Pith. sign in

REVIEW 5 major objections 4 minor 35 references

KM-UNet KAN Mamba UNet for medical image segmentation

T0 review · 5 major / 4 minor · reviewed 2026-08-10 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read KM-UNet claims that fusing KAN layers with Mamba-style selective scanning in a U-Net beats established U-Net variants on five medical segmentation benchmarks while staying small and fast.

desk verdict Plausible incremental architecture with promising efficiency numbers, but the empirical claims currently outrun the experimental reporting. read the letter →

arxiv 2501.02559 v1 pith:UMKIC25C submitted 2025-01-05 eess.IV cs.AIcs.CV

Yibo Zhang This is my paper · ORCID
classification eess.IVcs.AIcs.CV
keywords Kolmogorov-ArnoldNetworksMambastate-spacemodelsU-Netmedicalimagesegmentationselective-scanattentionefficientdeeplearninginterpretability
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

KM-UNet proposes a U-shaped segmentation network that fuses Kolmogorov-Arnold Networks (KANs) with Mamba-style state-space models, aiming to get long-range context and interpretable feature learning without Transformer-scale compute. The paper reports that this hybrid reaches the highest IoU and F1 scores on most of five medical benchmarks — ISIC17, ISIC18, CVC, BUSI, and GLAS — with an average IoU of 81.17, average F1 of 89.20, 7.35M parameters, and 17.66 Gflops. The author's point is that KAN-plus-SSM is a viable and efficient alternative to CNN and Transformer backbones for medical image segmentation. A sympathetic reader would care because it stakes out a concrete design point combining interpretability, long-range dependency, and low computational cost in one architecture.

What carries the argument

The load-bearing mechanism is the SEM (Selective-Scan Efficient Multi-scale) attention module paired with tokenized KAN blocks. SEM unfolds the feature map into sequences along four scan directions, passes each through an Mamba-derived S6 block, merges them back, then runs multi-scale attention through parallel 1×1 and 3×3 convolutions. The Tok-KAN phase replaces MLP linear transforms with learnable activation functions based on the Kolmogorov-Arnold representation theorem, using a residual connection of the form $Z' = \text{LN}(Z + \text{DwConv}(\Phi(Z)))$. This carries the argument because the four-direction selective scan supplies long-range dependency, while KAN supplies parameter-efficient nonlinear feature representation and the claimed interpretability.

What would settle it

Retrain U-Net, Att-UNet, UNet++, U-NeXt, Rolling-UNet, and U-Mamba using exactly KM-UNet's protocol (Adam, initial learning rate 1e-4 with cosine decay to 1e-5, batch size 8, 300 epochs, basic rotation and flipping augmentation, identical 80/20 splits) and average over at least three seeds; if the average IoU margins on ISIC17, ISIC18, CVC, BUSI, and GLAS shrink to within noise or reverse, the paper's central accuracy claim fails.

Watch

Extended reading notes

Core claim

The central claim is that KM-UNet, by replacing MLP layers in the U-Net bottleneck with tokenized KAN blocks and inserting the Selective-Scan Efficient Multi-scale (SEM) attention module in encoder and decoder, outperforms established U-Net variants (U-Net, Att-UNet, UNet++, U-NeXt, Rolling-UNet, U-Mamba) on most datasets while keeping cost low at 7.35M parameters and 17.66 Gflops. The paper also claims that the KAN layer improves explainability, with channel-activation heatmaps aligning better to ground-truth masks than without it, and that the rotation-style scanning strategy helps on single-object or simpler-shape tasks. The author frames this as the first medical-image segmentation framework integrating KANs and SSMs, providing a baseline and insights for efficient, interpretable segmentation.

Load-bearing premise

The ranking in Table 1 is taken to reflect architecture quality, but the paper reports KM-UNet's own optimizer, learning rate, batch size, augmentation, and 300-epoch schedule without stating matching protocols for the six baselines, and reports no seeds or error bars; if the baselines were trained under different conditions, the accuracy advantage could vanish.

Editorial extensions

If this is right

  • If the reported numbers hold, KAN-SSM hybrids become a credible middle ground between CNNs and Transformers for medical image segmentation.
  • The architecture's 7.35M parameters and 17.66 Gflops imply it could run in resource-constrained or real-time clinical settings where heavier SSM baselines such as U-Mamba (86.3M parameters, 2087 Gflops in Table 2) are impractical.
  • The KAN layer's claimed interpretability, if reproducible, would give clinicians a feature-activation map that tracks target boundaries more faithfully than an MLP bottleneck.
  • The rotation-from-outer-layers scanning strategy suggests a cheap accuracy lever for single-object and simple-shape segmentation tasks.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • If the reported margins are real, a natural next test is ablating the four-direction scan down to one direction on BUSI and ISIC to quantify exactly how much of the gain comes from the rotation strategy versus the KAN bottleneck.
  • The same KAN-SSM design could plausibly be lifted to 3D medical segmentation, where Mamba has already shown strength and where the 17.66 Gflops budget would matter even more.
  • Follow-up work will likely check whether the components are swappable — for instance, different KAN basis functions or different SSM scan orders — and whether the interpretability advantage survives outside the reported heatmaps.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

5 major / 4 minor

Summary. The paper proposes KM-UNet, a U-shaped segmentation network that integrates Kolmogorov-Arnold Networks (KANs), state-space models (SSMs) via a Mamba-inspired selective-scan module, and a multi-scale attention module. The approach is evaluated on five medical segmentation benchmarks (BUSI, GlaS, CVC-ClinicDB, ISIC17, ISIC18) against six baselines, reporting an average IoU of 81.17, average F1 of 89.20, 7.35M parameters, and 17.66 Gflops. The authors claim that KM-UNet achieves the highest IoU and F1 on most datasets and strikes an excellent accuracy-efficiency balance, and that this is the first medical image segmentation framework integrating KANs and SSMs.

Significance. If the reported results are reproducible, the architecture is a potentially valuable contribution: it combines KAN-based tokenized blocks with SSM-style selective scanning in a U-Net, and the claimed efficiency (17.66 Gflops, 7.35M parameters) is attractive. The paper also provides open-source code, and the interpretability discussion is a useful direction. However, the current experimental evidence does not yet separate architectural gains from training-protocol differences, because baseline protocols, error bars, and ablation results are missing.

major comments (5)
  1. [Section 4, Table 1] The implementation details state that 'all results were averaged over three independent runs,' but Table 1 reports only point estimates with no standard deviations, confidence intervals, or per-run values. The margins on several datasets are small (e.g., BUSI IoU 65.42 vs. 61.81 for U-Mamba; ISIC17 IoU 84.05 vs. 82.14 for Rolling-UNet), so without variance information it is impossible to assess whether the reported ranking is statistically meaningful. Please provide error bars or per-run results for all entries.
  2. [Section 4, Table 1] The manuscript gives the full training protocol for KM-UNet (Adam, learning rate 1e-4 with cosine schedule, batch size 8, 300 epochs, BCE+Dice loss, random rotation and flipping) but gives no comparable protocol for the baselines U-Net, Att-UNet, UNet++, U-NeXt, Rolling-UNet, and U-Mamba. Without evidence that baselines were trained under matched data splits, augmentation, and hyperparameters, the advantage in Table 1 could reflect training choices rather than architecture quality. Please describe the baseline implementations and, ideally, retrain all methods under a common protocol.
  3. [Section 5, Conclusions] The conclusions state that 'ablation studies show' the SEM module improves IoU and F1 by 2%-3% and that the cosine annealing schedule outperforms a fixed learning rate, but no ablation table or corresponding experiments appear in Section 4. These claims are currently unsupported. Please include the ablation experiments with concrete numerical results, or remove the claims.
  4. [Section 4, Main results and Table 2] There is a substantive numerical inconsistency: the main-results text reports an average IoU of 80.45% and an average F1 of 88.63%, while Table 2 reports 81.17% and 89.20%. Please correct the discrepancy and ensure all reported numbers are consistent. Additionally, the Experiments section opens by saying 'three distinct and heterogeneous datasets' but then lists and evaluates five datasets; this should be corrected.
  5. [Section 3, Equations (6)-(9)] The architecture description is not sufficiently detailed for reproducibility. The channel hyperparameters C1-C5 and D1-D5 are named but not given, the KAN grid size and spline order are not specified, and Eq. (6) uses Φ(Z) without defining Φ as the composition of KAN layers from Eqs. (7)-(9). Please provide a complete specification in the text or explicitly point to the released code with the exact configuration used for the reported results.
minor comments (4)
  1. [Keywords] The keyword 'Manba' appears to be a typo for 'Mamba'.
  2. [References] Several citations appear to point to the wrong papers: [16] is cited as the source of Mamba/S6 but is 'Hungry Hungry Hippos'; [17] is cited as S4 but is U-Mamba; [20] is cited as U-NeXt but is 'White-Box Transformers via Sparse Rate Reduction'. Please recheck all references against the text.
  3. [Section 4, Datasets] The GlaS dataset description states it 'comprises 612 standard-definition frames derived from 31 sequences' and then says 'we used 165 images following established protocols'; this is confusing because the standard GlaS benchmark contains 165 images. Please clarify the dataset split and the source of the 612-frame count.
  4. [Section 4, Explainability] The explainability experiment reports qualitative heatmap comparisons and refers to changes in IoU, but it does not provide the actual IoU values or a quantitative protocol. Please report the numbers or clarify that the comparison is qualitative.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: KM-UNet's reported results are empirical outcomes from training on public benchmarks against external baselines, not consequences of fitted constants or self-citations.

full rationale

KM-UNet's quantitative claims are produced by training the proposed architecture on five public datasets (BUSI, GlaS, CVC-ClinicDB, ISIC17, ISIC18) and comparing IoU/F1 metrics against U-Net, Att-UNet, UNet++, U-NeXt, Rolling-UNet, and U-Mamba. No equation in Sections 3-4 reduces a reported performance number to a fitted constant, to a self-authored citation, or to the claim being predicted. The KAN/SEM/S6 blocks are architectural constructions whose equations (Eqs. 1-9) define forward computations; they do not embed benchmark results. The novelty claim that KM-UNet is the first medical image segmentation framework integrating KANs and SSMs is a historical assertion about prior work, not a derivation from the benchmark outcomes. The explainability and ablation statements (e.g., 'Ablation studies show that the SEM module... with a 2%-3% increase') are not backed by an ablation table in the manuscript, and Table 1 omits matched training protocols and error bars for baselines; these are evidential and validity weaknesses that could change rankings under fair re-runs, but they are not circular. The paper does not rely on a self-citation chain as load-bearing support; reference [35] is related work on U-KAN and does not force the present outcome. Therefore no circular step can be exhibited; score 0.

Assumptions & free parameters 3 free parameters · 3 assumptions · 0 invented entities

The paper is an empirical architecture study, so its load is carried by dataset choice and training protocol rather than by mathematical axioms. I count the hand-set channel and schedule hyperparameters and the unproven effectiveness of KAN/SSM for 2D segmentation as the main unpaid assumptions.

free parameters (3)
  • Channel hyperparameters C1-C5 and D1-D5 = Not reported
    Section 3 defines these channel counts as hyperparameters but never gives their values; they determine capacity and are part of the reported 7.35M parameter count.
  • KAN grid size and spline order = Not reported
    The Tok-KAN blocks use learnable activation functions, but no grid size or spline parameters are specified, so the exact KAN variant is under-determined.
  • Training hyperparameters = Batch size 8; learning rate 1e-4 to 1e-5 cosine; 300 epochs; Adam; BCE plus Dice loss
    Chosen by hand in Section 4, with no sensitivity analysis and no indication that baselines were assigned the same configuration.
assumptions (3)
  • domain assumption Kolmogorov-Arnold representation theorem justifies replacing linear transformation matrices with learnable univariate activation functions in a U-Net without loss of representational power.
    Invoked in Section 3; the theorem concerns exact representation of continuous multivariate functions and does not by itself guarantee better segmentation performance or interpretability.
  • domain assumption Four-direction selective scanning with S6 blocks provides efficient long-range, 2D spatial modeling for medical images.
    Transferred from 1D SSM theory in Section 3; no 2D-specific proof or scaling study is provided.
  • domain assumption Random 80/20 splits with basic augmentation and 300 epochs give fair, representative comparisons to published baselines.
    Stated in Section 4, but no seeds, baseline training protocols, or variance estimates are given.

how reviews work

0 comments
Cite this review

Pith. "Pith review of KM-UNet KAN Mamba UNet for medical image segmentation." pith.science (2026). https://pith.science/paper/UMKIC25C

@misc{pith2026250102559,
  author       = {Pith},
  title        = {Pith review of: KM-UNet KAN Mamba UNet for medical image segmentation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/UMKIC25C}},
  note         = {Machine review of arXiv:2501.02559}
}
read the original abstract

Medical image segmentation is a critical task in medical imaging analysis. Traditional CNN-based methods struggle with modeling long-range dependencies, while Transformer-based models, despite their success, suffer from quadratic computational complexity. To address these limitations, we propose KM-UNet, a novel U-shaped network architecture that combines the strengths of Kolmogorov-Arnold Networks (KANs) and state-space models (SSMs). KM-UNet leverages the Kolmogorov-Arnold representation theorem for efficient feature representation and SSMs for scalable long-range modeling, achieving a balance between accuracy and computational efficiency. We evaluate KM-UNet on five benchmark datasets: ISIC17, ISIC18, CVC, BUSI, and GLAS. Experimental results demonstrate that KM-UNet achieves competitive performance compared to state-of-the-art methods in medical image segmentation tasks. To the best of our knowledge, KM-UNet is the first medical image segmentation framework integrating KANs and SSMs. This work provides a valuable baseline and new insights for the development of more efficient and interpretable medical image segmentation systems. The code is open source at https://github.com/2760613195/KM_UNet Keywords:KAN,Manba, state-space models,UNet, Medical image segmentation, Deep learning

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

35 extracted references · 23 canonical work pages

  1. [31]

    Journal of Medical Imaging 6(1), 014006– 014006 (2019)

    Alom, M.Z., Yakopcic, C., Hasan, M., Taha, T.M., Asari, V.K.: Recurrent residual u-net for medical image segmentation. Journal of Medical Imaging 6(1), 014006– 014006 (2019)

    work page 2019

  2. [33]

    In: Proc

    Landman, B., Xu, Z., Igelsias, J., Styner, M., Langerak, T., Klein, A.: Miccai multiatlas labeling beyond the cranial vault –workshop and challenge. In: Proc. MICCAI Multi-Atlas Labeling Beyond Cranial Vault—Workshop Challenge. vol. 5, p. 12 (2015)

    work page 2015

  3. [1]

    Deep Learning in Medical Image Analysis

    Shen, Dinggang, et al. “Deep Learning in Medical Image Analysis.” Annual Review of Biomedical Engineering, June 2017, pp. 221– 48, https://doi.org/10.1146/annurev-bioeng-071516-044442

    doi:10.1146/annurev-bioeng-071516-044442 2017

  4. [2]

    Few -Shot Medical Image Segmentation Using a Global Correlation Network with Discriminative Embedding

    Sun, Liyan, et al. “Few -Shot Medical Image Segmentation Using a Global Correlation Network with Discriminative Embedding.” Computers in Biology and Medicine, Jan. 2022,p.105067,https://doi.org/10.1016/j.compbiomed.2021.105067

    work page arXiv 2022

  5. [3]

    Hierarchical Deep Network with Uncertainty-Aware Semi-Supervised Learning for Vessel Segmentation

    Li, Chenxin, et al. “Hierarchical Deep Network with Uncertainty-Aware Semi-Supervised Learning for Vessel Segmentation.” Neural Computing and Applications, Feb. 2022, pp. 3151–64, https://doi.org/10.1007/s00521-021-06578-3

    work page doi:10.1007/s00521-021-06578-3 2022

  6. [5]

    Medical Image Segmentation Using Deep Learning: A Survey

    Wang, Risheng, et al. “Medical Image Segmentation Using Deep Learning: A Survey.” IET Image Processing, Apr. 2022, pp. 1243– 67, https://doi.org/10.1049/ipr2.12419

    doi:10.1049/ipr2.12419 2022

  7. [6]

    Unsupervised Anomaly Segmentation Using Image-Semantic Cycle Translation

    Li, Chenxin, et al. “Unsupervised Anomaly Segmentation Using Image-Semantic Cycle Translation.” arXiv: Image and Video Processing,arXiv: Image and Video Processing, Mar. 2021

    work page 2021

  8. [7]

    UNet++: A Nested U-Net Architecture for Medical Image Segmentation

    Zhou, Zongwei, et al. UNet++: A Nested U-Net Architecture for Medical Image Segmentation

    work page

Show all 35 references
  1. [8]

    V -Net: Fully Convolutional Neural Networks for Volumetric Medical Image Segmentation

    Milletari, Fausto, et al. “V -Net: Fully Convolutional Neural Networks for Volumetric Medical Image Segmentation.” 2016 Fourth International Conference on 3D Vision (3DV), 2016, https://doi.org/10.1109/3dv.2016.79

    2016 doi

  2. [9]

    Do Vision Transformers See Like Convolutional Neural Networks

    Raghu, Maithra, et al. “Do Vision Transformers See Like Convolutional Neural Networks.” arXiv: Computer Vision and Pattern Recognition,arXiv: Computer Vision and Pattern Recognition, Aug. 2021

    2021

  3. [10]

    Global Context Vision Transformers

    Hatamizadeh, Ali, et al. Global Context Vision Transformers. June 2022

    2022

  4. [11]

    TransUNet: Transformers Make Strong Encoders for Medical Image Segmentation

    Chen, Jieneng, et al. “TransUNet: Transformers Make Strong Encoders for Medical Image Segmentation.” Cornell University - arXiv,Cornell University - arXiv, Feb. 2021

    2021

  5. [12]

    Swin UNETR: Swin Transformers for Semantic Segmentation of Brain Tumors in MRI Images

    Hatamizadeh, Ali, et al. “Swin UNETR: Swin Transformers for Semantic Segmentation of Brain Tumors in MRI Images.” Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries,Lecture Notes in Computer Science, 2022, pp. 272 –84, https://doi.org/10.1007/978-3-03...

    2022 doi

  6. [13]

    An Image Is Worth 16x16 Words: Transformers for Image Recognition at Scale

    Dosovitskiy, Alexey, et al. “An Image Is Worth 16x16 Words: Transformers for Image Recognition at Scale.” arXiv: Computer Vision and Pattern Recognition,arXiv: Computer Vision and Pattern Recognition, Oct. 2020

    2020

  7. [14]

    Liu, X.; Guo, X.; Liu, Y.; and Yuan, Y. 2021. Consolidated domain adaptive detection and localization framework for cross -device colonoscopic images. Medical image analysis, 71: 102052

    2021

  8. [15]

    Training Data -Efficient Image Transformers & Distillation through Attention

    Touvron, Hugo, et al. “Training Data -Efficient Image Transformers & Distillation through Attention.” International Conference on Machine Learning,International Conference on Machine Learning, July 2021

    2021

  9. [16]

    Hungry Hungry Hippos: Towards Language Modeling with State Space Models

    Dao, Tri, et al. Hungry Hungry Hippos: Towards Language Modeling with State Space Models. Dec. 2022

    2022

  10. [17]

    U-Mamba: Enhancing Long-Range Dependency for Biomedical Image Segmentation

    Ma, Jun, et al. U-Mamba: Enhancing Long-Range Dependency for Biomedical Image Segmentation. Jan. 2024

    2024

  11. [18]

    RWKV: Reinventing RNNs for the Transformer Era

    Peng, Bo, et al. RWKV: Reinventing RNNs for the Transformer Era

  12. [19]

    Xing, Z.; Ye, T.; Yang, Y.; Liu, G.; and Zhu, L. 2024. Segmamba: Long -range sequential modeling mamba for 3d medical image segmentation. arXiv preprint arXiv:2401.13560

    2024 arXiv

  13. [20]

    White-Box Transformers via Sparse Rate Reduction

    Yu, Yaodong, et al. White-Box Transformers via Sparse Rate Reduction. June 2023

    2023

  14. [21]

    U -Net: Convolutional Networks for Biomedical Image Segmentation

    Ronneberger, Olaf, et al. “U -Net: Convolutional Networks for Biomedical Image Segmentation.” Lecture Notes in Computer Science,Lecture Notes in Computer Science, Jan. 2015

    2015

  15. [22]

    3D MRI Brain Tumor Segmentation Using Autoencoder Regularization

    Myronenko, Andriy. “3D MRI Brain Tumor Segmentation Using Autoencoder Regularization.” Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries,Lecture Notes in Computer Science, 2019, pp. 311–20, https://doi.org/10.1007/978-3- 030-11726-9_28

    2019 doi

  16. [23]

    Mamba: Linear-Time Sequence Modeling with Selective State Spaces

    Gu, Albert, and Tri Dao. Mamba: Linear-Time Sequence Modeling with Selective State Spaces. Dec. 2023

    2023

  17. [24]

    ECA-Net: EfficientChannel Attention for Deep Convolutional Neural Networks.In CVPR, 2020

    Qilong Wang, Banggu Wu, Pengfei Zhu, Peihua Li,Wangmeng Zuo, and Qinghua Hu. ECA-Net: EfficientChannel Attention for Deep Convolutional Neural Networks.In CVPR, 2020

    2020

  18. [25]

    Dataset of Breast Ultrasound Images

    Al-Dhabyani, Walid, et al. “Dataset of Breast Ultrasound Images.” Data in Brief, Feb. 2020, p. 104863, https://doi.org/10.1016/j.dib.2019.104863

    2020

  19. [26]

    Medical Transformer: Gated Axial-Attention for Medical Image Segmentation

    Valanarasu, Jeya Maria Jose, et al. “Medical Transformer: Gated Axial-Attention for Medical Image Segmentation.” Medical Image Computing and Computer Assisted Intervention – MICCAI 2021,Lecture Notes in Computer Science, 2021, pp. 36 –46, https://doi.org/10.1007/978-3-030-87193-2_4

    2021 doi

  20. [27]

    WM -DOVA Maps for Accurate Polyp Highlighting in Colonoscopy: Validation vs. Saliency Maps from Physicians

    Bernal, Jorge, et al. “WM -DOVA Maps for Accurate Polyp Highlighting in Colonoscopy: Validation vs. Saliency Maps from Physicians.” Computerized Medical Imaging and Graphics, July 2015, pp. 99 –111, https://doi.org/10.1016/j.compmedimag.2015.02.007

    2015 doi

  21. [28]

    arXiv preprint arXiv:1703.00523 (2017)

    Berseth, M.: Isic 2017-skin lesion analysis towards melanoma detection. arXiv preprint arXiv:1703.00523 (2017)

    2017 arXiv

  22. [29]

    arXiv preprint arXiv:1902.03368 (2019)

    Codella, N., Rotemberg, V., Tschandl, P., Celebi, M.E., Dusza, S., Gutman, D., Helba, B., Kalloo, A., Liopyris, K., Marchetti, M., et al.: Skin lesion analysis toward melanoma detection 2018: A challenge hosted by the international skin imaging collaborati on (isic). arXiv pre...

    2019 arXiv

  23. [30]

    In: 2022 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)

    Ruan, J., Xiang, S., Xie, M., Liu, T., Fu, Y.: Malunet: A multi -attention and lightweight unet for skin lesion segmentation. In: 2022 IEEE International Conference on Bioinformatics and Biomedicine (BIBM). pp. 1150–1156. IEEE (2022)

    2022

  24. [32]

    In: International Workshop on PRedictive Intelligence In MEdicine

    Azad, R., Heidari, M., Shariatnia, M., Aghdam, E.K., Karimijafarbigloo, S., Adeli, E., Merhof, D.: Transdeeplab: Convolution -free transformer-based deeplab v3+ for medical image segmentation. In: International Workshop on PRedictive Intelligence In MEdicine. pp. 91–102. Sprin...

    2022

  25. [34]

    Liu, Y.; Zhu, H.; Liu, M.; Yu, H.; Chen, Z.; and Gao, J. 2024d. Rolling-Unet: Revitalizing MLP's Ability to Effi - ciently Extract Long-Distance Dependencies for Medical Image Segmentation. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 38, 3819-3827

  26. [35]

    & Yuan, Y

    Li, C., Liu, X., Li, W., Wang, C., Liu, H., Liu, Y., ... & Yuan, Y. (2024). U -kan makes strong backbone for medical image segmentation and generation. arXiv preprint arXiv:2406.02918

    2024 arXiv

  27. [36]

    Vm -unet: Vision mamba unet for medical image segmentation

    Ruan, Jiacheng, and Suncheng Xiang. "Vm -unet: Vision mamba unet for medical image segmentation." arXiv preprint arXiv:2402.02491 (2024)

    2024 arXiv

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.