A Systematic Review of Machine Learning Approaches for Predicting Heat-Related Skin Diseases under Climate Change

Climate change has intensified the frequency, duration, and severity of extreme heat events, posing significant risks to human health, particularly through the increasing burden of heat-related skin diseases. Elevated temperatures, prolonged ultraviolet radiation exposure, air pollution, and changing environmental conditions have contributed to the rising incidence of conditions such as heat rash, sunburn, skin infections, dermatitis, and skin cancer, with children and older adults being especially vulnerable. Recent advances in machine learning (ML) have enabled the development of predictive models for disease diagnosis and health risk assessment; however, existing studies predominantly focus either on general heat-related health outcomes or on skin disease classification using clinical images, with limited integration of climatic, environmental, and demographic factors for forecasting heat-related skin disease risk under future climate scenarios. The survey presents a comprehensive review of the current literature on climate-driven heat-related skin diseases and the application of machine learning techniques for their prediction. The review examines major climatic and environmental determinants, vulnerable population groups, commonly used datasets, feature selection methods, and predictive algorithms, including Random Forest, Support Vector Machine, Decision Tree, Artificial Neural Networks, Gradient Boosting, and Deep Learning models. It further compares model performance, interpretability, and limitations. The survey highlights emerging research directions for developing robust and interpretable predictive frameworks capable of forecasting future heat-related skin disease risks under climate change.