Russian Scientists Have Found a Way to Identify COVID-19 in a Minute
OREANDA-NEWS. The Russian Sberbank Artificial Intelligence Laboratory has developed an algorithm that within a minute can determine the likelihood of coronavirus disease by voice, breathing and coughing. The new development is said in a message from the company.
The message specifies that the developed algorithm within 60 seconds can determine the likelihood of contracting coronavirus infection COVID-19. The presence of the virus is determined based on the results of a short symptom survey and three sound patterns - voice, breathing and cough.
Experts explain that sound files are converted into a spectrogram showing the energy of sound at different frequencies, and then analyzed using a deep convolutional neural network. To train the program, only open data were used - these are more than a thousand samples of breathing and coughing sounds collected from diagnosed patients in Russian clinics.
At the same time, in the near future Russian Sberbank is planning to create a special application with this algorithm, which will become available in the App Store and Google Play.
As the first deputy chairman of the board of Sberbank, Alexander Vedyakhin, noted, the developed model does not yet achieve the accuracy of biological PCR, but now it has comparable characteristics. At the same time, it makes it possible to make customizable sensitivity, much easier to use, more convenient and cheaper. The developers clarify that this is not a medical diagnostic tool, but rather a personal daily diagnostic tool, because taking the test and getting the result takes only 60 seconds.