mlsanityData to 7 Oct 2026

Mamba Goes HoME: Hierarchical Soft Mixture-of-Experts for 3D Medical Image Segmentation

arXiv 2507.06363 ↗First posted 8 Jul 2025eess.IV cs.CVcited by 3

On the map: Mixture-of-experts → Medical and biomedical MoE

also in 1 more
Abstract (arXiv)

In recent years, artificial intelligence has significantly advanced medical image segmentation. Nonetheless, challenges remain, including efficient 3D medical image processing across diverse modalities and handling data variability. In this work, we introduce Hierarchical Soft Mixture-of-Experts (HoME), a two-level token-routing layer for efficient long-context modeling, specifically designed for 3D medical image segmentation. Built on the Mamba Selective State Space Model (SSM) backbone, HoME enhances sequential modeling through adaptive expert routing. In the first level, a Soft Mixture-of-Experts (SMoE) layer partitions input sequences into local groups, routing tokens to specialized per-group experts for localized feature extraction. The second level aggregates these outputs through a global SMoE layer, enabling cross-group information fusion and global context refinement. This hierarchical design, combining local expert routing with global expert refinement, enhances generalizability and segmentation performance, surpassing state-of-the-art results across datasets from the three most widely used 3D medical imaging modalities and varying data qualities. The code is publicly available at https://github.com/gmum/MambaHoME.

Key facts. Cited by 3 papers in the map (data through 7 Oct 2026). Placed in Medical and biomedical MoE, one of 12 directions in Mixture-of-experts. Builds on 48 of its 63 references that are in the map. First posted on arXiv 8 Jul 2025.

Builds on

48 of 63 references.

Closest ancestors · 2

Same direction, newest first

  1. 2024SAM-Med3D-MoE: Towards a Non-Forgetting Segment Anything Model via Mixture of Experts for 3D Medical Image Segmentation6 Jul
  2. 2024M4oE: A Foundation Model for Medical Multimodal Image Segmentation with Mixture of Experts15 May

Other references · 46

  1. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  2. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
  3. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,604
  4. 2021Swin Transformer: Hierarchical Vision Transformer using Shifted Windowscited by 12,082
  5. 2016Layer Normalizationcited by 5,579
  6. 2016SGDR: Stochastic Gradient Descent with Warm Restartscited by 4,993
  7. 2023Mamba: Linear-Time Sequence Modeling with Selective State Spacescited by 4,616
  8. 2021ViViT: A Video Vision Transformercited by 1,610
  9. 2024Transformers are SSMs: Generalized Models and Efficient Algorithms Through Structured State Space Dualitycited by 1,283
  10. 2021Scaling Vision with Sparse Mixture of Expertscited by 740
  11. 2021The Medical Segmentation Decathloncited by 620
  12. 2022Mixture-of-Experts with Expert Choice Routingcited by 611
34 more
  1. 2022TotalSegmentator: robust segmentation of 104 anatomical structures in CT imagescited by 597
  2. 2019UNet++: Redesigning Skip Connections to Exploit Multiscale Features in Image Segmentationcited by 583
  3. 2021VLMo: Unified Vision-Language Pre-Training with Mixture-of-Modality-Expertscited by 424
  4. 2021Self-Supervised Pre-Training of Swin Transformers for 3D Medical Image Analysiscited by 418
  5. 2022AMOS: A Large-Scale Abdominal Multi-Organ Benchmark for Versatile Medical Image Segmentationcited by 347
  6. 2024xLSTM: Extended Long Short-Term Memorycited by 320
  7. 2024SegMamba: Long-range Sequential Modeling Mamba For 3D Medical Image Segmentationcited by 266
  8. 2023From Sparse to Soft Mixtures of Expertscited by 245
  9. 2022Multimodal Contrastive Learning with LIMoE: the Language-Image Mixture of Expertscited by 202
  10. 2022Medical Image Segmentation Review: The success of U-Netcited by 190
  11. 2015DeepOrgan: Multi-level Deep Convolutional Networks for Automated Pancreas Segmentationcited by 158
  12. 2023RAPHAEL: Text-to-Image Generation via Large Mixture of Diffusion Pathscited by 141
  13. 2022M3ViT: Mixture-of-Experts Vision Transformer for Efficient Multi-task Learning with Model-Accelerator Co-designcited by 97
  14. 2024VoCo: A Simple-yet-Effective Volume Contrastive Learning Framework for 3D Medical Image Analysiscited by 71
  15. 2025Transformers without Normalizationcited by 69
  16. 2024AbdomenAtlas: A Large-Scale, Detailed-Annotated, & Multi-Center Dataset for Efficient Transfer Learning and Open Algorithmic Benchmarkingcited by 64
  17. 2024TotalSegmentator MRI: Robust Sequence-independent Segmentation of Multiple Anatomic Structures in MRIcited by 58
  18. 2024CuMo: Scaling Multimodal LLM with Co-Upcycled Mixture-of-Expertscited by 56
  19. 2023AbdomenAtlas-8K: Annotating 8,000 CT Volumes for Multi-Organ Segmentation in Three Weekscited by 52
  20. 2023Training Like a Medical Resident: Context-Prior Learning Toward Universal Medical Image Segmentationcited by 32
  21. 2020Capturing implicit hierarchical structure in 3D biomedical images with self-supervised hyperbolic representationscited by 24
  22. 2022Fetal Brain Tissue Annotation and Segmentation Challenge Resultscited by 20
  23. 2023E2ENet: Dynamic Sparse Feature Fusion for Accurate and Efficient 3D Medical Image Segmentationcited by 16
  24. 2024Low-Rank Mixture-of-Experts for Continual Medical Image Segmentationcited by 16
  25. 2023CancerUniT: Towards a Single Unified Model for Effective Detection, Segmentation, and Diagnosis of Eight Major Cancers Using a Large Collection of CT Scanscited by 15
  26. 2024EM-Net: Efficient Channel and Frequency Learning with Mamba for 3D Medical Image Segmentationcited by 12
  27. 2024MoE Jetpack: From Dense Checkpoints to Adaptive Mixture of Experts for Vision Taskscited by 9
  28. 2024Tri-Plane Mamba: Efficiently Adapting Segment Anything Model for 3D Medical Imagescited by 9
  29. 2024Multi-Center Fetal Brain Tissue Annotation (FeTA) Challenge 2022 Resultscited by 8
  30. 2024MambaMIM: Pre-training Mamba with State Space Token Interpolation and its Application to Medical Image Segmentationcited by 7
  31. 2024Integrating Deep Metric Learning with Coreset for Active Learning in 3D Segmentationcited by 7
  32. 2022Placenta Segmentation in Ultrasound Imaging: Addressing Sources of Uncertainty and Limited Field-of-Viewcited by 6
  33. 2023Mixture of Tokens: Continuous MoE through Cross-Example Aggregationcited by 5
  34. 2024Let Me DeCode You: Decoder Conditioning with Tabular Datacited by 4

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 1; 2026: 2.

3 papers.

  1. 2025Scaling Artificial Intelligence for Multi-Tumor Early Detection with More Reports, Fewer Maskscited by 8
  2. 2026BiSegMamba: Efficient Bidirectional Tri-Oriented Mamba for 3D Medical Image Segmentationcited by 0
  3. 2026Merlin Plus: A Large-Scale, Multi-Cancer, Image-Mask-Report Datasetcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Patch-MoE Mamba: A Patch-Ordered Mixture-of-Experts State Space Architecture for Medical Image Segmentation
  2. 2026Dual-Adaptive SAM3: Hierarchical Routing over Low-Rank Expert Layers for Parameter-Efficient Medical Image Segmentation
  3. 2025HybridMamba: A Dual-domain Mamba for 3D Medical Image Segmentation
  4. 2025A Comprehensive Analysis of Mamba for 3D Volumetric Medical Image Segmentation
  5. 2026A Hybrid Mamba-SAM Architecture for Efficient 3D Medical Image Segmentation
  6. 2024MambaClinix: Hierarchical Gated Convolution and Mamba-Based U-Net for Enhanced 3D Medical Image Segmentation
  7. 2025TK-Mamba: Marrying KAN With Mamba for Text-Driven 3D Medical Image Segmentation
  8. 2025MambaVesselNet++: A Hybrid CNN-Mamba Architecture for Medical Image Segmentation