Scientists and researchers have developed an AI system that can successfully predict the prevalence of the coronavirus by the sound of a person’s cough. Researchers at Massachusetts Institute of Technology said that this new algorithm can detect viruses by analyzing the way people speak or cough as the virus causes muscle impairment which affects the quality of the sound.
In a paper published in the IEEE Journal of Engineering in Medicine and Biology, the researchers stated that they tested the sound recordings of 4,000 people who were forced to cough. This new AI system was able to identify someone who had tested positive for COVId-19, up to 98.5 accuracy. With asymptomatic persons, (those who have been confirmed to have the virus but report no symptoms), the accuracy rose to 100 percent.
Co-author Brian Subirana, a research scientist at MIT’s Auto-ID Laboratory told MIT News, “The effective implementation of this group diagnostic tool could diminish the spread of the pandemic if everyone uses it before going to a classroom, factory, or a restaurant.”
The team is collaborating with a company and hope to develop it as an app that can be used as a pre-screening tool for pandemics.
Source: MIT News, NY Post