NEW YORK: Researchers at the Massachusetts Institute of Technology have devised a method to identify people infected with COVID-19 by analysing the sound of their cough.
The algorithm was trained using tens of thousands of recordings – both coughs and spoken words – and was able to correctly identity 98.5 percent of those who were displaying symptoms and had confirmed cases of COVID-19.
Additionally, the algorithm identified 100 percent of COVID-19 carriers confirmed to have the virus but who were not displaying any symptoms.
The recordings used to train the artificial intelligence (AI) model were submitted by volunteers online and included forced-coughs from healthy volunteers as well as COVID-19 sufferers.
The researchers, in this regard, said, “Asymptomatic people infected with COVID-19 are difficult to detect if they have contracted the virus because they show no symptoms.”
“The difference between a healthy person’s cough and the COVID-19 victim’s cough is so slight that it’s imperceptible to the human ear,” the researchers added.
Explaining how this algorithm works, researchers said that one neural network gauges sounds associated with vocal cord strength, while another detects cues related to a person’s emotional state, such as frustration, which can produce a flat affect.
A third network listens for subtle changes in lung and respiratory performance. The team then combined all three models and overlaid them with an algorithm to detect muscular degradation.
However, the MIT scientists stressed that people should not use AI as a replacement for the COVID-19 test. They also stated that it was not built to diagnose people who are actively exhibiting covid-19 symptoms.
The new technology can still be very useful tool in detecting viruses and the team is reportedly working on developing a free app that can be used as a pre-screening tool for individuals who are not showing any symptoms.
