DeepHealth, the AI radiology subsidiary of NASDAQ-listed RadNet, has secured FDA 510(k) clearance for DeepHealth Breast Ultrasound — an AI-powered solution that automates lesion detection, characterization and reporting during diagnostic breast ultrasound exams. The clearance, announced July 30, 2026 from Somerville, Massachusetts, is being deployed across RadNet's national imaging network before year-end.
98% localization accuracy, 37% faster reads
In a multi-reader multi-case study submitted to the FDA under 510(k) K260303, DeepHealth Breast Ultrasound demonstrated greater than 98% accuracy in localizing suspicious soft-tissue lesions, an 8% sensitivity boost in breast cancer detection, and a 37% reduction in radiologist interpretation time. The tool automatically extracts lesion shape, orientation, margin, echo pattern and posterior features in line with the ACR BI-RADS lexicon, and generates draft radiology reports that clinicians finalize.
Fits RadNet's 700,000-exam breast ultrasound footprint
RadNet operates the largest freestanding outpatient imaging network in the U.S. and plans to roll out DeepHealth Breast Ultrasound across its centers by the end of 2026 — a deployment that could touch roughly 700,000 annual breast ultrasound studies eligible for reimbursement under an existing Category III CPT code for quantitative ultrasound tissue characterization. Dr. Jason McKellop, Medical Director of Women's Imaging for RadNet California, said the tool "helps achieve greater standardization of workflows, improving consistency while saving time for patients, sonographers and radiologists."
Expanding a comprehensive AI breast suite
The ultrasound clearance builds out DeepHealth's Breast Suite, which already includes SmartMammo, Breast Density, Safeguard Review, breast arterial calcification assessment, mammography quality analytics and image-based breast cancer risk prediction (the latter cleared only in Europe). DeepHealth's cloud-native DeepHealth OS unifies the portfolio and now sits at more than 300 clinical sites globally.
The clearance is the latest in a busy stretch for RadNet's DeepHealth unit, which recently launched the AI-generated Reporting Pro workspace in June and follows earlier 2026 medical imaging funding rounds, including Aidoc's $150M Series E and multiple regulatory wins for its Neuro and Prostate suites. The See-Mode Augmented Reporting Tool for Breast (SMART-B) device underlying the product is manufactured by See-Mode Technologies, a DeepHealth company.
Reporting based on coverage from GlobeNewswire, MedTech Dive, HIT Consultant and Radiology Business.