ABSTRACT
In this chapter, we aim to provide a biological, statistical, and mathematical framework to understand and formulate sensible models to study the spreading dynamics of COVID-19. First, we discuss the epidemiological and clinical features that make COVID-19 challenging-to-control in different scales and ways. We then describe the different error sources present in raw COVID-19 epidemiological data and the logistic limitations associated with non-pharmaceutical interventions (NPIs), like test-trace-and-isolate (TTI). By studying compartmental SIR and SIR-like mathematical models and their underlying hypotheses, we demonstrate the derivation of significant parameters for evaluating this pandemic’s progression, as the reproduction number Rt. Then, we provide the statistical basis for the correction of delay-induced errors in raw data through the “nowcasting” of infections and describe the Machine-Learning-based approaches to tackle significant challenges in modeling COVID-19. We end our chapter with several case studies, where we describe the modeling aspects as carefully as their results, providing the reader with fresh multi-disciplinary insights to inspire their own models. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.