Automated Detection of Coronary Artery Calcification via Machine Learning Analysis of 12-Lead Electrocardiograms

Background

Coronary artery calcification (CAC) is a critical marker of subclinical atherosclerosis, yet its primary diagnostic tool, coronary computed tomography (CT), is limited by cost and radiation. We aimed to develop and validate a robust machine learning framework to identify significant CAC (Agatston score ≥100) using features extracted from routine 12-lead electrocardiograms (ECGs), including legacy archives stored as PDFs.