A UK study is taking place into whether algorithms can use recordings of coughs, and talking, to detect COVID-19.

The study involves people recording themselves coughing within 72 hours of having a COVID-19 test and then uploading it to a special National Health Service-branded website.

The idea of the study is that "using voice sounds to identify if someone is more likely to need a test could help us find more cases, and help slow the spread of COVID-19."

READ MORE

The 294 COVID-19 vaccines being developed

The IPCC's climate report: The key points

A new temperature record for Europe?

Those taking part are asked to record the sounds of "a forced cough, breathing sounds and a defined sentence."

There is also a warning to be careful when doing the recordings as "one of the voice sounds we will ask you to record is a forced cough – coughing is a potential risk to others around you.

"Please make sure that you only submit your voice sounds after you have left the testing site, when you are alone in either your vehicle or a room. Do not submit your voice sounds if you are in a vehicle or room with any other persons."

The UK's Department of Health and Social Care says the findings of the study into whether artificial intelligence (AI) can detect cases of COVID-19 "will be published in due course."

A study by Massachusetts Institute of Technology academics published last year in the IEEE Open Journal of Engineering in Medicine and Biology concluded that AI techniques "can produce a free, non-invasive, real-time, anytime, instantly distributable, large-scale COVID-19 asymptomatic screening tool to augment current approaches in containing the spread of COVID-19."

It said that "practical use cases could be for daily screening of students, workers, and public as schools, jobs, and transport reopen, or for pool testing to quickly alert of outbreaks in groups."