Estimation of undetected asymptomatic infections of COVID-19: a mathematical modeling approach

Abstract

The accurate quantification of asymptomatic infections is crucial for understanding and mitigating the spread of COVID-19. However, estimating the true prevalence of asymptomatic cases remains a significant challenge. This study introduces a novel mathematical modeling approach to estimate key epidemiological parameters associated with asymptomatic infections, leveraging comprehensive COVID-19 data from the Republic of Korea. We develop a refined compartmental model that explicitly incorporates asymptomatic individuals and employs a trajectory matching method, integrating least-squares fitting with gradient matching, to align the model with confirmed cases and deaths. The model successfully reproduces the initial outbreak and subsequent epidemic waves, demonstrating strong agreement with observed data and validating the estimated parameters, which align with prior findings. We observe a progressive increase in both infectivity and the proportion of asymptomatic infections from the initial strain to the Delta and Omicron variants. Notably, across all phases, the proportion of asymptomatic cases is higher among vaccinated individuals compared to unvaccinated individuals. This study provides valuable insights into the hidden dynamics of the COVID-19 pandemic and offers a robust methodology that can be broadly applied to estimate critical epidemiological parameters for other infectious diseases, enhancing our capacity to manage future outbreaks.