© 2020 Texas Public Radio
Real. Reliable. Texas Public Radio.
Play Live Radio
Next Up:
0:00
0:00
Available On Air Stations

Detecting COVID-19 Through Sound

Images from researchers at Cambridge University show how algorithms look for signs of COVID-19 in audio samples of coughs. (Courtesy of Dimitris Spathis)
Images from researchers at Cambridge University show how algorithms look for signs of COVID-19 in audio samples of coughs. (Courtesy of Dimitris Spathis)

A rising tide of COVID-19 cases is putting pressure on supplies of tests for the novel coronavirus. Now, some scientists think they might have found a way to relieve some of that pressure. They say they can narrow down who should be tested by using sounds hidden in human vocal cords.

Brett Dahlberg with IEEE Spectrum reports.

Resources From The Segment:

  • Dimitris Spathis is a doctoral researcher at the University of Cambridge’s Department of Computer Science and Technology. He’s an author of a paper on AI to diagnose respiratory diseases using crowdsourced audio. They’re getting their audio samples from user submissions on an app called COVID-19 Sounds.
  • This fall, researchers put together an online conference focused on automated diagnosis of COVID-19 sounds. It featured eight research teams from three continents.
  • David Liu is the CEO of Sonde Health, which is using its app called Sonde One to detect signs of COVID-19 in people’s voices.
  • Heather Mattie lectures on biostatistics and directs the Health Data Science master’s program at Harvard University’s T.H. Chan School of Public Health.

This article was originally published on WBUR.org.

Copyright 2020 NPR. To see more, visit https://www.npr.org.