BrihanMumbai Municipal Corporation (BMC) has begun screening suspected COVID-19 patients with their voice samples through artificial intelligence (AI). Over 100 voice samples of suspected patients at the NESCO jumbo centre have been collected to run the study. Civic officials said it is non-invasive. If the results prove fruitful, then the civic body has plans to start such diagnosis in hospitals for faster results.
Suresh Kakani, additional municipal commissioner, BMC, said that the voice collection will be done in two phases. In the first phase, they will collect 500 samples. They will then run these samples through artificial intelligence and will formulate a specific voice wavelength depending on the analysis of the samples infected with COVID-19, which will help detect patients. Later, a confirmatory test through RT-PCR will be conducted. “The first 500 samples will be used as a confirmatory for analysing the next 1,500 voice samples. We will set the wavelength on which COVID-19 is diagnosed. All samples will be run on the same wavelength. Simultaneously, the RT-PCR of those samples will be done to check the authenticity of the AI-based technology used,” he said.
Voice analysis was used in some European nations, including France and Italy, to detect COVID-19. The organs used for speech —lungs, trachea, larynx (voice box), mouth and nose— are also used for breathing, which is why we sound raspy or ‘stuffed-up’ when we have a cold or flu. “If the pilot shows promise, we plan to use the program in other hospitals as well. It is non-invasive and can deliver results quickly,” he added.
While RT-PCR results take an average of 24 hours (from the collection of samples to results), BMC’s new tests, such as the antigen test, give results within 30 minutes. The rapid antibody blood tests available in the private sector too quickly reveal if the patient has been exposed to the virus. However, there has been a lot of debate over the efficiency of these tests.
A doctor, who is associated with the BMC, said that the voice analysis method isn’t very sensitive. “It could lead to a lot of positives,” he said. While BMC said that the whole voice analysis project was its own initiative, it is learned that an IIT team is helping out.
Dr Neelam Andrade, dean, Nair Dental Hospital and NESCO in-charge, said this would be the first time voice biomarkers will be used to identify COVID-19 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 positive, suspected and negative patients. “We will record the 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 positive for COVID-19. However, if it is above 0.08, it indicates a severe case, requiring immediate treatment,” she said.
Dr KK Agarwal, a cardiologist and former president of Indian Medical Association (IMA) said, “For centuries, doctors have relied on the sound beats of the heart to detect a person through a stethoscope. Now, the same technique can be used to screen suspected COVID-19 patients. This technology may be new, but is being widely used across the globe.”