Abstract
The accurate segmentation of coronary Digital Subtraction Angiography (DSA) images is essential for diagnosing and treating coronary artery diseases. Despite advances in deep learning-based segmentation, challenges such as low contrast, noise, overlapping structures, high intra-class variance, and class imbalance limit precise vessel delineation. To overcome these limitations, we propose the MSA-UNet3+: a Multi-Scale Attention enhanced UNet3+ architecture for coronary DSA image segmentation. The framework combined Multi-Scale Dilated Bottleneck (MSD-Bottleneck) with Contextual Attention Fusion Module (CAFM), which not only enhances multi-scale feature extraction but also preserve fine-grained details, and improve contextual understanding. Furthermore, we propose a new Supervised Prototypical Contrastive Loss (SPCL), which combines supervised and prototypical contrastive learning to minimize class imbalance and high intra-class variance by focusing on hard-to-classified background samples. Experiments carried out on a private coronary DSA dataset demonstrate that MSA-UNet3+ outperforms state-of-the-art methods, achieving a Dice coefficient of 87.73%, an F1-score of 87.78%, and significantly reduced Average Surface Distance (ASD) and Average Contour Distance (ACD). The developed framework provides clinicians with precise vessel segmentation, enabling accurate identification of coronary stenosis and supporting informed diagnostic and therapeutic decisions. The code will be released at the following GitHub profile link this https URL.
Abstract (translated)
冠状数字减影血管造影（DSA）图像的精确分割对于诊断和治疗冠状动脉疾病至关重要。尽管基于深度学习的分割技术取得了进展，但诸如对比度低、噪声大、结构重叠、高类内变化以及类别不平衡等问题仍然限制了对血管的准确描绘。为克服这些局限性，我们提出了一种改进的多尺度注意力增强UNet3+架构（MSA-UNet3+），专门用于冠状DSA图像分割。该框架结合了多尺度膨胀瓶颈模块（MSD-Bottleneck）与上下文注意力融合模块（CAFM），不仅增强了多尺度特征提取，还保留了精细细节，并提升了上下文理解能力。此外，我们提出了一种新的监督原型对比损失函数（SPCL），它将监督学习和原型对比学习相结合，通过关注难以分类的背景样本来最小化类别不平衡及高类内变化。 在私人冠状DSA数据集上的实验表明，MSA-UNet3+优于现有方法，在Dice系数上达到了87.73%，F1分数为87.78%，同时显著减少了平均表面距离（ASD）和平均轮廓距离（ACD）。开发的这一框架为临床医生提供了精确的血管分割结果，有助于准确识别冠状动脉狭窄，并支持诊断与治疗决策。相关代码将在以下GitHub链接发布：[此URL]。
URL
https://arxiv.org/abs/2504.05184
