LWT-ARTERY-LABEL: A Lightweight Framework for Automated Coronary Artery Identification

2025-08-09

AI Summary

Large anatomical variability in coronary arteries and poor generalizability of traditional knowledge-driven methods hinder clinical deployment of automated anatomical labeling in CT coronary angiography (CTCA). Meanwhile, existing deep learning models suffer from high computational cost and limited clinical interpretability. To address these challenges, we propose a lightweight, knowledge-guided automated annotation framework that integrates anatomical priors with rule-based topological constraints for robust vessel branch identification within a compact neural architecture; further, a rule-guided topological correction module is introduced to jointly enhance accuracy and interpretability. Evaluated on multiple public benchmark datasets, our method achieves state-of-the-art performance—improving average labeling accuracy by 3.2%, accelerating inference speed by 5.8×, and reducing GPU memory consumption by 64%. The proposed approach effectively balances accuracy, efficiency, and clinical trustworthiness, demonstrating strong practicality and broad translational potential.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Calibration & Uncertainty Quantification
• Constraint Satisfaction and Optimization: Satisfiability Modulo Theories

Application Category

• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Semantics and Knowledge: Representation, semantic annotation, enhancement, enrichments, access and/or integration of a variety of data on the Web
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Coronary artery disease (CAD) remains the leading cause of death globally, with computed tomography coronary angiography (CTCA) serving as a key diagnostic tool. However, coronary arterial analysis using CTCA, such as identifying artery-specific features from computational modelling, is labour-intensive and time-consuming. Automated anatomical labelling of coronary arteries offers a potential solution, yet the inherent anatomical variability of coronary trees presents a significant challenge. Traditional knowledge-based labelling methods fall short in leveraging data-driven insights, while recent deep-learning approaches often demand substantial computational resources and overlook critical clinical knowledge. To address these limitations, we propose a lightweight method that integrates anatomical knowledge with rule-based topology constraints for effective coronary artery labelling. Our approach achieves state-of-the-art performance on benchmark datasets, providing a promising alternative for automated coronary artery labelling.

Problem

• Automated coronary artery labeling for CTCA analysis
• Overcoming anatomical variability in coronary artery trees
• Balancing clinical knowledge and computational efficiency

Innovation

• Lightweight framework for artery identification
• Integrates anatomical and rule-based constraints
• Achieves state-of-the-art performance efficiently