Cardiovascular disease classification using radiomics and geometric features from cardiac CT

2025-06-27

AI Summary

To address the challenge of balancing clinical interpretability and classification accuracy in automated cardiovascular disease (CVD) diagnosis, this paper proposes a decoupled three-stage framework. First, cardiac anatomical structures are precisely segmented using an Atlas-based Iterative Shape Transformation Network (Atlas-ISTN) integrated with a foundational segmentation model. Second, deformable registration against a healthy anatomical atlas generates dense deformation fields, from which physiologically meaningful geometric features are extracted. Third, these deformation-derived features are fused with radiomic features for final CVD classification. Crucially, the method explicitly incorporates anatomical priors into the classification pipeline, endowing model decisions with clear physiological grounding. Evaluated on the ASOCA dataset, the approach achieves 87.50% classification accuracy—substantially outperforming an end-to-end baseline (67.50%)—while providing intuitive, clinically verifiable deformation visualizations to support diagnostic interpretation.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning

Application Category

• Semantics and Knowledge: Methods, algorithms and applications for the development of semantic models, knowledge graphs and other forms of structured data models with machine-interpretable semantics
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for heterogeneous, signed, attributed, multi-relational, temporal, higher-order, and annotated Web-related graphs
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Automatic detection and classification of Cardiovascular disease (CVD) from Computed Tomography (CT) images play an important part in facilitating better-informed clinical decisions. However, most of the recent deep learning based methods either directly work on raw CT data or utilize it in pair with anatomical cardiac structure segmentation by training an end-to-end classifier. As such, these approaches become much more difficult to interpret from a clinical perspective. To address this challenge, in this work, we break down the CVD classification pipeline into three components: (i) image segmentation, (ii) image registration, and (iii) downstream CVD classification. Specifically, we utilize the Atlas-ISTN framework and recent segmentation foundational models to generate anatomical structure segmentation and a normative healthy atlas. These are further utilized to extract clinically interpretable radiomic features as well as deformation field based geometric features (through atlas registration) for CVD classification. Our experiments on the publicly available ASOCA dataset show that utilizing these features leads to better CVD classification accuracy (87.50%) when compared against classification model trained directly on raw CT images (67.50%). Our code is publicly available: https://github.com/biomedia-mira/grc-net

Problem

• Classify cardiovascular disease using interpretable CT features
• Improve CVD classification accuracy via radiomics and geometric features
• Address clinical interpretability in deep learning-based CVD detection

Innovation

• Atlas-ISTN framework for segmentation and registration
• Radiomic and geometric features for interpretability
• Three-component pipeline for CVD classification