The algorithm can detect what is being termed as the ‘Covid Cough’ which is inaudible to the human ear.
One of the biggest problems in tackling coronavirus has been the rise in the number of asymptomatic cases across the globe. However, a group of scientists at MIT have developed an algorithm that detects the Covid-19 virus in humans based on how we cough.
With the help of artificial intelligence, the algorithm has reportedly been trained to identify individuals with Covid-19 based on their coughing patterns. According to reports, the algorithm was fed with tens of thousands of recordings which included not just coughs but also spoken words. The algorithm can now correctly identify individuals with Covid-19, with a proven success rate of 98.5 per cent.
The MIT has published its findings in the IEEE Journal of Engineering in Medicine and Biology.
According to researchers, even if a person suffering from Covid-19 does not show any symptoms, the person’s cough is characteristically different from someone who doesn’t have the virus. This crucial difference cannot be heard by humans, but the algorithm that has been developed has been trained to identify it.
Researchers are hoping that the algorithm can be used in public spaces as an alert system. "Practical use cases could be for daily screening of students, workers and public, at schools, jobs, and transport reopen, or for pool testing to quickly alert of outbreaks in groups," the report says.
According to a report in the BBC, researchers are waiting for regulatory approval to develop the algorithm into an app.