Artificial Intelligence-Based Objective Evaluation of Nerve-Sparing Status Combined With Preserved Urethral Length...

Area of Science:

• Urology
• Surgical Technology
• Artificial Intelligence in Medicine

Background:

• Robot-assisted radical prostatectomy (RARP) is a common treatment for prostate cancer.
• Achieving nerve-sparing during RARP is crucial for urinary continence recovery.
• Objective assessment of nerve-sparing status is challenging with current methods.

Purpose of the Study:

• To objectively evaluate nerve-sparing status using artificial intelligence (AI) on intraoperative images.
• To determine if combining AI-assessed nerve preservation with urethral length improves prediction of urinary continence recovery post-RARP.

Main Methods:

• Analysis of 195 intraoperative still images from RARP.
• A deep learning classification model was used to assess neurovascular bundle preservation.
• Patients were grouped based on AI-assessed nerve preservation and urethral length (≥26 mm).

Main Results:

• AI reclassified nerve preservation in 30.8% of cases, identifying surgeon over/underestimations.
• Patients meeting both AI nerve preservation and urethral length criteria showed significantly better continence recovery.
• The combined factor independently predicted continence recovery at 60, 90, and 365 days.

Conclusions:

• Objective AI evaluation of nerve-sparing status enhances prediction of urinary continence recovery.
• Combining AI nerve assessment with urethral length is a valuable predictor of early continence post-RARP.
• This AI-driven approach may serve as an indicator of surgical quality in RARP.