Maties machine-learning tool coughs up the truth about Covid-19

Every time someone in your family or social circle coughs, you wonder if they have Covid-19.

If that describes you, you’re not alone: one of the hardest things about the airborne coronavirus that has wreaked havoc in the world is that one of its main symptoms is exactly like that of a common cold or even just a tickle in the throat.

Now, researchers at the University of Stellenbosch have come up with a cough analyser so lay people are able to distinguish between a “normal cough” and a Covid-19 cough.

The machine-learning tool makes use of scientific data that has shown Covid-19 coughs are 15% to 20% shorter than non-Covid-19 coughs.

Prof Thomas Niesler, from the department of electrical and electronic engineering at Stellenbosch, said the tool can “distinguish Covid-19-positive coughs from Covid-19-negative coughs and healthy coughs that have been recorded on a smartphone”.

He and the team’s research was recently published in the journal Computers in Biology and Medicine, and they used global and national data sets. 

The two data sets consist of coughing sounds, gathered from all six continents, recorded during or after the acute phase of Covid-19. 

For both data sets, participants were asked to provide cough recordings via a web-based data collection platform using their smartphones.

“Our analysis of the recordings shows that Covid-19-positive coughs are 15% to 20% shorter than non-Covid-19 coughs,” said co-researcher Dr Madhurananda Pahar.

“Since this type of cough audio classification is cost-effective and easy to use, it is potentially a useful and viable means of non-contact Covid-19 screening.

“Data has been captured on smartphones, and our classifier can also be implemented on these devices. Furthermore, it could be applied remotely, thus avoiding contact with medical personnel.”

The researchers said the systems they describe in their study require more stringent validation on a larger data set, but the results are promising and indicate that Covid-19 screening based on automatic classification of coughing sounds is viable.

However, they added: “The technology being developed should not be seen as an official form of testing, nor will it replace testing being done at accredited testing sites.”