arXiv 2508.028442025-08-04cs.CV

RefineSeg：用于医学图像分割的双重粗到细学习

RefineSeg: Dual Coarse-to-Fine Learning for Medical Image Segmentation

School of Engineering, College of Engineering and Physical Sciences（工程学院，工程与物理科学学院）；University of Birmingham（伯明翰大学）；Oxford Center for Clinical Magnetic Resonance Research (OCMR)（牛津临床磁共振研究所以（OCMR））；University of Oxford（牛津大学）；John Radcliffe Hospital（约翰·拉德克利夫医院）；Johns Hopkins University School of Medicine（约翰霍普金斯大学医学院）；Queen Mary University London（伦敦女王学院）；Barts Health NHS Trust（巴特斯医疗信托）

Anghong Du, Nay Aung, Theodoros N. Arvanitis, Stefan K. Piechnik, Joao A C Lima, Steffen E. Petersen, Le Zhang

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

针对医学图像分割中高质量像素级标注成本高的问题，提出一种仅依赖粗标注的粗到细分割框架，利用转移矩阵建模噪声并联合训练多组粗标注以逐步细化输出，在多个心脏数据集上超越弱监督方法并接近全监督性能。

摘要：

高质量的医学图像像素级标注对于监督分割任务至关重要，但获取此类标注成本高昂且需要医学专业知识。为应对这一挑战，我们提出了一种新颖的粗到细分割框架，该框架完全依赖于粗粒度标注，包括目标和补充性绘制，尽管这些标注本身存在噪声。该框架通过引入转移矩阵来建模粗标注中的不准确和不完整区域。通过在多组粗标注上联合训练，它逐步细化网络输出并推断真实分割分布，通过基于矩阵的建模实现对精确标签的鲁棒近似。为验证所提方法的灵活性和有效性，我们在两个公开的心脏成像数据集ACDC和MSCMRseg上展示了结果，并进一步在UK Biobank数据集上评估了其性能。实验结果表明，我们的方法超越了最先进的弱监督方法，并与全监督方法的结果紧密匹配。

英文摘要：

High-quality pixel-level annotations of medical images are essential for supervised segmentation tasks, but obtaining such annotations is costly and requires medical expertise. To address this challenge, we propose a novel coarse-to-fine segmentation framework that relies entirely on coarse-level annotations, encompassing both target and complementary drawings, despite their inherent noise. The framework works by introducing transition matrices in order to model the inaccurate and incomplete regions in the coarse annotations. By jointly training on multiple sets of coarse annotations, it progressively refines the network's outputs and infers the true segmentation distribution, achieving a robust approximation of precise labels through matrix-based modeling. To validate the flexibility and effectiveness of the proposed method, we demonstrate the results on two public cardiac imaging datasets, ACDC and MSCMRseg, and further evaluate its performance on the UK Biobank dataset. Experimental results indicate that our approach surpasses the state-of-the-art weakly supervised methods and closely matches the fully supervised approach.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。