Prediction of Robotic Anastomosis Competency Evaluation (RACE) metrics during vesico-urethral anastomosis using...

Area of Science:

• Medical technology
• Surgical training
• Neuroscience

Background:

• Vesico-urethral anastomosis (VUA) is a critical step in robot-assisted radical prostatectomy (RARP).
• Objective assessment of VUA skills is crucial for surgical training and patient outcomes.
• Current assessment methods may lack objectivity and rely on subjective rater interpretation.

Purpose of the Study:

• To develop objective prediction models for Robotic Anastomosis Competency Evaluation (RACE) metrics.
• To utilize electroencephalogram (EEG) and eye-tracking data for performance prediction.
• To enhance the training of RARP anastomosis by providing objective feedback.

Main Methods:

• Collected EEG and eye-tracking data from 23 participants performing robot-assisted VUA on models and tissue.
• Extracted EEG and eye-tracking features and assessed subtask performance using the RACE tool.
• Developed Random Forest Regression (RFR) and Gradient Boosting Regression (GBR) models to predict RACE scores.

Main Results:

• Significant differences in performance scores were observed across skill levels (inexperienced, competent, experienced).
• High R-squared values were achieved in predicting RACE scores for plastic anastomoses (e.g., 0.80 for needle driving/tissue trauma).
• R-squared values for tissue anastomoses ranged from 0.62 to 0.76, demonstrating model efficacy.

Conclusions:

• Objective prediction models using EEG and eye-tracking data can accurately assess VUA performance.
• These models offer valuable, objective feedback for surgical trainees in RARP.
• The developed models have the potential to significantly improve RARP anastomosis training and outcomes.