Study: Smartphone App Detects Asymptomatic COVID-19 by Recording Forced Coughs

MIT team-developed app, derived from Alzheimer's research, expected to seek FDA approval

Massachusetts Institute of Technology (MIT)-based researchers have found the way that someone coughs can reliably determine if they have COVID-19, according to a recently published study.

Moreover, these coughs can be recorded by Web browsers and devices such as cell phones and laptops.

In the paper, published in the IEEE Journal of Engineering in Medicine and Biology, the research team describes an artificial intelligence model distinguishing asymptomatic COVID-19 individuals from a healthy person by the way they cough.

Gathering tens of thousands of samples of coughs, as well as spoken words, the research team trained the model. This model identified 98.5% of coughs from individuals who were verified to have COVID-19, including 100% of coughs from asymptomatic individuals – those who had tested positive for the virus, but reported they had no symptoms.

The next step, post-study, will be distributing an easy-to-use app. If approved by the Food and Drug Administration, it could potentially be a free, convenient, noninvasive prescreening tool, which could identify those who are likely to be asymptomatic for COVID-19.

App users could log in daily, cough into their phone, and get back information instantly on whether they might be infected. They could confirm the finding by taking a formal test.

"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," says Brian Subirana, PhD, a research scientist in MIT's Auto-ID Laboratory and a co-author of the study.

The algorithm originated in models the MIT team had previously developed to analyze forced-cough recordings to see if they could detect signs of Alzheimer's, one symptom of which is neuromuscular degradation such as weakened vocal cords.

Through a series of further tests, the team determined that the kinds of vocal degradation experienced by Alzheimer's patients was similar to the vocal degradation experienced by those with the COVID-19 virus.