Scientists Develop Artificial Intelligence That Predicts Pancreatic Cancer Risk Three Years Before Diagnosis
An artificial intelligence model developed by Mayo Clinic could help identify people at higher risk of pancreatic cancer up to three years before diagnosis, according to research to be presented at the American College of Surgeons Clinical Congress 2026.
Pancreatic cancer is relatively uncommon but highly deadly. Although it accounts for around 3% of new cancer cases, it causes about 8% of cancer deaths.
Because pancreatic cancer often develops over several years but produces few noticeable signs early on, universal screening is not currently practical. Researchers therefore developed an AI model to identify subtle signs of increased risk before diagnosis.
The model analysed patients’ long-term electronic health records along with results from routine laboratory tests. The study included 6,066 people who developed pancreatic cancer and 33,396 people without the disease. Their medical histories covered between 7.5 and 19 years.
When tested for its ability to predict pancreatic cancer three years before diagnosis, the model achieved an AUROC of 0.853, where 0.5 represents chance and 1.0 represents perfect discrimination. Its precision-recall score was 0.712, indicating an ability to identify people at higher risk while limiting false-positive results.
The model also showed good calibration, meaning its predicted risks were reasonably close to the outcomes observed. Among people given a predicted pancreatic cancer risk of more than 50%, 88% were diagnosed with the disease within one year.
Researchers are now testing the model prospectively to determine whether it works reliably in real-world clinical settings. They also plan to validate it outside the Mayo Clinic system and are developing newer machine-learning approaches.
However, the model is still a research tool and has not been established as a routine screening test. Further prospective and external validation will be needed to determine whether it can accurately identify high-risk patients and ultimately improve early pancreatic cancer detection and outcomes.
REFERENCE: Varghese C, et al. Enabling Digital Screening for Pancreatic Cancer using Artificial Intelligence Analysis of Disease Trajectories. Scientific Forum, American College of Surgeons (ACS) Clinical Congress 2026.
Disclaimer: This website is primarily for healthcare professionals. The content here does not replace medical advice and should not be used as medical, diagnostic, endorsement, treatment, or prescription advice. Medical science evolves rapidly, and we strive to keep our information current. If you find any discrepancies, please contact us at corrections@medicaldialogues.in. Read our Correction Policy here. Nothing here should be used as a substitute for medical advice, diagnosis, or treatment. We do not endorse any healthcare advice that contradicts a physician's guidance. Use of this site is subject to our Terms of Use, Privacy Policy, and Advertisement Policy. For more details, read our Full Disclaimer here.
NOTE: Join us in combating medical misinformation. If you encounter a questionable health, medical, or medical education claim, email us at factcheck@medicaldialogues.in for evaluation.