Spatio-temporal predictive modeling framework for infectious disease spread

TL;DR

This paper introduces a high-dimensional PDE-based framework for modeling infectious disease spread, incorporating societal factors and providing detailed spatio-temporal predictions, exemplified with Covid-19 data.

Contribution

It develops a novel high-dimensional PDE framework that captures complex societal and biological factors influencing disease spread, enabling detailed scenario analysis.

Findings

Covid-19 spread predictions using 6D PDE model

Scenario analysis reveals impact of policies and behaviors

Framework offers detailed insights into infection dynamics

Abstract

A novel predictive modeling framework for the spread of infectious diseases using high dimensional partial differential equations is developed and implemented. A scalar function representing the infected population is defined on a high-dimensional space and its evolution over all directions is described by a population balance equation (PBE). New infections are introduced among the susceptible population from non-quarantined infected population based on their interaction, adherence to distancing norms, hygiene levels and any other societal interventions. Moreover, recovery, death, immunity and all aforementioned parameters are modeled on the high-dimensional space. To epitomize the capabilities and features of the above framework, prognostic estimates of Covid-19 spread using a six-dimensional (time, 2D space, infection severity, duration of infection, and population age) PBE is presented. Further, scenario analysis for different policy interventions and population behavior is presented, throwing more insights into the spatio-temporal spread of infections across disease age, intensity and age of population. These insights could be used for science-informed policy planning.