XRlabs, the Surgical Intelligence company, announced at NVIDIA GTC that it has achieved the first-in-human use of real-time physical AI integrated with the ORBEYE™ 4K 3D exoscope during a neurosurgical procedure at Lenox Hill Hospital in New York.

The milestone marks the first clinical deployment of NVIDIA Jetson Thor™, NVIDIA's next-generation edge AI computing platform, and is believed to be the first documented use of a physical AI-enabled exoscope in neurosurgery.

XRlabs has developed the first device-agnostic Surgical Intelligence platform, which integrates real-time AI into active surgical environments to enhance visualization, situational awareness, and workflow while preserving established clinical systems. Running on NVIDIA Jetson Thor, the system processes the live surgical video intraoperatively, enabling computer vision-based instrument tracking, gaze and voice responsive automation with optimized AI inference and intraoperative scene recognition without modification to the ORBEYE™ hardware.

The procedure was performed by Dr. David Langer, MD, Chair of Neurosurgery at Lenox Hill Hospital and Dr. Randy D'Amico, MD, assistant professor of neurosurgery, Zucker School of Medicine.

"We always believed this would be the next way of operating, and now it is here," said Dr. Langer. "This was not incremental progress - this was a meaningful leap forward." Combining advanced exoscope imaging with immersive visualization on Apple Vision Pro, tool tracking and speech control begins to approach the holy grail of operative neurosurgery. Under the demands of the operating room, this technology delivered."

As the platform evolves, XRlabs will continue to refine these foundational capabilities, and adopt the enterprise-grade NVIDIA IGX Thor platform, which offers high performance and functional safety, to support scalable deployment in clinical environments, including:

Intelligent Exoscope Automation
Computer Vision-based tracking of surgical instruments with gaze and voice responsiveness to enable seamless framing, focus control and hands-free capabilities.

Real Time Surgical Video Integration
Edge-based processing of live surgical video on NVIDIA Jetson Thor, preserving the surgeon's native visualization experience while enabling real-time AI inference.

Intraoperative Scene Understanding
Instrument recognition and key anatomical regions to support emerging guidance, workflow optimization, and contextual awareness in the operating room.

Immersive Stereoscopic Viewing
Ultra-low-latency stereoscopic streaming to XR headsets, such as the Apple Vision Pro, for an immersive surgical viewing option.

"This milestone marks the emergence of Surgical Intelligence as a runtime capability inside the operating room," said Ali Haddad, founder and CEO of XRlabs. "By aligning with NVIDIA's three-computer framework for AI training, simulation and runtime, we are embedding intelligence directly into surgical workflows. Our vision is for Surgical Intelligence to become foundational infrastructure across modern surgery."