{"slug": "damamba-unet3d-a-parameter-efficient-mamba-state-space-u-net-with-dynamic-scan", "title": "DAMamba-UNet3D: A Parameter-Efficient Mamba State Space U-Net with Dynamic Adaptive Scan for 3D Medical Image Segmentation", "summary": "Researchers propose DAMamba-UNet3D, a parameter-efficient state-space model U-Net for 3D medical image segmentation that achieves a mean Dice of 0.815 on BraTS 2020 with ~5.3 million parameters, roughly 13 times fewer than SegMamba's ~70 million parameters. The model introduces Dynamic Adaptive Scan (DAS) for data-dependent reordering in 3D volumes, and a larger variant, DAMamba-L, reaches 0.829 Dice, surpassing SegMamba by 0.5 points.", "body_md": "arXiv:2607.22718v1 Announce Type: new\nAbstract: We propose parameter-efficient SSM-based U-Net architectures for 3D medical image segmentation. Convolutional U-Nets afford O(n) local mixing per layer but lack explicit global context; transformers provide global reasoning at O(n^2) cost in sequence length $n$. State-space models (SSMs), such as Mamba, offer $O(n)$ global propagation per block. Yet, existing medical SSM segmenters rely on fixed scan patterns and large parameter budgets. Dynamic Adaptive Scan (DAS), which learns data-dependent reordering before selective scan, has not been applied to medical imaging or extended to 3D volumes. We propose DAMamba-UNet3D, a hybrid encoder-decoder that integrates tri-plane 3D-DAS blocks at encoder stages E2-E4 while retaining convolutions elsewhere (~5.3M parameters). On BraTS 2020 five-fold cross-validation, DAMamba-UNet3D achieves mean Dice 0.815+/-0.013 (full-volume per-case evaluation) at ~13x lower parameter cost than SegMamba (0.824+\\-0.014, ~70M). At comparable scale, DAMamba-L (~70M), a wide DAS-native variant with encoder-only DAMamba and a convolutional bottleneck, reaches 0.829+\\-0.012, surpassing retrained SegMamba by 0.5pt. Component ablations show that encoder-only DAS placement is critical as bottleneck and decoder SSM blocks lower Dice. Together, the results suggest that learned tri-plane DAS in a hybrid U-Net is competitive with, and under our large-scale design may improve upon, SegMamba's fixed Tri-orientated Mamba (ToM) scanning on BraTS 2020. Code: https://github.com/marafathussain/DAMamba-UNet3D.", "url": "https://wpnews.pro/news/damamba-unet3d-a-parameter-efficient-mamba-state-space-u-net-with-dynamic-scan", "canonical_source": "https://arxiv.org/abs/2607.22718", "published_at": "2026-07-28 04:00:00+00:00", "updated_at": "2026-07-28 04:08:38.754732+00:00", "lang": "en", "topics": ["artificial-intelligence", "machine-learning", "computer-vision", "ai-research"], "entities": ["DAMamba-UNet3D", "SegMamba", "BraTS 2020", "Dynamic Adaptive Scan", "Mamba"], "alternates": {"html": "https://wpnews.pro/news/damamba-unet3d-a-parameter-efficient-mamba-state-space-u-net-with-dynamic-scan", "markdown": "https://wpnews.pro/news/damamba-unet3d-a-parameter-efficient-mamba-state-space-u-net-with-dynamic-scan.md", "text": "https://wpnews.pro/news/damamba-unet3d-a-parameter-efficient-mamba-state-space-u-net-with-dynamic-scan.txt", "jsonld": "https://wpnews.pro/news/damamba-unet3d-a-parameter-efficient-mamba-state-space-u-net-with-dynamic-scan.jsonld"}}