Integrating satellite-derived environmental variables and machine learning techniques in predicting malaria prevalence in Karagwe District, Tanzania

Malaria continues to place a heavy burden on public health systems in sub-Saharan Africa, despite substantial progress in disease control interventions. This study integrated remote sensing-derived environmental variables and machine learning techniques to predict Plasmodium falciparum parasite prevalence among children aged 2–10 years (PfPR 2–10 ), from Malaria Atlas Project, in Karagwe District, Tanzania. Environmental predictors were derived from Sentinel-2, MODIS, CHIRPS, SRTM DEM, and JRC Global Surface Water datasets for the period 2020-2024. Variables included the Normalized Difference Vegetation Index (NDVI), Normalized Difference Water Index (NDWI), Modified Normalized Difference Water Index (MNDWI), Normalized Difference Moisture Index (NDMI), Normalized Difference Built-up Index (NDBI), rainfall, land surface temperature (LST), elevation, slope, and distance to water bodies. Random Forest Regression and XGBoost models were developed to evaluate the relationship between environmental conditions and malaria prevalence. XGBoost showed moderate predictive performance under nested spatial cross-validation, achieving a mean coefficient of determination (R 2 ) of 0.551, root mean square error (RMSE) of 1.409, and mean absolute error (MAE) of 0.994. NDWI was identified as the most influential predictor with relative importance of 42.61% and, followed by elevation. Model interpretability also indicated hydrological variables and elevation had a large influence on the malaria prevalence prediction. The findings provide a scalable and cost-effective approach for malaria surveillance and environmental risk assessment in Tanzania and similar endemic settings.

Authors

Institutions

Publication Details

Journal
Frontiers in Artificial Intelligence
Published
2026-11-19
DOI
https://doi.org/10.3389/frai.2026.1952002
Primary Topic
Malaria Research and Control
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Integrating satellite-derived environmental variables and machine learning techniques in predicting malaria prevalence in Karagwe District, Tanzania

Elizabeth Sylvester Mkoba, Devotha Godfrey Nyambo, Joanitha Kaijage, Lawrence Mdegela
Frontiers in Artificial Intelligence
Malaria Research and Control
article

Integrating satellite-derived environmental variables and machine learning techniques in predicting malaria prevalence in Karagwe District, Tanzania

Elizabeth Sylvester Mkoba, Devotha Godfrey Nyambo, Joanitha Kaijage, Lawrence Mdegela
article en

Abstract

Malaria continues to place a heavy burden on public health systems in sub-Saharan Africa, despite substantial progress in disease control interventions. This study integrated remote sensing-derived environmental variables and machine learning techniques to predict Plasmodium falciparum parasite prevalence among children aged 2–10 years (PfPR 2–10 ), from Malaria Atlas Project, in Karagwe District, Tanzania. Environmental predictors were derived from Sentinel-2, MODIS, CHIRPS, SRTM DEM, and JRC Global Surface Water datasets for the period 2020-2024. Variables included the Normalized Difference Vegetation Index (NDVI), Normalized Difference Water Index (NDWI), Modified Normalized Difference Water Index (MNDWI), Normalized Difference Moisture Index (NDMI), Normalized Difference Built-up Index (NDBI), rainfall, land surface temperature (LST), elevation, slope, and distance to water bodies. Random Forest Regression and XGBoost models were developed to evaluate the relationship between environmental conditions and malaria prevalence. XGBoost showed moderate predictive performance under nested spatial cross-validation, achieving a mean coefficient of determination (R 2 ) of 0.551, root mean square error (RMSE) of 1.409, and mean absolute error (MAE) of 0.994. NDWI was identified as the most influential predictor with relative importance of 42.61% and, followed by elevation. Model interpretability also indicated hydrological variables and elevation had a large influence on the malaria prevalence prediction. The findings provide a scalable and cost-effective approach for malaria surveillance and environmental risk assessment in Tanzania and similar endemic settings.

Frontiers in Artificial IntelligenceVol. 9
Nelson Mandela African Institution of Science and Technology (TZ)
Malaria Research and Control
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon