- Mayo Clinic's REDMOD AI model detected 73% of pancreatic cancers on routine CT scans that radiologists originally read as normal.
- The AI flagged cancers at a median of 16 months before clinical diagnosis, nearly doubling specialist detection rates.
- A clinical trial called AI-PACED is now testing the model on high-risk patients in real care settings.
REDMOD identifies cancer signatures invisible to human specialists
Mayo Clinic published results in the British Society of Gastroenterology's journal Gut showing that its Radiomics-based Early Detection Model can identify pancreatic cancer on CT scans that were originally interpreted as completely normal. The model analyzed nearly 2,000 scans and flagged 73% of cases that would later become clinical diagnoses. Specialists reviewing the same images caught 39%.
The gap is not a failure of radiology. Pancreatic cancer at its earliest stages produces no visible mass. REDMOD works by measuring hundreds of quantitative imaging features that describe tissue texture and structure, capturing faint biological shifts as malignancy begins to form. These patterns exist below the threshold of human perception but within the reach of computational analysis.
According to the Mayo Clinic News Network, the model performed strongest on scans taken more than two years before diagnosis, identifying nearly three times as many early cancers compared to specialists in that window.
Why early detection reshapes pancreatic cancer survival math
Pancreatic cancer has a five-year survival rate below 13%, primarily because most cases are diagnosed after the disease has already spread. The median survival for late-stage diagnosis is measured in months. Every month of earlier detection shifts the probability of surgical eligibility, which remains the only curative intervention.
REDMOD does not require new imaging infrastructure. It analyzes CT scans already being performed for other clinical reasons, particularly in patients flagged by risk factors like new-onset diabetes. This means deployment does not depend on new screening protocols or additional patient burden.
Mayo Clinic has now launched the AI-PACED clinical trial to evaluate how clinicians can integrate REDMOD into real workflows for high-risk populations. The trial will measure early detection rates, false positive burden, and downstream clinical outcomes.
The model that catches cancer three years early on scans already being taken does not replace oncologists. It gives them something they have never had for this disease: time.
Santage is committed to independent, transparent journalism. This article is produced in accordance with Santage's Editorial Standards and aims to provide accurate and timely information. Readers are encouraged to verify information independently.