The Massachusetts Institute of Technology (MIT) has developed an algorithm to identify people infected with COVID-19. The algorithm listens to your step and understands whether you are infected or not.
The algorithm was trained using "tens of thousands" of recordings - both coughing and spoken words - and was able to recognize 98,5% of those who showed symptoms and were confirmed cases of COVID-19.
In addition, the algorithm detected 100% of the COVID-19 vectors that were confirmed to have the virus but showed no symptoms.
The recordings used to train the artificial intelligence (AI) model were submitted by volunteers on Internet and included forced coughing by healthy volunteers as well as patients with COVID-19. Over 70.000 samples have been collected so far and about 2.500 have been submitted by people confirmed to have COVID-19.
"People who are asymptomatic carriers of COVID-19 have a different cough from healthy people," the team said. "These differences are not detected by the human ear. But it turns out that they can be decrypted by Artificial Intelligence".
The MIT team, consisting of Brian Subirana, Jordi Laguarta, Ferran Hueto from his Auto-ID lab MIT, says they are now working in a user friendly application for mobile devices in which the algorithm will be integrated.
However, such an application requires approval and should not be considered an official diagnostic tool. Instead, the application could potentially serve as a “non-invasive tool preview”For users before confirming any suspicions with a diagnostic test. You should also keep in mind that an asymptomatic cough may be associated with the flu, colds or other situations.
Any tool that could possibly address the issue of asymptomatic transmission - in which those without symptoms could inadvertently spread virus - could be valuable in combating COVID -19.