Individual-based Multiscale Simulation Of COVID-19 Transmission

It has been more than two years since the outbreak of COVID-19 at the end of2019.The epidemic situation around the world still has the trend of local outbreak and pandemic.The close contact of modern human social life and production,the continuous variation of SARS-Co V-2,different social and geographical environments,health levels and prevention and control policies and measures have created the complexity of the transmission process of infectious diseases.The spatio-temporal transmission modeling of infectious diseases is an important method to understand the spatio-temporal characteristics and laws of the spread of COVID-19.It can provide quantitative scientific support for case resumption,epidemic prediction,prevention and control decision-making,etc.In view of the current research on the spread of COVID-19 in medical health geography,most of them focus on the statistical analysis of infectious disease data on the macro scale and the modeling calculation of treating the population as an individual "homogeneous",this paper mainly carries out the research on the spatio-temporal modeling of the spread of COVID-19 from different spatial scales and scenes,as well as the activity differences at the individual level.Based on the theoretical framework of spatio-temporal modeling of "human ground epidemic" infectious diseases,and according to the characteristics and needs of epidemic prevention and control,this paper puts forward the idea of multi-scale modeling of COVID-19 disease transmission based on individuals;At the same time,combined with network model,"flying point" model and agent model,the coupling modeling of human,scene and epidemic situation for multi-scale real scene is studied by using the spatial modeling method of "ground" oriented virtual geographic environment.From the three levels of urban,community and indoor scenes,this paper explores the modeling methods of scene social relationship network,urban geographical functional area,spatial distribution and individual attribute behavior model.Due to the repeated occurrence of the COVID-19,local outbreaks in China and the world have different characteristics in different regions,scales and different time points.This paper focuses on the domestic COVID-19 from the end of 2019 to the end of 2021,and selects the Wenzhou epidemic in Zhejiang caused by imported cases,the Xinfadi outbreak of cold chain transmission in Beijing and the clustered epidemic in Baodi department store in Tianjin as typical cases,build individual COVID-19 disease transmission models with different scale scenarios,promote the research of infectious disease transmission simulation on individual scale,and provide theoretical and methodological support for the analysis of temporal and spatial transmission characteristics of infectious diseases,the study of individual attribute behavior changes and the evaluation of epidemic prevention and control measures.The main achievements of this paper are as follows:(1)This paper analyzes the complexity of epidemic modeling from three aspects:disease characteristics,human activities and environmental characteristics that affect the spatio-temporal transmission of diseases,and defines the basic methods and concepts of spatio-temporal calculation of infectious disease transmission.The individual "human" is introduced into the modeling of COVID-19 as the minimum basic unit,and the modeling method of human land relationship under various spatial environmental scales is discussed.According to the key elements involved in the occurrence of the epidemic,the spatio-temporal transmission modeling of the disease is divided into three parts: spatio-temporal host(human),environment(ground)and event(epidemic).The "human" mainly includes three main types: groups with similar attributes(statistical homogeneous groups),groups classified by attributes(network group groups,spatial group groups)and multi-agent groups with individual differences(individual heterogeneous groups);" Ground " refers to different scales and scenes,such as the world,China,regions,cities,communities,indoor buildings,etc;"Epidemic" includes the biological characteristics of the virus,the transmission rate and the state changes of the infected host;Finally,the three are coupled through time to form a "human ground epidemic" theoretical analysis framework,which provides a theoretical basis for multi-scale spatial-temporal modeling of COVID-19 transmission.(2)For urban scale scenes,a crowd contact network model based on the superposition of multiple social scenes is proposed.According to the different contact types of the crowd,the crowd contact relationship network is divided into family network and work network with stable contact relationship,and temporary network with temporary random relationship.After modeling respectively,it is combined and superimposed to generate daily crowd contact network.On this basis,the SEIR infection probability model considering input and output and various prevention and control measures is established,which is integrated with the crowd contact network model to realize the simulation of epidemic transmission.Taking the epidemic situation in Wenzhou at the beginning of 2020 as a research case,the paper makes a modeling analysis,and analyzes the importance and sensitivity of relevant prevention and control policies to curb the epidemic situation based on the model.The results show that the travel restriction measure plays a vital role in curbing the epidemic situation in Wenzhou,and the sooner or later the measure is implemented is the most sensitive to the number of cases;In addition,rapid contact tracking and disease detection methods are great challenges faced in the early stage of the epidemic.Strengthening the construction of medical facilities can provide guarantee for the efficient disposal of public health emergencies in the future.(3)Facing the community scene,this paper puts forward a more refined modeling method than network simulation-the "flying point" COVID-19 transmission model integrating geospatial structure and individual social activities.Based on the "flying point" model of infectious diseases proposed by Li Xiaowen,this paper introduces geospatial pattern and individual schedule activities,and refines individual spatial activity behavior in disease transmission simulation.Taking the epidemic situation of Xinfadi agricultural products trading District in Beijing in June 2020 as an example,firstly,the geographical space is divided based on the geographical space-time data,and then combined with the regional population statistics and individual travel characteristics,the individual spatial distribution and geographical activity schedule are designed,the "flying point" movement model between different geospatial units is realized,and the scene location of the crowd at different times is simulated;Finally,for the short-term simulation of disease transmission,a susceptibility-exposed-infection quarantine-confirmed(SEIQC)model is proposed,which can easily evaluate the role of tracking isolation and nucleic acid detection measures in prevention and control.The case calculation test results of the model show that the important means to control the spread of large-scale population aggregation epidemic is to determine the source of infection as soon as possible and close contact isolation based on working relationship.(4)For indoor scenes,based on spatial data model and individual behavior model,COVID-19 space-time spread agent model for indoor public places is proposed.Taking the epidemic situation of department stores in Baodi District of Tianjin in early 2020 as an example,taking the indoor space of buildings as the main research area,a spatial model for agent interaction and path planning is constructed from two aspects of geometric space rendering and semantic mapping.Based on the three-dimensional space model,the agent model is constructed,and the avoidance method of the safe distance between individuals is proposed to realize the simulation of the mobile behavior of the agent in the indoor environment.Susceptibility infection(SE)model is used to simulate the spread of COVID-19 in indoor public places.In this paper,the walking direction and distance between agents are added to the calculation of agent contact probability,which makes the calculation of infection rate of different agents more precise,and better reflects the heterogeneity of individuals.Based on this model,the exposure of individuals in crowded areas and different social distances in shopping malls is evaluated.The results show that the counter dense area is an area with high population density,and the crowd guidance in this area should be strengthened or the layout should be re planned.(5)Based on the modeling of disease transmission in three scale scenarios of city,community and indoor,this paper puts forward three multi-scale fusion research ideas of parameter sharing,method sharing and model combination,and further analyzes the application scenarios and scope of the three simulation methods,as well as the advantages and disadvantages of the three modeling methods.Taking the COVID-19 in Yangzhou in July 2021 as an example,the application of parameter sharing method in nucleic acid detection strategy and indoor space layout was studied.The results showed that for outsiders in high-risk areas,requiring nucleic acid detection within 3 days was an appropriate strategy to control the disease,which could keep the number of infected people below 100;At the same time,measures such as increasing ventilation and the distance between seats were effective in controlling the transmission of novel coronavirus in chess and card rooms and other indoor places.