Predicting malaria outbreak in The Gambia using machine learning techniques
Problem
Malaria outbreaks in the districts of The Gambia are hard to anticipate, posing a growing threat to public health and economic growth.
Why it matters
Malaria is stated to be the most common cause of death among parasitic diseases in affected tropical and subtropical nations.
What the AI did
Eight machine learning algorithms were trained on historical meteorological data to predict malaria outbreaks per district, evaluated with 10-fold cross-validation repeated five times.
Results
Extreme gradient boosting and decision trees achieved 93.3% accuracy on the testing set
Impact
Evidence supports a validated predictive model on historical data; real-world deployment is not stated.
Record
| Venue | PLoS ONE |
|---|---|
| Published | 2024 |
| Work type | article |
| Authors | Ousman Khan, Jimoh Olawale Ajadi, M. Pear Hossain |
| Citations | 36 (OpenAlex, at import time) |
| References | 41 |
| Model or method | C5.0 decision trees, ANN, kNN, SVM (linear/radial), logistic regression, extreme gradient boosting, random forests |
| Organization | not stated in source |
| Primary source | https://doi.org/10.1371/journal.pone.0299386 |
| Open access | gold |
| Keywords | malaria, outbreak prediction, machine learning, XGBoost, The Gambia, meteorological data |
Importance rating
significant — thousands to millions affected
Abstract states malaria is the most common cause of death among parasitic diseases and a growing threat to public health and economic growth of tropical/subtropical nations; district-level prediction for The Gambia.