GLoG-CSUnet: Enhancing Vision Transformers with Adaptable Radiomic Features for Medical Image Segmentation

Hassan Bagher-Ebadian; Mohammad M. Ghassemi; Niloufar Eghbali; Tuka Alhanai

REVIEW 3 major objections 5 minor 21 references

GLoG-CSUnet: Enhancing Vision Transformers with Adaptable Radiomic Features for Medical Image Segmentation

T0 review · 3 major / 5 minor · reviewed 2026-08-10 · deepseek-v4-flash

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

Pith's one-line read Inserting 15 to 30 learnable Gabor and LoG filter parameters into a Swin-Unet patch embedding raises Dice scores on two medical segmentation benchmarks by 1.14 and 0.91 percentage points.

desk verdict A sensible but weakly supported incremental idea: learnable Gabor/LoG filters in a Swin-Unet patch embedding yield small Dice gains that need error bars and controlled baselines before they can be believed. read the letter →

arxiv 2501.02788 v2 pith:DU3S6LTJ submitted 2025-01-06 cs.CV cs.AIcs.LG

Niloufar Eghbali , Hassan Bagher-Ebadian , Tuka Alhanai , Mohammad M. Ghassemi This is my paper · ORCID

classification cs.CVcs.AIcs.LG

keywords medicalimagesegmentationvisiontransformerlearnableGaborfilterLaplacianofGaussianradiomicsSwin-UnetpatchembeddingDiceSimilarityCoefficient

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

The paper tries to show that a Vision Transformer for medical image segmentation can be made better at local detail simply by inserting a small bank of learnable classical filters into its patch-embedding stage. GLoG-CSUnet does this with Gabor filters for texture and orientation and Laplacian-of-Gaussian filters for edges and boundaries, adapting them during training and concatenating their outputs with the normal convolutional patch embedding before the Swin-Transformer blocks. On the Synapse multi-organ CT benchmark it reports a mean Dice of 83.36%, which is 1.14 percentage points above the best comparison model, and on the ACDC cardiac MRI benchmark it reports 92.28%, a gain of 0.91 points over the same baseline. The reason this would matter is that the whole addition costs only 15 or 30 extra parameters, suggesting that radiomics-style local features can be injected into Transformers without the parameter overhead of a hybrid CNN-Transformer design.

What carries the argument

The load-bearing mechanism is the learnable radiomic filter bank placed inside the Convolutional Patch Embedding. The Gabor layer applies 2D Gabor functions $G(x_\theta,y_\theta)=\exp(-(x_\theta^2+\gamma^2 y_\theta^2)/(2\sigma^2))\cos(2\pi x_\theta/\lambda+\psi)$ with trainable parameters $\lambda,\theta,\psi,\sigma,\gamma$; the LoG layer applies $\mathrm{LoG}(x,y)=-\frac{1}{\pi\sigma^4}(1-\frac{x^2+y^2}{2\sigma^2})e^{-(x^2+y^2)/(2\sigma^2)}$ with trainable $\sigma$. These filter responses are concatenated with the patch embedding features and then passed through convolutional layers, GELU activation, and layer normalization before the Convolutional Swin Transformer blocks process the patches. The claim is that this small set of adaptive scalars supplies precisely the local texture, edge, and boundary information the global self-attention mechanism lacks.

What would settle it

Re-run GLoG-CSUnet and CSUnet with identical preprocessing, augmentation, optimizer settings, and training budget for at least ten random seeds on the same Synapse and ACDC splits; if the distribution of Dice scores overlaps so that the difference between the two models is smaller than one standard deviation, the claimed advantage is not distinguishable from training noise.

Watch

Extended reading notes

Core claim

The central claim, on the paper's own terms, is that the local-information weakness of Vision Transformers in medical segmentation can be repaired at the embedding stage rather than by adding convolutional blocks throughout the network. In GLoG-CSUnet, the patch-embedding stage of the Convolutional Swin-Unet base is augmented with two adaptive filter units: a Gabor unit whose filters have learnable wavelength, orientation, phase, scale, and aspect ratio, and a LoG unit with a learnable scale, whose outputs are concatenated with the convolutional embedding before the transformer stages. With 2 Gabor and 5 LoG filters on Synapse and 5 of each on ACDC, the added trainable scalars number only 15 and 30; the reported mean Dice reaches 83.36% on Synapse and 92.28% on ACDC, with the largest per-organ gains on structures with fine or irregular boundaries such as the stomach and the right ventricle.

Load-bearing premise

The load-bearing premise is that a reported improvement of about one percentage point in Dice is larger than the ordinary run-to-run variation of these models, since the comparison pits a single run of the new model against baseline numbers quoted from earlier papers without error bars.

Editorial extensions

If this is right

If the reported gain is real, local feature extraction for Transformer segmentation can be added at the embedding stage rather than by deepening the convolutional component of a hybrid model.
With only 15 or 30 added scalars, the method would preserve the runtime and memory profile of the base model, making it usable in clinical settings with tight compute budgets.
The per-organ results suggest that the filter bank helps most on structures with weak or irregular boundaries, such as the stomach in CT and the right ventricle in cardiac MRI.
The paper's flexibility argument implies the same filter-bank insertion should transfer to other Transformer-based backbones, so the design is a plug-in rather than a full architecture rewrite.

Reading between the lines

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

A test the paper leaves implicit: running GLoG-CSUnet and CSUnet several times with different random seeds on the same data splits would show whether the one-point Dice difference exceeds seed-to-seed variability; the paper reports single runs.
The learned filter parameters themselves are not analyzed; inspecting them after training could reveal whether the model discovers organ-specific orientations or scales, and whether those values transfer across datasets.
The same embedding filter bank could be evaluated on natural-image segmentation or detection tasks, where boundary and texture cues also matter, to see whether the benefit is specific to small medical datasets.
Because the paper only trains from scratch, a natural extension is to test the filter bank on top of a pre-trained ViT backbone; if the gain disappears, the method may be substituting for features that large-scale pre-training already provides.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Desk Editor's Note

read the letter

Short version: this paper puts a few learnable Gabor and LoG filters into the patch embedding of a Swin-Unet variant and reports Dice gains of about 1 point on Synapse and ACDC. The idea is sensible and easy to test; the execution is clean enough to follow. But the reported gains are not backed by the kind of evidence needed to believe them: no error bars, no multiple seeds, and baselines are copied from earlier papers. As written, the central claim is plausible but not established.

What's genuinely new: I don't know of a prior Swin-based segmentation model that inserts learnable Gabor/LoG units into the patch embedding. That specific integration is a reasonable extension of existing hybrid transformer-CNN work. The paper does a decent job of describing the architecture, gives the filter equations, runs ablations, and points to code.

The soft spots are mostly evidentiary. The 1.14-point Synapse gain and 0.91-point ACDC gain are smaller than typical seed-to-seed variation for these datasets, and the paper gives no way to assess that variation. The baselines are quoted from their original papers, so training protocols differ; the paper even says pretrained weights were used for baselines 'if provided,' while their model is trained from scratch. The ablation on ACDC shows a 0.45-point gap between Gabor-only and combined, which is the same order as the headline gain, so a single run per variant cannot establish that the combination is systematic.

The 'minimal overhead' claim is also under-specified. The learnable scalars are indeed 15 and 30, but concatenating filter outputs into the patch embedding increases the input channels of the following conv layers, adding thousands of parameters. That doesn't kill the idea, but '15 and 30 additional parameters' overstates the resource argument.

There are smaller issues: the abstract says 0.99% for ACDC while Table II says 0.91; the discussion has a typo ('D of 23.02') and 'succsses'; and the authors miss prior deep-learning work on learnable Gabor filters. These are fixable.

Who benefits: anyone building transformer segmentation models on small medical datasets and wondering if radiomic-style inductive biases help. The paper is a reasonable starting point, not a finished statement.

Recommendation: send it to peer review, not desk reject. Ask the authors to report multiple seeds with error bars, re-run at least the strongest baselines under the same protocol, reconcile the parameter count with the actual architecture, and fix the abstract/table mismatch.

Referee Report

3 major / 5 minor

Summary. The paper proposes GLoG-CSUnet, an extension of the Convolutional Swin-Unet (CS-Unet) architecture that inserts learnable Gabor and Laplacian-of-Gaussian (LoG) filters into the patch-embedding stage of a Swin-Transformer-based segmentation model. The method is evaluated on the Synapse multi-organ CT and ACDC cardiac MRI benchmarks. The authors report Dice similarity coefficients of 83.36% on Synapse and 92.28% on ACDC, outperforming CS-Unet by about 1 percentage point, with a claimed overhead of only 15 and 30 additional parameters. An ablation on ACDC reports Gabor-only 92.01%, LoG-only 91.83%, and combined 92.28%. The paper argues that radiomics-inspired filters complement the global modeling of Transformers, particularly for small medical datasets.

Significance. If the reported gains are reproducible, the contribution is a simple and modular patch-embedding enhancement that yields consistent, albeit small, Dice improvements on two standard benchmarks at negligible parameter cost. The paper provides an ablation study and a public code repository, which are useful for verification. However, the evidence as presented is not yet sufficient to establish the central claim: all baseline numbers are quoted from prior publications, no error bars or multiple-seed results are reported, and the headline numbers in the abstract and body are internally inconsistent. The parameter-overhead claim also appears incompatible with the architecture description. The idea is worth publishing if these load-bearing issues are resolved with additional experiments and corrected reporting.

major comments (3)

[Section III, Tables I and II; Section II.E; Section III.A] The claimed Dice gains of about 1.14 and 0.91 percentage points over CS-Unet are not supported as 'significant improvements' because no method is evaluated with multiple random seeds or error bars. All baseline numbers in Tables I and II are copied from earlier publications, whereas GLoG-CSUnet is trained from scratch for 300 epochs with dataset-specific learning rates, augmentation, and a batch size of 24; differences in training protocol can easily produce differences of this magnitude. The ablation in Section III.A compounds the problem: the component gaps (Gabor-only 92.01 vs LoG-only 91.83 vs combined 92.28) are the same order of magnitude as the headline gain, so a single run per variant cannot establish that the effect is systematic. Please provide mean and standard deviation over at least three seeds for the proposed method, and ideally re-run CS-Unet under the identical training protocol.
[Section II.A, Figure 1(b), Section IV] The claim that GLoG-CSUnet adds only 15 parameters on Synapse and 30 on ACDC is not consistent with the architecture as described. In Figure 1(b), the outputs of the Gabor and LoG units are concatenated and then processed by 'multiple convolutional layers with GELU activation and layer normalization,' which introduces learnable convolution kernels and changes the input-channel dimension of the first convolutional layer. Counting only the scalar filter parameters (e.g., wavelength, orientation, sigma) omits these weights. Please report the total number of parameters and, if possible, FLOPs before and after the modification, and reconcile the count with Figure 1(b).
[Abstract vs. Table II] The abstract states that GLoG-CSUnet achieves a 0.99% increase in Dice score for ACDC, but Table II reports 92.28% for the proposed method versus 91.37% for CS-Unet, which is a difference of 0.91 percentage points; for Synapse, the table values give 83.36 - 82.21 = 1.15 percentage points, not 1.14. The central quantitative claim must be internally consistent; please correct the abstract or the tables and specify the exact rounding convention.

minor comments (5)

[Table I] The column header 'D' in Table I is undefined; it should be 'HD95' to match the metric described in Section II.D and the text in Section III.
[Section II.B.1, Eq. (1)] The symbol λ is used for both the wavelength of the sinusoidal component and the spatial aspect ratio in the equation; the aspect ratio is conventionally denoted γ. Please clarify the notation.
[Section II.E] The sentence fragment 'training dynamics.' appears in the experimental setup paragraph and should be removed or completed.
[Section IV and V] There are typos: 'sucsses' should be 'success' in Section IV, and 'a advancement' should be 'an advancement' in Section V.
[Section I.A] The related-work discussion would benefit from citing the original Swin Transformer paper alongside Swin-Unet, since the backbone is the Swin Transformer and the comparison is made against Swin-Unet.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the paper reports standard empirical benchmark results with held-out test evaluation, and no claimed result reduces to its own inputs.

full rationale

This is an empirical architecture paper, not a derivation. The Gabor and LoG equations are standard definitions cited to external sources, and the model is trained from scratch and evaluated on held-out test splits of the Synapse and ACDC datasets. The reported Dice improvements are measured outcomes, not quantities fitted to the test set or constructed from the evaluation metric. Baselines in Tables I and II are quoted from prior publications, but quoting baselines is not a circular step: the proposed model's numbers come from its own evaluation, and any concern about different training protocols or missing error bars is a reproducibility/validity issue, not circularity. The claim of only 15 and 30 additional parameters may be inconsistent with the architecture's concatenation-and-convolution design, but that is an accounting or correctness concern, not a reduction of the central result to its inputs. No self-citation chain, no imported uniqueness theorem, and no renamed-known-result pattern appears. The paper is self-contained in the sense that its contribution is an empirical comparison, so the circularity score is 0.

Assumptions & free parameters 8 free parameters · 3 assumptions · 0 invented entities

The paper's central claim is an empirical performance comparison, not a derivation. The listed free parameters are manually chosen hyperparameters and filter counts that affect the reported results; none were fitted to the test set. The axioms are background assumptions about benchmark comparability and the premise that Gabor/LoG features complement transformer context modeling. No new entities are introduced.

free parameters (8)

Gabor filter count (Synapse) = 2
Number of learnable Gabor filters used for the Synapse dataset; chosen by hand and affects the representation and the reported 15-parameter overhead.
Gabor filter count (ACDC) = 5
Number of learnable Gabor filters used for the ACDC dataset; chosen by hand and affects the reported 30-parameter overhead.
LoG filter count (both datasets) = 5
Number of learnable LoG filters used for both datasets; chosen by hand.
Learning rate (Synapse) = 1e-3
Dataset-specific learning rate for AdamW on Synapse; chosen by hand.
Learning rate (ACDC) = 1e-2
Dataset-specific learning rate for AdamW on ACDC; chosen by hand.
Weight decay = 2e-4
AdamW weight decay; chosen by hand.
Batch size = 24
Mini-batch size for training; chosen by hand.
Training epochs = 300
Number of training epochs from scratch; chosen by hand.

assumptions (3)

domain assumption The quoted baseline results from prior papers are accurate and are obtained under conditions comparable to the authors' training setup.
The paper does not retrain baselines; it cites numbers from Chen et al. [8], Cao et al. [10], Wang et al. [11], and Liu et al. [12]. Any difference in data preprocessing, augmentation, or training duration could explain the reported gains. Stated in Section II.E and Section III.
ad hoc to paper The Gabor and LoG filters, with only a few learnable scalars, provide complementary local features that improve segmentation on top of CSUnet.
This is the paper's central premise, not independently established; the ablation shows small gains but no statistical confidence. Stated in Section I and Section II.
domain assumption The two public datasets (Synapse, ACDC) and the standard split are representative enough that a ~1% Dice gain on them indicates a generally useful improvement.
Used as evidence for generalizable value. Stated in Section II.C.

how reviews work

0 comments

Cite this review

Pith. "Pith review of GLoG-CSUnet: Enhancing Vision Transformers with Adaptable Radiomic Features for Medical Image Segmentation." pith.science (2026). https://pith.science/paper/DU3S6LTJ

@misc{pith2026250102788,
  author       = {Pith},
  title        = {Pith review of: GLoG-CSUnet: Enhancing Vision Transformers with Adaptable Radiomic Features for Medical Image Segmentation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/DU3S6LTJ}},
  note         = {Machine review of arXiv:2501.02788}
}

read the original abstract

Vision Transformers (ViTs) have shown promise in medical image semantic segmentation (MISS) by capturing long-range correlations. However, ViTs often struggle to model local spatial information effectively, which is essential for accurately segmenting fine anatomical details, particularly when applied to small datasets without extensive pre-training. We introduce Gabor and Laplacian of Gaussian Convolutional Swin Network (GLoG-CSUnet), a novel architecture enhancing Transformer-based models by incorporating learnable radiomic features. This approach integrates dynamically adaptive Gabor and Laplacian of Gaussian (LoG) filters to capture texture, edge, and boundary information, enhancing the feature representation processed by the Transformer model. Our method uniquely combines the long-range dependency modeling of Transformers with the texture analysis capabilities of Gabor and LoG features. Evaluated on the Synapse multi-organ and ACDC cardiac segmentation datasets, GLoG-CSUnet demonstrates significant improvements over state-of-the-art models, achieving a 1.14% increase in Dice score for Synapse and 0.99% for ACDC, with minimal computational overhead (only 15 and 30 additional parameters, respectively). GLoG-CSUnet's flexible design allows integration with various base models, offering a promising approach for incorporating radiomics-inspired feature extraction in Transformer architectures for medical image analysis. The code implementation is available on GitHub at: https://github.com/HAAIL/GLoG-CSUnet.

Figures

Figures reproduced from arXiv: 2501.02788 by the authors.

**Figure 2.** Visualization of segmentation results for two test cases using our [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

21 extracted references · 17 canonical work pages

[1]

Towards robust general medical image segmentation,

Laura Daza, Juan C P ´erez, and Pablo Arbel ´aez, “Towards robust general medical image segmentation,” in Medical Image Computing and Computer Assisted Intervention–MICCAI 2021: 24th International Conference, Strasbourg, France, September 27–October 1, 2021, Pro- ceedings, Part III 24 . Springer, 2021, pp. 3–13

work page 2021
[2]

Medical image segmentation using deep learning: A survey,

Risheng Wang, Tao Lei, Ruixia Cui, Bingtao Zhang, Hongying Meng, and Asoke K Nandi, “Medical image segmentation using deep learning: A survey,” IET image processing, vol. 16, no. 5, pp. 1243–1267, 2022

work page 2022
[3]

Pet-guided delineation of radiation therapy treatment volumes: a survey of image segmentation techniques,

Habib Zaidi and Issam El Naqa, “Pet-guided delineation of radiation therapy treatment volumes: a survey of image segmentation techniques,” European journal of nuclear medicine and molecular imaging , vol. 37, pp. 2165–2187, 2010

work page 2010
[4]

An integrated visualization system for surgical planning and guidance using image fusion and an open mr,

David T Gering, Arya Nabavi, Ron Kikinis, Noby Hata, Lauren J O’Donnell, W Eric L Grimson, Ferenc A Jolesz, Peter M Black, and William M Wells III, “An integrated visualization system for surgical planning and guidance using image fusion and an open mr,” Journal of Magnetic Resonance Imaging: An Official Journal of the International Society for Magnetic R...

work page 2001
[5]

U-net: Con- volutional networks for biomedical image segmentation,

Olaf Ronneberger, Philipp Fischer, and Thomas Brox, “U-net: Con- volutional networks for biomedical image segmentation,” in Medical image computing and computer-assisted intervention–MICCAI 2015: 18th international conference, Munich, Germany, October 5-9, 2015, proceedings, part III 18 . Springer, 2015, pp. 234–241

work page 2015
[6]

Unet++: A nested u-net architecture for medical image segmentation,

Zongwei Zhou, Md Mahfuzur Rahman Siddiquee, Nima Tajbakhsh, and Jianming Liang, “Unet++: A nested u-net architecture for medical image segmentation,” in Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support: 4th Interna- tional Workshop, DLMIA 2018, and 8th International Workshop, ML- CDS 2018, Held in Conjunction w...

work page 2018
[7]

Unet 3+: A full-scale connected unet for medical image segmentation,

Huimin Huang, Lanfen Lin, Ruofeng Tong, Hongjie Hu, Qiaowei Zhang, Yutaro Iwamoto, Xianhua Han, Yen-Wei Chen, and Jian Wu, “Unet 3+: A full-scale connected unet for medical image segmentation,” in ICASSP 2020-2020 IEEE international conference on acoustics, speech and signal processing (ICASSP) . IEEE, 2020, pp. 1055–1059

work page 2020
[8]

Transunet: Transformers make strong encoders for medical image segmentation,

Jieneng Chen, Yongyi Lu, Qihang Yu, Xiangde Luo, Ehsan Adeli, Yan Wang, Le Lu, Alan L Yuille, and Yuyin Zhou, “Transunet: Transformers make strong encoders for medical image segmentation,” arXiv preprint arXiv:2102.04306, 2021

arXiv 2021

Show all 21 references

[9]

An image is worth 16x16 words: Transformers for image recognition at scale,

Alexey DOSOVITSKIY , “An image is worth 16x16 words: Transformers for image recognition at scale,” arXiv preprint arXiv:2010.11929, 2020

2010 arXiv
[10]

Swin-unet: Unet-like pure transformer for medical image segmentation,

Hu Cao, Yueyue Wang, Joy Chen, Dongsheng Jiang, Xiaopeng Zhang, Qi Tian, and Manning Wang, “Swin-unet: Unet-like pure transformer for medical image segmentation,” in European conference on computer vision. Springer, 2022, pp. 205–218

2022
[11]

Mixed transformer u-net for medical image segmentation,

Hongyi Wang, Shiao Xie, Lanfen Lin, Yutaro Iwamoto, Xian-Hua Han, Yen-Wei Chen, and Ruofeng Tong, “Mixed transformer u-net for medical image segmentation,” in ICASSP 2022-2022 IEEE international conference on acoustics, speech and signal processing (ICASSP) . IEEE, 2022, pp. 2390–2394

2022
[12]

Optimizing vision trans- formers for medical image segmentation,

Qianying Liu, Chaitanya Kaul, Jun Wang, Christos Anagnostopoulos, Roderick Murray-Smith, and Fani Deligianni, “Optimizing vision trans- formers for medical image segmentation,” in ICASSP 2023-2023 IEEE international conference on acoustics, speech and signal processing (ICASSP...

2023
[13]

Introduction to radiomics,

Marius E Mayerhoefer, Andrzej Materka, Georg Langs, Ida H ¨aggstr¨om, Piotr Szczypi ´nski, Peter Gibbs, and Gary Cook, “Introduction to radiomics,” Journal of Nuclear Medicine , vol. 61, no. 4, pp. 488–495, 2020

2020
[14]

Standardised convolutional filtering for radiomics,

Adrien Depeursinge, Vincent Andrearczyk, Philip Whybra, Joost van Griethuysen, Henning M ¨uller, Roger Schaer, Martin Valli`eres, and Alex Zwanenburg, “Standardised convolutional filtering for radiomics,” arXiv preprint arXiv:2006.05470, 2020

2006 arXiv
[15]

Interpretation of radiomics features–a pictorial review,

Ali Abbasian Ardakani, Nathalie J Bureau, Edward J Ciaccio, and U Rajendra Acharya, “Interpretation of radiomics features–a pictorial review,” Computer methods and programs in biomedicine , vol. 215, pp. 106609, 2022

2022
[16]

In search of a general picture processing operator,

Goesta H Granlund, “In search of a general picture processing operator,” Computer Graphics and Image Processing , vol. 8, no. 2, pp. 155–173, 1978

1978
[17]

Miccai multi-atlas labeling beyond the cranial vault–workshop and challenge,

Bennett Landman, Zhoubing Xu, J Igelsias, Martin Styner, Thomas Langerak, and Arno Klein, “Miccai multi-atlas labeling beyond the cranial vault–workshop and challenge,” in Proc. MICCAI Multi-Atlas Labeling Beyond Cranial Vault—Workshop Challenge , 2015, vol. 5, p. 12

2015
[18]

Deep learning techniques for automatic mri cardiac multi-structures segmentation and diagnosis: is the problem solved?,

Olivier Bernard, Alain Lalande, Clement Zotti, Frederick Cervenansky, Xin Yang, Pheng-Ann Heng, Irem Cetin, Karim Lekadir, Oscar Camara, Miguel Angel Gonzalez Ballester, et al., “Deep learning techniques for automatic mri cardiac multi-structures segmentation and diagnosis: is...

2018
[19]

Metrics for evaluating 3d medical image segmentation: analysis, selection, and tool,

Abdel Aziz Taha and Allan Hanbury, “Metrics for evaluating 3d medical image segmentation: analysis, selection, and tool,” BMC medical imaging, vol. 15, pp. 1–28, 2015

2015
[20]

Comparing images using the hausdorff distance,

Daniel P Huttenlocher, Gregory A. Klanderman, and William J Ruck- lidge, “Comparing images using the hausdorff distance,” IEEE Transactions on pattern analysis and machine intelligence , vol. 15, no. 9, pp. 850–863, 1993

1993
[21]

Decoupled weight decay regularization,

I Loshchilov, “Decoupled weight decay regularization,” arXiv preprint arXiv:1711.05101, 2017

2017 arXiv

Pith tools

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

[1] [1]

Towards robust general medical image segmentation,

Laura Daza, Juan C P ´erez, and Pablo Arbel ´aez, “Towards robust general medical image segmentation,” in Medical Image Computing and Computer Assisted Intervention–MICCAI 2021: 24th International Conference, Strasbourg, France, September 27–October 1, 2021, Pro- ceedings, Part III 24 . Springer, 2021, pp. 3–13

work page 2021

[2] [2]

Medical image segmentation using deep learning: A survey,

Risheng Wang, Tao Lei, Ruixia Cui, Bingtao Zhang, Hongying Meng, and Asoke K Nandi, “Medical image segmentation using deep learning: A survey,” IET image processing, vol. 16, no. 5, pp. 1243–1267, 2022

work page 2022

[3] [3]

Pet-guided delineation of radiation therapy treatment volumes: a survey of image segmentation techniques,

Habib Zaidi and Issam El Naqa, “Pet-guided delineation of radiation therapy treatment volumes: a survey of image segmentation techniques,” European journal of nuclear medicine and molecular imaging , vol. 37, pp. 2165–2187, 2010

work page 2010

[4] [4]

An integrated visualization system for surgical planning and guidance using image fusion and an open mr,

David T Gering, Arya Nabavi, Ron Kikinis, Noby Hata, Lauren J O’Donnell, W Eric L Grimson, Ferenc A Jolesz, Peter M Black, and William M Wells III, “An integrated visualization system for surgical planning and guidance using image fusion and an open mr,” Journal of Magnetic Resonance Imaging: An Official Journal of the International Society for Magnetic R...

work page 2001

[5] [5]

U-net: Con- volutional networks for biomedical image segmentation,

Olaf Ronneberger, Philipp Fischer, and Thomas Brox, “U-net: Con- volutional networks for biomedical image segmentation,” in Medical image computing and computer-assisted intervention–MICCAI 2015: 18th international conference, Munich, Germany, October 5-9, 2015, proceedings, part III 18 . Springer, 2015, pp. 234–241

work page 2015

[6] [6]

Unet++: A nested u-net architecture for medical image segmentation,

Zongwei Zhou, Md Mahfuzur Rahman Siddiquee, Nima Tajbakhsh, and Jianming Liang, “Unet++: A nested u-net architecture for medical image segmentation,” in Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support: 4th Interna- tional Workshop, DLMIA 2018, and 8th International Workshop, ML- CDS 2018, Held in Conjunction w...

work page 2018

[7] [7]

Unet 3+: A full-scale connected unet for medical image segmentation,

Huimin Huang, Lanfen Lin, Ruofeng Tong, Hongjie Hu, Qiaowei Zhang, Yutaro Iwamoto, Xianhua Han, Yen-Wei Chen, and Jian Wu, “Unet 3+: A full-scale connected unet for medical image segmentation,” in ICASSP 2020-2020 IEEE international conference on acoustics, speech and signal processing (ICASSP) . IEEE, 2020, pp. 1055–1059

work page 2020

[8] [8]

Transunet: Transformers make strong encoders for medical image segmentation,

Jieneng Chen, Yongyi Lu, Qihang Yu, Xiangde Luo, Ehsan Adeli, Yan Wang, Le Lu, Alan L Yuille, and Yuyin Zhou, “Transunet: Transformers make strong encoders for medical image segmentation,” arXiv preprint arXiv:2102.04306, 2021

arXiv 2021