Machine Learning Techniques for Predicting Malaria: Unpacking Emerging Challenges and Opportunities for Tackling Malaria in Sub-saharan Africa

Malaria resurgence significantly threatens progress made towards malaria elimination in the past years and consequently increases socioeconomic and public health burden, especially in developing countries. This is exacerbated by the lack of intelligent models for predicting, mapping, diagnosing, and detecting malaria to strengthen malaria prevention and control measures. Predicting malaria and understanding risk factors leading to malaria outbreaks can assist policymakers in re-strategizing and re-aligning malaria elimination strategies and optimizing resource allocation by prioritizing malaria-endemic areas. Therefore, this study provides a comprehensive review of machine learning techniques applied to predict malaria using various risk factors. The study revealed that despite the distribution of mosquito nets, indoor spraying of insecticides, community engagement programmes and awareness strategies, socioeconomic factors, climate and environmental conditions significantly contribute towards malaria outbreaks and remain underexploited and poorly understood. Socioeconomic factors such as lower income, living conditions with house type, distance to health facilities, availability, and use of mosquito nets influence malaria outbreaks. Climatic and environmental risk factors including land surface temperature, rainfall, humidity, enhanced vegetation index, normalized difference vegetation index, and normalized difference water index significantly influence malaria incidences. The study further revealed that machine learning models such as support vector machines, decision trees, random forests, Extreme Gradient Boosting, logistic regression, K-Nearest Neighbors, Naïve Bayes, and multilayer perceptron have been greatly used to predict malaria using socioeconomic, climatic and environmental data. Predicting malaria can assist to develop early malaria warning systems, redesign interventions, make informed decision-making and subsequently strengthening malaria prevention and control measures.