Grant: Master's Award: Canada Graduate Research Scholarships - Master's program

Background: Learning surgical skills requires deliberate practice and high-quality feedback. However, surgery residents are often limited by reduced operative exposure and fewer opportunities to receive feedback. Artificial intelligence (AI)-driven simulation platforms may offer a solution by providing real-time automated assessment and personalized feedback. These systems have primarily been applied to minimally invasive procedures, and their use in open surgery simulation remains limited. This is concerning, as open surgery remains the gold standard for many procedures. Our team has developed and gathered early validity evidence for the Open Surgery Tutor (OST), a proctorless platform that delivers assessment and feedback for open surgery skills. This study aims to compare the impact of OST feedback vs surgeon feedback on the development of open surgery skills in novice trainees. Methods: We will conduct a randomized controlled trial with 20 novice trainees who will be randomly assigned to receive either AI-driven OST feedback or surgeon feedback. Participants will practice on synthetic low-cost inguinal hernia repair models and receive feedback according to group allocation until they reach a predefined intermediate level of competence. Performance will be recorded and assessed using OST metrics and by blinded raters using established tools. The primary outcome is the time required to reach intermediate competence. Secondary outcomes include the number of practice attempts and the total time surgeons spend providing feedback. Significance: This study will evaluate whether OST feedback can replace surgeon feedback during open surgery simulation, offering a low-cost, evidence-based strategy to enhance resident training while reducing the burden on surgeons.