A groundbreaking collaboration between NVIDIA and academic researchers is set to transform the field of robotic surgery. The innovative simulation framework, ORBIT-Surgical, leverages cutting-edge AI and simulation technologies to train surgical robots, promising to enhance the capabilities of surgical teams and reduce cognitive load on surgeons. This exciting development is being showcased at the IEEE International Conference on Robotics and Automation (ICRA) in Yokohama, Japan.
The surgical robotics simulation market is growing due to factors like reduced medical errors, improved proficiency in high-risk procedures, and the increasing adoption of surgical robots. This has boosted demand for surgical robotics simulation products.
According to BIS Research, the global surgical robotics simulation market was valued at $427.9 million in 2023 and is expected to reach $2,132.3 million by 2033, growing at a CAGR of 17.31% between 2024 and 2033.
ORBIT-Surgical is the result of a collaborative effort involving researchers from the University of Toronto, UC Berkeley, ETH Zurich, Georgia Tech, and NVIDIA. The framework is designed to train surgical robots using a physics-based simulation, which supports over a dozen maneuvers inspired by laparoscopic surgery training curricula. These maneuvers include tasks such as grasping small objects like needles, passing them between robotic arms, and placing them with high precision.
The framework utilizes NVIDIA Isaac Sim, a robotics simulation platform, and NVIDIA Omniverse, a platform for developing advanced 3D applications. By employing these technologies, the researchers have created a highly realistic training environment where reinforcement learning and imitation learning algorithms are used to teach the robots. The training is conducted on NVIDIA GPUs, ensuring high-performance computing power for efficient learning processes.
One of the key features of ORBIT-Surgical is its ability to transfer training from a digital twin in simulation to a physical robot in a lab environment. This capability was demonstrated using the community-supported da Vinci Research Kit, provided by the Intuitive Foundation. The successful transfer of skills from simulation to real-world application highlights the effectiveness of the ORBIT-Surgical framework in preparing robots for surgical tasks.
The surgical framework includes support for various libraries for reinforcement learning and imitation learning. These AI agents are trained to mimic expert examples, allowing the robots to perform tasks with high accuracy and precision. ORBIT-Surgical introduces benchmark tasks for surgical training, such as inserting a shunt into a blood vessel, lifting a suture needle, and passing a threaded needle through a ring pole. These tasks are crucial for developing the dexterity and precision required for minimally invasive surgeries.
The use of GPU acceleration and parallelization in ORBIT-Surgical significantly enhances the speed of robot learning. The team has demonstrated that tasks such as inserting a shunt and lifting a suture needle can be completed in under two hours on a single NVIDIA RTX GPU. This rapid learning process is a significant improvement over existing surgical frameworks, which often require more time and computational resources.
In addition to speeding up the learning process, ORBIT-Surgical leverages the visual realism enabled by NVIDIA Omniverse to generate high-fidelity synthetic data. This data is essential for training AI models for perception tasks, such as segmenting surgical tools in real-world videos captured in operating rooms. The combination of simulation and real data has been shown to improve the accuracy of AI models, reducing the need for large and expensive real-world datasets.
The development of ORBIT-Surgical represents a significant advancement in the field of surgical robotics. By harnessing the power of AI, simulation, and high-performance computing, this framework has the potential to transform surgical training and improve patient outcomes.
Interested to know more about the growing technologies in your industry vertical? Get the latest market studies and insights from BIS Research. Connect with us at [email protected] to learn more.
.jpg)
