Sound controls

IISc researchers propose sound-based COVID-19 diagnostic tool

Coswara, a sound-based Covid-19 diagnostic tool developed by the Indian Institute of Science is based on breath, cough and speech sounds.

A team of researchers from the Indian Institute of Science (IISc) in Bengaluru has developed a COVID-19 diagnostic tool that can detect the virus within minutes. Coswara, a sound-based Covid-19 diagnostic tool developed by the Indian Institute of Science is based on breath, cough and speech sounds. According to a report in The Hindu, users can record their voice using a smartphone, tablet or computer on Coswara’s website. The tool will display the probability score once the data is uploaded (whether Covid-19 is positive or not). The probability score is between 0 and 1.

According to Sriram Ganapathy, associate professor in the Department of Electrical Engineering Learning and Acoustic Model Extraction at IISc, participants can use their personal smartphones to log into the website and check their Covid-19 probability score. Basic demographic, symptom and Covid-19 related information is requested on the Coswara website. In addition, the patient’s voice, breathing and cough are recorded. “The participant is redirected to a warning page once the data has been successfully submitted (an activity that takes about 5 minutes). The tool will display the probability score if the user accepts our terms and conditions. All A score above 0.5 indicates a higher chance of testing positive for Covid-19 The entire score calculation is done within one minute after the data is uploaded, resulting in a fast and reliable tool. the tool is offered free to users,” Sriram told the Hindu.

According to the Coswara website, machine learning and signal processing will be used to analyze the collected data. He goes on to say that the goal is to create mathematical models that make it easier to recognize “infection fingerprints” from voice samples. However, they are still working on this step as they are building a data set.

According to the Coswara team’s internal evaluation, the prediction rate is over 90%. To get the tool approved, the team plans to conduct additional clinical trials. The Indian Council of Medical Research (ICMR) and the Drug Controller General of India (DCGI) have not yet approved the tool. However, the data was collected ethically and the IISc approved the research. Knowing what they would be used for, the hospitals provided the project team with the necessary data, reports The Hindu.

The website clarifies that the tool will not replace chemical tests or imaging methods, but will simply supplement existing test methods with a cost-effective, fast and simple technique.