A model for predicting the GEARS score from virtual reality surgical simulator metrics

Area of Science:

• Robotic surgery
• Surgical education
• Medical simulation

Background:

• Surgical education heavily utilizes simulation for skill development.
• Robotic simulator assessments and Global Evaluative Assessment of Robotic Skills (GEARS) are validated tools.
• GEARS requires a human evaluator, limiting its use for immediate feedback.

Purpose of the Study:

• To develop a predictive model for GEARS scores using robotic simulator metrics.
• To establish a correlation between simulator performance and expert assessment scores.
• To enhance the feedback loop for surgical trainees.

Main Methods:

• Linear and multivariate logistic regressions were applied to existing data.
• Subjects performed tasks on dV-Trainer (dVT) and da Vinci Skills Simulator (dVSS).
• Performance was evaluated using simulator metrics and GEARS.

Main Results:

• Positive linear correlations were found between simulator overall scores and GEARS Total Score for each simulator and exercise.
• Specific simulator metrics (e.g., Excessive Instrument Force, Instrument Collision, Time to Complete) were identified as significant predictors.
• Linear models predicting GEARS Total Score from simulator Overall Score showed similar results to multivariate models.

Conclusions:

• Trainees can utilize robotic simulator Overall Score to predict their GEARS Total Score.
• Developed linear regression equations facilitate skill improvement.
• This predictive capability can enhance preparation for high-stakes surgical assessments.