MIT developing AI app that can identify COVID-19 through the sound of your cough

MIT developing AI app that can identify COVID-19 through the sound of your cough

Massachusetts Institute of Technology (MIT) has developed a "cough detector" for COVID-19 and it could be a game-changer.

In the ongoing battle against COVID-19, asymptomatic patients are one of the most dangerous factors of the pandemic.

Without showing the tell-tale signs of the deadly virus, asymptomatic patients are rarely required to get tested and yet can still infect thousands of people. However, due to recent development research by MIT, those who show no symptoms can be identified from healthy individuals through their cough.

Researchers have found that coughs from asymptomatic patients carry sonic differences. only recognised by an artificial intelligence (AI) model. Through voluntarily submitted samples of "forced-cough recordings", the AI model has learnt to pick up on these audible differences, resulting in identifying 98.5% of asymptomatic COVID-19 patients.

The research team is working on rolling with an approved mobile-app that can easily be adopted and used on a large scale. The app is said to allow users to "log in daily, cough into their phone, and instantly get information on whether they might be infected".

"The effective implementation of this group diagnostic tool could diminish the spread of the pandemic if everyone uses it before going to a classroom, a factory, or a restaurant,” said Brian Subirana, a MIT research scientist.