With the help of a new app developed by the researches of the University of Cambridge that will collect data to develop machine learning algorithms, soon, doctors will be able to find out if a person is coronavirus positive from their voice, breathing and coughing.
According to the European Research Council (ERC), since COVID-19 is a respiratory disease, the sounds made by people, their breathing and coughing will be very specific.
The users of the app will have to take a survey, punch in basic information such as location, medical history, etc. and the give their sound recordings, along with samples of breathing and coughing.
Once a large number of people use the app and there’s a pool of data, it will be shared amongst other researchers which will further analyse the dataset to infer any relationship between respiratory complications due to COVID-19 with a medical history.
The aim of the app is to collect a large amount of data, big enough for the machine to analyse and deduce a pattern if any.
Usually, machines need to be provided with a considerable amount of data on which the algorithms can be run.
Related news: Watch: Mumbai cops tell what they would do in quarantine
Other AI-based approaches to deal and study coronavirus include, COVID-19 Open Research Dataset (CORD-19) which provides information for free regarding all the research that is being done on COVID-19.
COVID-Net is also an open-access neural network, which is being trained by researchers to identify signs of coronavirus in chest x-rays using roughly 6,000 images taken from over 2,800 patients with various lung conditions, including COVID-19, bacterial infections and non-COVID-19 related viral infections.
However, all these AI tools are work in progress and their implementation in the field remains to be seen.