Mount Sinai researchers have developed an AI-powered surgical training model that guided researchers through a complex kidney cancer procedure — with no human instructor present. All 17 trainees in the study successfully completed the simulation.
Published in the Journal of Medical Extended Reality, the study introduced a system called ESIST, which combines deep learning algorithms with an extended-reality headset. Trainees practiced placing a clamp on the renal artery as part of a partial nephrectomy using a 3D-printed kidney model. While performing the simulated surgery, participants received real-time feedback and corrective prompts through the headset’s first-person camera.
Researchers say the approach offers a cost-effective and scalable alternative to traditional surgical instruction, which typically relies on the availability of experienced proctors. The model may also reduce surgical errors by allowing trainees to gain proficiency outside of the operating room.
Mount Sinai’s team is now working on building full-procedure simulations using the same AI framework. All trainees surveyed found the program valuable, and researchers believe it could help address workforce shortages while improving the consistency of resident training.
