Spotting Virus From Satellites: Modeling the Circulation of West Nile Virus Through Graph Neural Networks#

The article explores predicting the circulation of the West Nile virus using Deep Learning and satellite images. In particular, the spread of the virus is modeled through a graph neural network that considers the spatial and environmental relations between different locations. The proposed MultiAdjacency Graph ATtention network (MAGAT) outperforms other methods, especially when pretraining is used. The results show that the proposed model can be used to predict the spread of the virus with high accuracy.