AI app to tackle TB in Kashmir by listening to coughs of patients

For three decades, Dr Adfar Yaseen's fight against tuberculosis in India 's Jammu and Kashmir region has often been a frustrating one. At one point, it would take a month for a test result to arrive in the territory, the country's northernmost region, where climate and terrain are harsh. But a dramatic change in technology has hugely improved India's chances of eradicating the disease. Now a mobile app is expected to revolutionise early detection in potential patients using artificial intelligence (AI). The app, called Cough Against TB, screens for the disease by comparing a patient's cough with a dataset of nearly 15,000 TB-positive and TB-negative cough sounds. Dr Yaseen, who heads Kashmir’s tuberculosis treatment division, is now overseeing a programme to train doctors and healthcare workers in using the app, which she said is as effective as an X-ray in detecting the disease. “As a screening tool, it is going to be helpful because we have hard-to-reach places and at many remote places we do not have X-rays. So it is going to help us and refine our screening,” Dr Yaseen told The National . “This app is very efficient. As much as we could know through an X-ray, we would know through this app.” The audio inputs are converted to spectrograms used in a network that learns to predict the likelihood of a patient having the disease, according to the Wadhwani Institute for Artificial Intelligence, the company that developed the technology and an AI partner of the Indian government’s Central TB Division. “For every 2.8 TB cases prevalent in the community, one case gets notified and 1.8 cases get missed,” said Wadhwani, which is backed by the United States Agency for International Development. The app is just one part of India’s growing efforts to increase surveillance and testing for TB, a dangerous disease caused by a type of bacteria that spreads through the air when infected people cough, spit or sneeze. The death rate for untreated TB is about 50 per cent, and treatment includes a lengthy and regular medication process. About 1.3 million people died of the disease in 2022, said the World Health Organisation, which lists tuberculosis as the second leading infectious killer after Covid-19. India wants to eradicate TB by 2025 – a big challenge given that of 10.6 million people worldwide who contracted the disease in 2022, 23 per cent were from the country. Rouf Ahmad Tramboo, supervisor of the TB Centre in Kashmir’s Budgam district, said the app works as a first-line detection system, giving a presumptive diagnosis that can be confirmed by more tests. “This app is excellent for a place like Kashmir, where some places are so remote that there is not even electricity available,” he said. It will be distributed among grassroots healthcare workers who will be trained to use it, Mr Tramboo said. “The app generates the result whether the person is presumptive positive or not. At the end there is also an option which lets the healthcare worker decide, on his own analysis and diagnosis, whether a further test is needed or not,” he said. In Kashmir, a picturesque Himalayan valley where winters are bitterly cold and temperatures often plummet below freezing, respiratory ailments and coughs are common, which makes it difficult for people to make immediate self-reporting of tuberculosis. The region has achieved significant success in the fight against TB, with several districts designated free of the disease, while others are showing an improvement that doctors attribute to spacious living conditions and healthy dietary habits. “We have other respiratory ailments here, like chronic obstructive pulmonary disease or asthma, but we do not have tuberculosis to the extent as it is in other parts of the country,” Dr Yaseen said. She said effective screening is needed to further reduce the incidence in the region because the threat of a resurgence of infections is always present. “We just can’t sit back, we have to be vigilant,” she said.

AI app to tackle TB in Kashmir by listening to coughs of patients

Platform 'as effective as an X-ray' in early detection of deadly disease