Simulation of the potential impact of climate change on malaria incidence using artificial neural networks (ANNs).

Abstract

Climate change can increase the spread of infectious diseases and public health concerns. Malaria is one of the endemic infectious diseases of Iran, whose transmission is strongly influenced by climatic conditions. The effect of climate change on malaria in the southeastern Iran from 2021 to 2050 was simulated by using artificial neural networks (ANNs). Gamma test (GT) and general circulation models (GCMs) were used to determine the best delay time and to generate the future climate model under two distinct scenarios (RCP2.6 and RCP8.5). To simulate the various impacts of climate change on malaria infection, ANNs were applied using daily collected data for 12 years (from 2003 to 2014). The future climate of the study area will be hotter by 2050. The simulation of malaria cases elucidated that there is an intense increasing trend in malaria cases under the RCP8.5 scenario until 2050, with the highest number of infections occurring in the warmer months. Rainfall and maximum temperature were identified as the most influential input variables. Optimum temperatures and increased rainfall provide a suitable environment for the transmission of parasites and cause an intense increase in the number of infection cases with a delay of approximately 90 days. ANNs were introduced as a practical tool for simulating the impact of climate change on the prevalence, geographic distribution, and biological activity of malaria and for estimating the future trend of the disease in order to adopt protective measures in endemic areas.