ECR 2026: Poster Presentation
Akira Hasegawa, James Pike, Nikolaos Gkanatsios, Jonathan Go, Brittaney Everitt, Jun Ge, Yinhao Ren, Annie Ng, Jiye Kim, Bryan Haslam
Abstract
Purpose
Breast arterial calcifications (BAC) visible on mammography have been associated with cardiovascular and peripheral vascular disease. However, BAC reporting is inconsistent and subjective due to impractical manual quantification. This study assesses the standalone detection and localization performance of an AI-based BAC algorithm on full-field digital mammography (FFDM) and digital breast tomosynthesis (DBT).
Methods and Materials
A retrospective, multicenter study was conducted to assess the standalone performance of a BAC software on independent FFDM (n=850) and DBT (n=925) screening mammography cases collected from 37 U.S. clinical sites. Cases were randomly sampled from screening populations and were not used in algorithm training. Ground truth was established by three expert radiologists using a consensus and adjudication process. Detection performance was assessed by case-level sensitivity and specificity for BAC detection on each modality. Localization performance was evaluated using length coverage (LC), which is calculated as the length of BAC detected divided by the total length of BAC.
Results
The AI demonstrated strong detection performance across imaging modalities. For FFDM, sensitivity was 0.960 (95% CI: 0.929–0.991), specificity was 0.884 (95% CI: 0.861–0.908), and localization LC was 0.755 (95% CI: 0.724–0.783). For DBT, sensitivity was 0.905 (95% CI: 0.861–0.949), specificity was 0.906 (95% CI: 0.885–0.927), and LC was 0.694 (95% CI: 0.660–0.728).
Conclusion
The AI demonstrated reliable standalone detection and localization of BAC on both FFDM and DBT, supporting routine mammography for automated BAC assessment. This functionality enables consistent and quantitative BAC measurements and may facilitate opportunistic cardiovascular risk assessment without additional imaging burden.
Disclaimers:
Disclaimers: Breast Suite comprises multiple applications including ProFound Pro, Breast Density, Safeguard Review, Risk Assessment, and DeepHealth Viewer. DeepHealth Viewer is manufactured by eRAD, Inc. and distributed by DeepHealth, Inc. Any claims made about Breast Suite may reference claims associated with its individual components. Not all products and functionalities are commercially available in all countries.
Any claims made about Breast Suite may reference claims associated with its individual components.
Not all products and functionalities are commercially available in all countries. For clearance and commercial availability in your geography of functionalities listed and compatibility with other systems, please contact your account manager.