Speakers - 2026

Title: Spatio temporal diffusion of leishmaniasis: Enhanced assessment of environmental and biological drivers through statistical and machine learning frameworks

Abstract

Leishmaniasis is a vector-borne disease characterized by highly heterogeneous spatiotemporal patterns shaped by complex interactions among environmental, ecological, and anthropogenic factors. This study provides a strengthened and integrative assessment of the mechanisms driving disease diffusion by jointly analyzing vector activity, rodent mobility, climatic variability, and vegetation dynamics. Using a hybrid framework that combines classical statistical modeling with state-of-the-art machine learning techniques, we quantify the relative influence of vector density, radiotelemetry-tracked rodent behaviour, temperature, humidity, NDVI, precipitation, and associated diffusion coefficients. Random Forest models offer robust predictive capabilities, while SHAP (SHapley Additive exPlanations) and LIME (Local Interpretable Model-agnostic Explanations) provide high-resolution interpretability of individual predictors. The enhanced insights generated from this multi-factor analysis aim to support more effective, targeted, and spatially optimized public health interventions for leishmaniasis control.