NVIDIA open sources GPU framework for medical robot simulation

NVIDIA has introduced Medical Physics Simulation, an open source, GPU-accelerated framework within NVIDIA Isaac for Healthcare for developing medical robotics systems. The framework is designed to help teams model anatomy-device interaction, generate hard-to-capture scenarios, test in silico and train or evaluate robot policies before hardware-intensive testing.

The system combines anatomy and medical device behavior with sensor simulation and robot learning, allowing developers to build reusable simulation environments rather than custom scenes for each workflow. Powered by NVIDIA CUDA and built on NVIDIA Warp, Newton and Cosmos technologies, it can run hundreds of parallel simulation environments. Benchmarks show 8,192 robot-training environments running in parallel with GPU-native simulation cut training from over five hours to under two minutes.

Medical Physics Simulation supports workflows that connect vascular anatomy, flexible instruments such as catheters and guidewires, simulated X-ray imaging and reinforcement learning. It also combines classical physics simulation for contact, friction and motion with NVIDIA Cosmos-H Dreams, a real-time generative AI physics simulation capability that models visual scene dynamics learned from procedural data.

CMR Surgical, Cambridge Consultants, Johnson & Johnson MedTech, XCath, Inner Logic and Medtronic Structural Heart are among the organizations exploring or using the framework for surgical simulation, digital twins, endovascular autonomy, synthetic data and catheter navigation research.