SEIAQRDT model for the spread of novel coronavirus (COVID-19): A case study in India

COVID-19 is a global pandemic declared by WHO. This pandemic requires the execution of planned control strategies, incorporating quarantine, self-isolation, and tracing of asymptomatic cases. Mathematical modeling is one of the prominent techniques for predicting and controlling the spread of COVID-19. The predictions of earlier proposed epidemiological models (e.g. SIR, SEIR, SIRD, SEIRD, etc.) are not much accurate due to lack of consideration for transmission of the epidemic during the latent period. Moreover, it is important to classify infected individuals to control this pandemic. Therefore, a new mathematical model is proposed to incorporate infected individuals based on whether they have symptoms or not. This model forecasts the number of cases more accurately, which may help in better planning of control strategies. The model consists of eight compartments: susceptible (S), exposed (E), infected (I), asymptomatic (A), quarantined (Q), recovered (R), deaths (D), and insusceptible (T), accumulatively named as SEIAQRDT. This model is employed to predict the pandemic results for India and its majorly affected states. The estimated number of cases using the SEIAQRDT model is compared with SIRD, SEIR, and LSTM models. The relative error square analysis is used to verify the accuracy of the proposed model. The simulation is done on real datasets and results show the effectiveness of the proposed approach. These results may help the government and individuals to make the planning in this pandemic situation.