arXiv 2512.103532026-06-18cs.CV

混合Transformer-Mamba用于弱监督体积医学分割

Hybrid Transformer-Mamba for Weakly Supervised Volumetric Medical Segmentation

University of Western Australia（西澳大学）；Harry Perkins Institute of Medical Research（哈利·佩金斯医学研究所）；National Imaging Facility（国家成像设施）；Fiona Stanley Hospital（菲奥娜·斯蒂尔医院）；Victor Chang Cardiac Research Institute（维多利亚·张心脏研究中心）

Yiheng Lyu, Lian Xu, Coen Arrow, Mohammed Bennamoun, Farid Boussaid, Girish Dwivedi

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

提出TranSamba混合架构，通过跨平面建模捕获3D上下文，在弱监督下实现高效体积分割，在三个数据集上达到最优性能。

摘要：

弱监督分割使得模型能够从平面级标签进行训练。现有方法通常依赖2D编码器，忽略了医学数据的体积特性。我们提出TranSamba，一种混合Transformer-Mamba架构，旨在通过跨平面建模捕获3D上下文。TranSamba在Vision Transformer骨干网络基础上增加跨平面Mamba块，利用线性时间建模实现相邻平面间的高效信息交换。这种交换改善了平面内自注意力以及后续用于目标定位的注意力图。TranSamba在输入体积深度上保持线性时间复杂度和恒定空间复杂度。在涵盖不同模态和病理的三个数据集上的大量实验表明，TranSamba达到了最先进的性能，展示了跨平面建模的泛化有效性。代码可在以下网址获取：this https URL.

英文摘要：

Weakly supervised segmentation enables model training from plane-level labels. Existing methods often rely on 2D encoders, neglecting the volumetric nature of medical data. We propose TranSamba, a hybrid Transformer-Mamba architecture designed to capture 3D context via cross-plane modeling. TranSamba augments a Vision Transformer backbone with Cross-Plane Mamba blocks, leveraging linear-time modeling for efficient information exchange across neighboring planes. This exchange improves in-plane self-attention and subsequent attention maps for object localization. TranSamba maintains linear time complexity and constant space complexity with respect to the input volume depth. Extensive experiments on three datasets covering diverse modalities and pathologies show that TranSamba achieves state-of-the-art performance, demonstrating the generalizable efficacy of cross-plane modeling. Code is available at: https://github.com/YihengLyu/TranSamba.

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。