Respiratory infectious disease condition characteristic and transmission chain comprehensive visual analysis system

Abstract

The invention discloses a respiratory infectious disease condition characteristic and propagation chain comprehensive visual analysis system, which belongs to the technical field of medical data analysis and comprises a virtual reality simulation module, a data integration and analysis module, a GIS and propagation chain tracking module, a visualization and report generation module and a training and evaluation module. Wherein the virtual reality simulation module utilizes virtual reality to create an interactive training scene simulating a real environment and simulate different sites and conditions of illness conditions, and medical staff perform illness condition investigation, patient treatment and implementation of isolation measures in the virtual scene; according to the method, multi-source heterogeneous data such as clinical data, epidemiological survey data, laboratory detection results, environment monitoring data and social media information are integrated, and comprehensive data support is provided for analysis. Not only can a static propagation chain be displayed, but also a dynamic process of disease propagation can be updated and displayed in real time, and a decision maker is helped to quickly identify propagation key nodes and risk areas.

Description

Technical Field

[0001] The present invention relates to the technical field of medical data analysis, and specifically to a comprehensive visualization analysis system for disease characteristics and transmission chains of respiratory infectious diseases. Background Art

[0002] Respiratory infectious diseases are a class of diseases that are transmitted through the respiratory tract and are usually caused by pathogens such as viruses, bacteria, fungi or parasites. There are several visualization analysis systems for respiratory infectious diseases, each with different characteristics and functions.

[0003] However, the existing comprehensive visualization analysis system for disease characteristics and transmission chains has the following limitations in spatiotemporal data analysis: it lacks sufficient resolution in spatial analysis and cannot accurately display the distribution and spread of the disease in geographic space. It is also often unable to effectively capture changes in the disease over time and has difficulty showing the dynamic process of disease transmission.

[0004] Moreover, in terms of early warning capability, the ability to process real-time data streams is poor, resulting in delayed warnings and failure to respond to changes in the disease in a timely manner. To this end, a comprehensive visualization analysis system for the disease characteristics and transmission chain of respiratory infectious diseases is proposed. Summary of the invention

[0005] The present invention aims to solve at least one of the technical problems existing in the prior art. To this end, one object of the present invention is to provide a comprehensive visualization analysis system for the disease characteristics and transmission chain of respiratory infectious diseases.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A comprehensive visualization analysis system for the characteristics and transmission chain of respiratory infectious diseases, including: a virtual reality simulation module, a data integration and analysis module, a GIS and transmission chain tracking module, a visualization and report generation module, and a training and evaluation module;

[0008] The virtual reality simulation module uses virtual reality to create interactive training scenarios that simulate real environments, simulating different locations and situations where illnesses occur. Medical staff conduct disease investigations, treat patients, and implement isolation measures in virtual scenarios.

[0009] The data integration and analysis module collects disease data from various sources, cleans, converts and standardizes the data, and analyzes the data to predict disease trends and identify high-risk areas;

[0010] The GIS and transmission chain tracking module combines case data with geographic location information to analyze the contact network between individuals, track the disease transmission path, explore the changes of the disease over time and space, and reveal the transmission pattern;

[0011] The visualization and report generation module displays the analysis results and generates a comprehensive report;

[0012] The training and evaluation module provides training scenarios for emergency response to illnesses and evaluates the training effectiveness of medical staff.

[0013] As a further optimization scheme of the present invention, the virtual reality simulation module includes a scene construction unit, an interactive logic unit, a simulation control unit, a hardware integration unit, and a medical staff interface unit;

[0014] The data integration and analysis module includes data acquisition unit, data preprocessing unit, data storage and management unit, data analysis unit, data integration unit, data security and privacy protection unit, data visualization unit, and data exchange and sharing unit;

[0015] The GIS and transmission chain tracking module includes a spatial data management unit, a disease map display unit, a case geographic location unit, a transmission chain analysis unit, a geographic spatial analysis unit, a risk area assessment unit, a decision support unit, and an interactive query and reporting unit;

[0016] The visualization and report generation module includes data preprocessing unit, visualization design unit, interactive chart generation unit, map visualization unit, dashboard design unit, report template design unit, report generation and editing unit, and report export and sharing unit.

[0017] As a further optimization scheme of the present invention, in the case geographic positioning unit, the specific process steps of coordinate conversion are:

[0018] First, determine the source coordinate system (WGS 84) and the target coordinate system, obtain the parameters required for coordinate transformation, including three translation parameters, three rotation parameters and one scaling parameter, and read the WGS 84 coordinates from the data source.

[0019] Convert WGS 84 coordinates from geographic to Cartesian coordinates;

[0020] Apply the seven-parameter transformation formula to translate, rotate and scale X, Y and Z;

[0021] The converted Cartesian coordinates are converted back to the geographic coordinates of the target coordinate system, and the converted coordinates are stored back in the database.

[0022] As a further optimization scheme of the present invention, in the interactive logic unit, the interaction between medical staff and virtual objects is determined by a collision detection algorithm, the integrated physics engine simulates physical phenomena through a physics algorithm, and the behavior logic of virtual characters or objects is defined by designing a behavior tree algorithm.

[0023] Among them, Bullet is selected as the physical engine to implement the physical simulation of the virtual reality simulation module. The specific steps are:

[0024] Download the Bullet physics engine library and create btDiscreteDynamicsWorld, which is the main class in Bullet and is used to manage all physical entities and simulations. Set the gravity vector in the physics world through the setGravity method and create btDefaultCollisionConfiguration and btSequentialImpulseConstraintSolver.

[0025] Create btBoxShape and btSphereShape for the object, use the btRigidBody class to create a rigid body, set the mass, friction coefficient, and elastic coefficient properties, and add the rigid body to the physical world through the dynamicsWorld->addRigidBody method;

[0026] In each simulation step, the medical staff's input is processed, the dynamicsWorld->stepSimulation method is called to update the state of objects in the physical world, and the rendering scene is updated according to the results calculated by the physics engine;

[0027] Bullet automatically handles collision detection. Bullet calculates the collision response based on the physical properties of the object. At the end of the simulation, all created Bullet objects are deleted to release memory.

[0028] Compared with the prior art, the present invention has the following beneficial effects:

[0029] The present invention integrates multi-source heterogeneous data such as clinical data, epidemiological survey data, laboratory test results, environmental monitoring data, and social media information, providing comprehensive data support for analysis. It can not only display the static transmission chain, but also update and display the dynamic process of disease transmission in real time, helping decision makers to quickly identify key nodes and risk areas of transmission. In combination with the geographic information system, spatiotemporal data analysis of disease transmission is carried out to intuitively display the disease transmission trend in different regions and time periods, providing a basis for prevention and control strategies. Finally, by establishing a risk assessment model based on big data, it is possible to predict the disease transmission trend and issue an early warning when the threshold is reached, helping relevant departments to take prevention and control measures in a timely manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 It is a module block diagram of the visual analysis system of the present invention;

[0031] Figure 2This is a flow chart of using Moran's I to detect spatial autocorrelation of disease conditions in the present invention;

[0032] Figure 3 This is a flow chart for analyzing propagation trends through the autoregressive integrated moving average model in the present invention. DETAILED DESCRIPTION

[0033] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0034] See also Figure 1-Figure 3 , the present invention provides a technical solution:

[0035] A comprehensive visualization analysis system for the characteristics and transmission chain of respiratory infectious diseases, including: a virtual reality simulation module, a data integration and analysis module, a GIS and transmission chain tracking module, a visualization and report generation module, and a training and evaluation module;

[0036] The virtual reality simulation module uses virtual reality to create interactive training scenarios that simulate real environments, simulating different locations and situations where illnesses occur. Medical staff conduct disease investigations, treat patients, and implement isolation measures in virtual scenarios.

[0037] The data integration and analysis module collects disease data from various sources, cleans, converts and standardizes the data, and analyzes the data to predict disease trends and identify high-risk areas;

[0038] The GIS and transmission chain tracking module combines case data with geographic location information to analyze the contact network between individuals, track the disease transmission path, explore the changes of the disease over time and space, and reveal the transmission pattern;

[0039] The visualization and report generation module displays the analysis results and generates a comprehensive report;

[0040] The training and evaluation module provides training scenarios for emergency response to illnesses and evaluates the training effectiveness of medical staff.

[0041] The virtual reality simulation module includes a scene construction unit, an interactive logic unit, a simulation control unit, a hardware integration unit, and a medical staff interface unit;

[0042] The data integration and analysis module includes data acquisition unit, data preprocessing unit, data storage and management unit, data analysis unit, data integration unit, data security and privacy protection unit, data visualization unit, and data exchange and sharing unit;

[0043] Specifically, collect data from various data sources and clean, transform and standardize the data in preparation for analysis, store the pre-processed data in the database, and create indexes. Perform descriptive, exploratory and predictive analysis on the stored data. Merge and analyze data from different sources and establish connections between data. Exchange data with other systems through standardized interfaces.

[0044] The GIS and transmission chain tracking module includes a spatial data management unit, a disease map display unit, a case geographic location unit, a transmission chain analysis unit, a geographic spatial analysis unit, a risk area assessment unit, a decision support unit, and an interactive query and reporting unit;

[0045] Specifically, in the spatial data management unit, the specific steps of using the R-tree as the spatial index algorithm are:

[0046] Create an empty R-tree structure and determine the maximum and minimum capacity;

[0047] Starting from the root node, select the appropriate leaf node to insert the new spatial object. When the selected leaf node is full, perform node splitting, select the splitting axis with the widest distribution of spatial objects, redistribute the objects to two new nodes, create a bounding box for the new node, and update the bounding box of its parent node to include the new child node bounding box. When the node split causes the parent node to be full, repeat the splitting process until the root node is adjusted;

[0048] An axis is selected so that the projections of all objects on the axis are most widely distributed, and the objects are allocated to two new nodes so that the bounding box area of ​​each new node is reduced while satisfying the minimum capacity requirement of each node.

[0049] Starting from the root node, check whether the bounding box of each node intersects with the query range. When intersecting, recursively check the child nodes. In addition to checking the intersection of the bounding box, the distance to the query point is calculated at the same time, giving priority to nodes with closer distances.

[0050] Starting from the root node, recursively find the leaf node containing the object to be deleted, delete the object from the leaf node, and adjust the bounding box. When the deletion operation causes the node to fall below the minimum capacity, merge the node.

[0051] In the transmission chain analysis unit, a graph theory algorithm is used to analyze the transmission relationship between cases. The specific steps are as follows:

[0052] Each case is defined as a node in the network, and the edges between nodes are defined based on the contact information between cases. When case A has direct contact with case B, an edge is drawn from node A to node B in the network graph. When there is information about the frequency or duration of contact, weights are assigned to the edges to indicate the intensity of contact or the possibility of transmission.

[0053] Use Dijkstra algorithm to find the shortest path between any two nodes in the network. The specific steps are:

[0054] Select the starting node and set its distance to 0, and set the distances of all remaining nodes to infinity;

[0055] Sort all nodes by distance, with the node with the smallest distance at the front;

[0056] Take the node with the smallest distance from the priority queue and update the distance of its adjacent nodes;

[0057] Once a node is fetched and processed, it is marked as processed;

[0058] The algorithm ends when the priority queue is empty or the target node is removed.

[0059] Use community detection algorithms to identify closely connected groups of nodes in the network. The specific steps are:

[0060] Each node initially belongs to an independent community. The modularity of the network is optimized by iteratively moving nodes to other communities, that is, the difference between the density of edges within the community and the density of edges between communities. In each iteration, the communities are merged to form a new community structure, and the iteration continues until the community structure no longer changes significantly.

[0061] Analyze the betweenness centrality or closeness centrality indicators in the network, identify the key nodes in the network, and analyze the transmission pattern between cases based on the community structure and the shortest path.

[0062] The visualization and report generation module includes data preprocessing unit, visualization design unit, interactive chart generation unit, map visualization unit, dashboard design unit, report template design unit, report generation and editing unit, and report export and sharing unit.

[0063] The propagation trend is analyzed by the autoregressive integrated moving average model. The specific steps are as follows:

[0064] First, draw a time series graph to observe the trend of the time series and check whether there is a trend or seasonality. Use the ADF test to determine whether the time series is stationary. If the series is not stationary, perform a difference.

[0065] Determine the order of difference by taking multiple differences of the time series until the series becomes stationary. The order of difference is d. Determine the order of autoregression and use the autocorrelogram to determine the value of p, which is the lag number of the last significant autocorrelation. Determine the order of moving average and use the partial autocorrelogram to determine the value of q, which is the lag number of the last significant partial autocorrelation.

[0066] Use the statsmodels library to estimate the ARIMA model parameters and check whether the model residuals are white noise, that is, whether the residuals are independent and have constant variance. If they are not constant variance, readjust the model parameters.

[0067] The residuals of the model are analyzed, the predictive ability of the model is verified using rolling forecasts, and the fitted ARIMA model is used to make short-term forecasts of the number of cases in the next few days or weeks.

[0068] Analyze the prediction results and understand the transmission trends, and formulate corresponding prevention and control measures based on the prediction results.

[0069] In the risk area assessment unit, Kriging interpolation is used to predict the risk of unsampled areas. The specific steps are as follows:

[0070] First, the semivariogram of data points at different distance intervals is calculated to understand the spatial structure of the data and select an appropriate variation model based on the semivariogram function.

[0071] Determine the location of unsampled areas that require prediction and create spatial weights for each interpolation point and known sample point based on the semivariogram and variogram model.

[0072] For each interpolation point, a Kriging equation system is established and solved to find the optimal weight for each interpolation point. The obtained weights and the values ​​of the known sampling points are used to calculate the predicted values ​​of the unsampled areas.

[0073] Calculate the semivariogram, select different lag distances (distance intervals), calculate the semivariogram value at each distance, and calculate the formula for the semivariogram value:

[0074]

[0075] Among them, γ(h) is the semivariate value at the lag distance h, Z(x i ) and Z(x i +h) is the spatial position x i and x i +h, N(h) is the logarithm of the data at the lag distance h.

[0076] In the case geolocation unit, the specific process steps of coordinate conversion are:

[0077] First, determine the source coordinate system (WGS 84) and the target coordinate system, obtain the parameters required for coordinate transformation, including three translation parameters, three rotation parameters and one scaling parameter, and read the WGS 84 coordinates from the data source.

[0078] Convert WGS 84 coordinates from geographic coordinates (latitude and longitude) to Cartesian coordinates (X, Y, Z);

[0079] Apply the seven-parameter transformation formula to translate, rotate and scale X, Y and Z;

[0080] The converted Cartesian coordinates are converted back to the geographic coordinates of the target coordinate system, and the converted coordinates are stored back in the database.

[0081] In the transmission chain analysis unit, a graph theory algorithm is used to analyze the transmission relationship between cases. The specific steps are as follows:

[0082] Each case is defined as a node in the network, and the edges between nodes are defined based on the contact information between cases. When case A has direct contact with case B, an edge is drawn from node A to node B in the network graph. When there is information about the frequency or duration of contact, weights are assigned to the edges to indicate the intensity of contact or the possibility of transmission.

[0083] In the interactive logic unit, the collision detection algorithm is used to determine the interaction between medical staff and virtual objects. The integrated physics engine simulates physical phenomena through physical algorithms. The behavior tree algorithm is designed to define the behavior logic of virtual characters or objects.

[0084] The specific steps for determining the interaction between medical staff and virtual objects are as follows:

[0085] AABB (Axis-Aligned Bounding Box) collision detection step: For each virtual object, define an AABB bounding box determined by the minimum point (minX, minY, minZ) and the maximum point (maxX, maxY, maxZ). For each simulation step, compare the AABB of the medical staff agent with the AABB of each object in the environment.

[0086] Check whether the following conditions are met at the same time:

[0087] The maxX of the medical staff agent >= the minX of the object

[0088] The minX of the medical staff agent <= maxX of the object

[0089] The maxY of the medical staff agent >= the minY of the object

[0090] The minY of the medical staff agent <= maxY of the object

[0091] The maxZ of the medical staff agent >= the minZ of the object

[0092] The minZ of the medical staff agent <= the maxZ of the object

[0093] When all conditions are met, a collision occurs.

[0094] OBB (oriented bounding box) collision detection steps: For each virtual object, an OBB bounding box is defined and determined by the center point, three mutually perpendicular axes and half-side length. The medical staff agent is defined as an OBB bounding box.

[0095] Calculate the nine possible separation axes of the two OBBs, namely the three axes of each OBB and their cross product. For each separation axis, calculate the projection interval of the two OBBs on the axis. When the two projection intervals do not overlap on any separation axis, there is no collision.

[0096] Sphere collision detection steps: For each virtual object and medical staff agent, define a sphere determined by a center point and a radius, calculate the distance between the two sphere center points, and a collision occurs when the distance between the center points is less than or equal to the sum of the two sphere radii.

[0097] Collision response process: determine the specific point where the collision occurs, determine the normal direction of the collision point, calculate the collision force based on the material properties of the colliding objects, apply the collision force to the objects and change their motion state.

[0098] Select Bullet as the physics engine to implement the physics simulation of the virtual reality simulation module. The specific steps are:

[0099] Download the Bullet physics engine library and create btDiscreteDynamicsWorld, which is the main class in Bullet and is used to manage all physical entities and simulations. Set the gravity vector in the physics world through the setGravity method and create btDefaultCollisionConfiguration and btSequentialImpulseConstraintSolver.

[0100] Create btBoxShape and btSphereShape for the object, use the btRigidBody class to create a rigid body, set the mass, friction coefficient, and elastic coefficient properties, and add the rigid body to the physical world through the dynamicsWorld->addRigidBody method.

[0101] In each simulation step, the medical staff's input is processed, the dynamicsWorld->stepSimulation method is called to update the state of objects in the physical world, and the rendering scene is updated according to the results calculated by the physics engine.

[0102] Bullet automatically handles collision detection. Bullet calculates the collision response based on the physical properties of the object. At the end of the simulation, all created Bullet objects are deleted to release memory.

[0103] In the geospatial analysis unit, Moran's I is used to detect the spatial autocorrelation of the disease. The specific steps are as follows:

[0104] The neighbors of each region are determined based on distance and adjacency, and a spatial weight matrix is ​​created to reflect the spatial relationship between each region and its neighbors. The matrix is ​​symmetric, and the sum of the weights of each region is 1.

[0105] For each region, the disease indicators of the number of cases or incidence were extracted, and the average of the disease indicators for all regions was calculated.

[0106] For each region i, calculate the sum of the products of its attribute value and its neighbor's attribute value, the formula is:

[0107] Among them, w ij is an element in the spatial weight matrix, x i and x j are the attribute values ​​of regions i and j respectively, is the mean of the attribute values.

[0108] Calculate the sum of the squares of the differences between the attribute values ​​of all regions and the mean, the formula is:

[0109] Moran's I is calculated using the following formula:

[0110] Where n is the total number of regions.

[0111] The value of Moran's I is between -1 and 1, with values ​​close to 1 indicating strong positive spatial autocorrelation, values ​​close to -1 indicating strong negative spatial autocorrelation, and values ​​close to 0 indicating no significant spatial autocorrelation.

[0112] Create a Moran scatter plot to show the relationship between the standardized attribute values ​​for each region and its spatial lag. Use a LISA plot to show which regions have significant spatial autocorrelation.

[0113] Parts not involved in the present invention are the same as the prior art or can be implemented by the prior art. Although the embodiments of the present invention have been shown and described, it is understood by those skilled in the art that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention, and the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A comprehensive visualization analysis system for the characteristics and transmission chain of respiratory infectious diseases, characterized in that: include: Virtual reality simulation module, data integration and analysis module, GIS and transmission chain tracking module, visualization and report generation module, training and evaluation module; The virtual reality simulation module uses virtual reality to create interactive training scenarios that simulate real environments, simulating different locations and situations where illnesses occur. Medical staff conduct disease investigations, treat patients, and implement isolation measures in virtual scenarios. The data integration and analysis module collects disease data from various sources, cleans, converts and standardizes the data, and analyzes the data to predict disease trends and identify high-risk areas; The GIS and transmission chain tracking module combines case data with geographic location information to analyze the contact network between individuals, track the disease transmission path, explore the changes of the disease over time and space, and reveal the transmission pattern; The visualization and report generation module displays the analysis results and generates a comprehensive report; The training and evaluation module provides training scenarios for emergency response to illnesses and evaluates the training effectiveness of medical staff.

2. According to claim 1, a comprehensive visualization analysis system for respiratory infectious disease disease characteristics and transmission chain, characterized in that: The virtual reality simulation module includes a scene construction unit, an interactive logic unit, a simulation control unit, a hardware integration unit, and a medical staff interface unit; The data integration and analysis module includes data acquisition unit, data preprocessing unit, data storage and management unit, data analysis unit, data integration unit, data security and privacy protection unit, data visualization unit, and data exchange and sharing unit; The GIS and transmission chain tracking module includes a spatial data management unit, a disease map display unit, a case geographic location unit, a transmission chain analysis unit, a geographic spatial analysis unit, a risk area assessment unit, a decision support unit, and an interactive query and reporting unit; The visualization and report generation module includes data preprocessing unit, visualization design unit, interactive chart generation unit, map visualization unit, dashboard design unit, report template design unit, report generation and editing unit, and report export and sharing unit.

3. According to claim 2, a comprehensive visualization analysis system for respiratory infectious disease disease characteristics and transmission chain, characterized in that: In the case geolocation unit, the specific process steps of coordinate conversion are: First, determine the source coordinate system and the target coordinate system, obtain the parameters required for coordinate transformation, including three translation parameters, three rotation parameters and one scaling parameter, and read the WGS 84 coordinates from the data source. Convert WGS 84 coordinates from geographic to Cartesian coordinates; Apply the seven-parameter transformation formula to translate, rotate and scale X, Y and Z; The converted Cartesian coordinates are converted back to the geographic coordinates of the target coordinate system, and the converted coordinates are stored back in the database.

4. According to claim 2, a comprehensive visualization analysis system for respiratory infectious disease disease characteristics and transmission chain, characterized in that: In the interactive logic unit, the interaction between medical staff and virtual objects is determined by the collision detection algorithm. The integrated physics engine simulates physical phenomena through the physics algorithm. The behavior tree algorithm is designed to define the behavior logic of virtual characters or objects. Among them, Bullet is selected as the physical engine to implement the physical simulation of the virtual reality simulation module. The specific steps are: Download the Bullet physics engine library and create btDiscreteDynamicsWorld, which is the main class in Bullet and is used to manage all physical entities and simulations. Set the gravity vector in the physics world through the setGravity method and create btDefaultCollisionConfiguration and btSequentialImpulseConstraintSolver. Create btBoxShape and btSphereShape for the object, use the btRigidBody class to create a rigid body, set the mass, friction coefficient, and elastic coefficient properties, and add the rigid body to the physical world through the dynamicsWorld->addRigidBody method; In each simulation step, the medical staff's input is processed, the dynamicsWorld->stepSimulation method is called to update the state of objects in the physical world, and the rendering scene is updated according to the results calculated by the physics engine; Bullet automatically handles collision detection. Bullet calculates the collision response based on the physical properties of the object. At the end of the simulation, all created Bullet objects are deleted to release memory.

Images