Artificial Intelligence and Biparametric MRI in Prostate Cancer Detection

J. Twilt

• Promotor: J. Fütterer and H. Huisman
• Copromotor: M. de Rooij
• Graduation year: 2026
• Radboud University

Abstract

This thesis aims to address key gaps towards the clinical application of AI and bpMRI for prostate cancer diagnosis. First, it examines whether AI systems, when trained on large and high-quality datasets, can achieve performance non-inferior to, or exceeding, radiologists for csPCa diagnosis. Second, it investigates whether integrating such AI systems into clinical workflows can maintain or improve diagnostic accuracy while alleviating radiologist workload. Finally, it investigates whether bpMRI can achieve non-inferior performance compared to mpMRI in the diagnosis of csPCa, supporting its potential as a streamlined alternative in prostate cancer imaging.