VM-CAGSeg: A Vessel Structure-Aware State Space Model for Coronary Artery Segmentation in Angiography Images

Yuanqing He¹Zhenhuan Lyu¹Yayue Mai¹Si Li^1*Chen-kai Hu^2*

• ¹Guangzhou Institute of Science and Technology, Guangzhou, China
• ²Nanchang University Second Affiliated Hospital, Nanchang, China

Coronary artery segmentation in X-ray angiography is clinically critical for percutaneous coronary intervention (PCI), providing critical morphological guidance for stent deployment, stenosis assessment, and hemodynamic optimization. Nevertheless, inherent angiographic limitations, including complex vasculature, low contrast, and fuzzy boundaries persist as significant challenges. Current methodologies exhibit notable shortcomings in fragmented output continuity, noise susceptibility, and computational inefficiency. This study proposes VM-CAGSeg, a novel U-shaped architecture integrating vessel structure-aware state-space modeling to address these limitations. The framework introduces three key innovations: (1) A Vessel Structure-Aware State Space (VSASS) Block synergizing geometric priors from a Multiscale Vessel Structure-Aware (MVSA) module with long-range contextual modeling via Kolmogorov-Arnold State Space (KASS) Blocks. The Multiscale Vessel Structure-Aware (MVSA) module enhances tubular feature representation through Hessian eigenvalue-derived vesselness measures. (2) A Cross-Stage Feature Interaction Fusion (CSFIF) module replacing conventional skip connections with cross-stage feature fusion strategies to enhance the variability of learned features, preserving long-range dependencies and fine-grained details. (3) A unified architecture that integrates the Vessel Structure-Aware State Space (VSASS) Block and Cross-Stage Feature Interaction Fusion (CSFIF) module to achieve comprehensive vessel segmentation by synergizing multiscale geometric awareness, long-range dependency modeling, and cross-stage feature refinement. Experiments demonstrate that VM-CAGSeg achieves state-of-the-art performance, surpassing CNN-based (e.g., UNet++), Transformer-based (e.g., MISSFormer), and SSM-based (e.g., H_vmunet) methods with an 88.15% DSC, 79.19% mIoU, and 13.68mm HD95. The framework significantly improves boundary delineation, reducing HD95 by 49.8% compared to UNet++ (27.15mm) and 16.6% over TransUNet (15.85mm). While sensitivity (90.05%) is marginally lower than TransUNet (90.33%), its balanced performance in segmentation accuracy and edge precision confirms robustness. These findings validate the effectiveness of integrating multiscale vessel-aware modeling, long-range dependency learning, and cross-stage feature fusion, making VM-CAGSeg a reliable solution for clinical vascular segmentation tasks requiring fine-grained detail preservation. The proposed method is implemented as an open-source project on https://github.com/GIT-HYQ/VM-CAGSeg.