Modelling and Visualization of Spatial Data: A Case Study of COVID-19 Data in Maine, USA

• Published in: 2024 IEEE 3rd Conference on Information Technology and Data Science (CITDS) pp. 1 - 5
• Main Author: Abdi, Ismail H.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 26.08.2024
• Subjects: Autocorrelation; BYM model; conditional autoregressive model; COVID-19; Data models; Data visualization; disease mapping; Diseases; Information technology; INLA; Pandemics; Planning; Prevention and mitigation; relative risk; Spatial autocorrelation; Spatial databases
• DOI: 10.1109/CITDS62610.2024.10791361

The coronavirus pandemic has rapidly spread worldwide and significantly disrupted the daily lives of people globally. This paper presents the results of a detailed spatial analysis of the COVID-19 cases reported between March 2020 and December 2022 in Maine, USA. The aim of this study was to estimate and test the spatial autocorrelation of disease cases among the counties of the state, determine relative risks, and assess excess risk in each county. Using the adjacency matrix, Moran's I statistic was computed, and a significant positive spatial autocorrelation was found. The study employed Bayesian Conditional Autoregressive (CAR) model with Integrated Nested Laplace Approximation (INLA) to map the relative risk of COVID-19 cases, and identified that 56.25% of the counties in the state had a relative risk greater than one. The study further found clustering of disease cases in the state, which may indicate areas that need intervention measures in the future.