•  
  •  
 

Abstract

Since the inception of modern medicine, surgical trainees have benefitted from coaching and feedback from senior surgeons. In the high-demand operating room environment, granular feedback is often limited by time constraints. Artificial intelligence (AI) represents a novel frontier in surgical education in that it may be able to provide reliable, objective, and comprehensive feedback in an automated manner. While prior studies have given broad overviews of AI in surgical training, there has been no study to our knowledge that has codified the myriad applications and offered guidance for which applications best suit trainees at various levels of surgical education. We aim to describe four key domains under which surgical training AI applications can be classified. We define which applications would best suit junior, intermediate, and experienced surgeons by matching their educational needs. The four domains include: enhanced simulation (realistic and safe training environment), real time visual cues (guidance for safe planes of dissection and steps of the operation), automated performance metrics (analyzing kinematic data with immediate feedback and potential improvements in operative efficiency and surgical technique), and surgical decision-making (utilizing metacognition to develop surgical gestalt). Each of these tenants represents a key step in surgical trainees’ development, from practicing technical skills on high fidelity models as a junior trainee to more advanced surgical decision-making as an advanced practitioner. We aim to provide a comprehensive framework for surgical mentors, training programs, and trainees in order to maximize benefits with the use of AI technologies.

Download

Share

COinS