NVIDIA cuts medical robot training from 5 hours to 2 minutes
NVIDIA today open sourced the Medical Physics Simulation framework inside Isaac for Healthcare. It runs 8,192 robot training environments in parallel, cutting training from over five hours to under two minutes. CMR Surgical, J&J MedTech, and Medtronic are already using it.
Transcript
NVIDIA just open sourced a virtual training ground where medical robots learn surgery before touching a real patient.
Medical robots need huge amounts of varied training data, but capturing rare scenarios on real patients is slow, risky, and hard to reproduce.
The framework runs eight thousand one hundred ninety two robot training environments in parallel, cutting training from over five hours to under two minutes.
CMR Surgical, Johnson and Johnson MedTech, and Medtronic already use it to model surgical robots, kidney stone procedures, and catheter navigation.
It is open source so regulators can inspect the models, review the weights, and reproduce results.
Inboxsmith is the AI receptionist that never misses a business call. Please like and subscribe for more news.
Sources
Every claim in this video comes from the top ranking coverage of this topic. The claims and where each one came from:
- NVIDIA announced the open source Medical Physics Simulation framework within Isaac for Healthcare, built on CUDA, Warp, Newton, and Cosmos.(NVIDIA official announcement)
- GPU-native simulation ran 8,192 robot training environments in parallel and cut training from over five hours to under two minutes.(NVIDIA official announcement)
- Early adopters include CMR Surgical with Cambridge Consultants, Johnson and Johnson MedTech, XCath, Inner Logic, and Medtronic Structural Heart.(NVIDIA official announcement)