arXiv 2502.083472025-02-12cs.CV

Hi-End-MAE：分层编码器驱动的掩码自编码器是更强的医学图像分割视觉学习器

Hi-End-MAE: Hierarchical encoder-driven masked autoencoders are stronger vision learners for medical image segmentation

University of Science and Technology of China (USTC)（中国科学技术大学）；Suzhou Institute for Advanced Research, USTC（中国科学技术大学苏州高等研究院）；Stanford University（斯坦福大学）；State Key Laboratory of Precision and Intelligent Chemistry, USTC（中国科学技术大学精准与智能化学国家重点实验室）；Key Laboratory of Intelligent Information Processing of Chinese Academy of Sciences (CAS), Institute of Computing Technology, CAS（中国科学院计算技术研究所中国科学院智能信息处理重点实验室）

Fenghe Tang, Qingsong Yao, Wenxin Ma, Chenxu Wu, Zihang Jiang, S. Kevin Zhou

备注：: 19 pages, Code: https://github.com/FengheTan9/Hi-End-MAE

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

针对医学图像分割的标签稀缺问题，提出Hi-End-MAE预训练框架，通过编码器驱动重建和分层密集解码利用ViT不同层的丰富表示，在10K CT扫描数据集预训练后，于7个基准上实现更优迁移学习性能。

摘要：

医学图像分割因标签稀缺仍是一项艰巨挑战。通过在大规模无标签医学数据集上基于掩码图像建模（MIM）预训练视觉Transformer（ViT）是一种有前景的解决方案，能为各类下游任务提供计算效率和模型泛化能力。但当前基于ViT的MIM预训练框架主要强调输出层的局部聚合表示，未能利用不同ViT层的丰富表示——这些表示能更好捕捉更精确医学下游任务所需的细粒度语义信息。为填补上述空白，我们提出分层编码器驱动的MAE（Hi-End-MAE），一种简单有效的基于ViT的预训练方案，其核心创新有二：（1）编码器驱动的重建，鼓励编码器学习更具信息性的特征以指导掩码块重建；（2）分层密集解码，通过分层解码结构捕捉不同层的丰富表示。我们在含10K次CT扫描的大规模数据集上预训练Hi-End-MAE，并在7个公开医学图像分割基准上评估其性能。大量实验表明，Hi-End-MAE在各类下游任务中实现了更优的迁移学习能力，揭示了ViT在医学成像应用中的潜力。代码可在https://github.com/FengheTan9/Hi-End-MAE获取。

英文摘要：

Medical image segmentation remains a formidable challenge due to the label scarcity. Pre-training Vision Transformer (ViT) through masked image modeling (MIM) on large-scale unlabeled medical datasets presents a promising solution, providing both computational efficiency and model generalization for various downstream tasks. However, current ViT-based MIM pre-training frameworks predominantly emphasize local aggregation representations in output layers and fail to exploit the rich representations across different ViT layers that better capture fine-grained semantic information needed for more precise medical downstream tasks. To fill the above gap, we hereby present Hierarchical Encoder-driven MAE (Hi-End-MAE), a simple yet effective ViT-based pre-training solution, which centers on two key innovations: (1) Encoder-driven reconstruction, which encourages the encoder to learn more informative features to guide the reconstruction of masked patches; and (2) Hierarchical dense decoding, which implements a hierarchical decoding structure to capture rich representations across different layers. We pre-train Hi-End-MAE on a large-scale dataset of 10K CT scans and evaluated its performance across seven public medical image segmentation benchmarks. Extensive experiments demonstrate that Hi-End-MAE achieves superior transfer learning capabilities across various downstream tasks, revealing the potential of ViT in medical imaging applications. The code is available at: https://github.com/FengheTan9/Hi-End-MAE

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。