AI in Pancreatic Cancer: Moving From Promise to Practical Application

The full article for this interview is available on Oncology Data Advisor. Check it out here: https://bit.ly/4n5aU6K Artificial intelligence is generating excitement in oncology, but its real-world clinical impact remains inconsistent. In this episode of Exploring AI in Oncology, Dr. Ashish Manne, Gastrointestinal Medical Oncologist at The Ohio State University Comprehensive Cancer Center, shares his AACR presentation on AI-driven pathological subtyping in pancreatic cancer. Dr. Manne discusses how his team is moving AI from theoretical promise to practical application by using digital pathology to classify pancreatic tumors into classical and basal subtypes—categories with significant therapeutic implications. He explains how this accessible approach could democratize precision medicine globally, reducing reliance on expensive genomic testing. The conversation also addresses the barriers to AI adoption in clinical decision-making, including the "black box" problem of trust and transparency, and why current AI applications are primarily focused on workflow optimization. Dr. Manne advocates for a physician-led, specialty-specific approach to model development to ensure AI meets the real-world needs of clinicians and patients.