Covid-19, vaccinations remain crucial, equally so are tools that enable early diagnosis of the disease. On this front, there may soon be a decisive new ally.
A smartphone app that can detect infection by voice. The project is by an Australian company specializing in digital health care, but the first results seem so satisfactory that it convinced the pharmaceutical company Pfizer to invest 116 million to buy it.
Thanks to the use of a new Artificial Intelligence model, starting from simple voice recordings the Application is able to detect Covid infection with an 89% accuracy rate.
In addition, its use is faster and, most importantly, less expensive than rapid antigen texts, which are also very difficult to distribute in poorer countries.
A new app could accurately detect COVID-19 infection through people's voices with the help of artificial intelligence.
The project is by an Australian company specializing in digital health care, but the early results seem so satisfactory that it convinced the pharmaceutical company Pfizer to invest 116 million to buy it.
COVID-19 infection usually affects the upper respiratory tract and vocal cords, leading to changes in a person's voice. Aljbawi, Sami Simons, a pulmonologist at Maastricht University Medical Center, and Visara Urovi, also of the Institute of Data Science, decided to investigate whether it would be possible to use an AI to analyze voices to detect infection.
The two professionals used data from the University of Cambridge's COVID-19 Sounds app, which contains 893 audio samples from 4,352 healthy and unhealthy participants, 308 of whom tested positive for COVID-19.
Operation is extremely simple: once installed on the cell phone, the App requires you to enter some personal data and record some breathing sounds such as coughing, taking a deep breath, reading a short sentence. The developed algorithm will then do the rest.