Diseases spreading through individual based models with realistic mobility patterns

TL;DR

This paper introduces a novel mobile agents model with realistic human mobility patterns and demonstrates its equivalence to contact network models in simulating infectious disease spread.

Contribution

It presents a new mobile agents model based on empirical human mobility data and a method to derive equivalent contact networks, linking mobility patterns to network topology.

Findings

Small world properties emerge from truncated power-law mobility distributions.

The mobile agents model and contact network approach produce equivalent disease spread dynamics.

Empirical mobility data enhances the realism of infectious disease modeling.

Abstract

The individual-based models constitute a set of widely implemented tools to analyze the incidence of individuals heterogeneities in the spread of an infectious disease. In this work we focus our attention on human contacts heterogeneities through two of the main individual-based models: mobile agents and complex networks models. We introduce a novel mobile agents model in which individuals make displacements with sizes according to a truncated power-law distribution based on empirical evidence about human mobility. Besides, we present a procedure to obtain an equivalent weighted contact network from the previous mobile agents model, where the weights of the links are interpreted as contact probabilities. From the topological analysis of the equivalent contact networks we show that small world characteristics are related with truncated power-law distribution for agent displacements. Finally, we show the equivalence between both approaches through some numerical experiments for the spread of an infectious disease.