Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A Reverse Mamba Attention Network for Pathological Liver Segmentation

Jun Zeng, Debesh Jha, Ertugrul Aktas, Elif Keles, Alpay Medetalibeyoglu, Matthew Antalek, Robert Lewandowski, Daniela Ladner, Amir A. Borhani, Gorkem Durak, Ulas Bagci

TL;DR

RMA-Mamba introduces a Reverse Mamba Attention mechanism integrated with Vision Mamba to tackle pathological liver segmentation across MRI and CT. By combining a VMamba encoder with hierarchical VSS blocks and a reverse-attention decoder, it captures fine tissue details while modeling long-range context with efficient state-space-inspired computation. The approach achieves state-of-the-art Dice scores on CirrMRI600+ (92.08% with RMAMamba-S) and strong generalization on LiTS (Dice 92.94% with RMAMamba-T), validating the efficacy of SSM-based architectures in medical image analysis. The work also outlines a scalable path toward 3D extensions to better capture volumetric liver anatomy in future research.

Abstract

We present RMA-Mamba, a novel architecture that advances the capabilities of vision state space models through a specialized reverse mamba attention module (RMA). The key innovation lies in RMA-Mamba's ability to capture long-range dependencies while maintaining precise local feature representation through its hierarchical processing pipeline. By integrating Vision Mamba (VMamba)'s efficient sequence modeling with RMA's targeted feature refinement, our architecture achieves superior feature learning across multiple scales. This dual-mechanism approach enables robust handling of complex morphological patterns while maintaining computational efficiency. We demonstrate RMA-Mamba's effectiveness in the challenging domain of pathological liver segmentation (from both CT and MRI), where traditional segmentation approaches often fail due to tissue variations. When evaluated on a newly introduced cirrhotic liver dataset (CirrMRI600+) of T2-weighted MRI scans, RMA-Mamba achieves the state-of-the-art performance with a Dice coefficient of 92.08%, mean IoU of 87.36%, and recall of 92.96%. The architecture's generalizability is further validated on the cancerous liver segmentation from CT scans (LiTS: Liver Tumor Segmentation dataset), yielding a Dice score of 92.9% and mIoU of 88.99%. Our code is available for public: https://github.com/JunZengz/RMAMamba.

A Reverse Mamba Attention Network for Pathological Liver Segmentation

TL;DR

RMA-Mamba introduces a Reverse Mamba Attention mechanism integrated with Vision Mamba to tackle pathological liver segmentation across MRI and CT. By combining a VMamba encoder with hierarchical VSS blocks and a reverse-attention decoder, it captures fine tissue details while modeling long-range context with efficient state-space-inspired computation. The approach achieves state-of-the-art Dice scores on CirrMRI600+ (92.08% with RMAMamba-S) and strong generalization on LiTS (Dice 92.94% with RMAMamba-T), validating the efficacy of SSM-based architectures in medical image analysis. The work also outlines a scalable path toward 3D extensions to better capture volumetric liver anatomy in future research.

Abstract

We present RMA-Mamba, a novel architecture that advances the capabilities of vision state space models through a specialized reverse mamba attention module (RMA). The key innovation lies in RMA-Mamba's ability to capture long-range dependencies while maintaining precise local feature representation through its hierarchical processing pipeline. By integrating Vision Mamba (VMamba)'s efficient sequence modeling with RMA's targeted feature refinement, our architecture achieves superior feature learning across multiple scales. This dual-mechanism approach enables robust handling of complex morphological patterns while maintaining computational efficiency. We demonstrate RMA-Mamba's effectiveness in the challenging domain of pathological liver segmentation (from both CT and MRI), where traditional segmentation approaches often fail due to tissue variations. When evaluated on a newly introduced cirrhotic liver dataset (CirrMRI600+) of T2-weighted MRI scans, RMA-Mamba achieves the state-of-the-art performance with a Dice coefficient of 92.08%, mean IoU of 87.36%, and recall of 92.96%. The architecture's generalizability is further validated on the cancerous liver segmentation from CT scans (LiTS: Liver Tumor Segmentation dataset), yielding a Dice score of 92.9% and mIoU of 88.99%. Our code is available for public: https://github.com/JunZengz/RMAMamba.

Paper Structure

This paper contains 21 sections, 3 equations, 3 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. State Space Model
  4. Liver segmentation
  5. Method
  6. Overview of the architecture
  7. Encoder
  8. VSS block
  9. Reverse Mamba Attention Module
  10. Loss function
  11. Experiments and Results
  12. Datasets
  13. CirrMRI600+ dataset:
  14. LiTS dataset:
  15. Implementation
  16. ...and 6 more sections

Figures (3)

  • Figure 1: Overview of the proposed RMA-Mamba architecture.
  • Figure 2: Qualitative results of different methods on the CirrMRI600+ dataset.
  • Figure 3: Qualitative results of different methods on the LiTS dataset.