusing proxy trajectory data with fixed trajectory data, complete trajectory data of non-independent individuals is obtained. The complete trajectory data is then mapped to each sub-region of the area to be evaluated to obtain spatiotemporal activity data of non-independent individuals.

[0263] (III) Regional Related Unit 203.

[0264] The regional association unit 203 is used to determine the set of activity sub-regions associated with each individual based on spatiotemporal activity data.

[0265] (iv) Clustering unit 204.

[0266] Clustering unit 204 is used to cluster multiple individuals based on the region type of each sub-region, the individual type corresponding to the individual, and the set of active sub-regions to obtain multiple groups. Individuals in the same group have the same individual type and the same set of active sub-regions.

[0267] (v) Intake calculation unit 205.

[0268] The intake calculation unit 205 is used to calculate the average daily intake of heavy metals of each group under multiple preset exposure pathways, based on the soil heavy metal content of each activity sub-region in the group's activity sub-region set, the differential exposure parameters corresponding to the group, and the distribution characteristics of the group's residence time in each activity sub-region, and to obtain the group's health risk score based on the average daily intake of heavy metals.

[0269] In some embodiments, before calculating the average daily heavy metal intake of the population under multiple preset exposure pathways based on the soil heavy metal content of each activity sub-region in the population's activity sub-region set, the differential exposure parameters corresponding to the population, and the distribution characteristics of the population's residence time in each activity sub-region, the method further includes:

[0270] For any activity sub-region in the set of activity sub-regions of the group, calculate the average stay time of all individuals in the group in any activity sub-region.

[0271] Calculate the total average dwell time of all individuals in the group across all active sub-regions in the set of active sub-regions;

[0272] The ratio of the average stay time in any activity sub-region to the total average stay time is determined as the proportion of time the group spends in any activity sub-region.

[0273] The distribution characteristics of the duration of a group's stay in each activity sub-region are determined by the proportion of time the group spends in each activity sub-region.

[0274] In some embodiments, a health risk score for a population is obtained based on the average daily intake of heavy metals, including:

[0275] The carcinogenic slope factor for each of the various preset exposure pathways was obtained. The carcinogenic slope factor represents the probability of carcinogenesis caused by a unit of daily intake of heavy metals.

[0276] For the average daily intake of heavy metals in a group under each preset exposure route, the intake is multiplied by the carcinogenic slope factor corresponding to the preset exposure route to obtain the individual carcinogenic risk value of the group through the preset exposure route.

[0277] The total carcinogenic risk value of the population is obtained by summing up the individual carcinogenic risk values ​​of the population through all preset exposure pathways.

[0278] A health risk score for the population is determined based on the total carcinogenic risk value.

[0279] In some embodiments, before calculating the average daily intake of heavy metals of the population under multiple preset exposure pathways based on the soil heavy metal content of each activity sub-region in the population's activity sub-region set, the differential exposure parameters corresponding to the population, and the distribution characteristics of the population's residence time in each activity sub-region, the following steps are included:

[0280] Obtain a pre-set exposure parameter library, which stores baseline exposure parameters for different individual types and different regional types corresponding to various preset exposure pathways;

[0281] Based on the individual type corresponding to the group and the region type of each activity sub-region in its activity sub-region set, the corresponding baseline exposure parameters are matched and extracted from the exposure parameter library and determined as the differentiated exposure parameters corresponding to the group in each activity sub-region.

[0282] In some embodiments, the average daily intake of heavy metals by the population under multiple preset exposure pathways is calculated based on the soil heavy metal content in each activity sub-region of the population's activity sub-region set, the differential exposure parameters corresponding to the population, and the distribution characteristics of the population's residence time in each activity sub-region, including:

[0283] For each of the multiple preset exposure pathways, obtain the differential exposure parameters of the population in any of its activity sub-regions corresponding to the preset exposure pathway. The differential exposure parameters are determined based on the individual type of the population and the region type of any activity sub-region.

[0284] The soil heavy metal content of any active sub-region is multiplied with its corresponding differential exposure parameter to obtain the unit exposure intensity of the population in any active sub-region via a preset exposure pathway.

[0285] Based on the time proportion of any activity sub-region in the distribution characteristics of the group's stay time in each activity sub-region, the unit exposure intensity is multiplied by the time proportion to obtain the weighted exposure contribution value of the group in any activity sub-region via a preset exposure path.

[0286] The weighted exposure contribution values ​​corresponding to all activity sub-regions in the activity sub-region set are summed to obtain the average daily intake of heavy metals of the group through the preset exposure pathway.

[0287] (vi) Risk determination unit 206.

[0288] The risk determination unit 206 is used to spatially map and accumulate the health risk scores of each group according to their corresponding length of stay distribution characteristics to obtain the population health risk load value of each sub-region, and to determine the sub-region to be treated from multiple sub-regions based on the population health risk load value of each sub-region.

[0289] In some embodiments, the health risk scores of each group are spatially mapped and accumulated based on their corresponding length of stay distribution characteristics to obtain the population health risk load value for each sub-region, including:

[0290] For each group, the risk contribution value of the group to each activity sub-area is calculated based on the group's health risk score, the number of individuals in the group, and the distribution characteristics of their corresponding stay duration.

[0291] The risk contribution values ​​of all groups mapped to the same sub-region are summed to obtain the population health risk load value of the sub-region.

[0292] In some embodiments, the risk contribution value of the group to each activity sub-region is calculated based on the group's health risk score, the number of individuals in the group, and the corresponding length of stay distribution characteristics, including:

[0293] For any activity sub-region in the set of activity sub-regions of a group, the time proportion of any activity sub-region in the dwell time distribution characteristics is determined as the time allocation weight;

[0294] The risk contribution of the group to any activity sub-region is obtained by calculating the health risk score of the group, the product of the number of individuals in the group and the time allocation weight.

[0295] In practice, each of the above units can be implemented as an independent entity or can be arbitrarily combined to be implemented as the same or several entities. For the specific implementation of each of the above units, please refer to the previous method embodiments, which will not be repeated here.

[0296] Therefore, the embodiments of this application can accurately identify areas that pose significant health risks to specific populations and effectively avoid ignoring high-risk individuals due to regional average effects, thereby improving the fairness and targeting of soil heavy metal pollution remediation.

[0297] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be performed by instructions, or by instructions controlling related hardware. These instructions can be stored in a computer-readable storage medium and loaded and executed by a processor.

[0298] Therefore, embodiments of this application provide a computer-readable storage medium storing a plurality of instructions that can be loaded by a processor to execute steps in any of the soil heavy metal health risk prediction methods for population risk differentiation provided in embodiments of this application.

[0299] The storage medium may include: read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.

[0300] Since the instructions stored in the storage medium can execute the steps in any of the soil heavy metal health risk prediction methods for population risk differentiation provided in the embodiments of this application, the beneficial effects that any of the soil heavy metal health risk prediction methods for population risk differentiation provided in the embodiments of this application can achieve can be realized, as detailed in the preceding embodiments, and will not be repeated here.

[0301] According to one aspect of this application, a computer program product or computer program is provided, comprising a computer program / instructions stored in a computer-readable storage medium. A processor of an electronic device reads the computer program / instructions from the computer-readable storage medium and executes the computer program / instructions, causing the electronic device to perform the method provided in the above embodiments for predicting soil heavy metal health risks based on population risk differentiation.

[0302] The above provides a detailed description of a soil heavy metal health risk prediction method based on population risk differentiation provided by the embodiments of this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for predicting the health risk of heavy metals in soil based on population risk differentiation, characterized in that, The method includes: The area to be evaluated is divided into grids to obtain multiple sub-regions, and the region type and soil heavy metal content of each sub-region are obtained. Obtain the individual types corresponding to multiple individuals in the region to be evaluated, as well as the spatiotemporal activity data of each individual in each sub-region. The spatiotemporal activity data includes at least the duration of the individual's stay in each sub-region. Based on the aforementioned spatiotemporal activity data, determine the set of activity sub-regions associated with each individual; Based on the region type of each sub-region, the individual type corresponding to the individual, and the set of active sub-regions, the multiple individuals are clustered to obtain multiple groups. Individuals in the same group have the same individual type and the same set of active sub-regions. For each group, based on the soil heavy metal content of each activity sub-region in the group's activity sub-region set, the differential exposure parameters corresponding to the group, and the distribution characteristics of the group's residence time in each activity sub-region, the average daily intake of heavy metals of the group under multiple preset exposure pathways is calculated, and the health risk score of the group is obtained based on the average daily intake of heavy metals. The health risk scores of each group are spatially mapped and accumulated based on their corresponding length of stay distribution characteristics to obtain the population health risk load value of each sub-region. Based on the population health risk load value of each sub-region, the sub-region to be treated is determined from the multiple sub-regions.

2. The method as described in claim 1, characterized in that, The data is based on the soil heavy metal content in each activity sub-region of the population, the differential exposure parameters corresponding to the population, and the distribution characteristics of the population's residence time in each activity sub-region. Before calculating the average daily heavy metal intake of the group under multiple preset exposure pathways, the method further includes: For any activity sub-region in the set of activity sub-regions of the group, calculate the average stay time of all individuals in the group in that activity sub-region; Calculate the total average dwell time of all individuals in the group across all activity sub-regions in the set of activity sub-regions; The ratio of the average stay time in any activity sub-area to the total average stay time is determined as the proportion of time the group spends in any activity sub-area. Based on the proportion of time the group spends in each activity sub-region, the distribution characteristics of its stay duration in each activity sub-region are constituted.

3. The method as described in claim 1, characterized in that, The process of spatially mapping and accumulating the health risk scores of each group based on their corresponding length of stay distribution characteristics to obtain the population health risk load value for each sub-region includes: For each group, the risk contribution value of the group to each activity sub-area is calculated based on the group's health risk score, the number of individuals in the group, and the corresponding length of stay distribution characteristics. The risk contribution values ​​of all groups mapped to the same sub-region are summed to obtain the population health risk load value of the sub-region.

4. The method as described in claim 3, characterized in that, The data is based on the health risk score of the group, the number of individuals in the group, and the distribution characteristics of their corresponding length of stay. Calculating the risk contribution of the group to each activity sub-region includes: For any activity sub-region in the set of activity sub-regions of the group, the time proportion of any activity sub-region in the dwell time distribution characteristics is determined as the time allocation weight; The risk contribution value of the group to any activity sub-region is obtained by multiplying the health risk score of the group, the number of individuals in the group, and the time allocation weight.

5. The method as described in claim 1, characterized in that, Before obtaining the region type and soil heavy metal content of each sub-region, the following steps are also included: Obtain the land environmental characteristic parameters of each sub-region in the area to be evaluated, as well as the measured soil heavy metal content of some sub-regions; Based on the land environmental characteristic parameters of each sub-region and their corresponding measured soil heavy metal content, multiple heavy metal prediction models are constructed, wherein different heavy metal prediction models are constructed based on different combinations of land environmental characteristics. Using the multiple heavy metal prediction models, the predicted soil heavy metal content of the remaining sub-regions is predicted based on the land environmental characteristic parameters of each sub-region. The remaining sub-regions are the sub-regions in the region to be evaluated other than the aforementioned sub-regions. The predicted soil heavy metal content of the remaining sub-region is obtained by fusing multiple predicted soil heavy metal contents.

6. The method as described in claim 5, characterized in that, Based on the land environmental characteristic parameters of each sub-region and their corresponding measured soil heavy metal content, multiple heavy metal prediction models are constructed, including: Acquire multiple pre-set reference heavy metal prediction models, as well as target feature information characterizing the regional characteristics of the area to be evaluated; From the land environmental feature parameters, select the target land environmental feature parameters that match the target feature information; From the multiple reference heavy metal prediction models, a candidate reference heavy metal prediction model whose input variables include the target land environmental characteristic parameters is selected; The model parameters of the candidate reference heavy metal prediction model are calibrated by using the target land environmental characteristic parameters of the sub-regions and their corresponding measured soil heavy metal content, so as to obtain the calibrated heavy metal prediction model. The calibrated heavy metal prediction model, together with the remaining reference heavy metal prediction models (excluding the candidate reference heavy metal prediction model) among the multiple reference heavy metal prediction models, are determined as multiple heavy metal prediction models.

7. The method as described in claim 1, characterized in that, The health risk score of the group obtained based on the average daily intake of heavy metals includes: Obtain the carcinogenic slope factor corresponding to each of the multiple preset exposure pathways, wherein the carcinogenic slope factor characterizes the probability of carcinogenesis caused by a unit of daily heavy metal intake; For the average daily intake of heavy metals of the group under each preset exposure pathway, the intake is multiplied by the carcinogenic slope factor corresponding to the preset exposure pathway to obtain the individual carcinogenic risk value of the group through the preset exposure pathway. The total carcinogenic risk value of the group is obtained by summing the individual carcinogenic risk values ​​of the group through all preset exposure pathways. Based on the total carcinogenic risk value, a health risk score is determined for the group.

8. The method as described in claim 1, characterized in that, The data is based on the soil heavy metal content in each activity sub-region of the population, the differential exposure parameters corresponding to the population, and the distribution characteristics of the population's residence time in each activity sub-region. Before calculating the average daily heavy metal intake of the group under multiple preset exposure pathways, the following steps are included: Obtain a preset exposure parameter library, which stores baseline exposure parameters for different individual types and different regional types corresponding to multiple preset exposure pathways; Based on the individual type corresponding to the group and the region type of each activity sub-region in its activity sub-region set, the corresponding baseline exposure parameters are matched and extracted from the exposure parameter library and determined as the differentiated exposure parameters of the group in each activity sub-region.

9. The method as described in claim 1, characterized in that, The method is based on the soil heavy metal content of each activity sub-region in the set of activity sub-regions of the population, the differential exposure parameters corresponding to the population, and the distribution characteristics of the population's residence time in each activity sub-region. Calculate the average daily intake of heavy metals of the group under multiple preset exposure pathways, including: For each of the multiple preset exposure pathways, obtain the differential exposure parameters of the group corresponding to the preset exposure pathway in any of its activity sub-regions. The differential exposure parameters are determined based on the individual type of the group and the region type of the any activity sub-region. The soil heavy metal content of any active sub-region is multiplied with its corresponding differential exposure parameter to obtain the unit exposure intensity of the population in any active sub-region via the preset exposure pathway. Based on the time proportion of any activity sub-region in the distribution characteristics of the group's stay time in each activity sub-region, the unit exposure intensity is multiplied by the time proportion to obtain the weighted exposure contribution value of the group in any activity sub-region via the preset exposure path. The weighted exposure contribution values ​​corresponding to all active sub-regions in the set of active sub-regions are summed to obtain the average daily intake of heavy metals of the group through the preset exposure pathway.

10. The method as described in claim 1, characterized in that, Before obtaining the individual types corresponding to multiple individuals in the region to be evaluated, and the spatiotemporal activity data of each individual in each sub-region, the method further includes: Obtain basic attribute information of multiple individuals in the region to be evaluated; Based on the aforementioned basic attribute information, the multiple individuals are divided into independent individuals and non-independent individuals; For the individual, the spatiotemporal activity data of the individual is obtained by mapping the original trajectory data generated by the mobile terminal it holds within a preset time period to each sub-region of the area to be evaluated. For the non-independent individual, based on the guardianship relationship identifier in the basic attribute information, the associated independent individual bound to the non-independent individual is determined, and the original trajectory data of the associated independent individual within the preset time period, as well as the independent activity time period and corresponding fixed activity location of the non-independent individual are obtained. When the non-independent individual is in a non-independent activity period, the original trajectory data of the associated independent individual is used as the proxy trajectory data of the non-independent individual; When the non-independent individual is in an independent activity period, fixed trajectory data of the non-independent individual is generated based on the geographical location of the fixed activity location and the preset stay duration; The proxy trajectory data and the fixed trajectory data are fused in a time series to obtain the complete trajectory data of the non-independent individual. The complete trajectory data is then mapped to each sub-region of the region to be evaluated to obtain the spatiotemporal activity data of the non-independent individual.

Images