Johns Hopkins surgeon sees promise in AI cancer detection

Dr. Peter A. Najjar, a surgeon and clinical innovation vice president at Johns Hopkins Health System, said AI is beginning to reshape cancer care by helping researchers detect disease earlier, accelerate drug development and reduce administrative work for clinicians. He said the technology is still in its “early days,” but described the progress as “very exciting.”

Najjar pointed to pancreatic cancer research in which AI models found signs of the disease in CT scans before human readers. “Researchers were able to identify signs of pancreatic cancer up to 16 months ahead of human readers,” he said. Pancreatic cancer is often diagnosed after it has spread beyond the pancreas, and the American Cancer Society puts the overall five-year relative survival rate at 13%. When detected before spreading outside the pancreas, the five-year survival rate rises to 44%.

AI is also speeding drug discovery by letting scientists model how potential treatments might interact with cancer-related proteins before laboratory testing. Najjar said AI-powered medical scribes could offer a more immediate benefit by organizing records and documenting visits, giving doctors more time to focus on patients.