Machine Learning‐Based Prediction of Heatwave‐Related Hospitalizations: A Case Study in Matam, Senegal

This study analyzes the impact of heatwaves on hospital admissions in the Matam region, Senegal, by integrating climate modeling and advanced machine-learning techniques. Given the increasing frequency and severity of heatwaves, particularly in vulnerable regions such as the Sahel, it is critical to understand their health impacts to enhance public health responses. The primary objective of this research is to characterize the temporal dynamics of heatwave-induced hospitalizations and to identify key climatic and socio-demographic factors influencing these admissions. Our methodological approach involved detecting and characterizing heatwaves using climatic indices, such as daily maximum temperatures (TMAX) and maximum heat index (HI), and assessing their impacts on hospital admissions using Ourossogui Regional Hospital data from January 2017 to May 2022. Advanced predictive models, including Random Forest (RF), Extreme Gradient Boosting (XGB), and traditional Generalized Additive Models (GAM), were developed and compared to evaluate their effectiveness in capturing complex interactions between climatic and health variables. To ensure the robustness of our findings, a bootstrapping approach with 1000 iterations was applied, allowing for the estimation of confidence intervals and reducing dependence on a single training sample. The results revealed a significant delayed increase in hospitalizations occurring approximately three to five days after heatwave events, suggesting that physiological deterioration and behavioral factors, such as delayed health-seeking due to environmental constraints and social behaviors, contribute to this latency. Among the predictive models tested, RF achieved the best performance with an R² of [0.51; 0.72], clearly outperforming traditional statistical approaches. Bootstrapping further confirmed the stability of these predictions, reinforcing the reliability of machine learning models in climate-health studies. These findings highlight the critical need for enhancing heatwave monitoring and establishing robust early-warning systems. Integrating climate predictions into public health strategies could significantly improve preparedness and enable health systems to respond more effectively to extreme heat events. Targeted prevention measures, particularly focused on vulnerable groups such as the elderly, children, and outdoor workers, should be prioritized. In future research, incorporating additional environmental factors such as air pollution, wind, and others could further refine predictions. Moreover, expanding the study to include a greater number of healthcare facilities and integrating vulnerability factors such as the availability of healthcare structures, access to potable water, energy access, vegetation cover, hydrology, mobile network coverage, poverty dimension, and other socioeconomic indicators would enhance the analysis. The use of advanced Deep Learning techniques could also improve prediction accuracy by effectively modeling complex and nonlinear interactions. Additionally, it would be relevant to extend the study to other regions particularly vulnerable to extreme heat, to gain a more comprehensive understanding of its health and climatic impacts.