Introduction
The deadliest creature on earth isn’t a lion, snake, or shark; it’s the MOSQUITO, which is Responsible for over 600,000 malaria deaths every year, mostly in Africa, this tiny insect outkills every predator combined.
If we could predict malaria outbreaks before they occur, countless lives could be spared, and communities could prepare in time with prevention measures- This is achievable, not only with traditional control methods, but with Artificial Intelligence (AI) predictive models. AI can forecast outbreaks before they occur, offering health systems in Africa the opportunity to act swiftly and prevent avoidable deaths.
Causing an alarming number of deaths annually, Malaria is a life-threatening disease spread to humans by some types of mosquitoes and is mostly found in tropical countries, hence remains a global health burden(1). The World Health Organization’s Global Technical Strategy (GTS) for Malaria calls for ambitious reductions in both case incidence and mortality- at least 40% by 2020, 75% by 2025, and 90% by 2030(2). While promising work on AI Forecasting models is being carried out in multiple countries, this report highlights just a few examples from Kenya, Tanzania, and Nigeria.
AI Forecasting in Kenya
In February 2025 a study ‘Explainable AI (XAI) models for malaria risk prediction in Kenya’, utilized climatic, environmental, and epidemiological datasets to improve prediction accuracy while ensuring transparency in decision-making (3). To complement this, in August 2025 a malaria risk modeling workshop was hosted by Chuka University’s Center for Data Analytics and Modeling (CDAM) Kenya, which emphasized the potential of AI/ML techniques to guide public health interventions more effectively. These initiatives demonstrate Kenya’s leadership in leveraging AI for real-time disease forecasting and capacity building among researchers(4).
Tanzania: Using Artificial Intelligence, Machine Learning, and Advanced Statistical Modeling to Decode Malaria Patterns
A study published in 2024 identified the most influential drivers of malaria transmission using an Extreme Gradient Boosting (XGBoost) model combined with SHapley Additive exPlanations (SHAP) for interpretability. Using machine learning, this study revealed the interactions between climate factors and malaria transmission in southeastern Tanzania. The study (5). Another study in 2025, utilized the Standardized Incidence Ratio (SIR) to evaluate the regional distribution of malaria risk across Tanzania mainland regions and applied Bayesian spatio-temporal models to examine the influence of climatic factors and disease interventions on malaria incidence in two groups: children under five years of age and individuals aged five years (6).
Nigeria: Machine Learning for infection prediction and diagnostic transparency
A Nigerian study published in June 2025 identified the most effective machine learning algorithms for predicting malaria status in children under five (0–59 months). The findings revealed that Random Forest (RF) and Support Vector Machine (SVM) emerged as the most robust models. RF proved highly reliable and well-balanced across performance metrics, while SVM was particularly strong in recall, making it effective for identifying true malaria cases. Simple Logistic Regression (SLR) and Decision Trees (DT) showed promise but require further optimization to improve accuracy. In contrast, Naïve Bayes (NB) and K-Nearest Neighbors (KNN) demonstrated significant performance gaps, making them less suitable for malaria prediction in young children(7).
Another Nigerian study published in April 2025 leveraged patient data from the Federal Polytechnic Ilaro Medical Centre in Ogun State, covering 337 individuals aged 3–77 years over a four-week period (180 females and 157 males). After addressing class imbalance through oversampling, ensemble methods such as Random Forest, AdaBoost, Gradient Boost, XGBoost, and CatBoost were applied, with Explainable AI tools like LIME, SHAP, and Permutation Feature Importance enhancing interpretability and transparency. Together, these studies underscore the potential of advanced machine learning and explainable AI in improving malaria prediction and supporting evidence-based interventions in Nigeria(8).
These studies have shown, without doubt, that with AI, malaria prediction is no longer guesswork – it is a precise, actionable science. These innovations underscore the potential of AI to shift malaria control from reactive measures to proactive, data-driven strategies.
References
1. World Health Organization. Malaria. 2024 Dec; Available from: https://www.who.int/news-room/fact-sheets/detail/malaria
2. World Health Organization. World malaria report 2024. 2024; Available from: file:///C:/Users/ADMIN/Downloads/9789240104440-eng.pdf
3. Dennis Kariuki Muriith, Victor Wandera Lumumba, Olushina Olawale Awe, Daniel Mwangi Muriithi. Explainable Artificial Intelligence Models for Predicting Malaria Risk in Kenya. 2025 Feb; Available from: https://www.ej-ai.org/index.php/ejai/article/view/47
4. Chuka University Kenya. Empowering Malaria Control Through AI. 2025 Aug; Available from: https://www.chuka.ac.ke/empowering-malaria-control-through-ai/
5. Ijakara Scientists. MACHINE LEARNING: Leveraging AI to predict malaria outbreaks. 2024 Dec; Available from: https://www.ihi.or.tz/blog/news/machine-learning-leveraging-ai-to-predict-malaria-outbreaks/?utm
6. Lembris Laanyuni Njotto, Wilfred Senyoni, Ottmar Cronie, Anna-Sofie Stensgaard. Bayesian spatio-temporal modeling and prediction of malaria cases in Tanzania mainland (2016-2023): unveiling associations with climate and intervention factors. 2025; Available from: https://ij-healthgeographics.biomedcentral.com/articles/10.1186/s12942-025-00408-8?utm_
7. Sbongiseni Makhosonke Mthethwa, Sileshi Fanta Melesse. A Machine Learning Approach to the Prediction of Malaria in Under-five Children: Analysis of the 2021 Nigerian Malaria Indicator Survey. 2025 Jun; Available from: https://openpublichealthjournal.com/VOLUME/18/ELOCATOR/e18749445396163/FULLTEXT/?utm_
8. Olushina Olawale Awe, Peter Njoroge Mwangi, Samuel Kotva Goudoungou, Ruth Victoria Esho, Olanrewaju Samuel Oyejide. Explainable AI for enhanced accuracy in malaria diagnosis using ensemble machine learning models. 2025; Available from: https://bmcmedinformdecismak.biomedcentral.com/articles/10.1186/s12911-025-02874-3?utm_
