Flatten Wisely: How Patch Order Shapes Mamba-Powered Vision for MRI Segmentation

• URL: http://arxiv.org/abs/2507.13384v1
• Date: Tue, 15 Jul 2025 21:37:37 GMT
• Title: Flatten Wisely: How Patch Order Shapes Mamba-Powered Vision for MRI Segmentation
• Abstract summary: Vision Mamba models promise transformer-level performance at linear computational cost.<n>Their reliance on serializing 2D images into 1D sequences introduces a critical, yet overlooked, design choice: the patch scan order.<n>This paper presents the first systematic study of how scan order impacts MRI segmentation.
• Score: 4.9921974350575296
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Vision Mamba models promise transformer-level performance at linear computational cost, but their reliance on serializing 2D images into 1D sequences introduces a critical, yet overlooked, design choice: the patch scan order. In medical imaging, where modalities like brain MRI contain strong anatomical priors, this choice is non-trivial. This paper presents the first systematic study of how scan order impacts MRI segmentation. We introduce Multi-Scan 2D (MS2D), a parameter-free module for Mamba-based architectures that facilitates exploring diverse scan paths without additional computational cost. We conduct a large-scale benchmark of 21 scan strategies on three public datasets (BraTS 2020, ISLES 2022, LGG), covering over 70,000 slices. Our analysis shows conclusively that scan order is a statistically significant factor (Friedman test: $\chi^{2}_{20}=43.9, p=0.0016$), with performance varying by as much as 27 Dice points. Spatially contiguous paths -- simple horizontal and vertical rasters -- consistently outperform disjointed diagonal scans. We conclude that scan order is a powerful, cost-free hyperparameter, and provide an evidence-based shortlist of optimal paths to maximize the performance of Mamba models in medical imaging.

Related papers

• DAMamba-UNet3D: A Parameter-Efficient Mamba State Space U-Net with Dynamic Adaptive Scan for 3D Medical Image Segmentation [1.9414157088543738]
  DAMamba-UNet3D is a hybrid encoder-decoder that integrates tri-plane 3D-DAS blocks at encoder stages E2-E4 while retaining convolutions elsewhere.<n>On BraTS 2020 five-fold cross-validation, DAMamba-UNet3D achieves mean Dice 0.815 +/-0.013 at 13x lower parameter cost than SegMamba.
  arXiv  Detail & Related papers  (2026-07-21T16:42:48Z)
• BiSegMamba: Efficient Bidirectional Tri-Oriented Mamba for 3D Medical Image Segmentation [6.344197178127241]
  We propose BiSegMamba, an efficient tri-oriented Mamba network for 3D medical image segmentation.<n>BiSegMamba generalizes well across vascular, cardiac, brain tumor, and abdominal multi-organ segmentation tasks.<n>Compared with SegMamba-V2, BiSegMamba achieves slightly better performance on BraTS2023 and clear improvements on ACDC and the carotid dataset.
  arXiv  Detail & Related papers  (2026-05-29T08:13:22Z)
• TopoMamba: Topology-Aware Scanning and Fusion for Segmenting Heterogeneous Medical Visual Media [36.26611641287451]
  TopoMamba is a topology-aware scan-and-fuse framework for segmenting heterogeneous medical visual media.<n>We show that TopoMamba consistently improves segmentation quality over strong CNN, Transformer, and SSM baselines.
  arXiv  Detail & Related papers  (2026-04-28T12:11:58Z)
• Multimodal Visual Surrogate Compression for Alzheimer's Disease Classification [69.87877580725768]
  Multimodal Visual Surrogate Compression (MVSC) learns to compress and adapt large 3D sMRI volumes into compact 2D features.<n>MVSC has two key components: a Volume Context that captures global cross-slice context under textual guidance, and an Adaptive Slice Fusion module that aggregates slice-level information in a text-enhanced, patch-wise manner.
  arXiv  Detail & Related papers  (2026-01-29T13:05:46Z)
• Unified 3D MRI Representations via Sequence-Invariant Contrastive Learning [0.15749416770494706]
  We present a sequence-invariant self-supervised framework leveraging quantitative MRI (qMRI)<n> Experiments on healthy brain segmentation (IXI), stroke lesion segmentation (ARC), and MRI denoising show significant gains over baseline SSL approaches.<n>Our model also generalises effectively to unseen sites, demonstrating potential for more scalable and clinically reliable volumetric analysis.
  arXiv  Detail & Related papers  (2025-01-21T11:27:54Z)
• 2DMamba: Efficient State Space Model for Image Representation with Applications on Giga-Pixel Whole Slide Image Classification [40.10133518650528]
  We propose 2DMamba, a novel 2D selective SSM framework that incorporates the 2D spatial structure of images into Mamba.<n>Experiments on 10 public datasets for WSI classification and survival analysis show that 2DMamba improves up to 2.48% in AUC, 3.11% in F1 score, 2.47% in accuracy and 5.52% in C-index.
  arXiv  Detail & Related papers  (2024-12-01T05:42:58Z)
• V2M: Visual 2-Dimensional Mamba for Image Representation Learning [68.51380287151927]
  Mamba has garnered widespread attention due to its flexible design and efficient hardware performance to process 1D sequences. Recent studies have attempted to apply Mamba to the visual domain by flattening 2D images into patches and then regarding them as a 1D sequence. We propose a Visual 2-Dimensional Mamba model as a complete solution, which directly processes image tokens in the 2D space.
  arXiv  Detail & Related papers  (2024-10-14T11:11:06Z)
• MambaCSR: Dual-Interleaved Scanning for Compressed Image Super-Resolution With SSMs [14.42424591513825]
  MambaCSR is a framework based on Mamba for the challenging compressed image super-resolution (CSR) task.<n>We propose an efficient dual-interleaved scanning paradigm (DIS) for CSR, which is composed of two scanning strategies.<n>Results on multiple benchmarks have shown the great performance of our MambaCSR in the compressed image super-resolution task.
  arXiv  Detail & Related papers  (2024-08-21T16:30:45Z)
• Towards a Benchmark for Colorectal Cancer Segmentation in Endorectal Ultrasound Videos: Dataset and Model Development [59.74920439478643]
  In this paper, we collect and annotated the first benchmark dataset that covers diverse ERUS scenarios. Our ERUS-10K dataset comprises 77 videos and 10,000 high-resolution annotated frames. We introduce a benchmark model for colorectal cancer segmentation, named the Adaptive Sparse-context TRansformer (ASTR)
  arXiv  Detail & Related papers  (2024-08-19T15:04:42Z)
• Slice-Consistent 3D Volumetric Brain CT-to-MRI Translation with 2D Brownian Bridge Diffusion Model [3.4248731707266264]
  In neuroimaging, generally, brain CT is more cost-effective and accessible than MRI. Medical image-to-image translation (I2I) serves as a promising solution. This study is the first to achieve high-quality 3D medical I2I based only on a 2D DM with no extra architectural models.
  arXiv  Detail & Related papers  (2024-07-06T12:13:36Z)
• MHS-VM: Multi-Head Scanning in Parallel Subspaces for Vision Mamba [0.43512163406552]
  State Space Models (SSMs) with Mamba have shown great promise for long-range dependency modeling with linear complexity. To effectively organize and construct visual features within the 2D image space through 1D selective scan, we propose a novel Multi-Head Scan (MHS) module. The resulting sub-embeddings, obtained from the multi-head scan process, are then integrated and ultimately projected back into the high-dimensional space.
  arXiv  Detail & Related papers  (2024-06-10T03:24:43Z)
• Gamba: Marry Gaussian Splatting with Mamba for single view 3D reconstruction [153.52406455209538]
  Gamba is an end-to-end 3D reconstruction model from a single-view image. It completes reconstruction within 0.05 seconds on a single NVIDIA A100 GPU.
  arXiv  Detail & Related papers  (2024-03-27T17:40:14Z)
• VMamba: Visual State Space Model [98.0517369083152]
  We adapt Mamba, a state-space language model, into VMamba, a vision backbone with linear time complexity.<n>At the core of VMamba is a stack of Visual State-Space (VSS) blocks with the 2D Selective Scan (SS2D) module.
  arXiv  Detail & Related papers  (2024-01-18T17:55:39Z)
• MedMNIST v2: A Large-Scale Lightweight Benchmark for 2D and 3D Biomedical Image Classification [59.10015984688104]
  MedMNIST v2 is a large-scale MNIST-like dataset collection of standardized biomedical images. The resulting dataset consists of 708,069 2D images and 10,214 3D images in total.
  arXiv  Detail & Related papers  (2021-10-27T22:02:04Z)
• Self-Supervised Multi-Modal Alignment for Whole Body Medical Imaging [70.52819168140113]
  We use a dataset of over 20,000 subjects from the UK Biobank with both whole body Dixon technique magnetic resonance (MR) scans and also dual-energy x-ray absorptiometry (DXA) scans. We introduce a multi-modal image-matching contrastive framework, that is able to learn to match different-modality scans of the same subject with high accuracy. Without any adaption, we show that the correspondences learnt during this contrastive training step can be used to perform automatic cross-modal scan registration.
  arXiv  Detail & Related papers  (2021-07-14T12:35:05Z)
• DR-SPAAM: A Spatial-Attention and Auto-regressive Model for Person Detection in 2D Range Data [81.06749792332641]
  We propose a person detection network which uses an alternative strategy to combine scans obtained at different times. DR-SPAAM keeps the intermediate features from the backbone network as a template and recurrently updates the template when a new scan becomes available. On the DROW dataset, our method outperforms the existing state-of-the-art, while being approximately four times faster.
  arXiv  Detail & Related papers  (2020-04-29T11:01:44Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.