Automated assessment of breast arterial calcification: a new frontier in cardiovascular risk prediction.

Authors

Abstract

Breast Arterial Calcification (BAC) is a common incidental finding on mammograms, traditionally regarded as a benign degenerative change. However, a growing body of evidence indicates that BAC is associated with coronary artery disease (CAD), particularly in women with hypertension. Hypertension may contribute to BAC through persistent hemodynamic stress and vascular endothelial dysfunction, which promote the osteoblast-like transformation of medial vascular smooth muscle cells and calcium deposition in the extracellular matrix. The resulting increase in arterial stiffness is consistent with Mönckeberg's medial calcification, a process distinct from atherosclerotic plaque calcification. Epidemiological data indicate that individuals with BAC have a higher risk of CAD than those without BAC. This association may be stronger in individuals with elevated blood pressure and may provide information beyond traditional Framingham risk scores. In clinical practice, routine mammography can readily detect BAC. Emerging deep learning-based artificial intelligence (AI) algorithms may improve the consistency and sensitivity of BAC assessment, providing a non-invasive, cost-effective, and widely accessible approach to cardiovascular screening in hypertensive women. Current multicenter cohort studies are beginning to integrate BAC scores into cardiovascular risk prediction models and to examine their relationships with blood pressure control and anti-calcification therapies. Future studies should validate AI algorithms across equipment and populations, clarify the relationships of BAC with blood pressure variability, micro-inflammation, and myocardial perfusion, and evaluate interventions targeting medial calcification.

Tags

Topics