Automatically rating trainee skill at a pediatric laparoscopic suturing task.

Area of Science:

• Medical Education
• Surgical Technology
• Biomedical Engineering

Background:

• Pediatric minimally invasive surgery requires complex skills, posing challenges for trainee education.
• Current human-based skill assessment is subjective and time-consuming.
• Objective, automated feedback is needed for effective surgical training.

Purpose of the Study:

• To validate a novel motion-tracking system for automatically evaluating pediatric laparoscopic suturing performance.
• To compare automated performance metrics with expert human ratings.

Main Methods:

• 32 subjects performed suturing tasks in a box trainer, with motion tracked via magnetic sensors and flex sensors.
• Expert surgeons rated performance using Objective Structured Assessment of Technical Skill (OSATS).
• Machine learning models predicted OSATS scores from 280 motion-derived features.

Main Results:

• The best model, using all sensor data, achieved 71% accuracy in predicting scores within 2 points and a 0.85 correlation with human ratings.
• 59% of predictions were perfect, and 100% were within 1 point.
• The model utilized 87 features, primarily from tool tip motion and grip angle.

Conclusions:

• The developed system automatically and accurately assesses surgical skills, closely mirroring expert evaluations.
• This technology can enhance surgical training by providing frequent, objective feedback.
• The system offers potential insights into the core components of surgical proficiency.