Addressing Biases in Ambient PM2.5 Exposure and Associated Health Burden Estimates by Filling Satellite AOD Retrieval Gaps over India.

Ambient PM2.5 exposure statistics in countries with limited ground monitors are derived from satellite aerosol optical depth (AOD) products that have spatial gaps. Here, we quantified the biases in PM2.5 exposure and associated health burden in India due to the sampling gaps in AOD retrieved by a Moderate Resolution Imaging Spectroradiometer. We filled the sampling gaps and derived PM2.5 in recent years (2017-2022) over India, which showed fivefold cross-validation R2 of 0.92 and root mean square error (RMSE) of 11.8 µg m-3 on an annual scale against ground-based measurements. If the missing AOD values are not accounted for, the exposure would be overestimated by 19.1%, translating to an overestimation in the mortality burden by 93,986 (95% confidence interval: 78,638-110,597) during these years. With the gap-filled data, we found that the rising ambient PM2.5 trend in India has started showing a sign of stabilization in recent years. However, a reduction in population-weighted exposure balanced out the effect of the increasing population and maintained the mortality burden attributable to ambient PM2.5 for 2022 (991,058:798,220-1,183,896) comparable to the 2017 level (1,014,766:812,186-1,217,346). Therefore, a decline in exposure alone is not sufficient to significantly reduce the health burden attributable to ambient PM2.5 in India.

Application of machine learning approaches to predict the impact of ambient air pollution on outpatient visits for acute respiratory infections.

With a remarkable increase in industrialization among fast-developing countries, air pollution is rising at an alarming rate and has become a public health concern. The study aims to examine the effect of air pollution on patient's hospital visits for respiratory diseases, particularly Acute Respiratory Infections (ARI). Outpatient hospital visits, air pollution and meteorological parameters were collected from March 2018 to October 2021. Eight machine learning algorithms (Random Forest model, K-Nearest Neighbors regression model, Linear regression model, LASSO regression model, Decision Tree Regressor, Support Vector Regression, X.G. Boost and Deep Neural Network with 5-layers) were applied for the analysis of daily air pollutants and outpatient visits for ARI. The evaluation was done by using 5-cross-fold confirmations. The data was randomly divided into test and training data sets at a scale of 1:2, respectively. Results show that among the studied eight machine learning models, the Random Forest model has given the best performance with R2 = 0.606, 0.608 without lag and 1-day lag respectively on ARI patients and R2 = 0.872, 0.871 without lag and 1-day lag respectively on total patients. All eight models did not perform well with the lag effect on the ARI patient dataset but performed better on the total patient dataset. Thus, the study did not find any significant association between ARI patients and ambient air pollution due to the intermittent availability of data during the COVID-19 period. This study gives insight into developing machine learning programs for risk prediction that can be used to predict analytics for several other diseases apart from ARI, such as heart disease and other respiratory diseases.