PSMA-PET deep learning score plus clinical factors improves recurrence prediction after prostatectomy

A preoperative model integrating a PSMA-PET deep learning score with D'Amico classification and SUVmax improved prediction of biochemical recurrence-free survival after radical prostatectomy (C-index 0.846 training, 0.806/0.774 external validation).

• Retrospective multicohort study in 697 prostate cancer patients (training n=445, validation n=190 and 62) from a single healthcare network.
• The best deep learning network (VGG19) achieved 3-year recurrence AUCs of 0.834, 0.755, and 0.723 in training and two validation cohorts.
• Limitation: retrospective, single-network design with small external validation cohorts; no prospective validation reported.

RadPigeon summaries are original and for information only. They are not clinical advice.