AI Accelerates Breast Cancer Care from Scan to Treatment

Why it matters for GPU / AI infrastructure

Breast cancer is the most common cancer among women in the United States, yet many skip the recommended yearly mammogram, creating a large detection gap. A projected shortage of tens of thousands of radiologists by 2035 further strains the system, while weeks‑long waits for genomic test results delay treatment decisions.

iSono Health, an NVIDIA Inception startup, offers the ATUSA platform, a wearable 3D quantitative ultrasound that captures a complete breast volume in under two minutes per breast. Powered by NVIDIA GPU acceleration and trained on more than 1.5 million ultrasound frames, its AI delivers a 3D scan that is about 28 % more sensitive than conventional 2D handheld ultrasound.

Because the system acquires the whole breast uniformly, scans can be compared year over year, reducing operator dependence and variability. The platform is already deployed in partner clinics across several U.S. states and is preparing to integrate additional imaging modalities such as mammography, MRI and clinical data for a multimodal diagnostic pipeline.

The speed and scalability of NVIDIA GPUs enable real‑time AI inference on large imaging datasets, helping clinics overcome staffing shortages and shorten the path from screening to treatment. This infrastructure support is crucial for sustaining AI‑driven workflows that can improve outcomes for millions of patients.