Modelling transmission dynamics of Covid-19 during pre-vaccination period in Malaysia : a gui-based seird predictive model using streamlit

• Main Authors: Zulkarnain, Norsyahidah, Abdul Hadi, Muhammad Salihi, Mohammad, Nurul Farahain, Shogar, Ibrahim
• Format: Article
• Language: English
• Published: UKM Press 2024
• Subjects: QA Mathematics
• Online Access: http://irep.iium.edu.my/111523/7/111523_Modelling%20transmission%20dynamics%20of%20Covid-19.pdf; http://irep.iium.edu.my/111523/; https://www.ukm.my/jqma/wp-content/uploads/2024/03/Paper_3.pdf

Coronavirus disease (COVID-19) is a major health threat worldwide pandemic, first identified in Malaysia on 25 January 2020. This outbreak can be represented in the mathematical expressions of a non-linear system of ordinary differential equations (ODEs). With the lack of a predictive SEIRD model in terms of Graphical Users Interface (GUI) in Malaysia, this paper aims to model the COVID-19 outbreak in Malaysia during the pre-vaccination period using the Susceptible-Exposed-Infected-Recovered-Death (SEIRD) model with time-varying parameters, then develop a GUI-based SEIRD predictive model using Streamlit Python library. The GUI-based SEIRD predictive model called GUI-mSEIRDsr predictor considers various values for the proportion of the quarantine-abiding population (r) and three different decisions regarding the MCO lifting date to forecast the number of active cases (I) on 15 October 2020. This information provides insightful information not only to government agencies but public as well, in understanding the effects of population behaviour to COVID-19 spread. The mathematical model is solved using the Scipy odeint function, which uses the Livermore Solver for Ordinary Differential Equations with an Automatic method switching (LSODA) algorithm. The time-varying coefficients of the SEIRD model that best fit the real data of COVID-19 cases are obtained using the Nelder-Mead optimisation algorithm. This extended SIRD model with exposed (E) compartment becoming SEIRD, leads to a robust model. It adequately fitted two datasets of Malaysian COVID-19 indicated by the slightest average values of Root Mean Square Error (RMSE) as compared to other existing models. The results highlight that the larger the values of the proportion of the quarantine-abiding population (r) and the later the date of the lifted MCO, the lower the spread of COVID-19 which eventually helps Malaysia to reach disease-free equilibrium in a shorter duration.