Open-H-Embodiment dataset enables foundation models for medical robotics

Researchers have introduced Open-H-Embodiment, a large-scale dataset designed to advance foundation models in medical robotics. This dataset includes synchronized kinematic and video data from over 49 institutions and multiple robotic platforms, covering various surgical procedures. The dataset has enabled the development of GR00T-H, a vision-language-action model that achieved full end-to-end task completion on a suturing benchmark, and Cosmos-H-Surgical-Simulator, an action-conditioned world model for multi-embodiment surgical simulation. AI

影响 Enables development of foundation models for medical robotics, potentially improving surgical precision and access to care.

排序理由 This is a research paper introducing a new dataset and two associated models for medical robotics.

在 arXiv cs.AI 阅读 →

• CMR Versius
• Cosmos-H-Surgical-Simulator
• da Vinci Research Kit
• GR00T-H
• Intuitive Surgical
• Maestro
• MIRA
• Moon Surgical
• Open-H-Embodiment
• Rob Surgical BiTrack

• 论文
• 基础设施
• 模型发布

AI 生成摘要 · Google Gemini · 来自 1 个来源。 我们如何撰写摘要 →