Instrument Motion Metrics for Laparoscopic Skills Assessment in Virtual Reality and Augmented Reality.

ABSTRACT

OBJECTIVE: To determine the construct and concurrent validity of instrument motion metrics for laparoscopic skills assessment in virtual reality and augmented reality simulators. STUDY DESIGN: Evaluation study. SAMPLE POPULATION Veterinarian students (novice, n = 14) and veterinarians (experienced, n = 11) with no or variable laparoscopic experience. METHODS: Participants' minimally invasive surgery (MIS) experience was determined by hospital records of MIS procedures performed in the Teaching Hospital. Basic laparoscopic skills were assessed by 5 tasks using a physical box trainer. Each participant completed 2 tasks for assessments in each type of simulator (virtual reality bowel handling and cutting; augmented reality object positioning and a pericardial window model). Motion metrics such as instrument path length, angle or drift, and economy of motion of each simulator were recorded. RESULTS: None of the motion metrics in a virtual reality simulator showed correlation with experience, or to the basic laparoscopic skills score. All metrics in augmented reality were significantly correlated with experience (time, instrument path, and economy of movement), except for the hand dominance metric. The basic laparoscopic skills score was correlated to all performance metrics in augmented reality. The augmented reality motion metrics differed between American College of Veterinary Surgeons diplomates and residents, whereas basic laparoscopic skills score and virtual reality metrics did not. CONCLUSION: Our results provide construct validity and concurrent validity for motion analysis metrics for an augmented reality system, whereas a virtual reality system was validated only for the time score.