NVIDIA has unveiled an open-source simulation framework designed to accelerate surgical robotics development by enabling developers to build and train robots in virtual environments before physical lab testing1,2.
The framework, called the Medical Physics Simulation framework, combines anatomy, medical device behavior, sensor simulation, and robot learning to create reusable simulation environments. According to David Niewolny, Director of Business Development for Healthcare and Medical at NVIDIA, the framework allows teams to "create reusable simulation environments instead of rebuilding custom scenes for every workflow, saving developers time and bringing innovations to market faster".
The technology reportedly cuts surgical robot training time from hours to minutes3.
NVIDIA described one of the central challenges in healthcare robotics as the volume and variety of data needed to train, test, and improve robot behavior — particularly teaching robots to interpret physical resistance before they can operate safely in lab environments. Open-source simulators are positioned as a means to unlock these bottlenecks and accelerate industry-wide progress in surgical robotics.
ANALYSIS By releasing the framework as open-source, NVIDIA is making its simulation tooling available for broad adoption across the surgical robotics ecosystem rather than restricting it to proprietary partnerships.
NVIDIA is also extending its AI technology across other areas of healthcare. Johnson & Johnson MedTech is using the Medical Physics Simulation framework and Cosmos foundation models to develop digital twins of its MONARCH robotic platform. CMR Surgical and Cambridge Consultants are using Cosmos-H-Dreams within the framework to create patient-specific surgical simulations, with the collaboration contributing almost 500 hours of anonymized clinical data. Inner Logic is generating synthetic data for device validation using NVIDIA's platform, Medtronic Structural Heart is exploring simulated X-ray data for catheter navigation research, and XCath is training autonomous endovascular robots on the platform.
Beyond surgical robotics, NVIDIA's healthcare AI footprint continues to expand. Bristol Myers Squibb expanded its partnership with NVIDIA in July 2026 to build what it describes as the most powerful AI factory in life sciences. Novo Nordisk, announced as an early pharmaceutical partner for NVIDIA's Proteina generative AI model at NVIDIA GTC 2026 in March 2026, is using NVIDIA BioNeMo foundation models to accelerate structure-based drug discovery. HOPPR integrated NVIDIA's NV-Reason and NV-Generate foundation models into the HOPPR AI Foundry earlier this year and has introduced its own Chest CT Narrative Model.
ANALYSIS The breadth of partners — spanning surgical robotics firms, medtech companies, pharmaceutical giants, and imaging startups — positions the Medical Physics Simulation framework as a horizontal platform play across healthcare AI rather than a narrow surgical robotics tool.