Globally, Covid 19 pandemic will end by December 9, predicts a team of researchers from the Singapore University of Technology and Design. The researchers used an artificial intelligence algorithm which helped them to make a data-driven prediction on the end date of Covid-19 pandemic in 131 countries. As per the algorithm, 97 per cent of the total expected cases in India will be identified by May 21. The last new Covid case in India will be determined by July 26, say the research team.

Similar to other pandemics, Covid outbreak too has a life cycle pattern. The outbreak has an initial acceleration phase where we saw a huge increase in the cases followed by the inflexion point. At this stage, the curve begins to flatten. In the next step, which is called the deceleration phase, the pandemic curve comes down and eventually ends. 

To predict the pandemic course, the team used the latest infection data from each country and used the SIR model(a compartmental model describing the dynamics of infectious disease) to predict the curve. This model calculates the number of people that will likely become infected with an infectious illness in a population over a while. The estimate is based on the number of people susceptible to the disease (S), infected (I) and recovered (R) from the disease.

The forecasts present three alternative estimates of end dates — the date when the last expected case has been identified, 99 per cent of the expected total cases have been identified, and 97 per cent of the expected cases have been identified.

In the worst-hit nations like the US and Italy, the pandemic will end by the last week of August. Though the pandemic will end by December, countries like Bahrain and Qatar will see the last of its fresh cases in February 2021 says the Singapore researchers. 

That said, the model has some limitations too. This method has not considered the second wave of infection. According to the researchers, being overly optimistic on end dates “might be dangerous” and easing existing restrictions can cause the infection rates to change. The estimates may not be accurate for countries that may be under-reporting cases as the data taken for the prediction will not be correct, warn the researchers.