Generalized SIR (GSIR) epidemic model: An improved framework for the predictive monitoring of COVID-19 pandemic

Abstract

Novel coronavirus respiratory disease COVID-19 has caused havoc in many countries across the globe. In order to contain infection of this highly contagious disease, most of the world population is constrained to live in a complete or partial lockdown for months together with a minimal human-to-human interaction having far reaching consequences on countries’ economy and mental well-being of their citizens. Hence, there is a need for a good predictive model for the health advisory bodies and decision makers for taking calculated proactive measures to contain the pandemic and maintain a healthy economy. This paper extends the mathematical theory of the classical Susceptible–Infected–Removed (SIR) epidemic model and proposes a Generalized SIR (GSIR) model that is an integrative model encompassing multiple waves of daily reported cases. Existing growth function models of epidemic have been shown as the special cases of the GSIR model. Dynamic modeling of the parameters reflect the impact of policy decisions, social awareness, and the availability of medication during the pandemic. GSIR framework can be utilized to find a good fit or predictive model for any pandemic. The study is performed on the COVID-19 data for various countries with detailed results for India, Brazil, United States of America (USA), and World. The peak infection, total expected number of COVID-19 cases and thereof deaths, time-varying reproduction number, and various other parameters are estimated from the available data using the proposed methodology. The proposed GSIR model advances the existing theory and yields promising results for continuous predictive monitoring of COVID-19 pandemic.

A Physics-Based Data-Driven Approach for Finite Time Estimation of Pandemic Growth

“…A time-window based SIR model was proposed [7], for dynamic data analysis measuring the basic infection number and predicting the growth rate of the pandemic. A generalised SIR model [8] was presented that incorporated multiple waves of daily reported cases and provided continuous predictions & monitoring of the COVID-19 pandemic. Metapopulation Dynamics [9] [10] [11] was used to study and evaluate the effects of domestic and international travel limitations on the spread of COVID-19 [12].…”

A new hybrid prediction model of cumulative COVID-19 confirmed data

“…Epidemic models used to study COVID-19 are mainly divided into three types: susceptible infectious (SI) model ( Cong et al, 2020 ), susceptible infectious recovered (SIR) model ( Cooper et al, 2020 , Rafieenasab et al, 2020 , Singh and Gupta, 2021 ), and susceptible exposed infectious recovered (SEIR) model ( Annas et al, 2020 , Feng et al, 2021 ). Epidemic model can better predict the number of people infected and susceptible in the future, but their establishment process is complex, and the transmission coefficient needs to be obtained by a large number of experiments.…”

COVID-19 Cases Prediction in Saudi Arabia Using Tree-based Ensemble Models

“…The Statistical approaches used to estimate and predict the COVID-19 pandemic include the Bayesian approaches [7] and Kalman filtering [8]. Mathematical approaches included the parameter estimation on compartmental models such as Susceptible-Exposed-Infected-Removed (SEIR) model [9,10] or the SIR model [11,12]. Data scientists utilized various machine learning algorithms such as Deep Learning [13], Long Short-Term Memory (LSTM) [6,14], neuro-fuzzy inference [15,16], and decision tree-based algorithms [17].…”