From Mumbai’s slum clusters to the Mahakumbh Mela, Qure.ai has demonstrated what happens when AI reaches the frontlines: lives are saved, systems become smarter, and healthcare becomes more equitable. “We build AI solutions for medical imaging, impacting oncology, cardiology, tuberculosis, and multiple disease areas. We are at this summit to showcase the impact we have created for 40 million people across 100+ countries,” said Ankit Modi, chief strategy and growth officer, Qure.ai, in an interview with Sudhir Chowdhary. Excerpts:
What makes India uniquely positioned to scale AI-driven surveillance and early detection within its public health system?
India is uniquely positioned for AI applications in public health because of the sheer volume and need for such systems in disease detection. For instance, India conducts around 100 million chest X-rays every year. We apply our AI software on top of these X-rays to detect early signs of tuberculosis, cancer, and other diseases. We are not asking the government to buy new X-ray machines or make large investments — everything is already in place. Adding AI at the backend is what makes these X-rays create immense value.
What does it mean to embed AI into routine public health workflows at the ground level?
We use AI to read chest X-rays and make interventions at the right time. For example, when we deployed AI at the Mahakumbh last year, we observed that in a very small area, many people were crowded and undergoing X-rays — and 3% of those had signs of tuberculosis.
Similarly, in the Brihanmumbai Municipal Corporation, AI-based screening identified many incidental TB cases that would otherwise have been missed. In other words, 20–30% of patients who underwent an X-ray for another problem were found to have TB incidentally. We have demonstrated that lung cancer diagnosis time can be reduced by 50% when AI is applied at the right moment.
In the case of tuberculosis, diagnosis time has dropped from four weeks to just minutes across the world.
In Punjab, Qure.ai’s software detects strokes from CT scans in roughly two minutes, cutting critical waiting time. The AI-driven project is active in several district hospitals and has enabled severe stroke treatment, including complex clot-removal procedures for many patients.
The Indian Council of Medical Research (ICMR) evaluated our solution through a health technology assessment and found that using AI reduces the cost of diagnosing each TB patient by about ₹10,000.
How do you ensure accuracy, accountability, and trust in AI-led health screening?
First, we are very open to generating evidence. This is where peer-reviewed papers in The Lancet or Nature Medicine have helped create trust in our solutions. The second step is obtaining the right regulatory approvals — whether it’s FDA or CE clearance, or Central Drugs Standard Control Organisation (CDSCO) approval in India. You need the right processes in place.
Third, proper auditing for data security is essential — whether compliance with GDPR globally or DPDP in India. We also have systems to monitor model drift. About 5% of X-rays read by AI are also reviewed by radiologists within 24 hours, so we can check for any discrepancies. Continuous monitoring, evaluation, and improvement lie at the heart of our process.
What is the next phase of AI in India’s public health ecosystem?
If we wait for digitalisation to happen first and then think of deploying AI, it will be too late. AI and digitalisation must go hand in hand. In the coming years, governments are setting up electronic medical record (EMR) systems across hospitals, but there isn’t enough workforce to populate them because healthcare workers are already overburdened.
Here, AI can play a crucial role — it can listen to or interpret conversations so that patients and doctors capture context automatically, filling EMR systems seamlessly. This digitalisation will then enable a new wave of AI algorithms trained on vast datasets. Over the next 5–10 years, we will see many more AI solutions developed for India, tailored to its local context.