Mayo Clinic researchers have developed an artificial intelligence (AI) system that could change the way pancreatic cancer is detected by identifying subtle signs of the disease, usually years before it is diagnosed. The technology, known as the Radiomics-Based Early Detection Model (REDMOD), analyzes routine CT scans to detect subtle changes in pancreas tissue that are invisible to the human eye. Published in the journal Gut in April 2026, the landmark study found that AI could identify many future pancreatic cancer cases long before tumors became visible on imaging. Although REDMOD is still undergoing clinical evaluation, researchers believe it could become an important tool to diagnose one of the world’s deadliest cancers at a much earlier stage.
What is AI breakthrough in pancreas cancer detection ?
The breakthrough focuses on REDMOD, an AI model developed by Mayo Clinic researchers that analyzes routine contrast-enhanced CT scans for hidden radiomic signatures associated with pancreatic cancer. Unlike traditional imaging methods that rely on detecting existing tumors, REDMOD identifies small structural and textural changes within pancreatic tissue that may appear months or even years before cancer becomes visible.The findings, published in Gut, showed that AI correctly identified 73% of future pancreatic cancer cases, and detected the disease 16 months (475 days) before clinical diagnosis. In some patients, the AI detected warning signs up to three years before diagnosis, almost double the detection rate achieved by radiologists reviewing the same scans. Researchers believe it could eventually replace routine abdominal CT scans into an early warning system for pancreatic cancer, although the technology has not yet been approved for routine clinical use.
Why is pancreas cancer so difficult to detect?
Pancreatic cancer is one of the deadliest forms of cancer because it rarely causes symptoms during its early stages. The pancreas is located deep inside the abdomen, making small tumors difficult to detect using conventional imaging. By the time symptoms such as abdominal pain, jaundice or unexplained weight loss develop, the disease has often spread beyond the pancreas.Researchers estimate that more than 85% of patients are diagnosed when the cancer has already advanced, leaving few treatment options. Current five-year survival rates remain below 15%, making early diagnosis one of the greatest unmet needs in cancer care. Detecting the disease before symptoms appear can greatly improve the patient’s chances of receiving potentially curative treatment.
How the REDMOD AI model works
Unlike traditional computer-aided detection systems that search for visible tumors, REDMOD uses radiomics, a technology that converts medical images into hundreds of quantitative measurements describing tissue texture, density, size, and microstructural patterns. The AI automatically isolates the pancreas on routine contrast-enhanced CT scans before analyzing these hidden imaging features using machine-learning algorithms.Rather than looking for an existing tumor, REDMOD searches for biological changes occurring within the pancreatic tissue that occur long before the tumor is visible. According to the researchers, these microscopic imaging signatures represent early tissue remodeling associated with pancreatic cancer development.The biggest advantage of REDMOD is that it does not require patients to undergo special scans. Instead, it analyzes routine CT scans that people have already had for unrelated medical conditions, such as stomach pain, digestive disorders or kidney stones. Researchers believe this could allow hospitals to identify high-risk individuals without exposing them to additional radiation or imaging procedures.
Understanding Radiomics
Radiomics is an emerging field that transforms ordinary medical images into large amounts of measurable data. Instead of relying solely on what the radiologist can see, special computer software extracts hundreds or even thousands of quantitative features from each CT image.These features describe subtle differences in tissue texture, size, density, and spatial organization that cannot normally be seen by the human eye. Machine-learning algorithms then analyze these patterns to determine whether they resemble healthy tissue or early biological changes associated with disease.In the REDMOD study, researchers initially extracted about 1,000 radiomic features from each CT scan before selecting the most informative features to create the final AI model. The strongest predictive signals came from wavelet-filtered images, which enhance microscopic tissue features invisible during routine clinical review.
What the study found
The researchers trained and validated REDMOD using approximately 2,000 abdominal CT scans collected from multiple health care institutions. Many of these scans were originally reported normal but were from patients who were later diagnosed with pancreatic cancer.During validation, the AI correctly identified 73% of future pancreatic cancers, and detected them approximately 16 months (475 days) before clinical diagnosis. In scans obtained more than two years before diagnosis, REDMOD identified nearly three times as many cancers as radiologists reviewing the same scans without AI assistance.The researchers also found that REDMOD maintained consistent performance across CT scans acquired using different scanner manufacturers, imaging protocols, and health care institutions. This suggests that the model can be integrated into a wide range of hospitals rather than being limited to a single imaging system or clinical setting.
Why could this be a huge success?
The biggest challenge in treating pancreatic cancer has always been detecting the disease while it is still curable. By identifying subtle tissue changes before tumors become visible, REDMOD can provide doctors valuable time to investigate suspected cases and begin treatment earlier.Explaining the significance of the findings, Dr. Ajit Goenka, a Mayo Clinic radiologist and one of the study’s lead researchers, said the technology identifies imaging biomarkers before tumors are visible, providing an opportunity to intervene when treatment is likely to be far more effective.The researchers also emphasized that REDMOD is designed to support physicians rather than replace them. Instead of making the diagnosis itself, AI can help radiologists recognize subtle warning signs that might otherwise go unnoticed.The researchers further suggested that REDMOD could eventually act as an opportunistic screening tool, automatically analyzing routine CT scans performed for unrelated medical conditions and alerting physicians when hidden imaging features indicate an increased risk of pancreatic cancer.
How is it different from liquid biopsy?
Another promising strategy for early detection of pancreatic cancer is liquid biopsy, which analyzes blood samples for tumor DNA, RNA, proteins, and other cancer-related biomarkers released into the bloodstream.Unlike REDMOD, which analyzes medical images, liquid biopsy searches for molecular evidence of cancer. Researchers believe that the two technologies may complement each other rather than compete. Combining AI-powered radiomics with blood-based biomarkers may improve diagnostic accuracy and enable even earlier detection in the future.
Is AI available to patients?
not yet. Although the findings are highly encouraging, REDMOD remains an investigational technology and is not approved for routine clinical use or population screening.To determine whether AI improves patient outcomes in real-world health care, Mayo Clinic has launched the AI-PACEd (Artificial Intelligence for Pancreatic Cancer Early Detection) prospective clinical trial. The study will evaluate REDMOD among people at increased risk of pancreatic cancer, measuring its impact on diagnosis, treatment decisions, and long-term outcomes.
What did the researchers conclude?
The researchers concluded that REDMOD demonstrates a strong ability to identify pancreatic cancer during its pre-diagnostic stage by detecting radiomic signatures that remain invisible during routine CT interpretation. However, he stressed that prospective clinical trials are still needed to determine whether integrating the technology into everyday health care can improve survival and reduce deaths from pancreatic cancer.According to the Mayo Clinic, REDMOD has the potential to transform routine abdominal CT scans into an early warning system for pancreatic cancer by identifying hidden imaging biomarkers before tumors are visible. If validated through future clinical studies, this technology could allow doctors to screen suspected cases much earlier than currently.
A promising step towards early diagnosis
The REDMOD study highlights how artificial intelligence is moving beyond simply recognizing visible abnormalities and beginning to detect early biological fingerprints of disease. Rather than replacing radiologists, the technology has the potential to enhance routine imaging by revealing warning signs that human observers cannot see.While REDMOD is not yet part of standard clinical care, researchers believe it represents one of the most promising advances in pancreatic cancer detection in recent years. If ongoing clinical trials confirm the findings, routine CT scans performed for unrelated medical conditions could one day become powerful tools to identify pancreatic cancer before symptoms develop, giving thousands of patients a better chance of receiving life-saving treatment.