Conference Abstract | Volume 8, Abstract ELIC2025187 (Oral 109) | Published: 15 Aug 2025
1Department of Electrical and Electronics Engineering, Federal University of Technology, Minna, Nigeria, 2East Kent Hospital NHS Foundation Trust, Kent, United Kingdom, 3Department of Electrical and Electronics Engineering, Confluence University of Science and Technology, Osara, Kogi State, Nigeria, 4West African Science Service Centre on Climate Change and Adapted Land Use (WASCAL), Federal University of Technology, Minna, Nigeria, 5International Federation of the Red Cross and Red Crescent Societies (IFRC), Abuja, Nigeria
&Corresponding author: Michael Chukwuemeka Etuonuma, Department of Electrical and Electronics Engineering, Federal University of Technology, Minna, Nigeria. Email: michaeletuonuma01@gmail.com
Received: 12 Jun 2024, Accepted: 09 Jul 2025, Published: 15 Aug 2025
Domain: Infectious Disease Epidemiology
Keywords: Lassa fever, machine learning, XGBoost, digital health, early diagnosis, surveillance, predictive model
©Michael Chukwuemeka Etuonuma et al. Journal of Interventional Epidemiology and Public Health (ISSN: 2664-2824). This is an Open Access article distributed under the terms of the Creative Commons Attribution International 4.0 License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Cite this article: Michael Chukwuemeka Etuonuma et al., AI-driven early diagnosis of Lassa Fever: Development of an XGBoost-based predictive web application. Journal of Interventional Epidemiology and Public Health. 2025;8(ConfProc5):00109. https://doi.org/10.37432/JIEPH-CONFPRO5-00109
Lassa fever, a viral hemorrhagic disease endemic to West Africa, poses a serious public health threat due to high fatality rates, diagnostic delays, and nonspecific symptoms. In over 70% of confirmed cases, diagnosis occurs after Day 6 of symptom onset, often when complications have already developed (Nigeria Centre for Disease Control, 2021).
A simulation-based approach using supervised machine learning was applied. A synthetic dataset of 10,000 pseudopatients was generated, modeling real-world clinical symptoms and physiological indicators from Lassa fever-endemic populations. Each record was labeled as either ‘positive’ or ‘negative’ based on a predefined risk scoring algorithm. The dataset was split into training (80%) and testing (20%) subsets. Four Machine learning models: Logistic Regression, Random Forest, Support Vector Machine, and XGBoost were trained and evaluated using accuracy, precision, recall, and F1-score.
Out of 10,000 pseudopatients, 4,873 (48.73%) were classified as Lassa fever positive. Among all models, XGBoost demonstrated the best performance: 94.80% accuracy, 94.50% precision, 95.20% recall, and 94.85% F1-score. This model was selected for deployment in a web-based early diagnostic system.
Menu