‘Your surgeons are R2D2 and C3PO’
A recent collaboration between NVIDIA and academic researchers from the University of Toronto, UC Berkeley, ETH Zurich, and Georgia Tech has led to the development of ORBIT-Surgical, a simulation framework designed to train robots for surgical procedures.
The framework supports more than a dozen manoeuvres inspired by the training curriculum for laparoscopic procedures, also known as minimally invasive surgery. These manoeuvres include grasping small objects like needles, passing them from one arm to another, and placing them with high precision. The team demonstrated how training a digital twin in simulation, transfers to a physical robot in a lab environment.
The framework will hopefully aid surgeons hone their skills and adapt to the use of robotic assistants in the operating room.
The ability to eventually train robots to perform complex surgical manoeuvres could pave the way for increased automation in certain aspects of surgery. Further research and validation is obviously needed, to ensure that the skills learned by the robots in simulation, can be transferred to real-world surgical environments.
Additionally, there will be safety, regulatory and ethical considerations (at the very least) before AI-assisted robotic surgery becomes the norm.