NVIDIA Open Sources First GPU-Accelerated Medical Physics Simulation Framework
NVIDIA Blog · David Niewolny · 2026-07-22
NVIDIA has open-sourced a GPU-accelerated Medical Physics Simulation framework within its Isaac for Healthcare platform that runs 8,192 parallel robot-training environments and cuts training time from over five hours to under two minutes, with early adopters including CMR Surgical, Johnson & Johnson MedTech, and Medtronic Structural Heart.
Extraction
Topics: medical-roboticssimulationopen-sourcegpu-computinghealthcare-ai
Claims
- NVIDIA's Medical Physics Simulation framework runs 8,192 robot-training environments in parallel, reducing training time from over five hours to under two minutes.
- The framework is open source to provide transparency required for regulatory evidence, allowing developers to inspect models and weights and reproduce results across anatomies.
- It combines classical physics simulation with Cosmos-H Dreams, NVIDIA's real-time generative AI physics simulation, to model both deterministic physical rules and learned visual scene dynamics.
- CMR Surgical contributed nearly 500 hours of anonymized clinical data from its Versius Surgical Robotic System to an open embodiment dataset covering cholecystectomy, prostatectomy, hernia repair, and hysterectomy.
- Johnson & Johnson MedTech is using the framework to build digital twins of its MONARCH endoluminal platform for urology, modeling complex anatomy and kidney-stone scenarios.
Key quotes
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.
Benchmarks show 8,192 robot-training environments running in parallel with GPU-native simulation cut training from over five hours to under two minutes.
For robot builders, this turns simulation from a bespoke engineering project into reusable infrastructure.