Patch-MoE Mamba: 一种用于医学图像分割的基于补丁顺序的专家混合状态空间架构
Patch-MoE Mamba: A Patch-Ordered Mixture-of-Experts State Space Architecture for Medical Image Segmentation
AI总结：
本文提出了一种基于补丁顺序的专家混合状态空间架构Patch-MoE Mamba，以解决现有Mamba分割模型在像素级方向扫描破坏局部二维空间结构以及简单求和融合方向无法适应多样物体大小、形状和边界的问题。
摘要：
基于CNN和Transformer的架构在医学图像分割中已取得优异性能，但CNN在建模长距离依赖性方面存在限制，而Transformer则常面临二次计算和内存复杂度的问题。状态空间模型，尤其是基于Mamba的网络，提供了一种高效的替代方案，具有线性序列复杂度。然而，现有的Mamba分割模型仍面临两个限制：像素级方向扫描会破坏局部二维空间结构，而简单的求和融合方向无法适应多样化的物体大小、形状和边界。为了解决这些问题，我们提出了Patch-MoE Mamba，一种用于医学图像分割的基于补丁顺序的专家混合状态空间架构。它引入了一种分层的补丁顺序扫描机制，能够在保留局部空间邻域的同时捕捉多尺度上下文，并引入了基于MoE的方向融合模块，通过四个方向专家、一个可学习的连接专家和残差方向聚合，自适应地结合多个Mamba扫描器输出。在五个公开的息肉分割基准和ISIC 2017/2018皮肤病变分割数据集上的实验表明了Patch-MoE Mamba的有效性和通用性。
英文摘要：
CNN- and Transformer-based architectures have achieved strong performance in medical image segmentation, but CNNs are limited in modeling long-range dependencies, while Transformers often suffer from quadratic computational and memory complexity. State space models, especially Mamba-based networks, offer an efficient alternative with linear sequence complexity. However, existing Mamba segmentation models still face two limitations: pixel-wise directional scanning can disrupt local 2D spatial structure, and simple summation-based fusion of scan directions cannot adapt well to diverse object sizes, shapes, and boundaries. To address these issues, we propose \textit{Patch-MoE Mamba}, a patch-ordered mixture-of-experts state space architecture for medical image segmentation. It introduces a hierarchical patch-ordered scanning mechanism that preserves local spatial neighborhoods while capturing multi-scale context, and an MoE-based directional fusion module that adaptively combines multiple Mamba scanner outputs using four directional experts, a learnable concatenation expert, and residual directional aggregation. Experiments on five public polyp segmentation benchmarks and the ISIC 2017/2018 skin lesion segmentation datasets demonstrate the effectiveness and generality of Patch-MoE Mamba.
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。
