发表机构
Dalian University of Technology; Chongqing University of Posts and Telecommunications（大连理工大学; 重庆邮电大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对基于配准的小样本医学图像分割忽视解剖结构层次的问题，提出H2AL框架，通过H2I模块与梯度聚合算法提升配准和分割性能，实验验证其有效性。
AI 中文摘要
基于配准的小样本医学图像分割（RFMIS）旨在通过配准将已标记图像进行变形，从而为未标记图像生成伪标签。然而，现有方法主要在欧氏空间中执行像素级优化与推理，将解剖结构视为平坦且不相交的，这种对固有层次结构的忽视会降低伪标签质量，削弱模糊区域的区分度，进而限制分割性能。为应对该挑战，我们提出用于RFMIS的双曲层次感知聚合学习框架H2AL，可增强双任务学习的变形合理性与解剖区分度。具体而言，我们引入双曲层次感知注入（H2I）模块，利用双曲空间的层次建模能力，通过变换引导的监督双曲对比学习学习精确的层次感知表示，并通过门控注入块将此类层次先验注入欧氏空间，同时保留语义丰富性。此外，我们提出一种基于梯度聚合的端到端联合优化算法，其中将配准解码器与分割解码器的梯度（包含语义和层次线索）进行聚合，以更新共享编码器，促进跨任务的协作学习。在两个解剖区域的五种实验设置下开展的大量实验，证明了我们的方法在配准和分割任务上的有效性与效率，代码公开于该网址。
英文摘要
Registration-based Few-shot medical image segmentation (RFMIS) aims to generate pseudo-labels for unlabeled images by warping a labeled image through registration. However, existing methods primarily perform pixel-level optimization and inference in Euclidean space, treating anatomical structures as flat and disjoint. This neglect of inherent hierarchies degrades pseudo-label quality and weakens the discrimination of ambiguous regions, limiting the segmentation performance. To overcome this challenge, we propose a Hyperbolic Hierarchy-aware Aggregative Learning framework for RFMIS, termed H2AL, that enhances both deformation plausibility and anatomical discrimination for dual-task learning. Specifically, we introduce a Hyperbolic Hierarchy-aware Infusion (H2I) module, which leverages the hierarchical modeling capability of hyperbolic space to learn precise hierarchy-aware representations via transformation-guided supervised hyperbolic contrastive learning, and injects such hierarchical priors into Euclidean space through a gated infusion block while preserving semantic richness. Furthermore, we propose an end-to-end joint optimization algorithm by gradient aggregation, where the gradients from the registration and segmentation decoders, embedding semantic and hierarchical cues, are aggregated to update the shared encoder to promote collaborative learning across tasks. Extensive experiments on two anatomical regions, with five experimental settings, demonstrate the effectiveness and efficiency of our method in both registration and segmentation. The code is publicly available at https://github.com/JiamingCai469/H2AL.
Comments10 pages, 9 figures. Accepted at ACM Multimedia 2026 (MM '26)
