Novel AI-Based Quantification of Breast Arterial Calcification to Predict Cardiovascular Risk

TL;DR

This study demonstrates that automated, transformer-based quantification of breast arterial calcification (BAC) from routine screening mammograms can predict major adverse cardiovascular events (MACE) independently of traditional risk factors and ASCVD scores. BAC is categorized into four absolute severity levels based on BAC area in mm$^2$, with risk increasing markedly from Mild to Severe and a consistent signal even in younger women under 50. The analysis, conducted in internal and external large cohorts and validated against ASCVD estimates, shows that BAC adds prognostic value beyond ASCVD, enabling opportunistic CVD risk assessment during mammography without extra radiation or cost. The findings support integrating BAC quantification into routine screening workflows to improve early cardiovascular risk stratification, particularly for younger women, and to enhance cross-site comparability through absolute BAC measurements.

Abstract

Women are underdiagnosed and undertreated for cardiovascular disease. Automatic quantification of breast arterial calcification on screening mammography can identify women at risk for cardiovascular disease and enable earlier treatment and management of disease. In this retrospective study of 116,135 women from two healthcare systems, a transformer-based neural network quantified BAC severity (no BAC, mild, moderate, and severe) on screening mammograms. Outcomes included major adverse cardiovascular events (MACE) and all-cause mortality. BAC severity was independently associated with MACE after adjusting for cardiovascular risk factors, with increasing hazard ratios from mild (HR 1.18-1.22), moderate (HR 1.38-1.47), to severe BAC (HR 2.03-2.22) across datasets (all p<0.001). This association remained significant across all age groups, with even mild BAC indicating increased risk in women under 50. BAC remained an independent predictor when analyzed alongside ASCVD risk scores, showing significant associations with myocardial infarction, stroke, heart failure, and mortality (all p<0.005). Automated BAC quantification enables opportunistic cardiovascular risk assessment during routine mammography without additional radiation or cost. This approach provides value beyond traditional risk factors, particularly in younger women, offering potential for early CVD risk stratification in the millions of women undergoing annual mammography.

Figures (8)

• Figure 1: Examples of mammograms with mild, moderate and severe BAC quantified by our AI model.
• Figure 2: A, D: Kaplan-Meier curves of study participants across any MACE event and each MACE event by BAC severity; Further stratification by age groups 40-60 years (B, E) and 60-80 years (C, F). A, B, C are curves for the internal dataset while D, E and F are curves for the external dataset.
• Figure 3: The relationship between MACE events with the presence or absence of BAC. A. In patients <50years; B. In patient groups stratified by their ASCVD risk category.
• Figure A1: Patient selection flow diagram for internal and external cohort.
• Figure A2: Patient selection flow diagram for internal and external cohort.