Method, system, electronic device and storage medium for respiratory infectious disease transmission analysis

Abstract

This application discloses a method, system, electronic device, and storage medium for analyzing the transmission of respiratory infectious diseases. The method includes receiving multi-source heterogeneous infectious disease data, performing spatial alignment and micro-spatial gridding to generate multidimensional tensor data; dynamically extracting spatiotemporal features from the multidimensional tensor data to obtain hotspot grids and coldspot grids; using the hotspot grids and coldspot grids as datasets respectively, and employing a geographically weighted regression model to obtain multiple key driving factors; constructing feature vector matrices for each key driving factor in each grid, and then performing spatial clustering on all feature vector matrices; based on the results of the spatial clustering, outputting cluster labels and key driving factor weights, and mapping them to corresponding dynamic parameters for the transmission of respiratory infectious diseases. This application is used to identify key driving factors and regions for the transmission of respiratory infectious diseases.

Description

Technical Field

[0001] This application belongs to the technical field of public health, specifically relating to a method, system, electronic device, and storage medium for analyzing the transmission of respiratory infectious diseases. Background Technology

[0002] Respiratory infectious diseases, such as influenza, pose a persistent threat to global public health security. Accurately identifying hotspots and key influencing factors in epidemic prevention and control is crucial for effectively allocating medical resources and formulating precise prevention and control strategies. Currently, research on respiratory infectious diseases based on spatial epidemiology mainly relies on traditional GIS mapping, Moran's I spatial autocorrelation analysis, and heat maps based on administrative divisions. These methods overlay geospatial data to show the distribution relationship between high-incidence areas and environmental or demographic factors. However, these methods have certain limitations: (1) Coarse spatial resolution; Existing technologies usually use provinces, cities and counties as the analysis unit, which cannot reflect the fine-grained differences in propagation such as communities or streets, thus leading to the mosaic effect; (2) One-sided identification of driving factors; existing technologies only focus on statistical correlation analysis and lack in-depth explanation of the driving factors in causal mechanisms (such as aerosol propagation physics), making it difficult to translate the analysis results into practical intervention measures; (3) The regional classification is too simple; existing technologies mostly rely on the cumulative incidence rate to classify high and low risks in a single dimension, ignoring the spatiotemporal evolution characteristics of infectious disease transmission (such as the distinction between input and output areas), and lacking accurate positioning of the dynamic role of the region. Summary of the Invention

[0003] To address one or more of the aforementioned problems, the first aspect of this application provides a method for analyzing the spread of respiratory infectious diseases, the second aspect provides a system for analyzing the spread of respiratory infectious diseases, the third aspect provides an electronic device for implementing the first aspect, and the fourth aspect provides a storage medium for implementing the first aspect, for identifying key driving factors and regions of the spread of respiratory infectious diseases.

[0004] The technical solution of this application is as follows.

[0005] In the first aspect, this application provides a method for analyzing the transmission of respiratory infectious diseases, including: It receives multi-source heterogeneous infectious disease data containing confirmed case location data, meteorological data, economic and population data, and population flow data. After spatially aligning the multi-source heterogeneous infectious disease data in the same coordinate system, it performs micro-spatial gridding processing to generate multi-dimensional tensor data. Hot spot grids and cold spot grids are obtained by dynamically extracting spatiotemporal features from multidimensional tensor data; Using hot spot grids and cold spot grids as datasets, meteorological data, economic and population data, and population flow data are input into a geographic weighted regression model to obtain multiple key driving factors. After constructing feature vector matrices for multiple key driving factors in each grid, spatial clustering is performed on all feature vector matrices. Based on the results of spatial clustering, cluster labels and key driving factor weights are output and mapped to the corresponding dynamic parameters of respiratory infectious disease transmission.

[0006] Preferably, when performing micro-scale spatial meshing processing, the following are included: A microgrid of regular hexagons is set based on latitude and longitude rules; For point-structured data in multi-source heterogeneous infectious disease data, a point spatial statistical algorithm within polygons is used to calculate the number of point features contained in each microgrid. For areal structure data in multi-source heterogeneous infectious disease data, the inverse distance weighting method and / or Kriging algorithm are used to resample and assign values ​​to the center points of the corresponding microgrids.

[0007] Preferably, when performing dynamic extraction of spatiotemporal features, a spatial weight matrix is ​​constructed based on the corresponding epidemiological indicators and the spatial adjacency relationship between grids.

[0008] Furthermore, dynamic extraction of spatiotemporal features based on the spatial weight matrix includes: Determine whether the distribution of infectious diseases among multiple grids exhibits spatial autocorrelation based on the global Moran index; Grids where infectious disease clusters are identified based on the local Moran's index; The spatial autocorrelation test of the Moran index was performed using the Monte Carlo random permutation method. Grids with spatial autocorrelation indices below a preset correlation threshold are removed, and then the remaining grids are extracted and classified in the Moran scatter plot to obtain hot grids and cold grids.

[0009] Preferably, before running the geographic weighted regression model, meteorological data, economic and population data, and population flow data are constructed into an environmental variable matrix; When the geographically weighted regression model is running, the corresponding local influence weights are calculated based on the environmental independent variable matrix. After the geographically weighted regression model is run, the local influence weights are output, and then the significance test value is calculated.

[0010] Furthermore, when one or more variables have significance test values ​​greater than a preset significance threshold in multiple grids, and the number of multiple grids accounts for a greater than a preset percentage of the total number of grids, the one or more variables are determined to be key driving factors.

[0011] Preferably, the spatial clustering process is DBSCAN clustering and / or K-Means clustering.

[0012] Secondly, according to this application, a respiratory infectious disease transmission analysis system includes: The preprocessing module is used to receive multi-source heterogeneous infectious disease data containing confirmed case location data, meteorological data, economic and population data, and population flow data. After spatially aligning the multi-source heterogeneous infectious disease data in the same coordinate system, it performs micro-spatial gridding processing to generate multi-dimensional tensor data. The feature module is used to dynamically extract spatiotemporal features from multidimensional tensor data to obtain hot spot grids and cold spot grids; The Driving Factors module is used to take hot spot grids and cold spot grids as datasets, and input meteorological data, economic and population data, and population flow data into a geographic weighted regression model to obtain multiple key driving factors. The clustering module is used to construct feature vector matrices for multiple key driving factors in each grid, and then perform spatial clustering on all feature vector matrices. The output module is used to output cluster labels and key driving factor weights based on the results of spatial clustering and map them to the corresponding dynamic parameters of respiratory infectious disease transmission.

[0013] Thirdly, an electronic device according to this application includes a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program to implement the respiratory infectious disease transmission analysis method of the first aspect.

[0014] Fourthly, one storage medium of this application is a computer-readable storage medium storing computer instructions for causing a computer to implement the respiratory infectious disease transmission analysis method of the first aspect.

[0015] Compared with the prior art, the advantages of this application are as follows: This application breaks through the limitations of traditional administrative divisions by performing spatial alignment and micro-spatial gridding on multi-source heterogeneous infectious disease data under a unified coordinate system. It can be analyzed down to the fine grid level of communities and streets, overcoming the mosaic effect and thus improving the resolution and accuracy of epidemic transmission analysis. This application uses a geographically weighted regression model to calculate the local influence weights of variables such as meteorology, population, and economy on different grids, reflecting the impact of driving factors on transmission. Combined with statistical methods and quantitative analysis of local causality, the analysis results can be directly applied to the field of public health. This application constructs feature vectors based on the weights of identified key driving factors and applies a spatial clustering algorithm to divide geographical grids into categories with different transmission characteristics. It abandons the single static division mode based solely on incidence rates and can provide high-quality, heterogeneous dynamic parameters for downstream infectious disease dynamic prediction models, thereby providing a basis for formulating differentiated and dynamic infectious disease prevention and control strategies. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating a respiratory infectious disease transmission analysis method proposed in this application.

[0017] Figure 2 This is a schematic diagram of the structural framework of a respiratory infectious disease transmission analysis system according to this application.

[0018] Figure 3 This is a schematic diagram of the structural framework of an electronic device according to this application. Detailed Implementation

[0019] Referring to the illustrations, the principles of this application are illustrated by way of example implementation in a suitable operating environment. The following description is based on the illustrative specific embodiments of this application and should not be construed as limiting other specific embodiments not detailed herein, such as corresponding adjustments to the order of steps of the method of this application based on the principles of the technical concept.

[0020] Unless otherwise defined, all technical and scientific terms used in the embodiments of this application have the same meaning as commonly understood by one of ordinary skill in the art. The terminology used in the embodiments of this application is for descriptive purposes only and is not intended to limit the scope of this application.

[0021] Before providing a further detailed description of the embodiments of this application, the nouns and terms used in the embodiments of this application are explained, and the nouns and terms used in the embodiments of this application shall be interpreted as follows: (1) Multi-source heterogeneous infectious disease data: refers to a multi-dimensional data set with different sources and structures but related to the spread of infectious diseases; in this invention, it may include, but is not limited to: location data of confirmed cases with geographic coordinates and timestamps, meteorological data (such as temperature, humidity, wind speed), economic and demographic data (such as population density, GDP per capita, distribution of key points of interest POI), and population flow data (such as OD matrix). (2) Micro-spatial gridding: refers to the technical process of transforming irregular geographic spaces and discrete data points into unified and regular grid units for analysis; this process achieves spatial standardization and alignment of data by superimposing a virtual grid (such as a regular hexagon or rectangle) on the study area and assigning attribute values ​​of various data (points, lines, and surfaces) to each grid unit through spatial statistical algorithms (such as point counting and interpolation); (3) Multidimensional tensor data: refers to a data structure organized under a unified spatiotemporal framework after micro-spatial gridding; its dimensions usually include space (X and Y coordinates of the grid), time (time step) and multiple feature variables (such as number of cases, temperature, population density, etc.), forming a high-dimensional array, which is convenient for complex spatiotemporal data analysis. (4) Geographically Weighted Regression (GWR) Model: This is a local regression analysis method used to explore the spatial nonstationarity (i.e., spatial heterogeneity) of variable relationships. Unlike the traditional global regression model, the GWR model generates a set of independent regression coefficients for each geographic location (or grid), thereby revealing the possible differences in the influence of independent variables on dependent variables at different geographic locations. (5) Key driving factors: These are environmental or socioeconomic variables that have a statistically significant impact on the spread of infectious diseases after analysis using statistical methods such as geographically weighted regression (GWR) models, and whose impact is either widespread or strongly localized in space. (6) Dynamic parameters: These are coefficients used in infectious disease mathematical models (such as SIR and SEIR models) to describe the rate of virus transmission, population susceptibility, infection, and recovery. In this invention, they specifically refer to parameters that are customized for different regions based on spatial clustering results and the weights of key driving factors, and that can reflect their transmission characteristics, such as effective contact rate and external input rate.

[0022] like Figure 1 As shown in the figure, a respiratory infectious disease transmission analysis method in this embodiment includes the following steps.

[0023] S1. Receive multi-source heterogeneous infectious disease data, including confirmed case location data, meteorological data, economic and demographic data, and population movement data. Spatially align the multi-source heterogeneous infectious disease data in the same coordinate system, and then perform micro-spatial gridding processing to generate multi-dimensional tensor data. The purpose of this step is to integrate data from diverse sources and with varying formats into a unified analytical framework.

[0024] In this embodiment, the micro-scale spatial meshing process specifically includes: S11. A microgrid of regular hexagons is set based on latitude and longitude rules; S12. For point-structured data in multi-source heterogeneous infectious disease data, a point-in-polygon spatial statistical algorithm is used to calculate the number of point features contained in each microgrid. S13. For the areal structure data in the multi-source heterogeneous infectious disease data, the inverse distance weighting method (IDW) and / or Kriging algorithm are used to resample and assign values ​​to the center points of the corresponding microgrids.

[0025] In this embodiment, the location data of confirmed cases includes latitude and longitude coordinates and onset timestamps (obtained from various infectious disease data systems). The meteorological data includes daily average temperature, relative humidity, wind speed, and rainfall obtained through meteorological station open APIs. The economic and demographic data includes high-resolution gridded population density data, per capita GDP, and coordinates and kernel density of key POIs (such as medical institutions, transportation hubs, and large shopping malls) obtained through statistical yearbooks and geographic mapping service APIs. The population flow data includes the origin-destination matrix of cross-regional and intra-regional population flow within a specific time window obtained through a transportation big data platform.

[0026] In this embodiment, the coordinate system adopted is the WGS84 geographic coordinate system, and the size of the microgrid is set to a regular hexagon with a side length of 1 kilometer. The point-structured data includes confirmed case location data, key POIs, etc., while the area-structured data includes meteorological data, population density, etc.

[0027] S2. Dynamically extract hot spot grids and cold spot grids by spatiotemporal feature extraction from multidimensional tensor data.

[0028] In this embodiment, when performing dynamic extraction of spatiotemporal features, a spatial weight matrix is ​​constructed based on the corresponding epidemiological indicators and the spatial adjacency relationships between grids; the dynamic extraction of spatiotemporal features based on the spatial weight matrix specifically includes: S21. Determine whether the distribution of infectious diseases among multiple microgrids exhibits spatial autocorrelation based on the global Moran index. S22. Identify micro-grids where infectious disease clusters occur based on the local Moran's index; S23. Spatial autocorrelation test of Moran's I was performed using the Monte Carlo random permutation method; In step S23, the actual incidence rate of each microgrid is randomly shuffled into different microgrids and the Moran index is recalculated to generate the null hypothesis expected value distribution under random distribution. Then, the Moran index calculated in reality is compared with the expected value distribution, and the statistics Z-score and P-value are calculated respectively. S24. Remove grids whose spatial autocorrelation index is lower than the preset correlation threshold, and then extract and classify the other grids in the Moran scatter plot to obtain hot grids and cold grids. In step S24, the corresponding index of spatial autocorrelation is specifically the statistic P-value. When the statistic P-value is lower than the preset correlation threshold, it indicates that the statistics are not obvious. The extraction and classification are specifically based on the quadrant position of the Moran scatter plot. The grid located in the first quadrant of the Moran scatter plot (high observation value surrounded by high observation value) is identified as a hot spot, and the third quadrant of the Moran scatter plot (low observation value surrounded by low observation value) is identified as a cold spot. Isolated points in the Moran scatter plot are removed or marked separately.

[0029] S3. Using hot spot grids and cold spot grids as datasets, meteorological data, economic and population data, and population flow data are input into a geographic weighted regression model to obtain multiple key driving factors.

[0030] In this embodiment, before the geographically weighted regression model is run, meteorological data, economic and population data, and population flow data are constructed into an environmental variable matrix; when the geographically weighted regression model is run, the corresponding local influence weights are calculated based on the environmental variable matrix; after the geographically weighted regression model is run, the local influence weights are output, and then the significance test value is calculated.

[0031] In this embodiment, the geographic weighted regression model specifically uses a Gaussian kernel function and a bi-quare spatial kernel function to construct local influence weights. The obtained local influence weights include, but are not limited to, grid population density influence weights, average temperature influence weights, relative humidity influence weights, medical POI density influence weights, transportation hub POI density influence weights, and cross-regional net population inflow rate influence weights.

[0032] In this embodiment, when the significance test values ​​of one or more variables in multiple grids are all greater than the preset significance threshold, and the number of multiple grids accounts for more than the preset percentage of the total number of grids, the corresponding variables have a wide range of spatial non-stationary driving effects, and one or more variables can be identified as key driving factors. In this embodiment, the threshold value is set to be greater than the preset significance threshold, and the significance test value specifically needs to be significant at a 95% confidence level.

[0033] S4. After constructing the feature vector matrix of each key driving factor in each grid, perform spatial clustering on all feature vector matrices.

[0034] In this embodiment, based on the local influence weights of multiple key driving factors in each grid, along with the grid's economic and demographic data, personnel flow data, etc., are concatenated column-wise to construct a high-dimensional feature vector matrix for each grid. To eliminate the interference caused by multicollinearity between key driving factors that may exist in clustering, principal component analysis (PCA) is used to perform dimensionality reduction on the high-dimensional feature vector matrix. By calculating the covariance matrix and eigenvalue decomposition, multiple principal components with a cumulative variance contribution rate exceeding a set threshold (90%) are extracted, mapping the originally sparse and redundant high-dimensional space to a low-dimensional dense space, forming a corresponding low-dimensional feature vector matrix as the input for spatial clustering.

[0035] In this embodiment, the spatial clustering process is DBSCAN clustering and / or K-Means clustering.

[0036] In this embodiment, if DBSCAN clustering is used, the input is a low-dimensional feature matrix, a neighborhood radius parameter, and a minimum number of contained points. High-density clusters of arbitrary shapes are identified through density reachability calculation, and sparse data points are filtered as noise. If K-Means clustering is used, the input is a low-dimensional feature matrix. The optimal number of categories is determined using the silhouette coefficient and the elbow rule. The cluster centers and Euclidean distance are calculated iteratively to complete the grid division, and a classification label is assigned to each microgrid, such as: high-frequency input type, local dense inner loop type, meteorological condition susceptible type, etc.

[0037] S5. Based on the results of spatial clustering, output cluster labels, key driving factor weights, and map them to the corresponding dynamic parameters of respiratory infectious disease transmission.

[0038] In this embodiment, when outputting the clustering classification labels and local influence weights of key driving factors for all microgrids, a key-value structure parameter mapping dictionary is constructed. The grid ID (or the merged classification region ID) is used as the key, and the relevant data set on the microgrid is used as the value. The dictionary maps the clustering classification labels to dynamic parameters related to the respiratory infectious disease transmission dynamic prediction model through the built-in rule engine.

[0039] Compared to existing technologies, this embodiment breaks through the limitations of traditional administrative divisions by performing spatial alignment and micro-grid processing on multi-source heterogeneous infectious disease data under a unified coordinate system. It enables analysis down to the fine grid level of communities and streets, overcoming the mosaic effect and thus improving the resolution and accuracy of epidemic transmission analysis. Employing a geographically weighted regression model, it calculates the local influence weights of variables such as meteorology, population, and economy on different grids, reflecting the impact of driving factors on transmission. Combining statistical methods with quantitative analysis of local causality allows the analysis results to be directly applied to the public health field. Based on the weights of the identified key driving factors, feature vectors are constructed, and spatial clustering algorithms are applied to divide geographical grids into categories with different transmission characteristics. This abandons the single static classification mode based solely on incidence rates, providing high-quality, heterogeneous dynamic parameters for downstream infectious disease dynamic prediction models, thereby providing a basis for formulating differentiated and dynamic infectious disease prevention and control strategies.

[0040] like Figure 2 As shown, a respiratory infectious disease transmission analysis system includes: The preprocessing module is used to receive multi-source heterogeneous infectious disease data containing confirmed case location data, meteorological data, economic and population data, and population flow data. After spatially aligning the multi-source heterogeneous infectious disease data in the same coordinate system, it performs micro-spatial gridding processing to generate multi-dimensional tensor data. The feature module is used to dynamically extract spatiotemporal features from multidimensional tensor data to obtain hot spot grids and cold spot grids; The Driving Factors module is used to take hot spot grids and cold spot grids as datasets, and input meteorological data, economic and population data, and population flow data into a geographic weighted regression model to obtain multiple key driving factors. The clustering module is used to construct feature vector matrices for multiple key driving factors in each grid, and then perform spatial clustering on all feature vector matrices. The output module is used to output cluster labels and key driving factor weights based on the results of spatial clustering and map them to the corresponding dynamic parameters of respiratory infectious disease transmission.

[0041] The system described above is based on the corresponding respiratory infectious disease transmission analysis method in the foregoing embodiments and has the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0042] like Figure 3 As shown, based on the same inventive concept, corresponding to any of the above embodiments, this application also discloses an electronic device, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the above-mentioned respiratory infectious disease transmission analysis method.

[0043] Specifically, the device includes: a processor 1010, a memory 1020, an input / output interface 1030, a communication interface 1040, and a bus 1050. The processor 1010, memory 1020, input / output interface 1030, and communication interface 1040 are interconnected within the device via the bus 1050.

[0044] The processor 1010 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), GPU (Graphics Processing Unit), or one or more integrated circuits, to implement relevant programs and achieve the technical solutions provided in the embodiments of this specification.

[0045] The memory 1020 can be implemented in the form of ROM (Read-Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 1020 can store the operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 1020 and called by the processor 1010. The input / output interface 1030 is used to connect input / output modules to realize information input and output. Input / output modules can be configured as components in the device (not shown in the figure) or externally connected to the device to provide corresponding functions. Input devices may include keyboards, mice, touch screens, microphones, various sensors, etc., and output devices may include displays, projectors, speakers, vibrators, indicator lights, etc.

[0046] The communication interface 1040 is used to connect the communication module (not shown in the figure) to enable communication between this device and other devices. The communication module can communicate via wired means (such as USB (Universal Serial Bus), network cable, etc.) or wireless means (such as mobile network, WIFI (Wireless Fidelity), Bluetooth, etc.).

[0047] Bus 1050 includes a pathway for transmitting information between various components of the device, such as processor 1010, memory 1020, input / output interface 1030, and communication interface 1040.

[0048] It should be noted that although the above-described device only shows the processor 1010, memory 1020, input / output interface 1030, communication interface 1040, and bus 1050, in specific implementations, the device may also include other components necessary for normal operation. Furthermore, those skilled in the art will understand that the above-described device may only include the components necessary for implementing the embodiments of this specification, and not necessarily all the components shown in the figures.

[0049] The electronic devices described in the above embodiments are used to implement the corresponding respiratory infectious disease transmission analysis methods in any of the foregoing embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0050] Based on the same inventive concept, corresponding to any of the above embodiments, this application also discloses a computer-readable storage medium that stores computer instructions for enabling a computer to implement the above-described respiratory infectious disease transmission analysis method.

[0051] The computer-readable storage medium of this embodiment includes permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transfer medium, which can be used to store information accessible by a computing device. The computer instructions stored in the storage medium of the above embodiments are used to cause the computer to implement the respiratory infectious disease transmission analysis method as described in any of the above embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0052] The above description is merely a preferred embodiment and the technical principles employed in this application. This application is not limited to the specific embodiments or combinations thereof, and various obvious changes, readjustments, and substitutions that can be made by those skilled in the art will not depart from the scope of protection of this application. Therefore, although this application has been described in detail through the above embodiments, this application is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of this application, the scope of which is determined by the scope of the claims.

Claims

1. A method for analyzing the transmission of respiratory infectious diseases, characterized in that, include: Receive multi-source heterogeneous infectious disease data containing confirmed case location data, meteorological data, economic and population data, and population flow data; after spatially aligning the multi-source heterogeneous infectious disease data in the same coordinate system, perform micro-spatial gridding processing to generate multi-dimensional tensor data. The spatiotemporal features of the multidimensional tensor data are dynamically extracted to obtain hot spot grids and cold spot grids; Using the hot spot grid and the cold spot grid as datasets respectively, the meteorological data, economic and population data, and population flow data are input into a geographic weighted regression model to obtain multiple key driving factors; After constructing the multiple key driving factors as feature vector matrices in each grid, spatial clustering is performed on all the feature vector matrices. Based on the results of the spatial clustering process, cluster labels and key driving factor weights are output and mapped to the corresponding dynamic parameters of respiratory infectious disease transmission.

2. The method for analyzing the transmission of respiratory infectious diseases according to claim 1, characterized in that, The microscopic spatial meshing process includes: A microgrid of regular hexagons is set based on latitude and longitude rules; For point-structured data in multi-source heterogeneous infectious disease data, a point spatial statistical algorithm within polygons is used to calculate the number of point features contained in each microgrid. For areal structure data in multi-source heterogeneous infectious disease data, the inverse distance weighting method and / or Kriging algorithm are used to resample and assign values ​​to the center points of the corresponding microgrids.

3. The method for analyzing the transmission of respiratory infectious diseases according to claim 1, characterized in that, When performing dynamic extraction of spatiotemporal features, a spatial weight matrix is ​​constructed based on the corresponding epidemiological indicators and the spatial adjacency relationship between grids.

4. The method for analyzing the transmission of respiratory infectious diseases according to claim 3, characterized in that, The dynamic extraction of spatiotemporal features based on the spatial weight matrix includes: Determine whether the distribution of infectious diseases among multiple grids exhibits spatial autocorrelation based on the global Moran index; Grids where infectious disease clusters are identified based on the local Moran's index; The spatial autocorrelation test of the Moran index was performed using the Monte Carlo random permutation method. Grids with spatial autocorrelation indices below a preset correlation threshold are removed, and then the remaining grids are extracted and classified in the Moran scatter plot to obtain hot grids and cold grids.

5. The method for analyzing the transmission of respiratory infectious diseases according to claim 1, characterized in that, Before running the geographic weighted regression model, meteorological data, economic and population data, and population flow data are constructed into an environmental variable matrix. When the geographically weighted regression model is running, the corresponding local influence weights are calculated based on the aforementioned environmental independent variable matrix. After the geographically weighted regression model is completed, the local influence weights are output, and then the significance test value is calculated.

6. The method for analyzing the transmission of respiratory infectious diseases according to claim 5, characterized in that, When one or more variables have significance test values ​​greater than a preset significance threshold in multiple grids, and the number of these multiple grids accounts for a greater than a preset percentage of the total number of grids, the one or more variables are determined to be key driving factors.

7. The method for analyzing the transmission of respiratory infectious diseases according to claim 1, characterized in that, Spatial clustering is performed using DBSCAN clustering and / or K-Means clustering.

8. A respiratory infectious disease transmission analysis system, characterized in that, include: The preprocessing module is used to receive multi-source heterogeneous infectious disease data containing confirmed case location data, meteorological data, economic and population data, and population flow data. After spatially aligning the multi-source heterogeneous infectious disease data in the same coordinate system, it performs micro-spatial gridding processing to generate multi-dimensional tensor data. The feature module is used to dynamically extract spatiotemporal features from the multidimensional tensor data to obtain hot spot grids and cold spot grids; The driving factor module is used to take the hot spot grid and the cold spot grid as datasets respectively, and input the meteorological data, economic and population data and population flow data therein into a geographic weighted regression model to obtain multiple key driving factors; The clustering module is used to construct feature vector matrices for the multiple key driving factors in each grid, and then perform spatial clustering on all the feature vector matrices. The output module is used to output cluster labels, key driving factor weights, and map them to the corresponding dynamic parameters of respiratory infectious disease transmission based on the results of the spatial clustering process.

9. An electronic device, characterized in that, The method includes a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program to implement the respiratory infectious disease transmission analysis method according to any one of claims 1-7.

10. A storage medium, which is a computer-readable storage medium, characterized in that, The device stores computer instructions for causing the computer to implement the respiratory infectious disease transmission analysis method according to any one of claims 1-7.

Images