Patch-MoE Mamba: A Patch-Ordered Mixture-of-Experts State Space Architecture for Medical Image Segmentation

Authors

• Diego Adame, The University of Texas Rio Grande Valley
• Fabian Vazquez Jr., The University of Texas Rio Grande Valley
• Jose A. Nunez, The University of Texas Rio Grande Valley
• Huimin Li, The University of Texas Rio Grande Valley
• Jinghao Yang, The University of Texas Rio Grande ValleyFollow
• Erik Enriquez, The University of Texas Rio Grande ValleyFollow
• Dongchul Kim, The University of Texas Rio Grande ValleyFollow
• Haoteng Tang, The University of Texas Rio Grande Valley
• Bin Fu, The University of Texas Rio Grande ValleyFollow
• Pengfei Gu, The University of Texas Rio Grande Valley

Document Type

Conference Proceeding

Publication Date

2026

Abstract

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 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.

Comments

Copyright IEEE.

Recommended Citation

D. Adame, F. Vazquez, J. A. Nunez, et al. “Patch-MoE Mamba: A Patch-Ordered Mixture-of-Experts State Space Architecture for Medical Image Segmentation.” 2026 IEEE/ACM Conference on Connected Health: Applications, Systems and Engineering Technologies (CHASE), August 4, 2026, 276–80. https://doi.org/10.1109/CHASE69719.2026.00046.

Publication Title

2026 IEEE/ACM Conference on Connected Health: Applications, Systems and Engineering Technologies (CHASE)

DOI

10.1109/CHASE69719.2026.00046