ViG3D-UNet: Volumetric Vascular Connectivity-Aware Segmentation via 3D Vision Graph Representation

2025-04-18

AI Summary

To address fragmentation, missing endpoints, and poor connectivity in tree-like sparse coronary artery segmentation from coronary CT angiography (CCTA) volumes, this paper proposes a novel hybrid architecture integrating a 3D vision graph neural network (ViG3D) with U-Net. The key contributions are: (1) the ViG3D module, which employs 3D graph convolutional layers coupled with channel-wise attention to explicitly model voxel-level vascular connectivity and topological structure; and (2) a paperclip-shaped biased decoder that efficiently reconstructs sparse feature representations while preserving fine-grained anatomical details. Evaluated on the ASOCA and ImageCAS benchmarks, the method achieves state-of-the-art performance: endpoint recall improves by 12.7%, Dice coefficient increases by 1.9%, and both structural connectivity and segmentation accuracy surpass existing approaches.

Technology Category

• Computer Vision: Segmentation
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG

Abstract

Accurate vascular segmentation is essential for coronary visualization and the diagnosis of coronary heart disease. This task involves the extraction of sparse tree-like vascular branches from the volumetric space. However, existing methods have faced significant challenges due to discontinuous vascular segmentation and missing endpoints. To address this issue, a 3D vision graph neural network framework, named ViG3D-UNet, was introduced. This method integrates 3D graph representation and aggregation within a U-shaped architecture to facilitate continuous vascular segmentation. The ViG3D module captures volumetric vascular connectivity and topology, while the convolutional module extracts fine vascular details. These two branches are combined through channel attention to form the encoder feature. Subsequently, a paperclip-shaped offset decoder minimizes redundant computations in the sparse feature space and restores the feature map size to match the original input dimensions. To evaluate the effectiveness of the proposed approach for continuous vascular segmentation, evaluations were performed on two public datasets, ASOCA and ImageCAS. The segmentation results show that the ViG3D-UNet surpassed competing methods in maintaining vascular segmentation connectivity while achieving high segmentation accuracy. Our code will be available soon.

Problem

• Improves discontinuous vascular segmentation in 3D images
• Addresses missing endpoints in sparse tree-like vascular structures
• Enhances connectivity-aware segmentation for coronary diagnosis

Innovation

• 3D graph neural network for vascular connectivity
• Channel attention combines graph and convolutional features
• Paperclip-shaped offset decoder reduces redundant computations