Estimating the State of Epidemics Spreading with Graph Neural Networks

TL;DR

This paper proposes a Graph Convolutional Neural Network architecture to estimate epidemic states from limited data, leveraging social network structures, tested on COVID-19 scenarios including a generic and a city-specific model.

Contribution

It introduces a novel GCN-based method for epidemic state estimation that accounts for social network effects, addressing data limitations in epidemic monitoring.

Findings

Effective in modeling epidemic spread using social network data

Accurate state estimation in both generic and city-specific scenarios

Demonstrates potential for real-time epidemic monitoring

Abstract

When an epidemic spreads into a population, it is often unpractical or impossible to have a continuous monitoring of all subjects involved. As an alternative, algorithmic solutions can be used to infer the state of the whole population from a limited amount of measures. We analyze the capability of deep neural networks to solve this challenging task. Our proposed architecture is based on Graph Convolutional Neural Networks. As such it can reason on the effect of the underlying social network structure, which is recognized as the main component in the spreading of an epidemic. We test the proposed architecture with two scenarios modeled on the CoVid-19 pandemic: a generic homogeneous population, and a toy model of Boston metropolitan area.