用自监督方法提升少量数据下冠状动脉钙化去除效果
Self-Supervised Masked Autoencoders with Dense-Unet for Coronary Calcium Removal in limited CT Data
- 通过随机遮蔽血管腔体3D块，让Dense-Unet自学习重建
- 在少样本情况下，重建精度和狭窄评估更准
- 适合标注数据稀缺的医疗影像任务
冠状动脉钙化会在计算机断层扫描血管造影（CTA）中产生伪影，严重干扰管腔狭窄的诊断。尽管深度卷积神经网络（如Dense-Unet）在通过图像修复去除这些伪影方面展现出潜力，但通常需要大量标注数据，而医学领域此类数据稀缺。受三维点云掩码自编码器（MAE）进展的启发，我们提出一种新的自监督学习框架Dense-MAE，用于体数据。该方法随机遮蔽血管腔体的3D块，并训练Dense-Unet重建缺失几何结构，迫使编码器在无需人工标注的情况下学习动脉拓扑的高层特征。在临床CTA数据集上的实验表明，使用基于MAE预训练的权重初始化钙化去除网络，相比从零开始训练，在少样本场景下显著提升了修复精度与狭窄评估性能。
原文摘要 · Abstract (English)
Coronary calcification creates blooming artifacts in Computed Tomography Angiography (CTA), severely hampering the diagnosis of lumen stenosis. While Deep Convolutional Neural Networks (DCNNs) like Dense-Unet have shown promise in removing these artifacts via inpainting, they often require large labeled datasets which are scarce in the medical domain. Inspired by recent advancements in Masked Autoencoders (MAE) for 3D point clouds, we propose \textbf{Dense-MAE}, a novel self-supervised learning framework for volumetric medical data. We introduce a pre-training strategy that randomly masks 3D patches of the vessel lumen and trains the Dense-Unet to reconstruct the missing geometry. This forces the encoder to learn high-level latent features of arterial topology without human annotation. Experimental results on clinical CTA datasets demonstrate that initializing the Calcium Removal network with our MAE-based weights significantly improves inpainting accuracy and stenosis estimation compared to training from scratch, specifically in few-shot scenarios.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。