Tuberculosis kills around 1.5 million people every year in India and one of the reasons behind this is a late diagnosis of the infectious disease, reports reveal. To deal with this scenario, healthcare start-up Qure.ai has developed software — qXR — using a database of 2.5 million X-rays, which provides fast and accurate diagnosis of Tuberculosis (TB).
Qure.ai’s qXR is designed to screen and prioritise abnormal chest X-rays and individuals suspected of TB, says Prashant Warier, CEO and co-founder, Qure.ai, which was one of the recipients of the India Health Fund TB Quest Awards given for innovations in the area of TB elimination.
“The algorithm automatically identifies 19 most common chest X-ray abnormalities. A subset of these abnormalities, which can diagnose typical and atypical pulmonary TB, is combined to generate a ‘Tuberculosis screening’ algorithm within the product,” he said.
On being asked how it works, he said, the Qure.ai solution qXR automates the chest X-ray interpretation process. “When used as a point-of-care screening tool, followed by immediate bacteriological/NAAT confirmation, Qure.ai significantly enhances the on-site physician’s ability to treat the patient while he or she is still at the clinic. With the usual X-ray and test turnaround times, the patient is often gone — and ‘lost’ to the doctor — by the time the results come in. By that point, an infected patient may have spread the illness to family members and others in the community, not to mention worsening their prognosis with delayed treatment,” said Warier.
Founded in 2016, Qure.ai uses deep learning and Artificial Intelligence (AI) to bring access to quality healthcare, especially in remote areas. “We work closely with NITI Aayog and the Piramal Foundation’s Piramal Swasthya initiative, to leverage AI to enable faster diagnosis of TB in remote areas,” Warier said.
To fulfill the vision of Revised National Tuberculosis Control Programme’s national strategic plan, qXR can aid in Active Case Finding programmes and screening high-risk groups, he said, adding, “With about 10,000 practising radiologists in India, Qure.ai’s AI can help in mass screenings to achieve India’s TB 2025 agenda.”
Since deployment, qXR with the AI-enabled workflow has shortened treatment enrolment time from four days to one day. Previously, the referral time for sputum confirmation of presumptive was two days, which has now been reduced to two minutes.
Meanwhile, Qure.ai has collaborated with PATH, a non-profit global health organisation. The qXR TB solution was deployed at primary healthcare centres and it showcased a successful model for rapid and accurate TB screening. “To date, we have deployed our AI algorithms with digital chest X-rays. Now with the IHF grant award, we would be taking the AI-enabled TB screening workflow to centers with Analog X-ray systems, bringing AI interpretation of chest X-rays to analog X-ray films, a first from India and elsewhere, adding a new tool in the fight against Tuberculosis.”