Harnessing AI for Early Lung Cancer Detection: Qure.ai’s Innovations and Impact

The US Lung Cancer Screening Gap

Despite lung cancer being the leading cause of cancer-related deaths in the US, screening rates remain alarmingly low. “Only about 6% of eligible patients actually get a computed tomography (CT) scan for lung cancer screening, which means 94% are not being screened,” Dr. Shah noted. He attributed this gap to several factors: “Even though eligibility requirements have been relaxed, they’re still quite strict. The number of pack years smoked is high, and enrolling in a lung cancer screening program involves several hurdles.”

Dr. Shah also pointed out systemic challenges, “We’re generally not great at following up and making sure patients get into the right programs. Access is another issue. Some places have tried putting a CT scanner on a bus and bringing it to the community, which has helped, but overall, lung cancer screening rates are still very low compared to other cancers like breast or colon.”

AI-Driven Nodule Detection: Performance and Clinical Integration

Qure.ai’s approach to lung nodule detection is distinguished by its focus on both sensitivity and specificity. “If our AI flagged every rounded spot on an X-ray, radiologists would be overwhelmed. We’re already drowning in films, and we need technology that helps, not hinders,” Dr. Shah said. The goal is not just to identify nodules, but to detect early-stage lung cancers that are most treatable.

A key innovation in Qure.ai’s algorithm development was the use of a vast, diverse dataset—15 million X-rays from around the world—and the reliance on CT-proven nodules as the ground truth. “Instead of relying on radiology reports, which can be inconsistent, we used CT-proven nodules as our golden standard. This approach, which we call ‘super truth,’ makes our algorithm both highly sensitive and specific,” Dr. Shah explained.

Qure.ai also integrates large language models with electronic medical records (EMRs) to extract relevant patient information, such as smoking status and prior imaging findings. “We’ve developed our own large language model that connects with the EMR to pull in key clinical details, like whether the patient is a smoker. This context is crucial for accurate detection and helps radiologists focus on what matters most,” Dr. Shah said.

Enhancing Efficiency and Reporting in Radiology

Beyond X-ray analysis, Qure.ai has developed FDA-approved tools for CT-based lung nodule identification. These solutions automate the measurement of nodules, track their growth over time, and assess malignancy risk based on morphologic features and doubling time. “Our AI can measure every nodule, reconstruct the lungs, and compare nodules across all prior studies to calculate doubling time. This information is then presented in a clear, tabular report for the radiologist,” Dr. Shah described.

He emphasized the impact on workflow: “With this technology, radiologists can work up to 50% faster, and the reports are much more useful for oncologists and pulmonologists. It also helps ensure that slow-growing cancers aren’t missed just because prior studies weren’t reviewed.”

Key Takeaways

• AI can bridge critical gaps in lung cancer screening by identifying early-stage cancers on chest X-rays, especially in populations with low CT screening rates
• Qure.ai’s technology is built on a vast, diverse dataset and uses CT-proven nodules as ground truth, ensuring high sensitivity and specificity
• Integration with EMRs and large language models allows for more comprehensive and context-aware detection, improving clinical decision making
• FDA-approved CT tools from Qure.ai automate nodule measurement and growth tracking, enhancing radiologist efficiency and report quality
• Collaboration between clinicians and AI developers is essential to maximize the benefits of these technologies and improve patient outcomes

As Dr. Shah concluded, “AI isn’t here to replace clinical expertise, but to support it—helping us make better decisions, reduce errors, and ultimately save more lives.”

About Dr. Haque and Dr. Shah

Waqas Haque, MD, MPH, is a Hematology/Oncology Fellow at the University of Chicago. He recently completed his Internal Medicine Residency at New York University (NYU) in a Clinical Investigator Track. Dr. Haque has balanced his patient care work with a variety of research projects. During his fellowship training at the University of Chicago, he plans to further his work in innovative clinical trial design, value-based care delivery to cancer patients, and clinical investigation.

Samir Shah, MD, MMM, FACR, is the Chief Medical Officer at Qure.ai and a practicing radiologist with over two decades of experience. In addition to his medical training, he holds a Master of Medical Management from Carnegie Mellon University and a professional executive certificate in Artificial Intelligence from the Massachusetts Institute of Technology (MIT) Sloan School of Management. Dr. Shah is dedicated to advancing healthcare through technology and is passionate about integrating AI into radiology to improve patient care and outcomes.