Analysis Of COVID-19 Transmission Relationship And Development Of Visualization Platform At City Scale

Since the outbreak of the Corona Virus Disease 2019(COVID-19),it has continued to spread in various countries and regions,posing great challenges to the major public health event,especially with respect to factors such as individuals,geography,and time during the spread of the epidemic.This includes data from various sources and data types,such as patient trajectories,medical resources,and temporal case information.Due to the large amount of epidemic data and low degree of automated extraction,the analysis of epidemic transmission relationships has become a difficult task.This article focuses on the geographic and population transmission relationships in the identification of transmission relationships,respectively constructing multiple source COVID-19 data extraction and analysis models,patient spatiotemporal information knowledge graphs,and epidemic transmission relationship analysis and visualization platforms.The main research results and content are as follows.(1)Multi-source COVID-19 data extraction and analysis.In response to the challenges of low efficiency in key information extraction and poor data correlation in the multi-source data generated during the spread of COVID-19,a method for multi-source COVID-19 data extraction and analysis is proposed.Taking the epidemic transmission data of Yangzhou,Shijiazhuang,and Chongqing as examples,a BERT-Bi LSTM-CRF model based on adversarial perturbation is used to extract complex patient trajectories,and combined with a Cox proportional hazards model to construct a risk index for COVID-19 at the city level.The results show that the accuracy of the BERT-Bi LSTM-CRF model is increased by0.62% after adding an adversarial perturbation algorithm.The Pearson correlation coefficient and Cox proportional hazards model analysis show that the number of people over the age of 59,population mobility,total population,and number of medical institutions have a significant correlation with the spread of the epidemic in cities.Areas with a higher epidemic risk index exhibit characteristics such as higher population mobility and higher population density.(2)Construction of knowledge map of spatiotemporal information of patients.In order to solve the problem of not being able to accurately locate the transmission path of the epidemic,a method for constructing a knowledge graph of spatiotemporal information about patients is proposed.On the basis of parsing the COVID-19 case data,define the ontology of the knowledge map,and adopt a top-down construction method in the process of defining the ontology.Based on the epidemiological investigation data,case data is analyzed,entity recognition and data storage are completed,and the structure of the data layer is completed.The experimental verification is carried out through the analysis framework of the knowledge map dissemination relationship.The results show that the Jaccard similarity analysis,Overlap similarity analysis and correlation analysis are verified through the COVID-19 patient spatiotemporal information knowledge map.This method is more effective and has It must be feasible.(3)WebGIS-based visualization platform for analyzing epidemic transmission relationships.In response to the slow pace of epidemic transmission trend analysis and the low efficiency of emergency resource allocation,a visualization platform for analyzing epidemic transmission relationships has been developed.Using WebGIS and B/S architecture,and leveraging relational databases such as My SQL,graph databases such as Neo4 j,and spatial databases such as SDE,the platform takes a geographic and population-based perspective on the spread of the epidemic.Based on the results of multi-source COVID-19 data extraction and analysis and patient spatiotemporal information knowledge graphs,the platform provides four modules for epidemic data spatial visualization,key warning of high-risk areas for case transmission,case tracing,and epidemic decision-making support.These results can provide decision-making reference for epidemic prevention and control departments.