DeepHealth, a subsidiary of RadNet, Inc., has introduced the DeepHealth Breast Suite, a modular artificial intelligence platform intended to assist radiologists across the breast cancer screening and diagnostic workflow. According to DeepHealth, components of the suite now support more than 10 million mammograms annually, integrating detection, density assessment, short-term breast cancer risk estimation, and upcoming breast arterial calcification (BAC) evaluation tools.
“The launch of Breast Suite marks a pivotal step toward a new, AI-powered standard of care in breast cancer screening and diagnostic pathways,” said Kees Wesdorp, President and CEO of RadNet’s Digital Health Division, DeepHealth. “By embedding detection and risk intelligence with workflow tools, we give radiologists more capabilities to detect cancers earlier, with more confidence and to elevate patient care.”
The Breast Suite builds on DeepHealth’s internal development and technologies integrated from iCAD, creating an end-to-end set of interoperable tools designed to enhance consistency and efficiency across diverse imaging environments.
Clinical AI Tools for Detection and Risk Stratification
Breast Suite integrates a broad set of clinical AI applications, including:
- ProFound Pro AI-powered cancer detection uses prior data, automated region localization, and degree-of-suspicion scoring to support diagnostic accuracy.
- Automated density assessment provides 2D and 3D mammography to standardize classification.
- AI-powered risk assessment estimates short-term cancer risk within 1-2 years using mammogram calibration.
- Breast Arterial Calcification (BAC) assessment (in development) designed to flag breast arterial calcifications that may indicate cardiovascular disease risk.
The technologies have been evaluated in large real-world datasets. A study published in Nature Health examining more than 579,000 women across 100+ imaging sites found that Breast Suite applications were associated with a 21% increase in cancer detection. The benefits held across patient groups, including a 23% increase in cancers detected in women with dense breasts and 20% more cancers detected in Black, non-Hispanic women.
Further research in Science Translational Medicine—evaluating 154,000 women in Europe—showed that DeepHealth’s AI-based risk assessment identified short-term risk with high accuracy and suggested that supplemental screening for the highest-risk 10% of women could have enabled earlier detection of up to 44% of cancers, compared to 20% using the Tyrer-Cuzick traditional risk models.
Workflow Tools to Support Efficiency and Reader Performance
In addition to clinical AI, the Breast Suite incorporates tools intended to enhance workflow and reporting:
- Cloud-based multi-modality viewer supporting mammography, MRI, and ultrasound.
- Prioritized worklist that orders studies by suspicion level.
- Alerts for high-suspicion findings, enabling rapid follow-up.
- AI-powered Safeguard Review, providing a secondary review workflow to help reduce false negatives.
- Structured reporting tools with automated density pre-population and guideline alignment.
Built on DeepHealth’s operating system, the suite is designed for seamless integration with existing systems and offers remote access, and scalable deployment to accommodate evolving clinical needs.
Source: DeepHealth
