In India, lung cancer is often discovered only when the disease has already run its course — when the cough has lingered too long, when breathlessness finally becomes impossible to ignore. By then, the window for early treatment has usually closed. Subtle signs on routine X-rays are missed, overwhelmed by high patient volumes and fragmented referral pathways.
At Yashoda Hospital, Hitec City, an AI-enabled Lung Nodule Clinic, developed in collaboration with clinicians, Qure.ai, and AstraZeneca aims to change this reality by helping doctors identify abnormalities far earlier and far more consistently. The initiative marks a shift from reactive detection to proactive lung-health surveillance, particularly for high-risk and often-overlooked patients.
“Early lung-cancer detection in India has historically relied on specialist interpretation and multiple referral steps, which delay diagnosis and increase the chances of missed cancers. AI brings expert-level interpretation to the point of care within seconds, transforming the pathway from reactive to proactive,” says Sagar Sen, SVP, Global Life Sciences and Strategic Alliances at Qure.ai.
He explains that the AI model, trained on more than 10 million diverse X-rays — including data reflecting India’s varied comorbidities and imaging conditions — can spot subtle nodules, faint opacities, and irregularities that may escape even experienced eyes. He further adds, “One of our retrospective studies showed that AI could flag nodules years before a formal diagnosis was made. This is where AI becomes transformative — not by replacing clinicians, but by adding a systematic layer of vigilance.”
This systematic approach is exactly what the lung-nodule pathway at Yashoda Hospital aims to build. The clinic is jointly led by Dr Nagarjuna Maturu, Senior Consultant & Clinical Director of Pulmonary Medicine, and Dr Vipul Garg, Consultant, Pulmonary Medicine — both of whom emphasise that their coordinated workflow underpins the clinic’s success.
Dr Nagarjuna shares, “We’ve been working together on this structured pathway from day one, so every AI-flagged case is reviewed jointly. AI consistently highlights even faint nodular densities that might not be immediately visible on a standard X-ray. This additional layer of vigilance has made a significant difference.”
Over the last three months alone, their team has identified around 100 additional lung-abnormality cases because of AI alerts — several of which were later diagnosed as early malignancy, tuberculosis, sarcoidosis, interstitial lung disease, and lymphoma. Many of these findings were purely incidental, detected in patients who had come in for unrelated issues.
“We’ve analysed more than 17,000 X-rays since deployment. AI flagged 960 nodules, including 136 high-risk ones. Many of these patients wouldn’t have entered the cancer pathway at all if not for these incidental detections,” adds Dr Vipul.
For patients presenting with vague symptoms — persistent cough, fatigue, or mild chest discomfort — AI has emerged as an important first filter. Dr Nagarjuna emphasises, “Often, these X-rays appear normal at first glance. With AI, subtle abnormalities are flagged instantly. We receive an alert within a minute, and if needed, a CT scan is done the same day.”
This instant escalation has dramatically reduced diagnostic delays. The time from a flagged nodule to a final diagnosis now averages 48–72 hours, compared to the weeks or months these delays typically stretch across India.
For a high-volume hospital like Yashoda, where specialist time is limited and caseloads are heavy, the technology also brings structure. “AI provides prompts that direct attention to high-value findings. Radiologists, pulmonologists, and oncologists now begin discussions with a shared, risk-aligned starting point. It strengthens collaboration across the team,” highlights Dr Vipul.
Sagar points out that explainability was intentionally built into the system for exactly this reason. He adds, “We never provide a black-box output. Every finding has heatmaps, localisation, and confidence scores. Clinicians understand why the AI flagged a case, and they remain fully in control.”
However, both doctors stress that while AI strengthens consistency, reduces variability, and streamlines triage, it also has clear boundaries. “AI cannot replace clinical judgement. Poor-quality X-rays still pose limitations, and the real benefit comes only when hospitals invest in structured pathways,” highlights Dr Nagarjuna. Meanwhile, Dr Vipul, too, echoes this caution: “Without a strong follow-up protocol, even the best AI insights won’t translate into outcomes. The pathway matters as much as the technology.”
Yet, there is little doubt that AI-driven workflows are reshaping lung-health delivery in India. With rising med-tech investments, greater digitisation, and collaborative models like Yashoda’s nurse-navigator system, clinicians finally have tools to detect lung cancer far earlier than before.