Covid app can accurately diagnose virus by sound of a person’s cough, raising hopes of less invasive testing
2 min readScientists have created an tool that can accurately diagnose Covid-19 just by the sound of a person’s cough.
The software, named Deep Cough 3D, can listen to a person’s cough and predict whether they have the coronavirus with 98 per cent accuracy, researchers at the University of Essex have found.
The academics believe that tool could be rolled out in an app format and used as a less invasive alternative to eye-watering PCR swab tests.
In a study, Deep Cough listened 8,380 audio samples of people coughing in hospitals with different illnesses in Spain and Mexico since April 2020, close to when the virus first began.
The tool then classes how severe a person’s Covid case is, which researchers think will help doctors with ventilation equipment needs and beyond.
Doctors are known to listen to coughs to determine the likes of asthma and pneumonia, but a coronavirus coughs are not always discernible to the human ear.
“We believe the tool has the potential to significantly hamper the Covid-19 pandemic across the world,” the researchers wrote.
Lead researcher Dr Javier Andreu-Perez, of Essex’s Smart Health Technologies Group at the Centre for Computational Intelligence, said: “We are delighted with the promising results of this novel screening tool, which could prove a real game-changer and essential addition to our arsenal of tools to combat the pandemic as it is far less invasive than most other COVID-19 tests and also offers rapid results, paving the way to point-of-need pre-screening testing solutions.”
Similar algorithms have been used in studies in US universities, but this is the first time testing software has been fed clinical-grade sound samples.
“This work highlights new evidence that it is worth pursuing intelligent biometric systems, such us from coughing sounds analysis, as a valuable rapid preliminary screening tool for COVID-19,” Dr Andreu-Perez added.
