Student Presenter(s): Jennifer Guo, Kyle Gillani, and Chloe Chai
Faculty Mentor: Stephen DiRusso
School/College: Osteopathic Medicine, Old Westbury

Performed more than 750,000 each year in the US, laparoscopic cholecystectomies (LCs) are among the most common surgical procedures. A large part of a trainee’s education is learning to identify and avoid critical landmarks during the operation. Cholecystectomy AI/Surgeon’s JARVIS (JARVIS) takes advantage of the increasing reliability of artificial intelligence (AI) to help identify structures observed in the ‘critical view of safety’ (CVS), helping to avoid potentially devastating complications like bile duct injury (BDI). This study aims to determine the congruency between attending surgeon identification of critical landmarks and JARVIS identification of those landmarks during an LC. Surgeons will evaluate images taken of the CVS during LCs, then their responses will be compared with those of JARVIS on the same images. Data will be evaluated using analysis of variance and regression analyses. The researchers expect that the results will show high congruency between human and AI identification of landmarks in the CVS, bolstering the utility of AI in surgical education.