VR-based automated suturing skill assessment in pediatric robotic surgery.

Area of Science:

• Medical Simulation
• Robotic Surgery
• Pediatric Surgery

Background:

• Objective assessment of robotic surgical skills is crucial in pediatric surgery due to limited training case volumes.
• Virtual reality (VR) offers a promising avenue for surgical skills training and assessment.

Purpose of the Study:

• To present a VR-based framework for automated evaluation of robotic suturing skills in a neonatal surgical scenario.
• To investigate the agreement of this automated assessment with expert video-based evaluation.

Main Methods:

• A real-time VR simulator was created for neonatal robotic suturing using the SmartArm system.
• An automated skills assessment module used an 11-point checklist, with items converted to quantitative geometric and kinematic criteria.
• Automated scores were compared against expert pediatric surgeon evaluations using accuracy, precision, recall, and F1-score.

Main Results:

• The VR simulator facilitated stable real-time robotic suturing and metric extraction.
• The automated assessment achieved 67.3% accuracy, 0.933 precision, 0.560 recall, and 0.700 F1-score compared to expert evaluation.
• Higher agreement was found for objective metrics, with discrepancies mainly in visually subjective criteria.

Conclusions:

• VR-based automated assessment is a feasible method for evaluating robotic pediatric suturing skills.
• The framework provides objective, repeatable performance evaluation, translating checklist items into measurable simulation parameters.
• This approach offers a scalable alternative to manual video-based skills assessment in robotic surgery training.