tetano
Editor, Senior Moderator
Sci Rep
. 2022 Aug 4;12(1):13415.
doi: 10.1038/s41598-022-16496-6.
EKF-SIRD model algorithm for predicting the coronavirus (COVID-19) spreading dynamics
Abdennour Sebbagh[SUP] 1 [/SUP], Sihem Kechida[SUP] 2 [/SUP]
Affiliations
Abstract
In this paper, we study the Covid 19 disease profile in the Algerian territory since February 25, 2020 to February 13, 2021. The idea is to develop a decision support system allowing public health decision and policy-makers to have future statistics (the daily prediction of parameters) of the pandemic; and also encourage citizens for conducting health protocols. Many studies applied traditional epidemic models or machine learning models to forecast the evolution of coronavirus epidemic, but the use of such models alone to make the prediction will be less precise. For this purpose, we assume that the spread of the coronavirus is a moving target described by an epidemic model. On the basis of a SIRD model (Susceptible-Infection-Recovery- Death), we applied the EKF algorithm to predict daily all parameters. These predicted parameters will be much beneficial to hospital managers for updating the available means of hospitalization (beds, oxygen concentrator, etc.) in order to reduce the mortality rate and the infected. Simulations carried out reveal that the EKF seems to be more efficient according to the obtained results.
. 2022 Aug 4;12(1):13415.
doi: 10.1038/s41598-022-16496-6.
EKF-SIRD model algorithm for predicting the coronavirus (COVID-19) spreading dynamics
Abdennour Sebbagh[SUP] 1 [/SUP], Sihem Kechida[SUP] 2 [/SUP]
Affiliations
- PMID: 35927443
- DOI: 10.1038/s41598-022-16496-6
Abstract
In this paper, we study the Covid 19 disease profile in the Algerian territory since February 25, 2020 to February 13, 2021. The idea is to develop a decision support system allowing public health decision and policy-makers to have future statistics (the daily prediction of parameters) of the pandemic; and also encourage citizens for conducting health protocols. Many studies applied traditional epidemic models or machine learning models to forecast the evolution of coronavirus epidemic, but the use of such models alone to make the prediction will be less precise. For this purpose, we assume that the spread of the coronavirus is a moving target described by an epidemic model. On the basis of a SIRD model (Susceptible-Infection-Recovery- Death), we applied the EKF algorithm to predict daily all parameters. These predicted parameters will be much beneficial to hospital managers for updating the available means of hospitalization (beds, oxygen concentrator, etc.) in order to reduce the mortality rate and the infected. Simulations carried out reveal that the EKF seems to be more efficient according to the obtained results.