Spatial variation in lymphatic filariasis risk factors of hotspot zones in Ghana

Abstract Background Lymphatic Filariasis (LF), a parasitic nematode infection, poses huge economic burden to affected countries. LF endemicity is localized and its prevalence spatially heterogeneous. In Ghana, there exists differences in multiplicity of symptoms the country’s northern southern parts. Species distribution models (SDMs) have been utilized explore suite risk factors that influence transmission these geographically distinct regions. Methods Presence-absence records microfilaria ( mf ) cases were stratified into zones used run SDMs, while climate, socioeconomic, land cover variables provided explanatory information. Generalized Linear Model (GLM), Boosted (GBM), Artificial Neural Network (ANN), Surface Range Envelope (SRE), Multivariate Adaptive Regression Splines (MARS), Random Forests (RF) algorithms for both study also entire country comparison. Results Best model quality was obtained with RF GBM highest Area under Curve (AUC) 0.98 0.95, respectively. The predicted high suitable environments short grass savanna (northern) coastal (southern) areas Ghana. Mainly, socioeconomic such as proximity inland water bodies population density uniquely influenced south. At same time, poor housing distinctive factor north. Precipitation, temperature, slope, poverty common but subtle variations response values, which confirmed by countrywide model. Conclusions This has demonstrated different variable combinations occurrence lymphatic filariasis Thus, an understanding geographic distinctness required inform on development area-specific control systems towards elimination Ghana internationally.