Innovative AI and Extended Reality System Enhances Surgical Training for Residents

Researchers at Mount Sinai Hospital have developed a groundbreaking AI-powered training system that utilizes extended-reality (XR) headsets to teach complex surgical procedures without the need for an instructor's physical presence. In a recent study published in the Journal of Medical Extended Reality, 17 surgical trainees successfully learned and performed a partial nephrectomy—a procedure to remove a cancerous kidney tissue—using this innovative approach. The system employs a deep learning model linked to a specialized XR headset that streams real-time surgical instructions and visual feedback, allowing trainees to practice intricate tasks with guided support. The trainees used a 3D-printed kidney phantom modeled from real patient CT scans, filled with water-based polymers to replicate tissue, providing a realistic training environment. Throughout the procedure, the system's first-person camera monitors the user, assesses performance, and offers corrective prompts to ensure accuracy. This method addresses a significant challenge in surgical education—namely, the scarcity of experienced proctors—by effectively substituting human instructors with AI-driven guidance. The results showed that all participants achieved surgical success with an impressive accuracy rate of 99.9%. Dr. Nelson Stone, the study's lead author and a Clinical Professor of Urology at Icahn School of Medicine, emphasized that this technology could streamline skill acquisition, reduce training costs, and improve surgical proficiency outside the operating room, ultimately enhancing patient safety. The team plans to further develop this technology by creating more complex, procedural-wide training models using AI, with the goal to standardize and democratize surgical education in the future. The promising outcomes suggest that autonomous AI systems could revolutionize medical training, leading to a highly skilled future generation of surgeons and better treatment outcomes for patients.