In the Danger Zone - U-Net Driven Quantile Regression Predicts High-risk SARS-CoV-2 Regions - via Pollutant Particulate Matter and Satellite Imagery

Aug 2, 2026

Since the COVID-19 outbreak, policy makers have relied on non-pharmacological interventions, while air pollution has emerged as a plausible transmission vector requiring inclusion in intervention planning. The work presents a U-net driven quantile regression framework that predicts PM2:5 using readily obtainable satellite imagery. The method reconstructs PM2:5 concentrations against ground-truth data and produces plausible spatially distributed estimates, including locations lacking direct pollution measurements. Such PM2:5 predictions can support public policy strategies aiming to reduce COVID-19 transmission and lethality.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages6

Reading Time15 min

Abstract

Introduction

Transmission risk and vulnerability factors in SARS-CoV-2

Role of particulate matter (PM2:5) in transmission and severity

Limitations of sensor-based pollution measurements

Need for satellite-driven, up-to-date pollution prediction

Goal and paper structure

In the Danger Zone: U-Net Driven Quantile Regression can Predict High-risk SARS-CoV-2 Regions via Pollutant Particulate Matter and Satellite Imagery Jacquelyn A. Shelton 1 Przemyslaw Polewski 2 1 Wei Yao 1 Abstract Since the outbreak of COVID-19 policy makers have been relying upon non-pharmacological interventions to control the outbreak. With air pollution as a potential transmission vector there is need to include it in intervention strategies. We propose a U-net driven quantile regression model to predict PM2:5 air pollution based on easily obtainable satellite imagery. We demonstrate that our approach can reconstruct PM2:5 concentrations on ground-truth data and predict reasonable PM2:5 values with their spatial distribution, even for locations where pollution data is unavailable. Such predictions of PM2:5 characteristics could crucially advise public policy strategies geared to reduce the transmission of and lethality of COVID-19 . 1. Introduction Since the outbreak of the Severe Acute Respiratory Syndrome Corona Virus 2 (SARS-CoV-2), popularly referred to as COVID-19, many questions have been asked around the disease. Of particular interest are the questions of transmission methods, and characteristics that identify vulnerable populations. Studies into these properties are made more complex by the lack of information around the behavior of the virus, E.g. the percent of cases that are asymptomatic. However, as indicated by the inclusion in the CDC's vulnerable Population index, asthma and chronic lung disease seem to be a major factor in the severity of the cases. For example, in (Conticini et al., 2020) the connections between the lethality rate in Lombardy and Emilia Romagna, areas with high level of atmospheric pollution, are explored. In 1Department of Land Surveying and Geoinformatics, The Hong Kong Polytechnic University, Hong Kong SAR, China 2TomTom Location Technology Germany GmbH, Berlin, Germany. Correspondence to: Jacquelyn A. Shelton <jacque[lyn.ann.shelton@gmail.com](lyn.ann.shelton@gmail.com) >. Proceedings of the 37th International Conference on Machine Learning, Vienna, Austria, PMLR 108, 2020 . Copyright 2020 by the author(s) . particular, the paper studies the correlation between pollution, which is a known instigator of chronic lung disease even in young and other wise healthy subjects, and the lethality of SARS-CoV-2 . A similar result for the lethalityin the United States (Wu et al., 2020) using county level fatality rate, and county level long term air pollution, shows that, after adjusting for other known factors, that there is a strong correlation between the concentration of particulate matter 2:5 micrometers or less in diameter, or PM2:5, and the county level lethality. Speciﬁcally, a 1 􀀖mg~~3~~ increase in PM2:5 corresponds to an 8% increase in the fatality rate. However, as observed in (Coccia, 2020), there is also a correlation between particulate matter (PM) air pollution and the number of reported cases. This suggests that pollutionto-human may serve as another transmission dynamic for SARS-CoV-2 . These two results suggest a two factor vulnerability to SARS-CoV-2 caused by increased particulate matter in the air: On one hand it increases the likelihood of having a more sever reaction to infection, and the other it serves a transmission vector. Traditionally, PM2:5 concentration data can be obtained from ground sensors and measurement stations. However, the spatial resolution of these measurements is greatly limited by the sparsity of sensor networks. Thus, detailed pollution maps have been developed that integrate heterogeneous data sources (see e.g. van Donkelaar et al. (2019)), to create a database of estimated monthly pollutant concentrations over several countries. This data was used by Wu et al.(2020) linking PM2:5 concentrations to COVID-19 mortality rates. Although this historical data helped establish this causality relationship, up-to-date/live pollution inf

FAQs

Why should particulate matter (PM2:5) be considered in COVID-19 intervention strategies?

Air pollution can act as a transmission vector and is correlated with both severity and reported case counts. This motivates integrating pollution into COVID-19 planning rather than relying only on non-pharmacological measures.

What model is proposed to predict PM2:5 concentrations?

The document proposes a U-net driven quantile regression model that uses easily obtainable satellite imagery to estimate PM2:5.

How does the approach handle locations without ground pollution sensor data?

It reconstructs PM2:5 concentrations on ground-truth data and predicts reasonable PM2:5 values with spatial distribution even where pollution measurements are unavailable.

Aug 2, 2026

Upload by Olivia Brown