Machine-Learning Based Detection of Coronary Artery Calcification Using Synthetic Chest X-Rays

TL;DR

This study demonstrates that digitally reconstructed radiographs (DRRs) can effectively train deep learning models for coronary artery calcification detection, offering a scalable alternative to costly CT scans with promising results comparable to existing methods.

Contribution

First systematic evaluation of DRRs as surrogate training data for CAC detection, showing their effectiveness and potential for large-scale screening applications.

Findings

Lightweight CNNs outperform larger pretrained models.

Super-resolution and contrast enhancement improve detection accuracy.

Achieved mean AUC of 0.754, comparable to prior studies.

Abstract

Coronary artery calcification (CAC) is a strong predictor of cardiovascular events, with CT-based Agatston scoring widely regarded as the clinical gold standard. However, CT is costly and impractical for large-scale screening, while chest X-rays (CXRs) are inexpensive but lack reliable ground truth labels, constraining deep learning development. Digitally reconstructed radiographs (DRRs) offer a scalable alternative by projecting CT volumes into CXR-like images while inheriting precise labels. In this work, we provide the first systematic evaluation of DRRs as a surrogate training domain for CAC detection. Using 667 CT scans from the COCA dataset, we generate synthetic DRRs and assess model capacity, super-resolution fidelity enhancement, preprocessing, and training strategies. Lightweight CNNs trained from scratch outperform large pretrained networks; pairing super-resolution with contrast enhancement yields significant gains; and curriculum learning stabilises training under weak supervision. Our best configuration achieves a mean AUC of 0.754, comparable to or exceeding prior CXR-based studies. These results establish DRRs as a scalable, label-rich foundation for CAC detection, while laying the foundation for future transfer learning and domain adaptation to real CXRs.