DeepHealth's software (RadNet) obtained its 510(k) clearance in early August; the reading study reports +8% sensitivity and -37% interpretation time.
In early August 2026, the FDA authorized DeepHealth breast ultrasound software, a digital subsidiary of the RadNet imaging network. The tool tracks three usually separate tasks: locate the lesions of suspicious soft tissue, characterize its shape, orientation and margins, and then produce a structured reporting project with observations and impressions. The radiologist retains control of the final evaluation.
The evaluation is based on a multi-reader study involving sixteen radiologists in imaging centres and hospital facilities. The industry reports an increase of about 8% in cancer detection sensitivity and a 37% reduction in interpretation time. The full results have not yet been published in the peer-reviewed literature, which currently limits the scope of these figures.
The technology came from RadNet's acquisition of See-Mode Technologies in 2025. By the end of 2026, the company has announced a deployment on all of its more than 400 outpatient imaging centres.
Read the full article on the original source
Go to the source →