AI Model Detects Asymptomatic COVID Infections Through Coughs

Asymptomatic people infected with Covid-19 exhibit no discernible physical symptoms of the disease and are less likely to seek out testing for the virus, which could unknowingly spread the infection to others. MIT researchers have now found that asymptomatic people may differ from healthy individuals in the way that they cough. These differences are not decipherable to the human ear, but it turns out that they can be picked up by AI. The researchers trained the model on tens of thousands of samples of coughs and spoken words, reaching 98.5% accuracy rate of coughs from people who were confirmed to have Covid-19, including 100% of coughs from asymptomatic subjects. The team is working on incorporating the model into a user-friendly app, which could potentially be a free, convenient, non-invasive pre-screening tool. Before the pandemic’s onset, research groups already had been training algorithms on cell phone recordings of coughs to accurately diagnose conditions such as pneumonia and asthma. Similarly, the MIT team was developing AI models to analyse forced-cough recordings to see if they could detect signs of Alzheimer’s, a disease associated with not only memory decline but also neuromuscular degradation such as weakened vocal cords.