Eye Tracking As a Tool to Measure Laparoscopic Skill

Objective of the Study: To determine if eye tracking technology can be used to accurately assess laparoscopic skill. There remains a continued lack of objectively determining the skill level of a laparoscopic surgeon. We are investigating eye tracking as a means to objectively determine the laparoscopic skill level of the surgeon.

Methods and Procedures: The Eyelink II eyetracker (SR Research, Canada) was used to record 7 eye metrics, including blink rate (L/R), saccades (L/R), pupil diameter (L/R) and convergence/divergence of the eyes. Eye metric data was captured 250 times per second. Based on the 7 eye metrics, discriminant function analysis (DFA) was employed in order to determine if each single second of eye tracking data could be identified as that of “expert” surgeon. In addition, gaze tracings were compared between surgeons of different experience levels. Colon mobilization has been studied and we are currently evaluating cholecystectomy.

Results: We analyzed 14,966 seconds of eye tracking data captured during colon mobilization including novice (n=6686s), intermediate (n=5752s) and expert (n=2528s) surgeons. DFA correctly identified the experts’ eye metric data as that of an expert surgeon in 88 to 99% of seconds, intermediate eye metric data in 37-57% of seconds, and novice surgeon data in only 11 to 25% of seconds. Eye tracking reliably distinguished novice from expert eye metrics (p

Session: Poster

Program Number: P180

View Poster

« Return to SAGES 2008 abstract archive