A Stochastic Epidemiological SIRD-V Model With LSM-EKF Algorithm for Forecasting and Monitoring the Spread of COVID-19 Pandemic: Real Data

• Published in: IEEE access Vol. 12; pp. 62047 - 62058
• Main Authors: Sebbagh, Abdennour, Ennehar Bencheriet, Chemesse, Kechida, Sihem
• Format: Journal Article
• Language: English
• Published: Piscataway IEEE 2024; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Algorithms; Coronavirus; Coronaviruses; COVID-19; Decision support systems; Disease control; Disease transmission; Epidemics; Extended Kalman filter; Kalman filters; Long short term memory; LSM-EKF-SIRD-V algorithm; Mathematical models; Monitoring; Pandemics; Predictive models; SIRD-V model; Statistics; Stochastic processes; Vaccines; Viral diseases; Viruses (medical)
• ISSN: 2169-3536; 2169-3536
• DOI: 10.1109/ACCESS.2024.3395544

Predictive algorithms for the coronavirus epidemic are indispensable tools for monitoring the dynamic spread of COVID-19 and for implementing intervention and preparedness measures to mitigate the outbreak. Many of the existing mathematical models used for epidemic analysis are deterministic in nature, which may not fully capture the complex dynamics of disease transmission. In this paper, we introduce a novel stochastic predictive algorithm known as the LSM-EKF-SIRD-V algorithm. This algorithm combines a SIRD-V model, which accounts for susceptible, infected, recovered, deceased, and vaccinated cases, with the Least Square Method (LSM) and an Extended Kalman Filter (EKF). It provides daily dynamic predictions of the system's parameters and is employed to analyze the COVID-19 disease profile in Algeria from January 29, 2021, to October 2, 2022. The primary goal of this approach is to create a decision-support system that empowers governments and health authorities with future pandemic statistics. This information enables them to adapt and optimize hospitalization resources, allowing for more effective intervention and preparedness measures to control the spread of the pandemic. Simulation results demonstrate the effectiveness of the proposed algorithm in accurately predicting the future dynamics of coronavirus spread based on historical and current case data.