Real-time estimation and prediction of the mortality caused due to COVID-19 using particle swarm optimization and finding the most influential parameter

On March 11, 2020, the World Health Organization has declared the outbreak of COVID-19 as Pandemic, which is the massive challenges faced globally. Previous studies have indicated that the meteorological parameters can play a vital role in transmissibility and Mortality. In the present work, the influence of Comorbidity and meteorological parameters are investigated for Mortality caused due to COVID. For this, the most affected city by COVID-19 is considered, i.e., Mumbai, India, as a case study. It was found that Comorbidity is the most influential parameter on the Mortality of COVID-19. The Spearman correlation coefficient for meteorological parameters lies between 0.386 and 0.553, whereas for Comorbidity was found as 0.964. A regression model is developed using particle swarm optimization to predict the mortality cases for Mumbai, India. Further, the developed model is validated for the COVID-19 cases of Delhi, India, to emphasize the utility of the developed model for other cities. The measured and predicted curve shows a good fit with a mean percentage error of 0.00957% and a coefficient of determination of 0.9828. Thus, particle swarm optimization techniques demonstrate very high potential for the prediction of Mortality caused due to COVID-19. It is insisted that by providing constant health monitoring and adequate care for the comorbidity patients, the Mortality can be suppressed drastically. The present work can serve as an input to the policymakers to overcome the COVID-19 pandemic in India as well as other parts of the world.