A Hybrid SEIHCRDV-UKF Model for COVID-19 Prediction. Application on real-time data

The prevalence of COVID-19 has been the most serious health challenge of the 21th century to date, concerning national health systems on a daily basis, since December 2019 when it appeared in Wuhan City. Prediction of pandemic spread plays an important role in effectively reducing this highly contagious disease. Nevertheless, most of the proposed mathematical methodologies, which aim to describe the dynamics of the pandemic, rely on deterministic models that are not able to reflect the true nature of the spread of COVID. In this paper, we propose a SEIHCRDV model - an extension of the classic SIR compartmental model - which also takes into consideration the populations of exposed, hospitalized, admitted in intensive care units (ICU), deceased and vaccinated cases, in combination with an unscented Kalman filter (UKF), providing a dynamic estimation of the time dependent parameters of the system. Apparently, this new consideration could be useful for examining also other pandemics. We examine the reliability of our model over a long period of 265 days, where we observe two major waves of infection, starting in January 2021 which signified the start of vaccinations in Europe, providing quite encouraging predictive performance. Finally, special emphasis is given to proving the non-negativity of SEIHCRDV model, to achieve a representative basic reproductive number R0 and to investigating the existence and stability of disease equilibriums in accordance with the formula produced to estimate R0.