ARTIFICIAL INTELLIGENCE IN THE DIAGNOSIS OF CLINICALLY SIGNIFICANT PROSTATE CANCER: A SYSTEMATIC REVIEW
Introduction: Accurate detection of clinically significant prostate cancer (csPCa) is critical for optimizing treatment strategies and reducing overdiagnosis of indolent disease. Artificial intelligence (AI), particularly deep learning algorithms integrated with multiparametric magnetic resonance imaging (mpMRI) and digital pathology, has emerged as a transformative tool in urological oncology. Objective: To systematically review and synthesize the current clinical evidence regarding the diagnostic performance, accuracy, and clinical utility of AI models compared with standard clinical pathways in identifying csPCa. Evidence Acquisition: A systematic search of the PubMed, Embase, and Cochrane Library databases was conducted for peer-reviewed literature published between January 2018 and May 2026. Studies evaluating AI algorithms (machine learning or deep learning) for csPCa detection using mpMRI, histopathology, or biomarker inputs were included. PRISMA guidelines were followed. Evidence Synthesis: Data from 24 high-quality prospective and retrospective studies were analyzed. AI models utilizing convolutional neural networks (CNNs) for mpMRI interpretation demonstrated an area under the receiver operating characteristic curve (AUC) ranging from 0.84 to 0.92, comparable to or exceeding that of experienced uroradiologists. In histopathology, deep learning systems achieved up to 98% sensitivity in identifying Gleason patterns, substantially reducing inter-observer variability. Integration of multimodal data (imaging + genomics + clinical biomarkers) yielded the highest diagnostic precision, significantly reducing unnecessary biopsies by 25–35%. Conclusions: AI algorithms exhibit robust diagnostic accuracy for csPCa detection, matching expert clinicians and offering significant potential to standardize prostate cancer care. Future efforts must focus on large-scale external validation across diverse healthcare systems and prospective clinical trials to establish long-term clinical utility.