Sign In

Communications of the ACM

ACM TechNews

AI Model Detects Asymptomatic Covid-19 Infections Through Cellphone-Recorded Coughs

View as: Print Mobile App Share:

Massachusetts Institute of Technology researchers have found that people who are asymptomatic for Covid-19 may differ from healthy individuals in the way that they cough, which can be captured by artificial intelligence.

Credit: Christine Daniloff/MIT

Massachusetts Institute of Technology (MIT) researchers have developed an artificial intelligence model that differentiates between asymptomatic people infected with Covid-19 and healthy individuals via forced-cough recordings submitted through Web browsers, cellphones, and laptops.

The MIT team trained the model on cough samples and spoken words; it accurately identified 98.5% of coughs from people confirmed to have the virus (100% from those who are asymptomatic) when fed new cough recordings.

The researchers are incorporating the model into a user-friendly application which could potentially be a free, convenient, noninvasive prescreening tool to identify asymptomatic people infected with the virus. Users could log in daily, cough into their handset, and instantly receive information on whether they might be infected and confirm with a formal test.

From MIT News
View Full Article


Abstracts Copyright © 2020 SmithBucklin, Washington, DC, USA


No entries found

Sign In for Full Access
» Forgot Password? » Create an ACM Web Account