AI Assessment of Health Risk Based on Air Pollution: An Epidemiological Review

Globally, environmental pollution is leading to significant health challenges. From a vast set of pollutants, this discussion focuses on a subset: air pollution. Due to air pollution, many health consequences occur, encompassing severe cardiovascular, respiratory, neurological, and other systemic diseases, contributing to increased morbidity and mortality worldwide. These consequences are assessed with Health Impact Assessment Models (HIA). An approach is made to understand the framework for HIA with advanced methodologies, such as satellite remote sensing, AIML, DL, and Explainable AI. These techniques significantly enhance the resolution and accuracy of pollution monitoring, forecasting, and nonlinear relationships between pollutants and health, enabling more targeted and effective predictions. This paper reviews air pollution effect to human health for different population groups and health risk assessment model framework concepts using AIML techniques. When combined AI techniques are incorporated, it’s referred to as Hybrid AI. Data visualization techniques and Multiple metrics concepts are elaborated to perceive the features and prominent pollutants that most affect the body organs. Suggestions are also addressed for the processes that yield the best metrics, as identified in the selected studies, to inform further research in the environmental epidemiology domain.