A method and system for determining a regional suitability rating based on health risks

Abstract

The application provides a health risk-based regional suitability grade determination method and system, which comprises the following steps: collecting climate environment data and population health data of a target region; preprocessing the climate environment data and population health data to obtain an index feature set; constructing a climate risk relationship model based on the index feature set; calculating a suitability grade according to the climate risk relationship model, and correcting the suitability grade according to the population health risk, and outputting the corrected suitability grade. The application takes meteorological comfort as the evaluation background and corrects it in combination with the population health risk, overcomes the defects in the prior art that only meteorological comfort is considered without considering the population health, and thus scientifically and accurately determines the regional suitability grade, and provides precise data support for the development of health-care resources and local society.

Description

Technical Field

[0001] This invention relates to the field of big data analytics, specifically to a method and system for determining regional suitability levels based on health risks. Background Technology

[0002] With increasing health awareness, the impact of climate conditions on human health is receiving growing attention. Whether it's health tourism, healthy city planning, or public health early warning systems, scientific assessments of regional climate suitability are essential. Current technologies have the following main shortcomings:

[0003] First, the correlation with health risks is weak. Existing assessments are mostly based on physical comfort models (such as perceived temperature and comfort index), which only describe the subjective comfort of the human body to the meteorological environment. They lack a direct and quantitative correlation with real population health outcomes (such as the risk of morbidity or mortality from cardiovascular and cerebrovascular diseases), and the assessment results are not convincing enough in terms of health guidance.

[0004] Second, there is a lack of specificity for particular health-sensitive groups. Most methods are general comfort assessments applied to the general population, without addressing specific groups such as the elderly and patients with cardiovascular and cerebrovascular diseases who are sensitive to weather conditions. They fail to quantify the changes in disease risk with weather factors, making it difficult to provide targeted data for the health and wellness of these groups.

[0005] Third, the quantitative methods are subjective and lack precision. The classification of grades relies heavily on experience-based weighting or subjective thresholds, and lacks data-driven factor screening and exposure-response quantification based on health outcomes, resulting in strong subjectivity and low quantitative precision in the classification.

[0006] In summary, existing technologies lack a mechanism to directly and quantitatively link climate environment with the health outcomes of specific populations. This results in a high degree of subjectivity in grading, low quantitative accuracy, and insufficient health guidance significance. Consequently, it is impossible to automatically and quantitatively determine the suitability level of a region from the perspective of reducing population health risks. Summary of the Invention

[0007] The purpose of this invention is to overcome the problems of existing technologies that evaluate regional suitability based solely on physical models such as meteorological comfort, lack direct quantitative correlation with actual population health outcomes, resulting in highly subjective level classification, low quantitative accuracy, insufficient health guidance significance, and inability to automatically and quantitatively determine regional suitability levels from the perspective of reducing health risks.

[0008] To this end, in one aspect, embodiments of the present invention provide a method for determining regional suitability levels based on health risks, the method comprising the following steps:

[0009] Collect climate and environmental data and population health data for the target area;

[0010] Preprocessing of climate and environmental data and population health data yields a set of indicator features;

[0011] A climate risk relationship model is constructed based on the indicator feature set;

[0012] The suitability level is calculated based on a climate risk relationship model, and then corrected according to the health risk of the population, outputting the corrected suitability level. By constructing a climate risk relationship model, using meteorological comfort as the evaluation baseline, and using the exposure-response relationship between meteorological factors and population health outcomes for health risk correction, the climate environment is directly linked to population health risk. This allows for the discovery of meteorological advantages from the perspective of reducing health risk, thereby scientifically and accurately determining the regional suitability level.

[0013] Furthermore, climate and environmental data include: climate and environmental data, and population health data.

[0014] Furthermore, climate and environmental data and population health data are preprocessed to obtain an indicator feature set, including:

[0015] The climate and environmental data are cleaned and converted, and daily meteorological elements and their derived quantities are extracted as indicator feature values ​​of candidate meteorological factors.

[0016] Calculate the baseline level of weather comfort based on the tourism weather comfort index;

[0017] The indicator characteristic values ​​of candidate meteorological factors and the baseline level of meteorological comfort are included in the indicator characteristic set;

[0018] Quality control and daily statistics were performed on the population health data to obtain daily population health data corresponding to the time of the candidate meteorological factors, and these data were incorporated into the indicator feature set.

[0019] Furthermore, a climate risk relationship model is constructed based on the indicator feature set, specifically including:

[0020] Using the extreme gradient boosting algorithm, with candidate meteorological factors in the indicator feature set as independent variables and population health outcome data in the indicator feature set as dependent variables, we screened evaluation indicators that are strongly correlated with human health risks and removed multicollinearity factors to construct a health risk evaluation indicator system.

[0021] A distributed lag nonlinear model was adopted, with the selected evaluation indicators as exposure factors and population health outcome data as dependent variables, to fit the exposure-response relationship, obtain the relative risk and confidence interval of each evaluation indicator, and set the corresponding health risk threshold for each evaluation indicator accordingly.

[0022] Establish a nonlinear correlation mapping relationship between indicator characteristic values ​​and health risk levels;

[0023] Based on the baseline level of meteorological comfort with a set of indicator characteristics, and corrected according to the relative risk level and its confidence interval, a climate risk relationship model is constructed.

[0024] On the other hand, embodiments of the present invention provide a regional suitability level determination system based on health risk, comprising:

[0025] The data acquisition unit is used to collect climate and environmental data and population health data for the target area.

[0026] The preprocessing unit is used to preprocess climate and environmental data and population health data to obtain indicator feature sets;

[0027] A building block, used to construct climate risk relationship models based on indicator feature sets;

[0028] The system identifies a unit used to calculate the suitability level based on a climate risk relationship model, and then corrects the suitability level according to the health risks of the population, outputting the corrected suitability level. Through the coordinated efforts of various units, health risk corrections are made based on the baseline meteorological comfort level, directly linking the climate environment with the health risks of the population, thus achieving a systematic, scientific, and accurate determination of the regional health and wellness suitability level.

[0029] Furthermore, the climate and environmental data collected in the unit include meteorological data and environmental quality data.

[0030] Furthermore, the preprocessing unit includes:

[0031] The extraction module is used to clean and convert climate and environmental data, and extract daily meteorological elements and their derived quantities as indicator feature values ​​of candidate meteorological factors.

[0032] The comfort calculation module is used to calculate the baseline level of weather comfort based on the tourism weather comfort index.

[0033] The health data processing module is used to incorporate the indicator feature values ​​of candidate meteorological factors and the baseline level of meteorological comfort into the indicator feature set;

[0034] The aggregation module is used to perform quality control and daily statistics on population health data, obtain daily population health data corresponding to the time of candidate meteorological factors, and incorporate them into the indicator feature set.

[0035] Furthermore, the building blocks specifically include:

[0036] The first module is used to employ the extreme gradient boosting algorithm, with candidate meteorological factors in the indicator feature set as independent variables and population health outcome data in the indicator feature set as dependent variables, to screen evaluation indicators that are strongly correlated with human health risks, and to remove multicollinear factors to construct a health risk evaluation indicator system.

[0037] The second module is used to use a distributed lag nonlinear model, with the selected evaluation indicators as exposure factors and population health outcome data as dependent variables, to fit the exposure-response relationship, obtain the relative risk and confidence interval of each evaluation indicator, and set the corresponding health risk threshold for each evaluation indicator accordingly.

[0038] The third module is used to establish a non-linear correlation mapping relationship between indicator feature values ​​and health risk levels;

[0039] The fourth module is used to construct a climate risk relationship model by using the meteorological comfort baseline level of the indicator feature set as a benchmark and correcting it according to the relative hazard level and its confidence interval.

[0040] The above technical solution has the following beneficial effects:

[0041] This invention collects and preprocesses climate and environmental data and population health data to obtain an indicator feature set. Based on this feature set, a climate risk relationship model is constructed. Meteorological comfort is used as the evaluation baseline, and the health risk is corrected by utilizing the exposure-response relationship between meteorological factors and population health outcomes. This directly links the climate environment with population health risk, overcoming the shortcomings of existing technologies that only consider meteorological comfort without taking into account population health conditions. Thus, it enables the discovery of meteorological advantages from the perspective of reducing health risks, scientifically and accurately determines the regional suitability level, and provides precise data support for the exploration of health and wellness resources and local social development. Attached Figure Description

[0042] Figure 1 This is a flowchart of a method for determining regional suitability levels based on health risks, provided in an embodiment of the present invention.

[0043] Figure 2 This is a schematic diagram of a regional suitability level determination system based on health risk provided in an embodiment of the present invention. Detailed Implementation

[0044] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0045] In embodiments of the present invention, such as Figure 1 This paper provides a method for determining regional suitability levels based on health risks, which includes the following steps:

[0046] S101: Collect climate and environmental data and population health data for the target area;

[0047] Specifically, climate and environmental data characterize the external physical exposure conditions of the target area, while population health data characterizes health outcomes such as disease incidence or death under these conditions. By collecting data on the external environment and population health outcomes simultaneously, an indispensable joint data source is provided for subsequent exploration of meteorological advantages from the perspective of reducing health risks, avoiding the one-sidedness of a single meteorological perspective.

[0048] S102: Preprocess the climate and environmental data and population health data to obtain an indicator feature set; including:

[0049] The climate and environmental data are cleaned and converted to extract daily meteorological elements and their derivatives as indicator feature values ​​for candidate meteorological factors; the baseline level of meteorological comfort is calculated based on the tourism meteorological comfort index; the indicator feature values ​​of the candidate meteorological factors and the baseline level of meteorological comfort are included in the indicator feature set; the population health data is subjected to quality control and daily statistics to obtain daily population health data corresponding to the time of the candidate meteorological factors, and these data are included in the indicator feature set.

[0050] Specifically, this step follows the raw data input from step S101, performing logical transformations of data cleaning and feature extraction. The raw data often contains differences in dimensions, noise interference, or redundant information, making it impossible for the model to effectively process it directly. The preprocessing process is essential for transforming the complex raw data into a standardized feature set that the model can handle. Its output indicator feature set serves as the direct input to step S103, ensuring the purity of the data dimensions and the accuracy of the feature representation.

[0051] S103: Construct a climate risk relationship model based on the aforementioned indicator feature set. Specifically, the climate risk relationship model is not a simple meteorological statistical model, but a quantitative carrier that directly links climate environment with health risk. It receives the indicator feature set output from step S102 and, through its inherent mapping logic, reveals and solidifies the driving relationship between climate factor fluctuations and changes in population health risk. This allows the present invention to further evaluate the suitability of the climate environment from the underlying mechanisms of health risk, building upon the baseline of meteorological comfort, thus providing solid model support for subsequent level calculations.

[0052] S104: Calculate the suitability level based on the climate risk relationship model, and correct the suitability level according to the health risk of the population, outputting the corrected suitability level. Specifically, this step follows the model output of step S103: First, the baseline meteorological comfort level is used as the benchmark for the suitability level; then, based on the relative hazard and confidence interval of each strongly correlated meteorological factor given by the climate risk relationship model, the benchmark level is corrected for health risk—when the relative hazard of a factor is significantly greater than 1 (the lower limit of the confidence interval is greater than 1), the level is adjusted towards unsuitability; when it is significantly less than 1 (the upper limit of the confidence interval is less than 1), it is adjusted towards suitability; no adjustment is made when the confidence interval crosses 1; after the adjustment amounts of each factor are summed, a suitability level that takes into account both meteorological comfort and population health risk is obtained and output. This correction ensures that the final level no longer rests on subjective comfort but incorporates statistically tested disease risk information, thus more objectively reflecting the health suitability of the climate environment for a specific population.

[0053] As one implementation method, climate and environmental data includes meteorological data, environmental quality data, and geographic information data. Specifically, meteorological data includes time-series parameters reflecting the physical state of the atmosphere, such as temperature (including daily maximum and minimum temperatures), precipitation, wind speed, air pressure, humidity, and sunshine duration. Further parameters characterizing drastic changes in meteorological conditions, such as diurnal temperature range, 24-hour temperature variation, diurnal air pressure range, and 24-hour air pressure variation, can be derived. Environmental quality data covers indicators directly characterizing air pollution exposure levels, such as PM2.5 concentration, PM10 concentration, sulfur dioxide content, and ozone concentration. It should be understood that although the above common data types are listed here, in other embodiments, meteorological data may also include sunshine duration or ultraviolet index, and environmental quality data may also include water quality parameters or noise decibel values, as long as these parameters can objectively characterize external physical exposure conditions. This embodiment breaks through the limitations of relying solely on a single meteorological element by introducing multi-dimensional climate and environmental data. Its necessity lies in the fact that the health risks of the population are not determined by a single temperature, but are the result of the synergistic effect of multiple factors such as sudden temperature changes and air pressure fluctuations. Multi-dimensional data provides a joint data source for subsequent models to capture cross-dimensional synergistic health impacts, thereby improving the accuracy and reliability of suitability level determination.

[0054] As one implementation method, climate and environmental data and population health data are preprocessed to obtain an indicator feature set, including: cleaning and converting the climate and environmental data, extracting daily meteorological elements and their derivatives (such as daily minimum temperature, 24-hour temperature variation, 24-hour pressure variation, daily temperature range, and daily pressure range) as indicator feature values ​​for candidate meteorological factors; calculating the baseline meteorological comfort level by weighting the temperature and humidity index, wind efficiency index, and clothing index according to preset weights (such as 0.6, 0.3, and 0.1) based on the tourism meteorological comfort index; performing quality control and daily statistics on population health data (such as checking for duplicate patient numbers, removing abnormalities, and summarizing the number of morbidities or deaths daily) to obtain daily population health outcome data corresponding to the time of the candidate meteorological factors; and incorporating the above-mentioned candidate meteorological factor indicator feature values, the baseline meteorological comfort level, and the daily population health outcome data into the indicator feature set. The raw multidimensional meteorological data and population health records have only a loose spatiotemporal correlation. Directly inputting them into the model without processing will lead to inconsistencies in dimensions, the curse of dimensionality, and semantic distortion. This step, through standardized feature extraction and quality control, obtains candidate meteorological factor features for screening and a baseline comfort level for adjustment. It also yields a daily health outcome sequence as the regression target (dependent variable), providing a clean and accurate input for subsequent construction of a climate risk relationship model. It should be understood that the specific factors and weights of the comfort index, as well as the statistical caliber of health outcomes (morbidity or mortality), can be adjusted according to the scenario. The specific calculation formula for the local meteorological comfort level is as follows:

[0055] Z = 0.6X THI + 0.3X K +0.1K ICL (1)

[0056] Among them, X THI X K X ICL The temperature and humidity index, wind efficiency index, and clothing index were assigned values ​​respectively, and their classification is shown in Table 1.

[0057] Table 1. Tourism Weather Comfort Levels

[0058] 7≤Z≤9 Very comfortable 1 5≤Z＜7 Comfort 2 3≤Z＜5 More comfortable 3 1≤Z＜3 rather uncomfortable 4 Z＜1 discomfort 5

[0059] The construction of a climate risk relationship model based on an indicator feature set specifically includes: using an extreme gradient boosting algorithm, with candidate meteorological factors in the indicator feature set as independent variables and population health outcome data in the indicator feature set as dependent variables, screening evaluation indicators that are strongly correlated with human health risks, and removing multicollinearity factors to construct a health risk evaluation indicator system.

[0060] A distributed lag nonlinear model was adopted, with the selected evaluation indicators as exposure factors and the health outcome data of the population as dependent variables, to fit the exposure-response relationship, obtain the relative risk and confidence interval of each evaluation indicator, and set the corresponding health risk threshold for each evaluation indicator accordingly.

[0061] Establish a nonlinear correlation mapping relationship between indicator characteristic values ​​and health risk levels;

[0062] Using the baseline meteorological comfort level in the aforementioned indicator feature set as a benchmark, and correcting it based on the relative hazard level and its confidence interval, the climate risk relationship model is constructed. Specifically, evaluation indicators strongly correlated with human health risks are selected from the indicator feature set to construct a health risk evaluation indicator system. This embodiment preferably uses the Extreme Gradient Boosting (XGBoost) algorithm: a gradient boosting tree model is trained using each candidate meteorological factor in the indicator feature set as the independent variable and the daily population health outcome data (number of morbidities or deaths) as the dependent variable. The factors are ranked according to their importance scores (gains), and the ones with higher scores are selected as strongly correlated indicators. For example, for elderly people with cardiovascular and cerebrovascular diseases, daily minimum temperature, 24-hour temperature variation, and 24-hour pressure variation are highly important, while the direct driving effect of precipitation is weak and is not included. Further, a multicollinearity test is performed on the selected indicators to remove highly correlated redundant factors—such as the correlation coefficient between 24-hour temperature variation and 24-hour pressure variation, which is approximately -0.55. Only one of the two is retained, and finally, daily minimum temperature and 24-hour pressure variation are used as the core evaluation indicators. It should be understood that for people with respiratory diseases, PM2.5 and other pollutants can replace temperature as the primary strongly correlated indicator. This step, through XGBoost importance ranking and collinearity removal, constructs a concise, efficient, statistically and pathologically supported evaluation index system. The specific formulas for the above method are as follows:

[0063] (2)

[0064] Then, a distributed lag nonlinear model was used, with the selected evaluation indicators as exposure factors and the health outcome data of the population as the dependent variable, to fit the exposure-response relationship, obtain the relative risk and confidence interval of each evaluation indicator, and set the corresponding health risk threshold for each evaluation indicator accordingly. Specifically, a distributed lag nonlinear model was used to fit the exposure-response relationship of each evaluation indicator to classify risk attributes and set health risk thresholds. Using the selected evaluation indicators as exposure factors and daily population health outcomes as dependent variables, a distributed lag nonlinear model (DLNM) with Poisson regression as the connection function and a maximum lag of 30 days was used, while controlling for time trends and SO2, NO2, and PM2.5 levels. 10 Confounding factors such as confounding conditions were considered. The relative risk of each indicator to the health outcome and its 95% confidence interval were then obtained through fitting the model. The specific formula for the above ensemble model is as follows:

[0065] (3)

[0066] In the formula, g is a family of connection functions; y t Let be the expected number of cases on day t; α be the intercept. It is a hysteresis-response function; f·x t-1 Let f be the exposure-response function; f represents the various basis functions of the independent variable. L is the lag time; L is the maximum lag time, which is 30 days in this application; X is the exposure factor, which is a meteorological factor in this application; g(t) is a spline function, and for short-term acute effects such as emergency room visits and outpatient visits, the degrees of freedom are chosen to be 6 / year; μ k This is a confounding effect; γ k These are the parameters corresponding to the confounding factors. When calculating the confounding factors, each meteorological factor was selected based on the principle of minimizing the Akaike information criterion value, with all degrees of freedom set to 3. K represents the total number of dummy variables in the confounding variables.

[0067] Finally, using the baseline meteorological comfort level as a benchmark, and adjusting it based on the relative hazard level and its confidence interval, a climate risk relationship model is constructed. The system starts with the baseline comfort level and adjusts the level for each core indicator as follows: if the relative hazard level is significantly greater than 1 (lower limit > 1), the level is adjusted one level towards unsuitable; if it is significantly less than 1 (upper limit < 1), the level is adjusted one level towards suitable. No adjustment is made if the confidence interval crosses 1. The adjustments for each indicator are summed to obtain the corrected suitability level. The resulting model is a quantitative calculation model with a clear input interface (receiving meteorological indicator characteristic values) and deterministic output logic (outputting a suitability level that considers both comfort and health risk). The specific calculation method is as follows:

[0068] (4)

[0069] (5)

[0070] J (6)

[0071] (7)

[0072] In the formula, HWMG represents the meteorological suitability level for CVD populations, and L... i T respectively min The correction values ​​corresponding to dp24, RR, L, and H are respectively T min The relative risk, lower limit of the 95% confidence interval, and upper limit of the 95% confidence interval corresponding to the exposure response value of dp24.

[0073] The HWMG grading system is shown in Table 2.

[0074] Table 2 Classification of Meteorological Suitability for Health and Wellness

[0075] Suitability level description Very suitable suitable More suitable Less suitable Inappropriate

[0076] In embodiments of the present invention, such as Figure 2 It also provides a regional suitability level determination system based on health risk, including:

[0077] Data acquisition unit 21 is used to collect climate and environmental data and population health data of the target area;

[0078] Preprocessing unit 22 is used to preprocess the climate environment data and population health data to obtain an indicator feature set;

[0079] Construction unit 23 is used to construct a climate risk relationship model based on the indicator feature set;

[0080] The determination unit 24 is used to calculate the suitability level according to the climate risk relationship model, and to correct the suitability level according to the health risk of the population, and output the corrected suitability level.

[0081] A regional suitability level determination system based on health risk adopts the aforementioned method for determining regional suitability levels based on health risk. Its principle and process are the same as those of the method for determining regional suitability levels based on health risk, and will not be repeated here.

[0082] Example:

[0083] This invention provides a method for determining regional suitability levels based on health risks (taking the climate health and wellness suitability evaluation of people with cardiovascular and cerebrovascular diseases in Guizhou Province as an example), including the following steps: 1) Calculate the baseline level of meteorological comfort based on the tourism meteorological comfort index: The temperature and humidity index, wind efficiency index, and clothing index are weighted and summed at 0.6, 0.3, and 0.1 respectively to obtain the comfort index, and the daily meteorological comfort level is divided into several levels as the baseline level of suitability evaluation;

[0084] 2) The Extreme Gradient Boosting (XGBoost) algorithm is used. Each candidate meteorological factor (daily maximum temperature, daily minimum temperature, daily temperature range, 24-hour temperature variation, daily average air pressure, 24-hour air pressure variation, daily air pressure range, etc.) is used as the independent variable, and the number of daily cardiovascular and cerebrovascular disease morbidity or mortality is used as the dependent variable. Meteorological factors strongly associated with health risks (such as daily minimum temperature, 24-hour temperature variation, and 24-hour air pressure variation) are screened. Then, multicollinearity factors are removed (the correlation coefficient between 24-hour temperature variation and 24-hour air pressure variation is about -0.55, and only one of them is retained). Finally, daily minimum temperature and 24-hour air pressure variation are retained as core correction factors.

[0085] 3) A distributed lag nonlinear model (DLNM) with Poisson regression as the link function and a maximum lag of 30 days was adopted. The daily minimum temperature and 24-hour pressure variation were used as exposure factors, and the number of daily morbidities or deaths was used as dependent variables. The time trend and confounding factors such as SO2, NO2, and PM10 were controlled. The nonlinear exposure-response relationship curves of each factor were fitted to obtain the relative risk and its 95% confidence interval.

[0086] 4) Based on the baseline meteorological comfort level, the relative risk and confidence interval of each correction factor are adjusted accordingly. When the relative risk of a factor is greater than 1 and the lower limit of the confidence interval is greater than 1, it is adjusted one level towards unsuitable; when it is less than 1 and the upper limit of the confidence interval is less than 1, it is adjusted one level towards suitable; otherwise, no adjustment is made. After the adjustment amounts of each factor are added together, a meteorological health and wellness suitability level that takes into account both meteorological comfort and health risks is obtained, and it is divided into five levels: very suitable, suitable, relatively suitable, relatively unsuitable, and unsuitable.

[0087] 5) Calculate the above-mentioned meteorological health and wellness suitability levels based on daily meteorological data within a climate statistics period (e.g., 30 years). Combine the three levels of very suitable, suitable, and relatively suitable into "suitable", and combine the two levels of relatively unsuitable and unsuitable into "unsuitable". Calculate the number of suitable health and wellness days per year in each region, and determine the health and wellness climate suitability level and the distribution of suitable health and wellness areas in each region.

[0088] Furthermore, to verify the effectiveness of the classification, the less suitable level was used as the reference group, and the relative risk of death from cardiovascular and cerebrovascular diseases (SS) (the ratio of the average daily number of deaths in each level to the average daily number of deaths in the reference group) was calculated: SS less than 1 indicates that the level has a health protection effect and is conducive to health care, while SS greater than 1 indicates that the risk is relatively high; the results show that the more suitable the level, the lower the risk of death, which verifies the rationality of the classification.

[0089] The above method uses the tourism meteorological comfort index to characterize the livability baseline, and then uses the daily minimum temperature and 24-hour pressure change health risk, which are screened by XGBoost and quantified by DLNM, to correct the baseline level, thereby outputting a meteorological health and wellness suitability level that takes into account both comfort and health.

[0090] This invention can target the goal of health and wellness by starting with meteorological factors that reduce the risk of morbidity and mortality. It can add health risk corrections based on the baseline of meteorological comfort by superimposing strong correlation factors such as daily minimum temperature and 24-hour pressure variation, identify the threshold range of meteorological factors suitable for health and wellness in various regions, formulate a stratified strategy for different health and wellness climate suitability levels in various regions, determine the health and wellness climate suitability level based on the climate characteristics of each region, and ensure human health and safety.

[0091] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for determining regional suitability levels based on health risk, characterized in that, The method includes the following steps: Collect climate and environmental data and population health data for the target area; The climate and environmental data and population health data are preprocessed to obtain an indicator feature set; A climate risk relationship model is constructed based on the aforementioned indicator feature set; The suitability level is calculated based on the climate risk relationship model, and then the suitability level is corrected according to the health risk of the population, and the corrected suitability level is output.

2. The method for determining regional suitability levels based on health risk according to claim 1, characterized in that, The climate and environmental data include: climate and environmental data, and population health data.

3. The method for determining regional suitability levels based on health risk according to claim 1, characterized in that, The preprocessing of the climate and environmental data and population health data to obtain an indicator feature set includes: The climate and environmental data are cleaned and converted, and daily meteorological elements and their derived quantities are extracted as indicator feature values ​​of candidate meteorological factors. Calculate the baseline level of weather comfort based on the tourism weather comfort index; The index feature values ​​of the candidate meteorological factors and the baseline level of meteorological comfort are included in the index feature set; The health data of the population is subjected to quality control and daily statistics to obtain daily population health data corresponding to the time of the candidate meteorological factors, and these data are included in the indicator feature set.

4. The method for determining regional suitability levels based on health risk according to claim 3, characterized in that, The construction of the climate risk relationship model based on the indicator feature set specifically includes: Using the extreme gradient boosting algorithm, with candidate meteorological factors in the indicator feature set as independent variables and population health outcome data in the indicator feature set as dependent variables, evaluation indicators strongly correlated with human health risks are screened, and multicollinearity factors are removed to construct a health risk evaluation indicator system. A distributed lag nonlinear model was adopted, with the selected evaluation indicators as exposure factors and the health outcome data of the population as dependent variables, to fit the exposure-response relationship, obtain the relative risk and confidence interval of each evaluation indicator, and set the corresponding health risk threshold for each evaluation indicator accordingly. Establish a nonlinear correlation mapping relationship between indicator characteristic values ​​and health risk levels; Based on the baseline level of meteorological comfort in the set of indicator features, and corrected according to the relative hazard and its confidence interval, the climate risk relationship model is constructed.

5. A system for determining regional suitability levels based on health risk, characterized in that, include: The data acquisition unit is used to collect climate and environmental data and population health data for the target area. The preprocessing unit is used to preprocess the climate and environmental data and population health data to obtain an indicator feature set; The construction unit is used to construct a climate risk relationship model based on the indicator feature set; The determination unit is used to calculate the suitability level according to the climate risk relationship model, and to correct the suitability level according to the health risk of the population, and output the corrected suitability level.

6. A regional suitability level determination system based on health risk according to claim 5, characterized in that, The climate and environmental data in the acquisition unit includes meteorological data and environmental quality data.

7. A regional suitability level determination system based on health risk according to claim 5, characterized in that, The preprocessing unit includes: The extraction module is used to clean and convert the climate and environmental data, and extract daily meteorological elements and their derived quantities as indicator feature values ​​of candidate meteorological factors. The comfort calculation module is used to calculate the baseline level of weather comfort based on the tourism weather comfort index. The health data processing module is used to incorporate the indicator feature values ​​of the candidate meteorological factors and the baseline level of meteorological comfort into the indicator feature set; The aggregation module is used to perform quality control and daily statistics on the population health data, obtain daily population health data corresponding to the time of the candidate meteorological factors, and incorporate them into the indicator feature set.

8. A regional suitability level determination system based on health risk according to claim 7, characterized in that, The building unit specifically includes: The first module is used to employ the extreme gradient boosting algorithm, with candidate meteorological factors in the indicator feature set as independent variables and population health outcome data in the indicator feature set as dependent variables, to screen evaluation indicators that are strongly correlated with human health risks, and to remove multicollinearity factors to construct a health risk evaluation indicator system. The second module is used to use a distributed lag nonlinear model to fit the exposure-response relationship with the selected evaluation indicators as exposure factors and the health outcome data of the population as the dependent variable, to obtain the relative risk and confidence interval of each evaluation indicator, and to set the health risk threshold corresponding to each evaluation indicator accordingly. The third module is used to establish a non-linear correlation mapping relationship between indicator feature values ​​and health risk levels; The fourth module is used to construct the climate risk relationship model by using the meteorological comfort baseline level in the indicator feature set as a benchmark and correcting it according to the relative hazard level and its confidence interval.

Images