AI-based software to detect COVID-19 from sound of cough developed

Vijay Mohan

Today News Online Service

Mohali, May 12

A novel online tool based on artificial intelligence (AI) that can detect COVID-19 infection in an individual by analysing the sound of coughing has been developed by the Centre for Development of Advanced Computing (C-DAC) in Mohali.

The tool, christened ‘Vaani Mitr’, only requires people to submit a brief recording of coughing through a mobile phone or a laptop, which would then be processed to determine their status in real time. It can be used by anyone at anytime from anywhere.

A large data base is fundamental for AI tools to be effective. For the purpose, a data bank for coughing sounds of both, COVID-19 positive as well as COVID-19 negative persons is being built up by C-DAC, against which the samples are compared.

“COVID-19 primarily affects the respiratory system, which in turn causes subtle changes in the sounds produced during respiration, coughing or speaking,” Dr PK Khosla, Executive Director C-DAC, said. “The acoustic signature of COVID-positive individuals shows some particular characteristics and traits that can be identified and tracked,” he added.

The indigenously developed automated AI system analyses the acoustic signatures to detect the possibility of COVID-19 infection in a human.

So far, C-DAC’s data base for ‘Vaani Mitr’ comprises of over a thousand recordings from COVID-19 negative persons and over 300 recordings from COVID-19 positive persons. The data base has been built with assistance from Nation Health Mission, Haryana and the University of Cambridge, United Kingdom.

“We have to build up our data base further. Large the data base, more accurate the test will be.” Dr Khosla said. As of now, the tool is being looked at as a means of preliminary screening for suspected cases while sitting at home rather than a clinical diagnosis.

C-DAC, which functions under the Ministry of Electronics and Information Technology, is looking for volunteers who are either COVID-19 positive or have been cured to provide their voice and cough samples to build up its data base. This can be done online along with providing some basic medical details.

Use of AI in the healthcare sector, including areas such as hospital information, drug management tele-medicine mobile healthcare and decision support, is among the niche areas C-DAC is working on. The eSanjeevani, a nationwide tele-consultation service developed by it during the lockdown period last year had won the Digital India Award – 2020 in the category of “Innovation in Pandemic”. 

Be the first to comment on "AI-based software to detect COVID-19 from sound of cough developed"

Leave a comment

Your email address will not be published.


%d bloggers like this: