Beyond Volume Overlap: Surface Matching for Topology-Aware Coronary Artery Segmentation

2026-09-22

AI Summary

This study addresses the limitation of conventional volumetric metrics, such as the Dice coefficient, in evaluating the topological connectivity of fine tubular coronary arteries, where distal vessel omissions are frequently overlooked. To overcome this, we propose a topology-aware segmentation method based on surface matching. By introducing a bipartite graph matching metric with local radius tolerance, the approach enables decoupled evaluation of false positives and false negatives. Furthermore, a differentiable surface loss function is incorporated to fine-tune nnU-Net, SwinUNETR, and NexToU backbone architectures. Experimental results on the Image-CAS and ASOCA benchmarks demonstrate that the proposed method significantly improves the surface F1 score and effectively recovers missing distal vascular structures.

Abstract

Accurate coronary artery segmentation on coronary computed tomography angiography (CCTA) is essential for diagnosing coronary artery disease. Deep networks are conventionally trained and evaluated with the Dice coefficient, but volume-overlap metrics are poorly suited to thin, tubular anatomy: since most voxels belong to a few thickproximal segments, a missing distal branch barely affects Dice despite severely disrupting the connectivity required for clinical use. We introduce a surface metric that matches predicted and reference surface points via bipartite assignment under a localized, vessel-radius tolerance, reporting precision, recall, and F1 with decoupled false positives (spurious branches) and false negatives (missed branches) a distinction the symmetric Dice cannot make. With it we show that a strong Dice-trained baseline omits far more vessel surface than it hallucinates, an asymmetry its high Dice hides. Building on this, we propose a differentiable surface loss that simultaneously suppresses spurious mass and recovers absent structure, validated by fine-tuning three backbones (nnU-Net, SwinUNETR, NexToU) on two public benchmarks (Image-CAS, ASOCA). Against a matched-epoch control, it significantly improves surface F1 by recovering missed distal vessels at comparable Dice. Our findings argue for measuring and optimizing the vessel surface, not the volume it overlaps. Code

Problem

• Coronary Artery Segmentation
• Volume Overlap Metrics
• Surface Matching
• Topology Connectivity
• Distal Vessel Detection

Innovation

• Surface Matching
• Differentiable Surface Loss
• Coronary Artery Segmentation
• Topology-Aware
• Bipartite Assignment