Does Air Quality Really Impact COVID-19 Clinical Severity: Coupling NASA Satellite Datasets with Geometric Deep Learning

Abstract

Given that persons with a prior history of respiratory diseases tend to demonstrate more severe illness from COVID-19 and, hence, are at higher risk of serious symptoms, ambient air quality data from NASA's satellite observations might provide a critical insight into which geographical areas may exhibit higher numbers of hospitalizations due to COVID-19, how the expected severity of COVID-19 and associated survival rates may vary across space in the future, and most importantly how given this information, health professionals can distribute vaccines in a more efficient, timely, and fair manner. Despite the utmost urgency of this problem, there yet exists no systematic analysis on linkages among COVID-19 clinical severity, air quality, and other atmospheric conditions, beyond relatively simplistic regression-based models. The goal of this project is to glean a deeper insight into sophisticated spatio-temporal dependencies among air quality, atmospheric conditions, and COVID-19 clinical severity using the machinery of Geometric Deep Learning (GDL), while providing quantitative uncertainty estimates. Our results based on the GDL model on a county level in three US states, California, Pennsylvania and Texas, indicate that AOD attributes to COVID-19 clinical severity in 39, 30, and 132 counties out of 58, 67, and 254 total counties, respectively. In turn, relative humidity is another important factor for understanding dynamics of clinical course and mortality risks due COVID-19, but predictive utility of temperature is noticeably lower. Our findings do not only contribute to understanding of latent factors behind COVID-19 progression but open new perspectives for innovative use of NASA's datasets for biosurveillance and social good.