Diagnostic performance of bpMRI versus mpMRI and AI-assisted bpMRI in prostate cancer detection: a multi-reader study.
Authors
Affiliations (4)
Affiliations (4)
- Fondazione Policlinico Universitario A. Gemelli IRCCS, Largo Agostino Gemelli 8, 00168, Rome, Italy. [email protected].
- Fondazione Policlinico Universitario A. Gemelli IRCCS, Largo Agostino Gemelli 8, 00168, Rome, Italy.
- Università Cattolica del Sacro Cuore, Rome, Italy.
- Lucida Medical Ltd, Cambridge, UK.
Abstract
To compare biparametric MRI (bpMRI) and multiparametric MRI (mpMRI) for detecting clinically significant prostate cancer (csPCa), and to assess the impact of artificial intelligence (AI)-assisted bpMRI on diagnostic performance and biopsy-related outcomes in readers with different expertise. In this retrospective multi-reader study, 173 men referred for prostate mpMRI were evaluated by five radiologists (two experts, three basic readers) who scored bpMRI and mpMRI using PI-RADS v2.1. After a 45-day wash-out, the same readers reinterpreted bpMRI with concurrent AI decision support (available for 127 cases). Diagnostic performance for csPCa (ISUP ≥ 2) and benefit-to-harm ratios were compared across protocols and reader groups. Mean area under the receiver operating characteristic curve was 0.820 for bpMRI and 0.819 for mpMRI (difference 0.001), indicating comparable diagnostic performance. In the AI subset, AI increased mean specificity from 59.7% to 67.8% while reducing sensitivity from 90.2% to 84.5%. Biopsy selectivity and efficiency improved (3.9 to 4.6 and 1.7-1.9), particularly in basic readers. The effect was significant in basic readers (McNemar p = 0.012 for sensitivity, p < 0.001 for specificity) but not in experts. bpMRI showed comparable performance to mpMRI for csPCa detection in routine practice. AI-assisted bpMRI improved biopsy efficiency in less-experienced readers at the cost of reduced sensitivity, warranting careful consideration of the trade-off between missed cancers and avoided unproductive biopsies.