Public health policy assessment method and system fusing environmental factors

Abstract

The invention relates to the technical field of policy assessment methods, in particular to a public health policy assessment method and system fused with environmental factors, and the method comprises the steps: obtaining data related to the environmental factors and public health intervention measures in historical public health data of an assessment object; acquiring data of a reference object having correlation of environmental factors and public health intervention measures in similar historical public health data with the evaluation object; establishing a public health assessment model based on the public health data of the assessment object; establishing a reference model based on the public health data of the reference object; taking the established reference model as a reference of an evaluation object model; environment factors in the target time period are evaluated, and the influence of the environment factors on public health is evaluated according to a public health evaluation model; according to the method, the evaluation result of the influence of the environmental factors on public health is output, a dynamic comparison benchmark is established by introducing a plurality of reference objects, and the reliability and the universality of the evaluation result are greatly improved.

Description

Technical Field

[0001] The present invention relates to the technical field of policy evaluation methods, and more particularly, to a public health policy evaluation method and system integrating environmental factors. Background Art

[0002] With the increasingly severe environmental problems, their impact on public health has become the focus of global attention. Traditional public health policy evaluation methods often ignore the complex impact of environmental factors, resulting in the lack of comprehensiveness and foresight in policy formulation. In recent years, although some studies have begun to focus on the relationship between the environment and public health, there are still many deficiencies in the existing evaluation methods.

[0003] Currently, the closest prior art usually uses a single indicator or a simple linear model to evaluate the impact of environmental factors on public health. Although these methods are simple and intuitive, it is difficult to capture the complex non-linear relationship between the environment and health. For example, some studies only consider the direct relationship between the air quality index and the incidence of respiratory diseases, ignoring the modulating effects of meteorological conditions, population structure and other factors.

[0004] Another type of method attempts to introduce multiple environmental indicators, but often uses a simple weighted average method for comprehensive evaluation, failing to fully reflect the interaction between different environmental factors. This method may underestimate or overestimate the impact of some environmental factors, resulting in policy-making biases.

[0005] In addition, most of the existing evaluation methods are limited to static analysis and lack the ability to predict future trends. In a rapidly changing environment and public health situation, this limitation seriously restricts the foresight and adaptability of policies.

[0006] Another common problem of the existing methods is the lack of an effective reference system. Many evaluations are based only on data from a single region or country, making it difficult to determine whether the observed phenomena are universal and impossible to learn from the successful experiences of other regions.

[0007] Finally, most of the existing methods ignore the key factor of economic cost. In the case of limited resources, evaluation results that only consider health benefits and ignore economic feasibility are often difficult to implement.

[0008] In view of the above problems, there is an urgent need for an evaluation method that can comprehensively consider environmental factors, public health conditions and economic costs to support more scientific and effective public health policy formulation. Summary of the Invention

[0009] The present invention aims to solve the above problems existing in the prior art, and provides a method and system for evaluating public health policies integrating environmental factors. By introducing multiple reference objects, adopting advanced machine learning algorithms, and combining time series prediction and economic cost analysis, this method realizes a comprehensive evaluation of the complex relationship between environmental factors and public health.

[0010] The present invention provides a method for evaluating public health policies integrating environmental factors, including:

[0011] An acquisition step, including:

[0012] Acquire data related to environmental factors and public health intervention measures in the historical public health data of the evaluation object;

[0013] Acquire data of reference objects with the correlation of environmental factors and public health intervention measures in the historical public health data similar to that of the evaluation object;

[0014] A processing step, including:

[0015] Based on the public health data of the evaluation object, establish a public health evaluation model;

[0016] Based on the public health data of the reference object, establish a reference model;

[0017] Use the established reference model as a reference for the evaluation object model;

[0018] Evaluate the environmental factors in the target period, and evaluate their impact on public health according to the public health evaluation model;

[0019] An output step, including:

[0020] Output the evaluation result of the impact of environmental factors on public health.

[0021] Preferably, the environmental factors include air quality and water quality; the public health intervention measures include measures to improve air quality and water quality.

[0022] Preferably, the acquisition step further includes acquiring data related to economic costs in the historical public health data of the evaluation object and the reference object; in the processing step, the public health evaluation model further includes an economic cost module.

[0023] Preferably, the number of the reference objects is greater than or equal to 2.

[0024] Preferably, the method for establishing the reference model specifically includes:

[0025] For each set of public health data serving as a reference object, establish an independent model;

[0026] For a public health impact assessment model that needs to process input values and output values, the input values of the public health data set are input through the input layer of the neural network. In the model structure, the output layer outputs the evaluation results of the public health impact degree, and the learning of the network is realized by establishing the hidden layer of the neural network.

[0027] Preferably, the output value of the output layer of the public health data set is updated according to the established network structure by using the BP algorithm and the stochastic gradient descent method to obtain the weights and thresholds of each hidden layer.

[0028] Preferably, the processing steps further include:

[0029] Determine whether there are new public health intervention measures;

[0030] If there are new public health intervention measures, optimize the evaluation model on the basis of the original evaluation model;

[0031] The system feeds back and adjusts the optimized evaluation model.

[0032] Preferably, the public health evaluation model includes a time series prediction module for environmental factors and / or a public health impact evaluation module for environmental factors.

[0033] Preferably, the processing steps further include:

[0034] Calculate the comprehensive indicators of different evaluation objectives for multiple evaluation models of the evaluation object;

[0035] Rank the evaluation objects according to the comprehensive indicators.

[0036] The public health policy evaluation system integrating environmental factors for implementing the method includes:

[0037] An acquisition module, used for:

[0038] Acquire data related to environmental factors and public health intervention measures in the historical public health data of the evaluation object;

[0039] Acquire data of a reference object with the relevance of environmental factors and public health intervention measures in the historical public health data similar to that of the evaluation object;

[0040] A processing module, used for:

[0041] Based on the public health data of the evaluation object, establish a public health evaluation model;

[0042] Based on the public health data of the reference object, establish a reference model;

[0043] Use the established reference model as a reference for the evaluation object model.

[0044] Evaluate the environmental factors during the target period and evaluate their impact on public health according to the public health assessment model;

[0045] An output module, configured to:

[0046] Output the evaluation result of the impact of environmental factors on public health.

[0047] The beneficial effects of the present invention are mainly reflected in the following aspects:

[0048] First of all, by introducing multiple reference objects, the method establishes a dynamic comparison benchmark, greatly improving the reliability and universality of the evaluation results. This enables decision-makers to better understand the particularities of the local area while drawing on the successful experiences of other regions.

[0049] Secondly, the present method adopts advanced machine learning algorithms, especially neural network models, to effectively capture the non-linear relationship between environmental factors and public health. This method can handle complex multi-variable interactions and provide more accurate evaluation results than traditional linear models.

[0050] Furthermore, the method of the present invention integrates a time series prediction function, which can not only evaluate the current situation but also predict future trends. This forward-looking analysis provides an important basis for formulating long-term public health strategies and helps improve the adaptability and continuous effectiveness of policies.

[0051] In addition, the present method incorporates economic costs into the evaluation system, achieving a balance between health benefits and economic feasibility. This comprehensive consideration makes the evaluation results more practical and helps formulate policies that can both improve public health and conform to economic reality.

[0052] Finally, the method of the present invention has strong adaptability and scalability. Through a dynamic adjustment mechanism and feedback optimization, the method can respond in a timely manner to new public health intervention measures and continuously improve the accuracy and applicability of the model. This flexibility enables the method to cope with rapidly changing environments and public health challenges.

[0053] In summary, the public health policy evaluation method and system integrating environmental factors provided by the present invention provide a powerful decision-making support tool for formulating scientific, effective, and sustainable public health policies through multi-dimensional, dynamic, and forward-looking analysis. This innovative method is expected to significantly improve the accuracy and effectiveness of public health policies, thereby better protecting public health, enhancing the quality of people's lives, and promoting sustainable development. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 It is a flowchart of the method of the present invention.

[0055] Figure 2 This is the logic block diagram of the acquisition module of the present invention.

[0056] Figure 3 This is the logic block diagram of the processing module of the present invention.

[0057] Figure 4 This is the logic block diagram of the output module of the present invention.

[0058] Figure 5 This is the logic block diagram of the feedback adjustment module of the present invention. Detailed implementation manners

[0059] Please refer to Figures 1-5 , the present invention provides a method and system for evaluating public health policies integrating environmental factors. By comprehensively considering the impact of environmental factors on public health, this method provides a powerful decision-making support tool for formulating more scientific and reasonable public health policies.

[0060] Specifically, the method of the present invention includes the following steps:

[0061] First, in the acquisition step, this method acquires the historical public health data of the evaluation object, which are related to environmental factors and public health intervention measures. At the same time, this method also acquires the data of reference objects with similar historical public health data to the evaluation object, and these reference objects are relevant to the evaluation object in terms of environmental factors and public health intervention measures.

[0062] Next, in the processing step, this method establishes a public health evaluation model based on the public health data of the evaluation object. At the same time, this method also establishes a reference model based on the public health data of the reference objects. It should be noted that this method uses the established reference model as a reference for the evaluation object model, which helps to improve the accuracy and reliability of the evaluation. Subsequently, this method evaluates the environmental factors in the target period and evaluates their impact on public health according to the public health evaluation model.

[0063] Finally, in the output step, this method outputs the evaluation results of the impact of environmental factors on public health. These results can provide important reference information for decision-makers and help formulate more effective public health policies.

[0064] Preferably, in an embodiment of the present invention, the environmental factors include air quality and water quality. These two factors have important impacts on public health. For example, air quality indicators may include the concentrations of pollutants such as PM2.5, PM10, ozone, sulfur dioxide, etc. Water quality indicators may include turbidity, dissolved oxygen, pH value, heavy metal content, etc.

[0065] Correspondingly, the public health intervention measures include measures to improve air quality and water quality. For example, for air quality, possible measures may include restricting industrial emissions, promoting the use of clean energy, implementing motor vehicle tailpipe controls, etc. For water quality, possible measures may include strengthening sewage treatment, protecting water sources, controlling agricultural non-point source pollution, etc.

[0066] Furthermore, the method of the present invention also includes obtaining data related to economic costs in the historical public health data of the evaluation object and the reference object. These economic cost data may include medical expenditures, environmental governance investments, productivity losses, etc. In the processing step, the public health assessment model also includes an economic cost module. This enables the method to simultaneously consider the health benefits brought about by environmental improvement and the required economic inputs, thereby helping decision-makers make optimal policy choices under limited resources.

[0067] When establishing the assessment model, this method adopts advanced machine learning techniques. Specifically, for the public health impact assessment model that needs to process input values and output values, the input values of the public health data set are input through the input layer of the neural network, and the output layer in the model structure outputs the assessment results of the public health impact degree, and the learning of the network is realized by establishing the hidden layer of the neural network.

[0068] Taking the impact of air quality on the incidence of respiratory diseases as an example, we can construct the following neural network model:

[0069]

[0070] Among them, y is the incidence of respiratory diseases, x i is each air quality index (such as PM2.5, ozone, etc.), w i is the corresponding weight, b is the bias term, and f is the activation function (such as the sigmoid function).

[0071] During the model training process, this method uses the BP (backpropagation) algorithm and the stochastic gradient descent method to update the parameters and obtain the weights and thresholds of each hidden layer. This method can effectively capture the complex non-linear relationship between environmental factors and public health.

[0072] It should be noted that an important feature of this method is the introduction of a reference object. By comparing and analyzing the situations of other regions or countries with similar historical data, the impact mechanism of environmental factors on public health can be better understood, and the accuracy of the assessment can be improved. For example, when evaluating the effect of a certain air quality improvement measure, the experiences of other cities that have implemented similar measures under similar conditions can be referred to.

[0073] The method of the present invention has many advantages. First, it comprehensively considers environmental factors, public health conditions, and economic costs, providing a comprehensive evaluation framework. Second, by introducing reference objects and using advanced machine learning techniques, the accuracy and reliability of the evaluation are improved. Finally, the method has strong adaptability and scalability, and can continuously optimize and adjust the evaluation model according to actual needs.

[0074] Generally speaking, the public health policy evaluation method integrating environmental factors provided by the present invention provides a powerful tool for decision-makers, helps to formulate more scientific and effective public health policies, thereby better protecting public health and improving the quality of people's lives. In a preferred embodiment of the present invention, the number of the reference objects is greater than or equal to 2. This design helps to improve the reliability and accuracy of the evaluation. By introducing multiple reference objects, the method can more comprehensively capture the complex relationship between environmental factors and public health, and at the same time reduce the deviation that may be brought by a single reference object.

[0075] For example, when evaluating the impact of a certain air quality improvement measure on the incidence of cardiovascular diseases, the method may select 3-5 cities with similar climatic conditions, population structures, and economic development levels as reference objects. These cities may include cities that have implemented similar measures and control group cities that have not implemented the measures. By comparing and analyzing the data of these cities, the health benefits that may be brought by implementing the measure in the target city can be evaluated more accurately.

[0076] Preferably, the method of the present invention adopts a unique method when establishing a reference model. Specifically, for each public health data set used as a reference object, the method establishes an independent model. This approach can make full use of the characteristic information of each reference object and avoid information loss that may be caused by simply merging the data of different reference objects.

[0077] For the public health impact evaluation model that needs to process input values and output values, the method adopts a neural network structure. In this structure, the input values of the public health data set are input through the input layer of the neural network, and the evaluation results of the public health impact degree are output through the output layer in the model structure, and the learning of the network is realized by establishing the hidden layer of the neural network.

[0078] Taking the evaluation of the impact of air quality on the incidence of multiple chronic diseases as an example, a multi-layer neural network model can be constructed as follows:

[0079]

[0080] Among them, h j is the hidden layer neuron; x i is the input layer neuron, representing each air quality index (such as PM2.5 , O 3 , NO 2 , etc.); y k is the output layer neuron, representing the incidence rates of various chronic diseases (such as cardiovascular diseases, respiratory diseases, etc.). are the connection weights from the input layer to the hidden layer and from the hidden layer to the output layer respectively are the corresponding bias terms; f and g are activation functions, and functions such as ReLU and sigmoid can be selected.

[0081] In an embodiment of the present invention, the output value of the output layer of the public health data set is updated with parameters by using the BP algorithm and the stochastic gradient descent method according to the established network structure, and the weights and thresholds of each hidden layer are obtained. This method can effectively handle non - linear relationships and has strong generalization ability.

[0082] Specifically, the core idea of the BP algorithm is to calculate the partial derivatives of the loss function with respect to each parameter, back - propagate the error signal, and thus continuously adjust the network parameters. Assume that the mean squared error is used as the loss function:

[0083]

[0084] where m is the number of samples, y i is the true value, is the model prediction value.

[0085] Then, use the stochastic gradient descent method to update the parameters:

[0086]

[0087] where α is the learning rate, usually taking values between 0.01 and 0.1.

[0088] The method of the present invention also includes a dynamic adjustment mechanism. During the processing step, the method will determine whether there are new public health intervention measures. If there are new intervention measures, the evaluation model will be optimized based on the original evaluation model. This design enables the method to respond to policy changes in a timely manner and maintain the timeliness of the evaluation.

[0089] Preferably, the method of the present invention also includes a system feedback adjustment mechanism. After optimizing the evaluation model, the system will feedback and adjust the optimized evaluation model. This feedback mechanism can continuously improve the accuracy and applicability of the model. For example, the system may compare the model prediction results with the actual observed data regularly (such as monthly or quarterly). If the error exceeds a preset threshold (such as 10%), the retraining process of the model will be triggered.

[0090] In another embodiment of the present invention, the public health assessment model includes a time series prediction module for environmental factors and / or a public health impact assessment module for environmental factors. The time series prediction module can predict the change trend of environmental factors over a period of time in the future, while the public health impact assessment module evaluates the possible health impacts brought about by these changes.

[0091] For example, the time series prediction module may adopt the ARIMA (AutoRegressive Integrated Moving Average) model to predict the future Air Quality Index (AQI):

[0092]

[0093] AQI t =f(X t )

[0094] where X t is a time series, represents the d-th difference, c is a constant term, φ i and θ j are the autoregressive and moving average parameters respectively, ε t is white noise, and f is a function that converts the prediction result into AQI.

[0095] In this way, the method of the present invention can not only evaluate the impact of current environmental factors on public health, but also predict the possible future change trends, providing a scientific basis for formulating long-term public health policies. In another preferred embodiment of the present invention, the processing step further includes calculating a comprehensive index of different evaluation objectives for multiple evaluation models of the evaluation object, and sorting the evaluation objects according to the comprehensive index. This design enables the method to comprehensively consider the evaluation results from multiple dimensions and provide clearer and more intuitive policy recommendations for decision-makers.

[0096] Specifically, the calculation of the comprehensive index may involve multiple evaluation objectives, such as health benefits, economic costs, environmental improvement degree, etc. Each evaluation objective may have different weights, reflecting its relative importance in the decision-making process. For example, the following weighted summation method can be used to calculate the comprehensive index:

[0097]

[0098] where CI is the comprehensive index, w i is the weight of the i-th evaluation objective, satisfying is the normalized index value of the i-th evaluation objective. Where I i is the original index value.

[0099] The normalization process can adopt the min-max scaling method:

[0100]

[0101] Among them, I i is the original index value.

[0102] Preferably, the present method can adopt the Analytic Hierarchy Process (AHP) to determine the weights of each evaluation objective. This method constructs a judgment matrix and calculates the eigenvector through the way of expert scoring, so as to obtain the weights of each evaluation objective.

[0103] According to the calculated comprehensive index, the present method can rank different policy options in order of priority. This ranking can help decision-makers quickly identify the most potential policy options and optimize resource allocation. For example, a descending order can be adopted, and the policy option with the highest comprehensive index will be given the highest priority.

[0104] The present invention also provides a public health policy evaluation system integrating environmental factors corresponding to the above method. The system includes an acquisition module 1, a processing module 2 and an output module 3.

[0105] The acquisition module 1 is used to acquire the data related to environmental factors and public health intervention measures in the historical public health data of the evaluation object, and acquire the data of the reference object related to the correlation between environmental factors and public health intervention measures in the historical public health data similar to that of the evaluation object. The acquisition module 1 may include a data interface sub-module 11 and a data preprocessing sub-module 12. The data interface sub-module 11 is responsible for connecting and communicating with various data sources, such as environmental monitoring stations, medical institutions, statistical departments, etc. The data preprocessing sub-module 12 is responsible for cleaning, standardizing and format converting the original data to ensure data quality and consistency.

[0106] The processing module 2 is the core part of the system and is used to perform complex data analysis and model construction tasks. Specifically, the processing module 2 establishes a public health evaluation model based on the public health data of the evaluation object; establishes a reference model based on the public health data of the reference object; uses the established reference model as a reference for the evaluation object model; evaluates the environmental factors in the target time period of evaluation, and evaluates its impact on public health according to the public health evaluation model.

[0107] The processing module 2 may include multiple sub-modules, such as a model construction sub-module 21, a parameter optimization sub-module 22 and an evaluation execution sub-module 23. The model construction sub-module 21 is responsible for constructing an appropriate evaluation model according to the input data, and may adopt various machine learning algorithms such as neural networks and decision trees. The parameter optimization sub-module 22 is responsible for adjusting the model parameters to improve the model performance, and may use methods such as grid search and Bayesian optimization. The evaluation execution sub-module 23 is responsible for executing the actual evaluation process and calculating the impact of environmental factors on public health.

[0108] The output module 3 is used to output the evaluation results of the impact of environmental factors on public health. The output module 3 may include a result presentation sub-module 31 and a report generation sub-module 32. The result presentation sub-module 31 is responsible for presenting the evaluation results in intuitive forms such as charts and maps, facilitating understanding and interpretation by decision-makers. The report generation sub-module 32 is responsible for generating a detailed evaluation report, including data analysis, model interpretation, policy recommendations, and other contents.

[0109] Preferably, the system may further include a feedback adjustment module 4, which is used to continuously optimize and adjust the evaluation model according to the differences between the actual observed data and the model prediction results. The feedback adjustment module 4 may include a difference analysis sub-module 41 and a model update sub-module 42. The difference analysis sub-module 41 is responsible for calculating and analyzing the deviation between the predicted value and the actual value, while the model update sub-module 42 is responsible for making necessary adjustments and re-training to the model according to the analysis results.

[0110] Through this modular design, the system of the present invention can flexibly meet different evaluation requirements and is easy to maintain and upgrade. Each module can be independently optimized and updated, improving the scalability and adaptability of the system. For example, if new machine learning algorithms or data analysis methods emerge in the future, only the corresponding sub-module needs to be updated without changing the overall structure of the system.

[0111] Generally speaking, the public health policy evaluation method and system integrating environmental factors provided by the present invention, by comprehensively considering environmental factors, public health conditions, and economic costs, and combining advanced data analysis techniques, provide a powerful support tool for formulating scientific and effective public health policies. The application of this method and system is expected to significantly improve the accuracy and effectiveness of public health policies, thereby better protecting public health and enhancing the quality of people's lives.

[0112] It should be noted that the above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the principles of the present invention shall be included within the protection scope of the present invention.

Claims

1. A public health policy evaluation method integrating environmental factors, characterized by: include: The acquisition steps include: Obtain data related to environmental factors and public health intervention measures from the historical public health data of the assessment object; Obtain data on reference subjects with similar historical public health data on environmental factors and public health interventions as the subject of the assessment; Processing steps include: Establish a public health assessment model based on the public health data of the assessment object; Establish a reference model based on the public health data of the reference object; Use the established reference model as a reference for the evaluation object model; Evaluate environmental factors during the target period and assess their impact on public health based on the public health assessment model; Output steps include: Output the assessment results of the impact of environmental factors on public health.

2. The public health policy evaluation method integrating environmental factors according to claim 1 is characterized in that: The environmental factors include air quality and water quality; the public health intervention measures include measures to improve air quality and water quality.

3. The public health policy evaluation method integrating environmental factors according to claim 1 is characterized in that: The acquisition step also includes acquiring data related to economic costs from historical public health data of the evaluation object and the reference object; in the processing step, the public health evaluation model also includes an economic cost module.

4. The public health policy evaluation method integrating environmental factors according to claim 1 is characterized in that: The number of the reference objects is greater than or equal to 2.

5. The public health policy evaluation method integrating environmental factors according to claim 1 is characterized in that: The method for establishing a reference model specifically includes: For each of the public health data sets used as reference objects, establishing an independent model; For the public health impact assessment model that needs to process input values ​​and output values, the public health data set input value is input through the input layer of the neural network, and the output layer in the model structure outputs the assessment result of the public health impact degree, and the network learning is achieved by establishing the hidden layer of the neural network.

6. The public health policy evaluation method integrating environmental factors according to claim 5 is characterized in that: The output layer output value of the public health data set is updated according to the established network structure using the BP algorithm and the stochastic gradient descent method to obtain the weights and thresholds of each hidden layer.

7. The public health policy evaluation method integrating environmental factors according to claim 1 is characterized in that: The processing steps also include: Determine whether new public health interventions are available; If there are new public health intervention measures, the evaluation model should be optimized based on the original evaluation model; The system feedback adjusts the optimized evaluation model.

8. The public health policy evaluation method integrating environmental factors according to claim 1 is characterized in that: The public health assessment model includes a time series prediction module for environmental factors and / or a public health impact assessment module for environmental factors.

9. The public health policy evaluation method integrating environmental factors according to claim 1 is characterized in that: The processing steps also include: Calculate comprehensive indicators of different evaluation objectives for multiple evaluation models of the evaluation object; The evaluation objects are prioritized according to the comprehensive indicators.

10. A public health policy evaluation system integrating environmental factors and implementing the method according to any one of claims 1 to 9, characterized in that: include: Get modules for: Obtain data related to environmental factors and public health intervention measures from the historical public health data of the assessment object; Obtain data on reference subjects with similar historical public health data on environmental factors and public health interventions as the subject of the assessment; Processing modules for: Establish a public health assessment model based on the public health data of the assessment object; Establish a reference model based on the public health data of the reference object; Use the established reference model as a reference for the evaluation object model; Evaluate environmental factors during the target period and assess their impact on public health based on the public health assessment model; Output modules for: Output the assessment results of the impact of environmental factors on public health.

Images