A second reading for every ultrasound.

Challenge

In medical imaging, ultrasound is one of the essential tools for identifying tumours — and one of the hardest to read. Accurate identification and evaluation demand sustained concentration, careful observation, years of accumulated experience, and continuous training. Every scan depends on the person holding the probe.

Solution

YUAN's diagnostic assistance system gives that person a second reader. Built on the VPP6N0-S NX, it combines the low-latency 4K60 HDMI 2.0 M.2 capture card, 70/100 TOPS of computing power, and an AI module designed specifically for medical use. The model was trained on more than 6,000 images annotated by medical experts, and assists physicians in distinguishing normal from abnormal tissue with accuracy of up to 87 per cent — with analysis results delivered in real time, during the examination.

Key benefits

The system recognises tumours as the scan happens, identifying and differentiating various benign and malignant types through YUAN's dedicated medical model. Accuracy improves at the point that matters most: preliminary ultrasound detection pinpoints potential issues quickly and precisely. Risk falls with it — less need for contrast agents means fewer allergy and safety concerns for patients. Early intervention becomes the default: when the model flags an abnormality, follow-up examination happens promptly, and timely detection and treatment measurably reduce disease mortality. Underneath it all, the VPP6N0-S NX's 70/100 TOPS keeps the AI model running at full efficiency.