Clinical Report: Mammography AI Spots Vascular Signals

Overview

Artificial intelligence-based quantification of breast arterial calcification (BAC) on screening mammography can independently predict major adverse cardiovascular events (MACE) and mortality. The study involving 123,762 women demonstrated a clear dose-response relationship between BAC severity and cardiovascular outcomes.

Background

Breast cancer remains a leading cause of cancer-related mortality among women, making effective screening crucial. Recent advancements in artificial intelligence (AI) have the potential to enhance risk assessment in breast cancer screening by integrating cardiovascular risk factors. Understanding the relationship between BAC and cardiovascular outcomes can improve prognostic evaluations and patient management strategies.

Data Highlights

Key Findings

• 16.1% of women in the Emory cohort and 20.6% in the Mayo cohort had detectable BAC.
• Each 1 mm² increase in BAC area was associated with a 1% to 2% increase in cardiovascular risk.
• Adding BAC to the PREVENT model improved discrimination, increasing the concordance index from 0.71 to 0.73 in the internal cohort.
• In women under 50 years with moderate to severe BAC, event-free survival was lower compared to those with zero BAC.
• The AI model demonstrated 0.91 sensitivity and 0.95 specificity for BAC detection.

Clinical Implications

Automated BAC quantification from routine mammography may serve as an effective cardiovascular risk assessment tool for women, providing insights without additional radiation exposure. Clinicians should consider integrating BAC findings into cardiovascular risk evaluations, especially in younger women.

Conclusion

The study underscores the potential of AI in enhancing cardiovascular risk assessment through mammography, highlighting the need for further integration of BAC into clinical practice and guidelines.

References

1. Dapamede T., et al., European Heart Journal, 2026 -- Artificial intelligence–based quantification of breast arterial calcifications to predict cardiovascular morbidity and mortality
2. European Radiology — The Role of Artificial Intelligence as a Supplementary Reader in Breast Cancer Screening Diagnostics
3. European Radiology (Springer) — From pixels to pathology: how artificial intelligence mammographic risk scores capture tumor biology through imaging
4. The ASCO Post — RSNA Challenge AI Models Enhance Mammography Detection of Invasive Breast Cancer
5. European Radiology — Key Insights on AI Utilization in Breast Imaging: Guidelines from the European Society of Breast Imaging
6. 2026 Guideline on the Management of Dyslipidemia - Professional Heart Daily | American Heart Association
7. Artificial intelligence–based quantification of breast arterial calcifications to predict cardiovascular morbidity and mortality | European Heart Journal | Oxford Academic
8. Breast arterial calcification on mammography and cardiovascular outcomes in women: a meta-analysis. | Radiology AI Research Hub