通过分层监督提升医学图像分割的上下文学习效率
Scaling In-Context Segmentation with Hierarchical Supervision
- 采用分块选择与多级监督，只关注关键解剖区域
- 在512×512分辨率下计算量减少44%且精度相当
- 对OCT、皮肤镜等局部病灶主导模态效果更优
上下文学习（ICL）使医学图像分割模型能用少量样本适应新解剖结构，降低临床标注负担。然而，传统ICL方法依赖密集全局交叉注意力，随图像分辨率提升计算开销剧增。虽有研究引入局部注意力，但缺乏对选择过程的显式监督，导致非信息区域产生冗余计算。本文提出PatchICL，一种结合选择性图像分块与多级监督的分层框架，可主动识别并仅关注最相关信息区域。相较于强基线UniverSeg，PatchICL在同域CT分割任务中达到相当精度，同时在512×512分辨率下降低44%计算量；在35个跨域数据集上，覆盖多种成像模态，于13类模态中的6类表现更优，尤其在以局部病灶为主的OCT和皮肤镜模态中优势显著。训练与评估代码已开源。
原文摘要 · Abstract (English)
In-context learning (ICL) enables medical image segmentation models to adapt to new anatomical structures from limited examples, reducing the clinical annotation burden. However, standard ICL methods typically rely on dense, global cross-attention, which scales poorly with image resolution. While recent approaches have introduced localized attention mechanisms, they often lack explicit supervision on the selection process, leading to redundant computation in non-informative regions. We propose PatchICL, a hierarchical framework that combines selective image patching with multi-level supervision. Our approach learns to actively identify and attend only to the most informative anatomical regions. Compared to UniverSeg, a strong global-attention baseline, PatchICL achieves competitive in-domain CT segmentation accuracy while reducing compute by 44\% at $512\times512$ resolution. On 35 out-of-domain datasets spanning diverse imaging modalities, PatchICL outperforms the baseline on 6 of 13 modality categories, with particular strength on modalities dominated by localized pathology such as OCT and dermoscopy. Training and evaluation code are available at https://github.com/tidiane-camaret/ic_segmentation
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。