NVIDIA announced Medical Physics Simulation, an open source framework within Isaac for Healthcare that enables medical r

The Medical Physics Simulation framework addresses a critical bottleneck in healthcare robotics: obtaining the varied data needed to train, test and improve robot behavior in real-world conditions. The framework helps developers model how anatomy, medical devices, and sensors interact, generate scenarios that are difficult to capture in practice, test robots in silico, and train or evaluate robot policies before expensive hardware testing.

The framework combines anatomy and medical device behavior with sensor simulation and robot learning, allowing teams to create reusable simulation environments instead of rebuilding custom scenes for each workflow. Because Medical Physics Simulation is open source, developers can inspect the framework, adapt it to their own devices and workflows, and build on a GPU-accelerated foundation that integrates with the broader NVIDIA stack.

Open source is particularly important in healthcare because teams need transparency into the data, models and weights that shape system behavior. Access to open models and weights helps developers reproduce results, evaluate performance across different anatomies and scenarios, identify limitations and build evidence for regulatory review.

The framework powers physical AI development by enabling robots to operate properly even when anatomy changes, devices behave differently, or conditions shift unexpectedly. Medical Physics Simulation lets developers simulate anatomy, device contact, friction and sensor inputs, then test interactions and environments to evaluate how robots perform across these variations. Powered by NVIDIA CUDA, Isaac for Healthcare, NVIDIA Warp, Newton and Cosmos technologies, the framework can run hundreds of parallel simulation environments. Benchmarks show that 8,192 robot-training environments running in parallel with GPU-native simulation reduced training time from over five hours to under two minutes.

The framework combines classical physics simulation with generative AI physics simulation. Classical simulation models known physical rules like device contact, friction and motion. NVIDIA Cosmos-H Dreams, a real-time generative AI physics simulation capability, learns visual scene dynamics from procedural data to model complex interactions.

Medical robotics leaders are already applying the framework. CMR Surgical and Cambridge Consultants contributed nearly 500 hours of anonymized clinical data from the Versius Surgical Robotic System to support procedures including cholecystectomy, prostatectomy, hernia repair and hysterectomy. CMR Surgical CTO Chris Fryer said, "Open source models allow us to build on shared knowledge, accelerating responsible innovation and, ultimately, gives us the potential to deliver more consistent care and better outcomes for patients worldwide."

Johnson & Johnson MedTech is using Medical Physics Simulation to build digital twins of its endoluminal MONARCH platform for urology. XCath is using the framework for endovascular autonomy policy training. Inner Logic is using Medical Physics Simulation with synthetic data to validate device mechanics and produce evidence to support regulatory pathways. Medtronic Structural Heart is exploring Medical Physics Simulation with simulated X-ray sensing for catheter navigation research.

As a modular capability within Isaac for Healthcare, Medical Physics Simulation works on its own or alongside digital twin pipelines, medical sensor simulation, the NVIDIA Isaac Lab robot-learning framework and NVIDIA open models and policies. Developers can explore the open source framework, review reference workflows and start building simulation environments for their own devices and healthcare robotics applications.