Chapter 8 - Learning-Based Identification of Viral Infection Dynamics

Abstract

In history, viral infectious diseases have caused high levels of morbidity and mortality worldwide. Mathematical models have served as a central instrument to predict the kinetics of different infections. Unfortunately, the development of mechanistic models and the corresponding parameter estimation are difficult tasks. As an alternative, recurrent high-order neural networks (RHONNs) trained with an algorithm based on the extended Kalman filter (EKF) are presented to identify infectious diseases such as influenza A virus (IAV) and HIV dynamics. We consider within-host mathematical models of IAV and HIV as unknown signals to the RHONNs. A negligible identification error is reported for both identifiers, showing that RHONNs are able to identify the within-host dynamics. Simulation results indicate promising directions towards the problem of model identification of infectious diseases, serving for future model-based control strategies of infections.