The study has given hope (Image: yacobchuk via Getty Images)
US researchers have claimed that AI has outperformed radiologists in detecting early-stage pancreatic cancer during trials, with specialists declaring “this is what it looks like when AI adoption is done properly”. A paper published in Gut, part of the British Medical Journal (BMJ), asserts that an AI system is now capable of identifying pancreatic cancer years before any human can spot it on a scan.
Mayo Clinic researchers have unveiled a validated AI model known as Radiomics-based Early Detection Model (REDMOD), which detected pancreatic cancer on routine CT scans with 73% sensitivity, at a median of 475 days prior to clinical diagnosis. Radiologists examining the same scans identified the cancer with just 39% sensitivity.
For scans taken more than two years before diagnosis, the AI proved nearly three times more accurate. The study analysed close to 1,500 scans across several hospitals.
Pancreatic cancer claims the lives of around 92% of those diagnosed in the UK within five years, yet no population screening programme exists for the disease. The Medicines and Healthcare products Regulatory Agency’s (MHRA) dedicated AI medical device framework remains in development, with publication confirmed for 2026, but has yet to be delivered.
The study’s own authors concede that REDMOD has yet to be tested prospectively or across ethnically diverse populations, which will inevitably cause delays. It also requires further investigation for high-risk patients, including those experiencing unexplained weight loss and newly-diagnosed diabetes, before it can be widely adopted in clinical environments.
Colette Mason, author and AI consultant at London-based Clever Clogs AI, welcomed the news: “This is what it looks like when AI adoption is done properly. This is not an ‘AI Booster’ press release about productivity gains or a demo that falls apart in production.
More work is needed (Image: undefined)
Read more: Property lawyer explains when you can cut neighbour’s overhanging branches
Read more: Doctor warns of ‘significant’ lifestyle habit driving cancer increase
“This is a peer-reviewed clinical study, validated across multiple hospitals, tested head-to-head against the professionals it’s designed to support, with the authors themselves being clear and publishing what it can’t do yet, rather than hype.
“REDMOD needs prospective trials in diverse populations before it reaches patients. That takes time, and for a disease with 8% five-year survival, time is the one thing people don’t have.
“But the alternative, rushing an unvalidated tool into screening pathways because the headlines are exciting, is how you get AI systems that erode the clinical trust they depend on. Every organisation deploying AI should be looking at this study and asking why their own adoption process has fewer safeguards than a cancer screening tool that hasn’t even reached patients yet.”
Katrina Young, digital transformation strategist at KYC Digital, cautioned that the trials remained at an early stage and required further development.
She added: “Pancreatic cancer kills over 90% of patients within five years, largely due to late detection. REDMOD changes that on paper. A model identifying cancer around 475 days earlier, with 73% sensitivity versus 39% for radiologists, is not marginal progress.
“But earlier detection does not equal earlier treatment. REDMOD has not yet been prospectively validated or tested across diverse populations.
“The issue is no longer whether AI can detect earlier. It is whether the system can absorb the consequences. Layering AI into those pathways before full validation creates an accountability gap. The MHRA framework is still in development. The breakthrough is technical, but the constraint is structural.”
