CardioSyntax: End-to-End SYNTAX Score Prediction - Dataset, Benchmark and Method

2024-07-29

IEEE Workshop/Winter Conference on Applications of Computer Vision

AI Summary

This study addresses the inefficiency and subjectivity of manual SYNTAX Score assessment in coronary artery disease severity evaluation. We propose the first end-to-end deep learning framework for automated SYNTAX Score prediction. Leveraging a large-scale dataset of 3,018 multi-angle dynamic coronary angiography videos, we design a hybrid architecture integrating 3D convolutional neural networks with temporal modeling and a Transformer-based encoder. To enhance anatomical fidelity, we introduce a vessel-structure-aware loss function and jointly optimize regression and binary classification (zero-score vs. non-zero) tasks. Our model achieves an R² of 0.51 on SYNTAX Score regression and 77.3% accuracy in zero-score classification—significantly outperforming conventional semi-automatic approaches. This work establishes the first clinically deployable deep learning solution for fully automated SYNTAX Scoring, advancing intelligent decision support in coronary interventional therapy.

Technology Category

• Machine Learning: Deep Neural Architectures and Foundation Models
• Computer Vision: Learning & Optimization for CV
• Constraint Satisfaction and Optimization: Satisfiability Modulo Theories

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

The SYNTAX score has become a widely used measure of coronary disease severity, crucial in selecting the optimal mode of the revascularization procedure. This paper introduces a new medical regression and classification problem - automatically estimating SYNTAX score from coronary angiography. Our study presents a comprehensive CardioSYNTAX dataset of 3,018 patients for the SYN-TAX score estimation and coronary dominance classification. The dataset features a balanced distribution of individuals with zero and nonzero scores. This dataset includes a first-of-its-kind, complete coronary angiography samples captured through a multi-view X-ray video, allowing one to observe coronary arteries from multiple perspectives. Furthermore, we present a novel, fully automatic end-to-end method for estimating the SYNTAX. For such a difficult task, we have achieved a solid coefficient of determination R2 of 0.51 in score value prediction and 77.3% accuracy for zero score classification.

Problem

• Automatically predict SYNTAX score from coronary angiography
• Introduce comprehensive CardioSYNTAX dataset for score estimation
• Develop end-to-end method for accurate SYNTAX score prediction

Innovation

• End-to-end SYNTAX score prediction method
• Multi-view X-ray video dataset
• Automatic coronary angiography analysis