MEI243/PS-CTA-Plaque-Atlas: Initial release of the plaque atlases

This is the first official release (v1.0.0) of the codebase for constructing plaque atlases, associated with the study:

Yang X, et al. "Deciphering Age- and Sex-Specific Patterns of Coronary Artery Atherosclerosis from a Large Chinese Cohort."

Overview

This repository provides the official implementation of plaque atlas construction from coronary computed tomography angiography (CCTA) data. The study aims to advance the understanding of the prevalence, progression, and characteristics of coronary atherosclerosis across different age and sex groups. Using an AI-based CCTA analysis system, we conducted a retrospective, multi-center cohort study involving 16,300 patients in China and performed multi-level plaque characterization (patient-, segment-, and point-level).

Contents

• Source code for constructing three-dimensional (3D) coronary plaque atlases.
• Example data for testing and demonstration.
• Statistical data derived from the study cohort (N=16,300), including:
  • 3D point set of the mean shape of the coronary artery tree.
  • Frequency of each coronary segment (AHA-18 standard) stratified by age and sex.
  • Atlases revealing the fine-grained spatial distribution of plaques, stenosis, and high-risk features by age and sex groups.

This release is archived on Zenodo to ensure reproducibility and long-term accessibility.