Surgery at a distance, under 50 milliseconds.

Challenge

Healthcare resources are unevenly distributed: surgical skill and equipment concentrate in major hospitals, while patients do not. As 3D medical robots spread, telesurgery — operating through the remote control of a robot — has become one of medicine's most closely watched frontiers. It has one unforgiving constraint. For remote surgery to feel like in-person surgery, the delay from camera to surgeon's eyes must stay within 70 milliseconds — and holding that number is the field's central engineering challenge.

Solution

YUAN's integrated solution is built on NVIDIA's Jetson Orin platform. Low-latency HDMI/SDI DMA capture brings the 3D camera feed into the system with as little as 15 microseconds of delay. The SDK integrates NVIDIA's AV1 encoding and decoding to raise image quality while conserving bandwidth, and YUAN's SkyLink X ultra-low-latency network transport carries the stream — bringing overall glass-to-glass latency well below 50 milliseconds.

Compute is the other half of the story. The Jetson AGX Orin platform supplies up to 275 TOPS of AI performance, used both to assist physicians in assessing local tissue in real time and to run instant image depth calculations — sharpening the performance of 3D robotic arms and, with them, the quality of the procedure.

Key benefits

The pipeline reduces development cost outright: capture, encoding, streaming, and display arrive as one complete workflow, saving developers substantial time and investment. SkyLink X makes ultra-low-latency transmission real, letting surgical teams receive and manipulate images within 50 milliseconds — the threshold at which telesurgery becomes practicable. Integrated AI adds precision and safety through real-time image evaluation at the surgeon's side. And the geography problem dissolves: patients reach high-quality surgical care without travelling to it.