Diagnostic challenge in pancreatic cancer

Professor Maxime Ronot, head of the medical imaging, radiodiagnosis, and interventional radiology department at Beaujon Hospital in Clichy (AP-HP), highlights that the main difficulty in diagnosing pancreatic cancer is the absence of a specific risk factor that could help identify high-risk individuals.

AI's role at the Francophone Radiology Days

At the Francophone Radiology Days (JFR), held in Paris from October 8 to 11, discussions focused on recent technical advancements in medical imaging. However, artificial intelligence (AI) did not achieve a satisfactory result in detecting pancreatic cancer, despite the growing incidence of the disease.

Early detection remains a challenge

Because there is no clear risk factor to guide targeted screening, early detection of pancreatic cancer through CT scans or MRI remains largely dependent on chance. This makes it difficult to implement widespread, effective early detection programs.

AI's obstacles in pancreatic cancer imaging

AI-enhanced imaging techniques face multiple challenges in diagnosing pancreatic cancer. These obstacles are not only technical but also include issues related to the complexity of the disease and the variability of its presentation.

Types of pancreatic tumors

Pancreatic tumors are primarily composed of ductal adenocarcinoma, which accounts for 90 percent of cases, with neuroendocrine tumors making up the remaining 10 percent. This distinction is important for understanding the different approaches to diagnosis and treatment.