AI-driven Climate–Health Risk Mapping in Bangladesh: A Proposed Framework Integrating Remote Sensing and Machine Learning for Future Early Warning Systems

Abstract

Climate change is feared to significantly alter the distribution of climate-sensitive diseases in Bangladesh. However, there remains a lack of integrated, data-driven frameworks capable of linking climate hazards with public health risks. This paper proposes a conceptual and methodological framework for AI-driven climate–health risk mapping by integrating remote sensing (RS), Geographic Information Systems (GIS), and machine learning (ML). The proposed framework combines satellite-derived environmental variables, socioeconomic indicators, andhealth data to construct a composite Climate–Health Risk Index (CHRI) based on hazard, exposure, vulnerability, and adaptive capacity. A Light Gradient Boosting Machine (LightGBM) model is proposed for predictive risk estimation. Although the framework is not yet empirically implemented, it provides a scalable and operational roadmap for future research and policy applications in Bangladesh. The proposed system has the potential to support early warning, targeted interventions, and climate-resilient health planning. 

Medivision J. of Med. and Health Sci. Vol 1(1), Jul 2026; p 59-61