Fig. 3: Establishment of the DL-nomogram, a nomogram integrating NAFNet-classifier and clinical variables.

a Forest plot shows the multivariate logistic regression analyses constructed based on internal set for predicting adverse pathology events. NAFNet-classifier was treated as continuous variable. b A nomogram combining the NAFNet-classifier, biopsy ISUP and clinical T stage. Abbreviations: DL deep learning, PI-RADS Prostate ImagingāReporting and Data System, ISUP International Society of Urological Pathology, PSA prostate specific antigen, OR odds ratio, CI confidence interval, AP adverse pathology.