Bayesian and Machine Learning Methods for Infectious Disease Forecasting in Resource-Limited Settings: A Systematic Review of Predictive Performance, Data Requirements, and Implementation Challenges

This systematic review aims to synthesize evidence on the application of Bayesian and machine learning approaches for infectious disease forecasting in resource-limited settings. The review will examine the types of models used, diseases and geographic contexts studied, predictive performance metrics, data and computational requirements, and factors influencing implementation and scalability. By critically evaluating existing forecasting methods and their operational feasibility in low-resource environments, the review seeks to identify methodological strengths, limitations, and research gaps. The findings will inform the development of effective, data-driven forecasting systems to support epidemic preparedness, surveillance, and public health decision-making in low- and middle-income countries and other resource-constrained settings.