MIT’s AI Detects Asymptomatic COVID-19 Cases With Phone Recorded Coughs

“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 in a statement co-author Brian Subirana, a research scientist in MIT’s Auto-ID Laboratory.

The scientists designed the new AI-based app on a previous model they were using to detect Alzheimer's in speech patterns. Tweaking it to analyze coughs rather than speech turned out to be actually quite simple.

“The sounds of talking and coughing are both influenced by the vocal cords and surrounding organs. This means that when you talk, part of your talking is like coughing, and vice versa. It also means that things we easily derive from fluent speech, AI can pick up simply from coughs, including things like the person’s gender, mother tongue, or even emotional state. There’s in fact sentiment embedded in how you cough,” Subirana said.

“So we thought, why don’t we try these Alzheimer’s biomarkers [to see if they’re relevant] for COVID.”

To this date, the team has collected more than 70,000 recordings to create a model that identifies 98.5% of coughs from people confirmed with COVID-19. Not bad!