The U.S. Food and Drug Administration has granted De Novo authorization to CLAIRITY BREAST, an AI platform developed by Clairity Inc., for predicting five-year breast cancer risk using only a screening mammogram.

The Boston-based startup says the tool is designed to integrate with existing clinical systems, allowing providers to identify women at elevated risk without relying on traditional factors such as family history or genetic background. The platform analyzes subtle imaging features in mammograms to generate a personalized risk score, which can be used to guide follow-up care and screening strategies.

The AI model, marketed under the name Allix5, was trained on millions of mammographic images and validated across more than 77,000 scans collected from five U.S. screening centers. The sites included both hospital-based and outpatient facilities serving diverse patient populations. Validation was based on five-year outcomes data, according to the company.
“For more than 60 years, mammograms have saved lives by detecting early-stage cancers,” said Dr. Connie Lehman, founder of Clairity and a breast imaging specialist at Mass General Brigham. “Now, advancements in AI and computer vision can uncover hidden clues in the mammograms — invisible to the human eye — to help predict future risk.”

Traditional breast cancer risk models have limitations, particularly for women without a family history of the disease or those from underrepresented racial and ethnic backgrounds. Clairity cites studies indicating that up to 85% of breast cancer patients lack a family history, and nearly half have no identifiable risk factors.

Experts see potential in risk-based screening to improve early detection and reduce disparities. “By integrating AI models that assess individual risk, we can better identify women at higher risk, and those who may benefit from supplemental screening methods, such as MRI,” said Dr. Robert A. Smith, senior vice president of early cancer detection science at the American Cancer Society.

Clairity plans to launch the platform across major health systems through the remainder of 2025.

Artificial Intelligence Homepage