[论文] Can Unsupervised Methods Outperform Supervised Deep Learning When Grou...

论文概要

研究领域: CV
作者: Anna Mrukwa, Marek Socha, Aleksandra Suwalska et al. (12 authors)
发布时间: 2026-08-17
arXiv: 2608.16855

中文摘要

背景：肺癌仍是全球致死率最高的癌症，因为常被诊断过晚。有效治疗依赖早期筛查阶段的发现。然而，患者数量增长和放射科医生有限导致诊断等待时间延长。在极早期肺癌中，结节可见性进一步被相邻血管和气道壁降低，因为结节常与这些结构相连或由它们供血。因此，支气管血管束的任务特异性分析对高效结节检测很重要，其移除可提高肺癌筛查的诊断潜力。方法：我们使用广泛应用的低剂量CT数据集，包括Duke肺癌筛查（DLCS）数据集和Pomeranian肺癌筛查试点项目。提出的RONALD管道在CT图像上运行，返回位于肺实质内的血管和支气管二值掩膜。结果：该管道在低剂量CT扫描中分割支气管血管束，同时提高结节保留率——在DLCS中从93.98%和90.36%提高到100%，在Pomeranian数据集中从83.16%和62.36%提高到99.92%。结论：所得分割可改善极早期肺癌的肺结节检测。

原文摘要

Background Lung cancer remains the deadliest cancer worldwide because it is often diagnosed too late. Effective treatment depends on detection at an early screening stage. However, the growing number of patients and the limited number of radiologists lead to prolonged diagnostic waiting times. In very early stage lung cancer, nodule visibility is further reduced by adjacent blood vessels and airway walls, because nodules are often connected to or supplied by these structures. Task-specific analysis of the bronchovascular bundle is therefore important for efficient nodule detection, and its removal can increase the diagnostic potential of lung cancer screening. Materials and Methods To assess the efficacy of the proposed method, we used series from widely utilized LDCT datasets, including t...

自动采集于 2026-08-19

#论文 #arXiv #CV #小凯