「AI Helps Close Breast Cancer’s Deadliest」 — 「Treatment Plan」 — 「From Scan」 — NVIDIA Highlights Breast-Cancer AI Applications; iSono Health’s ATUSA Scans Each Breast in About Two Minutes

NVIDIA’s overview of AI in breast-cancer care describes companies in its NVIDIA Inception startup program working across imaging, risk assessment and treatment decisions. The post says about 40 million mammograms are performed in the United States each year. It also points to screening and radiologist-capacity challenges, and says genomic assays sent to outside labs can take weeks to process.

One example is iSono Health’s ATUSA, which NVIDIA describes as a wearable, automated 3D quantitative ultrasound system. The company says it captures a standardized breast volume in about two minutes per breast, compared with up to 45 minutes for conventional handheld ultrasound. NVIDIA says the system’s AI was trained on thousands of full-breast scans comprising more than 1.5 million ultrasound frames, and automates image acquisition using NVIDIA GPU acceleration and open-source medical imaging technology.

NVIDIA attributes to iSono Health the claim that ATUSA’s 3D scan is 28% more sensitive than handheld 2D ultrasound. The post also says the system captures the whole breast repeatably, which could help clinicians compare tissue changes across scans. NVIDIA reports that ATUSA is commercially available through partner clinics in California, Texas, Georgia, Tennessee and Washington, D.C. FDA records list the ATUSA Automated 3D Breast Ultrasound System under iSono Health, with a 510(k) decision dated April 22, 2022.