Statistical Analysis of Spatial Network Characteristics in Relation to COVID-19 Transmission Risks in US Counties
Statistical Analysis of Spatial Network Characteristics in Relation to COVID-19 Transmission Risks in US Counties
复制标题
美国各县与 COVID-19 传播风险相关的空间网络特征统计分析
DOI：
10.1109/embc46164.2021.9629892
复制
发表时间：
2021
期刊：
影响因子：
--
通讯作者：
Yang, Hui
中科院分区：
文献类型：
--
作者：
Zhang, Siqi;Yang, Sihan;Yang, Hui
自2020年1月新冠肺炎大流行以来，世界经历了巨大的社会经济变化。为了控制病毒的传播，已经进行了几项研究，以研究与新冠肺炎传播风险相关的促成因素。然而，很少有人研究人类在空间网络上的活动与病毒传播和传播之间的关系。本文通过统计分析检验了空间网络特征与美国各县新冠肺炎累积案例之间的关系。具体地说，既考虑了县级交通概况(例如，通勤工人总数、货运铁路路线里程)，也考虑了美国各县的公路网特征。然后，利用套索回归模型识别出对新冠肺炎案例响应变量敏感的显著变量的稀疏集合。最后，建立固定效应模型，以捕捉选定的预测值集合与响应变量之间的关系。这项工作有助于从空间网络特征和交通概况中识别和确定显著特征，从而提高对新冠肺炎传播动态的理解。这些重要变量也可用于开发模拟模型，用于预测病毒传播的实时位置和优化干预策略。
Since the pandemic of COVID-19 began in January 2020, the world has witnessed drastic social-economic changes. To harness the virus spread, several studies have been done to study contributing factors that are pertinent to COVID-19 transmission risks. However, little has been done to investigate how human activities on the spatial network are correlated to the virus transmission and spread. This paper performs a statistical analysis to examine interrelationships between spatial network characteristics and cumulative cases of COVID-19 in US counties. Specifically, both county-level transportation profiles (e.g., the total number of commute workers, route miles of freight railroad) and road network characteristics of US counties are considered. Then, the lasso regression model is utilized to identify a sparse set of significant variables that are sensitive to the response variable of COVID-19 cases. Finally, the fixed-effect model is built to capture the relationship between the selected set of predictors and the response variable. This work helps identify and determine salient features from spatial network characteristics and transportation profiles, thereby improving the understanding of COVID-19 spread dynamics. These significant variables can also be utilized to develop simulation models for the prediction of real-time positions of virus spread and the optimization of intervention strategies.
影响因子：
4.6
作者：
Chen Y;Yang H
通讯作者：
Yang H
影响因子：
4.8
作者：
Liu, Gang;Yang, Hui
通讯作者：
Yang, Hui