A Spatial Epidemiology Case Study of Coronavirus (COVID-19) Disease and Geospatial Technologies

Abstract

Spatiotemporal pattern analysis provides a new dimension for data interpretation due to new trends in computer vision and big data analysis. The main aim of this study was to explore the recent advances in geospatial technologies to examine the spatiotemporal pattern of COVID-19 at the Public Health Unit (PHU) level in Ontario, Canada. The spatial autocorrelation results showed that the incidence rate (no. of confirmed cases per 100,000 population–IR/100K) was clustered at the PHU level and found a tendency of clustering high values. Some PHUs in Southern Ontario were identified as hot spots, while Northern PHUs were cold spots. The space-time cube showed an overall trend with a 99% confidence level. Considerable spatial variability in incidence intensity at different times suggested that risk factors were unevenly distributed in space and time. The study also created a regression model that explains the correlation between IR/100K values and potential socioeconomic factors.