The Brihanmumbai Municipal Corporation (BMC) has now decided to conduct a 'non-invasive voice sample analysis' of positive and suspected Covid patients, which is expected to detect the SARS-CoV-2 virus in 30 seconds.
Civic officials have proposed to conduct this study, which involves the use of artifcial intelligence (AI), on the 2,000 patients admitted to the NESCO Centre in Worli. According to the dean of the centre, this process entails the use of vocal health research applications to record voices and will prove useful in identifying asymptomatic patients at airports and railway stations.
Dr Neelam Andrade, Dean, Nair Dental Hospital and NESCO in-charge, said this would be the first time voice biomarkers would be used to identify Covid-positive patients within 30 seconds. Currently, this study will be done across India with a sample size of 10,000, of which 2,000 are from Mumbai. The three groups selected for this study will include Covid-positive, suspected and negative patients. “We will record patients' voices by asking them to count from 50 to 70 and these samples will be analysed using artificial intelligence. If the patient's audio point is 0.05, they are covid-positive, but if it is above 0.08, it indicates a severe case, requiring immediate treatment,” she said.
“There is a voice app which can be installed on mobiles or smart laptops. The patient will be asked to recite numbers in front of a breathalyser-like device. The voice samples will automatically get synced with the main server of the provider. Then, through artificial intelligence, the result will be procured within 30 seconds,” Dr Andrade explained.
This study is based on the hypothesis that the infection may affect the voice of individuals, given the interdependence between the respiratory and speech systems in the body. "The organs used for speech — the lungs, trachea, larynx (voice box), the mouth and nose — are also used for breathing, which is why we sound raspy or ‘stuffed-up’ when we have a cold or flu," she said.
“To speak, air from the lungs is pushed out and shaped by our vocal apparatus into specific sounds as it moves toward our mouths in a coordinated process. To simplify somewhat, we can measure the acoustic signatures of distinct vowels, consonants, pitch, and other related aspects in the soundwaves of our emitted speech and, in turn, make predictions about the shape and state of the vocal tract that produced it,” she explained.
Moreover, these voices will be run through wavelengths, which will show three colours -- red, yellow and green -- based on which patients will be categorised as negative, moderate and severe cases. However, for confirmation, the gold standard RT-PCR (reverse transcription polymerase chain reaction) will also be carried out. “As this technology is in the initial stage, we need to cross-check results. So, even if a person has tested negative for the virus, he/she will have to undergo voice analysis to authenticate the technology. It will take 2-3 months to conclude the study,” said Additional Municipal Commissioner Suresh Kakani.