Climate change and risk of cutaneous leishmaniasis: an artificial neural network-based approach

Climate change significantly impacts the prevalence of vector-borne diseases, including cutaneous leishmaniasis (CL). This study evaluates the impact of climate change on CL by artificial neural network (ANN) in the central part of Iran. The Gamma was employed to estimate the least mean squared error and optimal lag time for predictions. General circulation models (GCMs) are essential tools for simulating future climate conditions with two pathways: RCP2.6 and RCP8.5. ANNs were used to simulate the various impacts of climate change on CL infection. We revealed that temperature and precipitation are crucial factors affecting CL incidence, with a 100-day lag time being most effective for predictions. ANN modeling analysis demonstrate that maximum temperature and rainfall are the most significant predictors of CL outbreaks. The study area's climate is projected to become warmer by the year 2050. Forecasts indicate an increase in CL cases by 2050 under the RCP8.5 scenario, but remain stable under RCP2.6. Seasonal trends in CL cases are predicted to remain unchanged. This study demonstrates that ANN modeling can predict disease trends and high-risk areas more efficiently. These insights are crucial for formulating effective public health interventions and improving disease control measures in the context of climate change.