Researchers at the Mayo Clinic have developed an artificial intelligence system capable of detecting signs of pancreatic cancer up to three years before a clinical diagnosis is made, offering new hope in the fight against one of the world\u2019s deadliest cancers.<\/p>\n
The breakthrough, published in the medical journal Gut<\/em>, could significantly improve survival rates by identifying the disease before tumors become visible on routine CT scans and while curative treatment remains possible.<\/p>\n The AI model, known as the Radiomics-based Early Detection Model (REDMOD)<\/strong>, was tested using nearly 2,000 abdominal CT scans from multiple healthcare institutions. Many of the scans had originally been interpreted as normal but were later linked to patients who eventually developed pancreatic cancer.<\/p>\n According to the study, REDMOD successfully identified 73 percent of these pre-diagnostic cancer cases an average of 16 months before they were clinically diagnosed. Researchers said this represented nearly double the detection rate achieved by medical specialists reviewing the same scans without AI assistance.<\/p>\n The technology proved particularly effective in identifying cancer at even earlier stages. In scans performed more than two years before diagnosis, the AI detected nearly three times as many cancers as traditional reviews.<\/p>\n Pancreatic cancer remains one of the most challenging forms of cancer to diagnose early because symptoms often do not appear until the disease has advanced. More than 85 percent of patients are diagnosed after the cancer has already spread, contributing to a five-year survival rate of less than 15 percent.<\/p>\n Dr. Ajit Goenka, the study\u2019s senior author and a Mayo Clinic radiologist, said the technology has the potential to transform early cancer detection.<\/p>\n \u201cThe greatest barrier to saving lives from pancreatic cancer has been our inability to see the disease when it is still curable,\u201d Goenka said. \u201cThis AI can now identify the signature of cancer from a normal-appearing pancreas, and it can do so reliably over time and across diverse clinical settings.\u201d<\/p>\n The system works by analyzing hundreds of subtle imaging characteristics that describe tissue texture and structure, enabling it to detect biological changes associated with developing cancer before a visible tumor forms.<\/p>\n