A Neural Network Model to Predict the Effect of Climate Change on West Nile Virus (WNV) Epidemiology

West Nile virus (WNV) is a mosquito-borne disease. The virus is transmitted cyclically between mosquitoes and avian hosts. It is influenced by a diverse array of environmental parameters. The prediction of WNV epidemic is challenging since any change in climate conditions and vector ecosystems affect the epidemiology of the WNV. In this study, a neural network (NN) model was developed to capture the non-linear effect of environmental parameters on WNV neuro-invasive disease by using historical disease data for four major cities in the USA. This NN model uses statistical and machine learning techniques to forecast spatial and temporal variation of WNV in other USA cities. This artificial intelligence framework was used further to quantify the correlation between various climate change parameters such as temperature, rainfall, season, and land coverage on WNV. This study addresses key questions on how projected climate change will affect the spatial and temporal dynamics of WNV disease epidemics which is critical to managing spread of the disease.