A study on geospatially assessing the impact of COVID-19 in Maharashtra, India

Abstract

The emergence of 2019 novel corona virus disease (COVID-19) raised global health concerns throughout the world. It has become a major challenge for healthcare personnel and researchers throughout the world to efficiently track and prevent the transmission of this virus. In this paper, the role of geographic information system (GIS) based spatial models for tracking the spread of COVID-19 and discovery of testing centres in Maharashtra, India was studied. The datasets collected from diverse sources were geocoded to make it geospatially compatible. A three-tiered framework was proposed to practically realize the impact of COVID-19 in a cartographic fashion. Initially, choropleth maps labeled with testing centres, number of confirmed cases and casualties were visualized in a district-wise manner. Heatmaps for visualizing the spatial density of confirmed cases and casualties were presented. The visualization of spatial K-means clustering for optimal value of “k” estimated using the heuristic-based Elbow method was provided along with zonal analysis of the districts. Map showing spatial autocorrelation was also presented to identify spatial hotspots and coldspots. The districts of Pune and Thane reported respective $z$-scores of 3.424 and 3.347 along with $p$-values of 0.006 and 0.001 respectively. It was inferred from the generated results that Pune and Thane districts in Maharashtra were identified as COVID-19 hotspots. Based upon this analysis, certain effective mitigation strategies can be devised in order to check the uncontrolled spread of COVID-19 in the identified hotspot areas.