When experts are oceans apart: comparing expert performance values for proficiency-based laparoscopic simulator...

Area of Science:

• Medical Education
• Surgical Simulation
• Virtual Reality Training

Background:

• Surgical skills training is shifting from operating rooms to simulation-based facilities.
• Proficiency-based training courses utilize expert performance data for assessment.
• A lack of published expert datasets and standardized methods hinders progress.

Purpose of the Study:

• To investigate the impact of different expert performance datasets on simulator training outcomes.
• To compare two existing expert value datasets for the LapSim laparoscopic virtual-reality simulator.
• To assess how these datasets affect training data from surgical residents.

Main Methods:

• Compared two published expert performance datasets (van Dongen et al. and Heinrichs et al.) for the LapSim simulator.
• Applied both datasets to LapSim training data from 16 surgical and gynecology residents.
• Analyzed differences in performance metrics like motion efficiency, duration, and damage control.

Main Results:

• Experts in the van Dongen dataset showed better motion efficiency but not duration or damage control compared to Heinrichs.
• Using the Heinrichs expert values resulted in residents completing training in an average of 1.5 fewer sessions.
• The choice of dataset influences the assessment of skills level and training duration.

Conclusions:

• The selection of proficiency values critically affects training length, skills assessment, and costs in simulator training.
• Standardized and well-controlled methods are essential for generating valid and reliable expert values.
• Valid expert values are crucial for effective surgical simulation training and research.