Clinical Applications of Quantitative Imaging and Artificial Intelligence for Pancreatic Cance

Abstract

Pancreatic cancer remains as a leading cause of cancer death in the United States due to the disease’s deadly combination of evasiveness to detection, aggressive biology, and resistance to treatment. Quantitative imaging and artificial intelligence (AI) methods are emerging as promising and innovative techniques to combat the extensive challenges facing the clinic in the diagnosis and treatment of pancreatic ductal adenocarcinoma. These methods extract data from the fabric of clinical images that allow for earlier diagnosis, improved prognostication, automation of treatment planning, and increased reliability for response assessment. This review examines quantitative imaging techniques from 2013 to 2025 and summarizes them into three parts: differential diagnosis for pancreatic disease, grading and staging of pancreatic tumors, and treatment response assessment and prognosis prediction. We outline key challenges specific to pancreatic cancer and potential mitigations for future direction. We also highlight developing areas such as MRI-guided adaptive radiotherapy, automated target delineation, and integrated radiomic-omics tools that may help incorporate quantitative imaging into routine care of pancreatic cancer. Altogether, the current investigation suggests that quantitative imaging will become an integral tool for this disease across the oncologic journey of a patient.