NHS study probes whether algorithms can detect Covid from cough recordings

Government seeks volunteers for research

Credit: Jernej Furman/CC BY 2.0

The government is conducting research that seeks to find out whether coronavirus infection could be detected by algorithms to examine coughing and speech recordings.

Via the NHS Test and Trace programme, volunteers are being sought to use their smartphone to record the sound of a “forced cough, breathing sounds and a defined sentence”.

“We will explore how algorithms can use voice data, including cough recordings, to detect Covid-19,” the government said. “By identifying if someone is more likely to need a test by using voice sounds, we can find more cases, help slow the spread of COVID-19 and protect our health service.”

Participants must have arranged to take a Covid-19 test at one of a selected number of test sites, and the recording must then be made and submitted within 48 hours of the swab being taken. Anyone interested in taking part is asked to use a smartphone or tablet – not a laptop or desktop computer – to visit www.ciab2021.co.uk.


Related content


After being asked to accept a privacy notice and answer a few questions concerning “things can affect how your voice sounds”, such as other health issues or first language, users will then be directed on how to make and submit their recordings.

Participants are advised that the study will not provide them with any medical, nor is any payment offered. 

“However, the information you contribute may be useful in the fight against Covid-19,” the government added.

Recordings should only be made in solitude in a closed room or vehicle, and participants are reminded that “coughing is a potential risk to others around you”.

The study has been approved by the Data Ethics Committee overseen by National Statistician Sir Ian Diamond, as well as by the Cambridge South NHS Research Ethics Committee. 

“Findings of the study will be published in due course.”

 

Sam Trendall

Learn More →

Leave a Reply

Your email address will not be published. Required fields are marked *

Processing...
Thank you! Your subscription has been confirmed. You'll hear from us soon.
Subscribe to our newsletter
ErrorHere