Medical News Observer

Key Points:

• The AI system for surgical training automatically evaluates medical students’ proficiency in laparoscopic surgery using a live video feed by offering immediate and personalized feedback to enhance learning outcomes.
• Utilizing the “You Only Look Once” (YOLO) model for object detection, the program provides an accuracy of 78.6% in predicting students’ performance, demonstrating its effectiveness in surgical education.
• The project, which received a Rutgers Medical Educator Innovator award for its innovative approach, aims to integrate it into the surgical training curriculum at Robert Wood Johnson Medical School by 2025, marking a significant advancement in the use of educational technology in healthcare.

AI System for Surgical Training

An artificial intelligence (AI) program is helping the way surgical trainees receive instruction, offering cutting-edge, real-time feedback during surgical exercises. This program helps to automatically assess medical students’ proficiency in laparoscopic surgery using a Fundamentals of Laparoscopic Surgery (FLS) kit. 

Development

Developed by researchers at the Ying Wu College of Computing’s Department of Data Science at New Jersey Institute of Technology, in collaboration with experts from Robert Wood Johnson Medical School (RWJ) and Robust AI, this system scans a live video feed as medical students conduct surgical tasks, providing immediate and personalized feedback. 

The initiative aims to integrate this technology into the RWJ curriculum by 2025. Recognized with a Rutgers Medical Educator Innovator award the project addresses a critical gap in surgical training. Traditional simulation-based training, crucial for developing laparoscopic skills, lacks real-time evaluation and depends heavily on manual feedback. 

YOLO Computer Vision Model

Using the “You Only Look Once (YOLO)” computer vision model for object detection, the AI-powered system automatically assesses students’ performance in tasks such as transferring rings between pegs under time constraints. This AI program provides a dynamic, interactive learning experience that significantly enhances surgical education. Tested on videos of junior medical residents, the AI demonstrated an accuracy of 78.6% in predicting pass-or-fail outcomes. 

Implications

This system stands out by offering real-time performative feedback without needing future video frames, allowing students to learn from their mistakes on the spot. This innovative AI teacher, one of the first of its kind, represents a significant step forward in educational technology. It offers a fully automated evaluation process that enhances learning by immediately informing students of their errors, all with a setup requiring a single camera. It promises to extend its benefits beyond healthcare into various other industries.

References

• Xue, Yunzhe, Andrew Hu, Rohit Muralidhar, Justin W. Ady, Advaith Bongu, and Usman Roshan. “An AI System for Evaluating Pass Fail in Fundamentals of Laparoscopic Surgery from Live Video in Realtime with Performative Feedback.” In 2023 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), 4167–71, 2023. https://doi.org/10.1109/BIBM58861.2023.10385428.
• “Robust and Autonomous AI.”  https://robustai.tech/about/