There are several ways that COVID-19 can be detected. The usual methods come in the form of medical tests involving swaps, but there are also less invasive methods that one can apply from the comfort of their own home, such as this app developed by researchers at MIT which attempts to analyze the sound of a person’s cough to determine if the person has COVID-19.
The app uses AI which was trained using tens of thousands of cough samples as well as spoken words, and according to the researchers, when they put the model through its paces, it managed to have a successful detection rate of 98.5% from those who had previously been confirmed to have the virus, and a 100% success rate from those who were deemed asymptomatic.
The app isn’t available to the public yet, but the researchers are currently working on turning it into a more user-friendly app and pending FAA approval, could roll it out to the public. This is particularly important because wide-scale testing is key in helping reduce the amount of transmissions, where early detection could limit the number of people the virus spreads to.
As to whether or not this is a more effective method compared to more traditional forms of COVID-19 testing remains to be seen, and chances are it probably shouldn’t be used as a medical diagnostics test. Instead, it could act as a form of a warning, where if it detects you have the virus, you could get yourself tested to be sure, instead of walking around potentially infecting other people.