Bengaluru: Indian Institute of Science’s (IISc) team Project Coswara is in the process of building a diagnostic tool for COVID-19 based on cough, respiratory and speech sounds.
The tool requires users to use a recording of breathing sounds, cough sounds, vowel sounds and count for five to seven minutes.
The team has so far collected over 1,100 voice samples. The Indian Council of Medical Research (ICMR) had in May reviewed the protocol and asked them to return after collecting samples for validation of the tool.
Signal processing and machine learning techniques are used to analyse the collected data. Then COVID-19 positive and negative patients are distinguished by the sharpness and frequency of the voice.
The team aims to launch the diagnosis tool as a web or mobile application that gives a score to show the probability of infection. The researchers are on a lookout for volunteers — both healthy and COVID-19 positive — for voice samples through the web tool https://coswara.iisc.ac.in.
The team has set target of collecting over 2,000 samples. Healthy individuals’ data will help the team set the right threshold.
IISc assistant professor Dr Sriram Ganapathy, who is leading the project, said the production of sounds is compromised in COVID-19 patients.
“We look for distinctive patterns in cough. Regular patients’ cough is harmonic in nature,” Ganapathy was quoted as saying by Deccan Herald.
He said that the cough has a lot of vibrations in COVID-19 patients. “We analyse signals with computer algorithms and create a model that identifies these vibrations to detect Covid-19,” Ganapathy added.
“A paper published by the University of Oklahoma, University of Michigan and a facility in Ukraine showed accuracy of 95%,” the professor said.