Hybrid Explainable AI-Based Predictive Modeling Approach for Spatial–Temporal Analysis of Malaria Risk in Kenya for Public Health Decision Support
Keywords:
Explainable Artificial Intelligence, Malaria Prediction, Spatial–Temporal Modeling, Machine LearningAbstract
Malaria remains one of the most persistent vector-borne diseases affecting sub-Saharan Africa, with Kenya experiencing significant spatial and temporal variability in disease burden. Accurate prediction of malaria risk is critical for timely intervention and resource allocation. This study proposes a hybrid explainable artificial intelligence (XAI)-based predictive modeling framework for spatial–temporal malaria risk analysis in Kenya. The framework integrates machine learning models, including gradient boosting techniques, with explainability mechanisms such as SHAP-based interpretation to enhance transparency and trust in predictive outcomes.
The proposed system leverages environmental, climatic, demographic, and epidemiological variables to construct a robust predictive architecture capable of identifying high-risk malaria zones. Existing studies confirm the effectiveness of machine learning in malaria prediction tasks (Adamu, 2021; Muriithi et al., 2024), while global health evidence highlights the continued burden of malaria and the need for advanced predictive systems (CDC, 2023; WHO, 2023).
The study also incorporates imbalance-aware learning strategies to handle skewed malaria datasets and integrates spatial–temporal modeling to capture seasonal transmission patterns. Experimental synthesis from literature indicates that hybrid AI models outperform traditional statistical approaches in predictive accuracy and interpretability (Chen & Guestrin, 2016; Adib et al., 2022). Furthermore, explainability techniques improve model transparency, which is essential for public health decision-making.
The proposed framework contributes to advancing AI-driven epidemiological modeling by combining predictive performance with interpretability, enabling health authorities in Kenya to deploy proactive malaria control strategies. The findings suggest that hybrid XAI systems can significantly improve early warning systems for malaria outbreaks, particularly in resource-constrained settings.
Downloads
References
1. Adamu YA. Malaria prediction model using machine learn-ing algorithms.Turk J Comput Math Educ (TURCOMAT).2021;12(10):7488–96.
2. Adib A, Yu X, Basu S. Machine learning in malaria prediction: asystematic review.J Artif Intell Healthc. 2022;12(4):345–59.
3. Brain Informatics. Interpreting artificial intelligence models:a systematic review on the application of LIME and SHAPin Alzheimer’s disease detection. 2024;11(1):1–13.
4. Centers for Disease Control and Prevention (CDC). Malaria. 2023.Available from:www.cdc.gov/malaria.
5. Chen T, Guestrin C. XGBoost: a scalable tree boosting system.Proceedings of the 22nd ACM SIGKDD International Conferenceon Knowledge Discovery and Data Mining, pp. 785–94, 2016.
6. Chinweuba AI, Eze PC, Okeke P. Fatigue as a symptom ofmalaria: clinical implications.Afr Health Sci. 2022;22(4):987–95.
7. Danioko S, Tapo AA, Tembine H, Traore A. Machine intelli-gence in Africa: a survey. 2024.
8. Gething PW, Casey DC, Weiss DJ, Bisanzio D, Bhatt S, CameronE,et al.MappingPlasmodium Falciparummortality in Africabetween 1990 and 2015.New Engl J Med. 2020;374(24):2432–44.
9. Hamonis JA, Kusuma SAF, Rukayadi Y, Hasanah AN. Exploringbiomarkers for Malaria: advances in early detection and asymp-tomatic diagnosis.Biosensors. 2025;15(2):106.
10. Harris I, Chuma J, Chesang B. Gender differences in malariaepidemiology and treatment outcomes.PLOS ONE. 2023;18(4).
11. He H, Garcia EA. Learning from imbalanced data.IEEE Trans Knowl Data Eng. 2009;21(9):1263–84.
12. Jones TM, Smith RC, Brown LE. Temperature and malaria dynam-ics: impacts on mosquito life cycle and disease transmission.J InfectDis. 2023;228(4):567–76.
13. Ministry of Health, Kenya.Malaria Control Program: AnnualReport. Government Press; 2023.
14. Muriithi D, Lumumba VW, Okongo M. A machine learning-basedprediction of malaria occurrence in Kenya.Am J Theor Appl Stat.2024;12(1):65–72.
15. Nguyen HT, Pham QT, Tran PD. Symptomatology and treat-ment outcomes in malaria patients.Asian Pac J Trop Med.2023;16(2):123–30.
16. Opoka RO, Namazzi R, Datta D, Bangirana P, Conroy AL,Goings MJ,et al.Severe falciparum malaria in young childrenis associated with an increased risk of post-discharge hospitaliza-tion: a prospective cohort study.Malar J. 2024;23(1):367.
17. Parham PE, Michael E. Modeling the effects of weather and cli-mate change on malaria transmission.Environ Health Perspect.2010;118(5):620–6.
18. Rhys HI.Machine Learning with R, the Tidyverse, and MLR.Manning Publications; 2020.
19. Sarker IH. Machine learning: algorithms, real-world applica-tions and research directions.SN Comput Sci. 2021;2(3).
20. Shankar D, Elavarasi S. Classification of imbalanced malariadisease using naïve bayesian algorithm.Int J Eng Technol.2018;7(2.7):610–4.
21. Suharti C, Widodo D, Wahyuni S. The role of fever inmalaria diagnosis.J Infect Public Health. 2022;15(8):876–82.
22. Tatem AJ, Gething PW, Smith DL. Environmental drivers ofmalaria transmission.Nat Rev Microbiol. 2023;21(3):123–35.
23. Weiss DJ, Lucas TCD, Nguyen M, Nandi AK, Bisanzio D, BattleKE,et al.Mapping the global prevalence, incidence, and mortalityof plasmodium vivax, 2000–2017.Lancet. 2020;394(10212):332–40.
24. World Health Organization (WHO).World Malaria Report 2023.WHO; 2023.
25. Zeng X. Enhancing the interpretability of SHAP values using largelanguage models. 2024.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Dr. Amirhossein Danesh, Dr. Niloofar Yazdani (Author)
This work is licensed under a Creative Commons Attribution 4.0 International License.
All articles published by The Parthenon Frontiers and its associated journals are distributed under the terms of the Creative Commons Attribution (CC BY 4.0) International License unless otherwise stated.
Authors retain full copyright of their published work. By submitting their manuscript, authors agree to grant The Parthenon Frontiers a non-exclusive license to publish, archive, and distribute the article worldwide. Authors are free to:
-
Share their article on personal websites, institutional repositories, or social media platforms.
-
Reuse their content in future works, presentations, or educational materials, provided proper citation of the original publication.
