Abstract
In this paper, we consider an extended SEIR compartmental model that incorporates young and old interacting subpopulations, allowing for cross-group transmission dynamics. Implicit behavioral changes are included to determine the influence of social behavior on coronavirus transmission dynamics. The basic reproduction number, the average number of secondary cases of infection produced by a single primary case, is derived for both the explicit and implicit model using the next-generation matrix method. We solve the associated differential equation systems and estimate useful parameters in the explicit model using physics-informed neural networks (PINNs). Our results point to how the PINNs approach offers an effective framework to predict the unique parameters of our model, forecast disease progression, and determine the impact of behavioral modifications on the reproduction number and transmission dynamics. Lastly, we have created an interactive dashboard where users can manipulate certain parameters and view the resulting graphs and the reproductive number.
Recommended Citation
Aubry-Romero, Naima; Ogueda-Oliva, Alonso; and Seshaiyer, Padmanabhan
(2025)
"Modeling, Analysis, and Prediction of COVID-19 Dynamics with Interacting Subpopulations and Implicit Behavior Using Physics-Informed Neural Networks,"
Spora: A Journal of Biomathematics: Vol. 11, 84–99.
DOI: https://doi.org/10.61403/2473-5493.1103
Available at:
https://ir.library.illinoisstate.edu/spora/vol11/iss1/8
Included in
COVID-19 Commons, Data Science Commons, Epidemiology Commons, Numerical Analysis and Computation Commons, Ordinary Differential Equations and Applied Dynamics Commons
