Pith. sign in

REVIEW 6 major objections 5 minor 1 cited by

S2S2: Semantic Stacking for Robust Semantic Segmentation in Medical Imaging

T0 review · 6 major / 5 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read A training-time consistency loss between real and synthetic images sharing the same segmentation map improves both in-domain accuracy and out-of-domain robustness in medical image segmentation.

desk verdict A plausible plug-in regularizer with a novel framing, but the headline gains are confounded by asymmetric early stopping and per-method hyperparameter selection. read the letter →

arxiv 2412.13156 v1 pith:BGNKLMFW submitted 2024-12-17 cs.CV

Yimu Pan , Sitao Zhang , Alison D. Gernand , Jeffery A. Goldstein , James Z. Wang This is my paper · ORCID
classification cs.CV
keywords semanticsegmentationdomaingeneralizationfeatureconsistencylosssyntheticmedicalimagesgenerativeaugmentationdenoisingbystackingdiffusionmodelsrobustness
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 proposes semantic stacking, a training-time add-on for medical image segmentation that improves both accuracy on the training domain and robustness to unseen domains. The idea is to generate several synthetic images from each ground-truth segmentation map and then add a loss that pushes the network's feature representations for real and synthetic images with the same map together. The authors argue this mimics image stacking in denoising: pooling many noisy semantic features yields an estimate closer to the true features, and they derive a bound showing two images per iteration suffice. Experiments across CT, MRI, and RGB polyp datasets show consistent Dice-score gains over several base architectures and augmentation baselines. The method is domain-agnostic, needing no assumption about modality or target distribution.

What carries the argument

The central object is the 'semantic stack': a set of images produced by a fine-tuned generative model (Stable Diffusion with ControlNet) conditioned on the same segmentation map, together with the pool of their encoder and decoder features. The mechanism is Bayesian-averaged pooling of these noisy semantic features to approximate the true feature map, and the operational identity is the upper bound $$D\bigl(t_i, E[\hat t]\bigr) \le \frac{\$sigma^{2}$}{\$sigma^{2}$ + n\$sigma_0^{2}$}\,D(t_i, t_0) + \frac{\$sigma_0^{2}$}{\$sigma^{2}$ + n\$sigma_0^{2}$}\sum_{j\neq i}D(t_i, t_j),$$ which justifies sampling only two images per iteration and minimizing their feature distance.

What would settle it

Train the same architecture with S2S2 but swap the generator for one with a known systematic bias, for example always rendering one organ with reversed intensity while keeping masks correct; if Dice still rises over baseline, the unbiased-mean assumption is not necessary for the gain, and if it falls, the assumption is load-bearing. Alternatively, compute the feature-space mean of generated versus real images on a validation set and compare the distance to the ground-truth features.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central claim is that minimizing a pairwise feature-consistency loss between a real image and a synthetic image that share a ground-truth segmentation map reduces bias and variance in the learned semantic feature space, thereby improving both in-domain segmentation and single-source domain generalization. The theoretical anchor is the inequality in Eq. (4), which upper-bounds the distance from any one feature map to the pooled posterior mean by a weighted sum of pairwise distances, so minimizing pairwise distances effectively minimizes distance to the denoised semantic representation. The loss is $L_{\mathrm{sc}} = D(F(x_i), F(x_j))$ with $D$ a triangle-inequality metric (implemented as $1 - \mathrm{CosSim}$), applied to encoder and decoder features, added to the segmentation loss.

Load-bearing premise

The whole benefit rests on the assumption that the mean of the generated images' semantic features equals the true feature mean of real images; if a generator produces biased appearances, the consistency loss pulls the network toward artifacts and can hurt instead of help (the paper states this as $t_g \approx t$ in Section 3.3 and never directly measures it).

Editorial extensions

If this is right

  • If the claim holds, any existing segmentation pipeline can be upgraded by adding synthetic-image feature consistency without altering architecture or augmentation, giving a plug-in robustness boost.
  • Domain-specific augmentation methods like SLAug, which already exploit CT/MRI intensity priors, still gain a small additional improvement when S2S2 is stacked on top, and the combination transfers to RGB where SLAug alone underperforms.
  • The method's benefit grows with the number of segmentation classes, suggesting the consistency loss is most valuable in complex multi-organ tasks.
  • Because the method requires a fine-tuned generative model per dataset, it is best suited to small-data medical settings; in data-rich regimes the extra cost may outweigh the gains (the paper's own caveat in the discussion).

Reading between the lines

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

  • If feature-space averaging is the true mechanism, the same loss should work with any generator that produces diverse but class-consistent appearance, including simpler augmentation chains; a direct test would compare synthetic-image stacks against photometric augmentation stacks with the same two-image loss.
  • The paper's bound is loose, so the claim that two images suffice is a sufficient condition rather than a proven optimum; measuring the gap between the bound and the actual distance could yield a principled way to choose the stack size.
  • The paper's Table 12 shows two settings where S2S2 does not beat the SLAug baseline, so the robustness claim should be read as conditional on how well a domain-specific baseline already captures target variation, not as a universal improvement.
  • A practical extension would be to make the loss adaptive to generator quality, for example weighting the consistency term by a per-dataset estimate of synthetic-feature bias.
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

6 major / 5 minor

Summary. This paper proposes S2S2, a training-time add-on for medical image segmentation that adds a semantic consistency loss between a real image and a synthetic image generated from the same ground-truth segmentation map. Synthetic images are produced by a per-dataset fine-tuned Stable Diffusion model with ControlNet. The method is motivated by an analogy to denoising by image stacking: pooling features from multiple synthetic views is claimed to estimate a denoised 'ground-truth semantic feature map.' A Bayesian-updating argument is used to reduce the loss to pairwise feature distances between two images. Experiments cover CT, MRI, and RGB polyp datasets across several architectures, reporting in-domain Dice gains and mixed out-of-domain gains, with ablations of the loss weights and of synthetic-image usage.

Significance. If the empirical claims were fully substantiated, S2S2 would be a convenient plug-in for improving segmentation robustness without domain-specific augmentation, and the two-image approximation would make it cheap to train. The paper's strengths are its breadth of modalities and models, the public code release, and the explicit attempt to connect the loss to a statistical stacking argument. However, the current evidence does not establish the claimed superiority because of confounded training schedules, outcome-dependent hyperparameter choices, and an absence of variance estimates; furthermore, the theoretical justification rests on a metric assumption that the implemented loss violates. After addressing these issues the method could be a useful contribution.

major comments (6)
  1. [Appendix C (Evaluation Methodology)] The comparison between SLAug and SLAug+S2S2 uses different training durations: the text states that the authors 'opt for an early stopping of the training process at 1,100 epochs for our method, as opposed to extending to the full 2,000 epochs,' while applying early stopping to the SLAug baseline 'negatively affects performance.' This makes the gains in Tables 3, 4, 10, and 12 uninterpretable as effects of the semantic-consistency loss, because the two methods are evaluated at different points in training. Please re-run the baseline under the same early-stopping rule, or report results at matched epochs (e.g., 1,100 and 2,000) to show that the conclusions are invariant.
  2. [Appendix C and Sec. 5 (Implementation Details)] The consistency-loss weights are chosen per model and dataset: αenc=1 for TransUNet, αenc=0.4 for FCBFormer, and αenc=0.1 for SLAug, and the FCBFormer configuration is described as 'optimizing for balanced performance enhancement.' Without a pre-specified selection protocol (e.g., fixed weights or validation-based selection before seeing test results), the favorable α values are a potential confound. Please provide a systematic tuning protocol or report performance across a grid of weights with the chosen value justified on validation data.
  3. [Tables 1-5 and 8-13] All experimental results are single runs without standard deviations, multiple seeds, or significance tests, despite differences as small as +0.15 mean Dice in Table 4 and negative differences in Table 12. The claim that S2S2 'significantly elevates' in-domain performance is therefore not statistically supported. Please report at least three seeds per condition together with confidence intervals or a paired significance test.
  4. [Sec. 3.2, Eq. (4), and Implementation Details] The derivation that minimizing D(ti, E[ˆt]) 'effectively requires minimizing D(ti,tj)' is only an upper-bound argument: a decrease in the bound does not imply a corresponding decrease in the minimized quantity, so the logical step is not established. Moreover, the loss is implemented as D(ti,tj) = 1 - CosSim(ti,tj), which does not satisfy the triangle inequality that the derivation explicitly assumes. Please either prove the implication for a metric that is actually used (e.g., angular distance), or state clearly that the pairwise loss is a heuristic inspired by the bound.
  5. [Sec. 3.3] The assumption that the mean feature map of synthetic images equals the ground-truth mean, tg ≈ t, is load-bearing: if the generative model is biased, the consistency loss pulls the segmenter toward generator-specific artifacts. The paper provides no measurement supporting this assumption and concedes that 'empirically validating this condition may be challenging.' Please provide empirical evidence about the feature statistics of real versus synthetic images (e.g., mean and variance of encoder features over a validation set), or otherwise characterize the failure mode.
  6. [Tables 4 and 12, Appendix D] The out-of-domain results are direction-dependent: SLAug+S2S2 is worse than SLAug on CT-MRI (−0.80) and bSSFP-LGE (−0.36) in Table 12, and the mean gain in Table 4 is +0.15. The main text's statement that S2S2 yields 'marginal improvements' and the abstract's claim of superiority overstate the evidence. Please present the per-direction results in the main text and either explain when S2S2 is expected to help or soften the universal claim.
minor comments (5)
  1. [Fig. 7 (pseudocode)] In line 23 of the pseudocode, the decoder consistency loss is computed with the function `enc_consist`; this is likely a typo for `dec_consist`.
  2. [Appendix A] 'Housefield values' should be 'Hounsfield values'.
  3. [Sec. 5, In-domain Performance] 'CT/MRT' appears to be a typo for 'CT/MRI'.
  4. [Sec. 3.2, Eq. (3)] The symbols σ and σ0 are used without prior definition; please define them as feature-noise variance and prior variance, respectively.
  5. [Appendix D] The sentence 'SLAug+CT (or SLAug+bSSFP) may better address the variations introduced by MRI (or LGE)' is confusing because the source-target direction is not stated explicitly; please rephrase as 'when CT is the source and MRI the target'.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the S2S2 consistency loss is an independent training objective and the reported Dice gains are measured on held-out test data; flagged assumptions and the asymmetric early-stopping rule are limitations, not circular reductions.

full rationale

The central claim—that adding L_sc = D(F(x_i), F(x_j)) between real and synthetic images sharing a segmentation map improves in-domain and out-of-domain Dice—is not an input to the loss or a rename of the evaluation metric. The derivation in Sec. 3.2 is a genuine upper bound: Eq. (4) bounds D(t_i, E[t-hat]) by a weighted sum of pairwise distances, and Eq. (5) then minimizes the pairwise term; this is a sufficient condition derived from the triangle inequality, not an identity that presupposes the reported improvement. The practical two-image reduction follows from that inequality rather than from any fitted target result. The paper's 'we posit that tg approximately equals t' (Sec. 3.3) is an unverified bias assumption about the generative model, and 'theoretical guarantees ensure its validity as n increases' is conditional on that assumption; both are support gaps, not circularity. The only self-citation (Pan et al. 2023, AI-SAM) appears in a contextual list of interactive segmentation tools and is not load-bearing. Appendix C's asymmetric early stopping ('we opt for an early stopping of the training process at 1,100 epochs for our method... applying early stopping to the SLAug baseline negatively affects performance') is a legitimate confound that threatens the empirical comparison, but it does not make the derivation circular: the consistency loss is still an independent objective and the test Dice is not optimized directly. Per-model choices of alpha (e.g., alpha_enc=1, 0.4, 0.1) are hyperparameter selections reported without error bars; this weakens the evidence but does not reduce the prediction to a fit of the predicted quantity. Overall, no step exhibits a self-definitional reduction, a fitted-input-as-prediction, or a load-bearing self-citation chain.

Assumptions & free parameters 4 free parameters · 4 assumptions · 1 invented entities

The central claim rests on a latent-variable model of features, an unbiased-generator assumption, and a triangle-inequality condition that the actual loss violates. The only fitted numbers are loss weights and training length, chosen per model after ablation. No new physical entities are introduced.

free parameters (4)
  • alpha_enc = 1.0 (TransUNet), 0.4 per encoder (FCBFormer), 0.1 (SLAug)
    Weight of encoder consistency loss; selected per model based on ablation experiments and affects final reported performance.
  • alpha_dec = 1.0 (TransUNet), 0.0 (FCBFormer final, SLAug)
    Weight of decoder consistency loss; chosen per model, with FCBFormer and SLAug final runs using alpha_dec = 0.
  • Stack size n = 16 synthetic images per mask
    Number of synthetic images generated per ground-truth mask; chosen without ablation and used in all experiments.
  • Early stopping epoch = 1100 for S2S2, 2000 for SLAug baseline
    S2S2 training is stopped early because the method achieves high in-domain performance; applying early stopping to the baseline 'negatively affects performance', making the comparison uneven.
assumptions (4)
  • domain assumption Feature maps of images sharing the same segmentation map are i.i.d. Gaussian samples around a single ground-truth feature map t.
    Invoked in Sec. 3.1 Eq. 2; the entire noise-reduction analogy and the pooled variance reduction depend on this distributional assumption, which is not verified.
  • ad hoc to paper Fine-tuned generative model yields synthetic images whose feature-map mean equals the true mean (tg approximately equals t).
    Stated as 'we posit that tg approximately equals t' in Sec. 3.3; this is load-bearing for the benefit of semantic stacking and is not validated empirically.
  • ad hoc to paper The distance D used in the loss is a metric satisfying the triangle inequality.
    The derivation in Eq. 4 relies on the triangle inequality, but the implemented loss uses 1 - CosSim, which is not a metric and does not satisfy the triangle inequality.
  • domain assumption A ground-truth semantic feature map t exists and the classifier H maps it to the segmentation label map y.
    Sec. 3.1 defines y = H(t); this latent-variable structure is assumed to justify the semantic stacking objective.
invented entities (1)
  • Ground-truth semantic feature map t
    purpose: Latent variable that all views of the same anatomy share; the proposed stacking theory assumes feature maps are noisy samples around t.
    Core to the theory but never observed or independently measured; its existence and the Gaussian noise model are postulates of the paper.

how reviews work

0 comments
Cite this review

Pith. "Pith review of S2S2: Semantic Stacking for Robust Semantic Segmentation in Medical Imaging." pith.science (2026). https://pith.science/paper/BGNKLMFW

@misc{pith2026241213156,
  author       = {Pith},
  title        = {Pith review of: S2S2: Semantic Stacking for Robust Semantic Segmentation in Medical Imaging},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/BGNKLMFW}},
  note         = {Machine review of arXiv:2412.13156}
}
read the original abstract

Robustness and generalizability in medical image segmentation are often hindered by scarcity and limited diversity of training data, which stands in contrast to the variability encountered during inference. While conventional strategies -- such as domain-specific augmentation, specialized architectures, and tailored training procedures -- can alleviate these issues, they depend on the availability and reliability of domain knowledge. When such knowledge is unavailable, misleading, or improperly applied, performance may deteriorate. In response, we introduce a novel, domain-agnostic, add-on, and data-driven strategy inspired by image stacking in image denoising. Termed ``semantic stacking,'' our method estimates a denoised semantic representation that complements the conventional segmentation loss during training. This method does not depend on domain-specific assumptions, making it broadly applicable across diverse image modalities, model architectures, and augmentation techniques. Through extensive experiments, we validate the superiority of our approach in improving segmentation performance under diverse conditions. Code is available at https://github.com/ymp5078/Semantic-Stacking.

Figures

Figures reproduced from arXiv: 2412.13156 by the authors.

Figure 1
Figure 1. An illustration of the proposed semantic stacking [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Illustration of the proposed S2S2 framework. A [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
Figure 3
Figure 3. Visualization of the improvement achieved by ap [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (5 more)
Figure 4
Figure 4. Figure 4: Visualization of the improvement achieved by ap [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]
Figure 5
Figure 5. Figure 5: Ablation study results using FCBFormer with the [PITH_FULL_IMAGE:figures/full_fig_p007_5.png]
Figure 6
Figure 6. Figure 6: Unique color assignments for classes in medical image segmentation datasets. [PITH_FULL_IMAGE:figures/full_fig_p011_6.png]
Figure 7
Figure 7. Figure 7: Pseudocode for S2S2 training. plicity in variable adjustment. When incorporating FCB￾Former (Sanderson and Matuszewski 2022), which features dual encoders, we extend the application of the semantic similarity loss across both encoders and the decoder for comprehensive …
Figure 8
Figure 8. Figure 8: Visualization of synthetic images generated. [PITH_FULL_IMAGE:figures/full_fig_p013_8.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. IntelliCardiac: An Intelligent Platform for Cardiac Image Segmentation and Classification

    eess.IV 2025-05 conditional novelty 4.0 of 10

    IntelliCardiac is a web platform that segments cardiac MRI structures with a 3D U-Net and classifies five cardiac conditions with a random forest plus SVM pipeline, reporting 92.6% Dice and 98% accuracy on ACDC.

Reference graph

Works this paper leans on

72 extracted references · 59 canonical work pages · cited by 1 Pith paper

  1. [1]

    , " * write output.state after.block = add.period write newline

    ENTRY address archivePrefix author booktitle chapter edition editor eid eprint howpublished institution isbn journal key month note number organization pages publisher school series title type volume year label extra.label sort.label short.list INTEGERS output.state before.all mid.sentence after.sentence after.block FUNCTION init.state.consts #0 'before.a...

  2. [2]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 global.max substring 't := if while FUNCTION word.in bbl.in capitalize " " * FUNCT...

  3. [3]

    S.; Karthikesalingam, A.; King, D.; Ashrafian, H.; and Darzi, A

    Aggarwal, R.; Sounderajah, V.; Martin, G.; Ting, D. S.; Karthikesalingam, A.; King, D.; Ashrafian, H.; and Darzi, A. 2021. Diagnostic accuracy of deep learning in medical imaging: a systematic review and meta-analysis. NPJ Digital Medicine, 4(1): 65

    work page 2021

  4. [4]

    High-resolution medical image synthesis using progressively grown generative adversarial networks

    Beers, A.; Brown, J.; Chang, K.; Campbell, J. P.; Ostmo, S.; Chiang, M. F.; and Kalpathy-Cramer, J. 2018. High-resolution medical image synthesis using progressively grown generative adversarial networks. arXiv preprint arXiv:1805.03144

    work page Pith review arXiv 2018

  5. [5]

    J.; Fern \'a ndez-Esparrach, G.; Gil, D.; Rodr \' guez, C.; and Vilari \ n o, F

    Bernal, J.; S \'a nchez, F. J.; Fern \'a ndez-Esparrach, G.; Gil, D.; Rodr \' guez, C.; and Vilari \ n o, F. 2015. WM-DOVA maps for accurate polyp highlighting in colonoscopy: Validation vs. saliency maps from physicians. Computerized Medical Imaging and Graphics, 43: 99--111

    work page 2015

  6. [6]

    Bernard, O.; Lalande, A.; Zotti, C.; Cervenansky, F.; Yang, X.; Heng, P.-A.; Cetin, I.; Lekadir, K.; Camara, O.; Ballester, M. A. G.; et al. 2018. Deep learning techniques for automatic MRI cardiac multi-structures segmentation and diagnosis: is the problem solved? IEEE Transactions on Medical Imaging, 37(11): 2514--2525

    work page 2018

  7. [7]

    Cai, Y.; Fan, L.; and Fang, Y. 2023. SBSS : Stacking-based semantic segmentation framework for very high-resolution remote sensing image. IEEE Transactions on Geoscience and Remote Sensing, 61: 1--14

    work page 2023

  8. [8]

    M.; D'Innocente, A.; Bucci, S.; Caputo, B.; and Tommasi, T

    Carlucci, F. M.; D'Innocente, A.; Bucci, S.; Caputo, B.; and Tommasi, T. 2019. Domain generalization by solving jigsaw puzzles. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2229--2238

    work page 2019

Show all 72 references
  1. [9]

    E.; Alsaadi, F

    Chai, L.; Wang, Z.; Chen, J.; Zhang, G.; Alsaadi, F. E.; Alsaadi, F. E.; and Liu, Q. 2022. Synthetic augmentation for semantic segmentation of class imbalanced biomedical images: A data pair generative adversarial network approach. Computers in Biology and Medicine, 150: 105985

    2022

  2. [10]

    L.; and Zhou, Y

    Chen, J.; Lu, Y.; Yu, Q.; Luo, X.; Adeli, E.; Wang, Y.; Lu, L.; Yuille, A. L.; and Zhou, Y. 2021. TransUNet : Transformers make strong encoders for medical image segmentation. arXiv preprint arXiv:2102.04306

    2021 arXiv

  3. [11]

    D.; Zoph, B.; Mane, D.; Vasudevan, V.; and Le, Q

    Cubuk, E. D.; Zoph, B.; Mane, D.; Vasudevan, V.; and Le, Q. V. 2019. AutoAugment : Learning augmentation policies from data. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

    2019

  4. [12]

    D.; Zoph, B.; Shlens, J.; and Le, Q

    Cubuk, E. D.; Zoph, B.; Shlens, J.; and Le, Q. V. 2020. RandAugment : Practical automated data augmentation with a reduced search space. In Proceedings of the IEEE/CVF conference on Computer Vision and Pattern Recognition Workshops, 702--703

    2020

  5. [13]

    L.; Chintala, S.; Fergus, R.; et al

    Denton, E. L.; Chintala, S.; Fergus, R.; et al. 2015. Deep generative image models using a laplacian pyramid of adversarial networks. Advances in Neural Information Processing Systems, 28

    2015

  6. [14]

    F.; and Lassila, T

    Deo, Y.; Dou, H.; Ravikumar, N.; Frangi, A. F.; and Lassila, T. 2023. Shape-guided conditional latent diffusion models for synthesising brain vasculature. In International Conference on Medical Image Computing and Computer-Assisted Intervention, 164--173. Springer

    2023

  7. [15]

    DeVries, T.; and Taylor, G. W. 2017. Improved regularization of convolutional neural networks with cutout. arXiv preprint arXiv:1708.04552

    2017 arXiv

  8. [16]

    Dosovitskiy, A.; Beyer, L.; Kolesnikov, A.; Weissenborn, D.; Zhai, X.; Unterthiner, T.; Dehghani, M.; Minderer, M.; Heigold, G.; Gelly, S.; Uszkoreit, J.; and Houlsby, N. 2021. An Image is Worth 16 16 Words: Transformers for Image Recognition at Scale. In International Confere...

    2021

  9. [17]

    Esteva, A.; Robicquet, A.; Ramsundar, B.; Kuleshov, V.; DePristo, M.; Chou, K.; Cui, C.; Corrado, G.; Thrun, S.; and Dean, J. 2019. A guide to deep learning in healthcare. Nature Medicine, 25(1): 24--29

    2019

  10. [18]

    Fan, D.-P.; Ji, G.-P.; Zhou, T.; Chen, G.; Fu, H.; Shen, J.; and Shao, L. 2020. PraNet : Parallel reverse attention network for polyp segmentation. In International Conference on Medical Image Computing and Computer-Assisted Intervention, 263--273. Springer

    2020

  11. [19]

    Fernandez, V.; Pinaya, W. H. L.; Borges, P.; Tudosiu, P.-D.; Graham, M. S.; Vercauteren, T.; and Cardoso, M. J. 2022. Can segmentation models be trained with fully synthetically generated data? In International Workshop on Simulation and Synthesis in Medical Imaging, 79--90. Springer

    2022

  12. [20]

    Goodfellow, I.; Pouget-Abadie, J.; Mirza, M.; Xu, B.; Warde-Farley, D.; Ozair, S.; Courville, A.; and Bengio, Y. 2020. Generative adversarial networks. Communications of the ACM, 63(11): 139--144

    2020

  13. [21]

    Guo, X.; Liu, J.; and Yuan, Y. 2024. Infproto-Powered Adaptive Classifier and Agnostic Feature Learning for Single Domain Generalization in Medical Images. International Journal of Computer Vision, 1--24

    2024

  14. [22]

    He, K.; Zhang, X.; Ren, S.; and Sun, J. 2016. Deep residual learning for image recognition. In Proceedings of the IEEE conference on Computer Vision and Pattern Recognition, 770--778

    2016

  15. [23]

    D.; Zoph, B.; Gilmer, J.; and Lakshminarayanan, B

    Hendrycks, D.; Mu, N.; Cubuk, E. D.; Zoph, B.; Gilmer, J.; and Lakshminarayanan, B. 2020. AugMix : A Simple Data Processing Method to Improve Robustness and Uncertainty. In International Conference on Learning Representations

    2020

  16. [24]

    Ho, J.; Jain, A.; and Abbeel, P. 2020. Denoising diffusion probabilistic models. Advances in Neural Information Processing Systems, 33: 6840--6851

    2020

  17. [25]

    Hu, S.; Liao, Z.; and Xia, Y. 2023. Devil is in Channels: Contrastive Single Domain Generalization for Medical Image Segmentation. arXiv preprint arXiv:2306.05254

    2023 arXiv

  18. [26]

    P.; and Huang, D

    Huang, Z.; Wang, H.; Xing, E. P.; and Huang, D. 2020. Self-challenging improves cross-domain generalization. In European Conference on Computer Vision, 124--140. Springer

    2020

  19. [27]

    F.; Kohl, S

    Isensee, F.; Jaeger, P. F.; Kohl, S. A.; Petersen, J.; and Maier-Hein, K. H. 2021. nnU-Net : a self-configuring method for deep learning-based biomedical image segmentation. Nature Methods, 18(2): 203--211

    2021

  20. [28]

    H.; Riegler, M

    Jha, D.; Smedsrud, P. H.; Riegler, M. A.; Halvorsen, P.; Lange, T. d.; Johansen, D.; and Johansen, H. D. 2020. Kvasir-seg: A segmented polyp dataset. In International Conference on Multimedia Modeling, 451--462. Springer

    2020

  21. [29]

    E.; Gezer, N

    Kavur, A. E.; Gezer, N. S.; Bar s , M.; Aslan, S.; Conze, P.-H.; Groza, V.; Pham, D. D.; Chatterjee, S.; Ernst, P.; \"O zkan, S.; et al. 2021. CHAOS challenge-combined (CT-MR) healthy abdominal organ segmentation. Medical Image Analysis, 69: 101950

    2021

  22. [30]

    K.; Heidari, M.; Azad, R.; Fayyaz, M.; Hacihaliloglu, I.; and Merhof, D

    Kazerouni, A.; Aghdam, E. K.; Heidari, M.; Azad, R.; Fayyaz, M.; Hacihaliloglu, I.; and Merhof, D. 2023. Diffusion models in medical imaging: A comprehensive survey. Medical Image Analysis, 88: 102846

    2023

  23. [31]

    Khader, F.; M \"u ller-Franzes, G.; Tayebi Arasteh, S.; Han, T.; Haarburger, C.; Schulze-Hagen, M.; Schad, P.; Engelhardt, S.; Bae ler, B.; Foersch, S.; et al. 2023. Denoising diffusion probabilistic models for 3D medical image generation. Scientific Reports, 13(1): 7303

    2023

  24. [32]

    C.; Lo, W.-Y.; et al

    Kirillov, A.; Mintun, E.; Ravi, N.; Mao, H.; Rolland, C.; Gustafson, L.; Xiao, T.; Whitehead, S.; Berg, A. C.; Lo, W.-Y.; et al. 2023. Segment anything. In Proceedings of the IEEE/CVF International Conference on Computer Vision, 4015--4026

    2023

  25. [33]

    A.; et al

    Krishna, R.; Zhu, Y.; Groth, O.; Johnson, J.; Hata, K.; Kravitz, J.; Chen, S.; Kalantidis, Y.; Li, L.-J.; Shamma, D. A.; et al. 2017. Visual Genome : Connecting language and vision using crowdsourced dense image annotations. International Journal of Computer Vision, 123: 32--73

    2017

  26. [34]

    Liao, S.; Peng, T.; Chen, H.; Lin, T.; Zhu, W.; Shi, F.; Chen, X.; and Xiang, D. 2024. Dual-Spatial Domain Generalization for Fundus Lesion Segmentation in Unseen Manufacturer's OCT Images. IEEE Transactions on Biomedical Engineering

    2024

  27. [35]

    Lin, T.-Y.; Maire, M.; Belongie, S.; Hays, J.; Perona, P.; Ramanan, D.; Doll \'a r, P.; and Zitnick, C. L. 2014. Microsoft COCO : Common objects in context. In European Conference on Computer Vision, 740--755. Springer

    2014

  28. [36]

    E.; Setio, A

    Litjens, G.; Kooi, T.; Bejnordi, B. E.; Setio, A. A. A.; Ciompi, F.; Ghafoorian, M.; Van Der Laak, J. A.; Van Ginneken, B.; and S \'a nchez, C. I. 2017. A survey on deep learning in medical image analysis. Medical Image Analysis, 42: 60--88

    2017

  29. [37]

    Liu, Z.; Mao, H.; Wu, C.-Y.; Feichtenhofer, C.; Darrell, T.; and Xie, S. 2022. A ConvNet for the 2020s. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 11976--11986

    2022

  30. [38]

    Lyu, J.; Zhang, Y.; Huang, Y.; Lin, L.; Cheng, P.; and Tang, X. 2022. AADG : automatic augmentation for domain generalization on retinal image segmentation. IEEE Transactions on Medical Imaging, 41(12): 3699--3711

    2022

  31. [39]

    Ma, J.; He, Y.; Li, F.; Han, L.; You, C.; and Wang, B. 2024. Segment anything in medical images. Nature Communications, 15(1): 654

    2024

  32. [40]

    Milletari, F.; Navab, N.; and Ahmadi, S.-A. 2016. V-Net : Fully convolutional neural networks for volumetric medical image segmentation. In International Conference on 3D Vision, 565--571. Ieee

    2016

  33. [41]

    Nguyen, D. M. H.; Pham, T. N.; Diep, N. T.; Phan, N. Q.; Pham, Q.; Tong, V.; Nguyen, B. T.; Le, N. H.; Ho, N.; Xie, P.; et al. 2023. On the Out of Distribution Robustness of Foundation Models in Medical Image Segmentation. arXiv preprint arXiv:2311.11096

    2023 arXiv

  34. [42]

    Ouyang, C.; Chen, C.; Li, S.; Li, Z.; Qin, C.; Bai, W.; and Rueckert, D. 2022. Causality-inspired single-source domain generalization for medical image segmentation. IEEE Transactions on Medical Imaging, 42(4): 1095--1106

    2022

  35. [43]

    Ozbulak, U.; Van Messem, A.; and De Neve, W. 2019. Impact of adversarial examples on deep learning models for biomedical image segmentation. In Medical Image Computing and Computer-Assisted Intervention, 300--308. Springer

    2019

  36. [44]

    D.; Goldstein, J

    Pan, Y.; Zhang, S.; Gernand, A. D.; Goldstein, J. A.; and Wang, J. Z. 2023. AI-SAM : Automatic and Interactive Segment Anything Model. arXiv preprint arXiv:2312.03119

    2023 arXiv

  37. [45]

    H.; Zhao, Q.; and Pohl, K

    Peng, W.; Adeli, E.; Bosschieter, T.; Park, S. H.; Zhao, Q.; and Pohl, K. M. 2023. Generating realistic brain mris via a conditional diffusion probabilistic model. In International Conference on Medical Image Computing and Computer-Assisted Intervention, 14--24. Springer

    2023

  38. [46]

    Perez, F.; Vasconcelos, C.; Avila, S.; and Valle, E. 2018. Data augmentation for skin lesion analysis. In OR 2.0 Context-Aware Operating Theaters, Computer Assisted Robotic Endoscopy, Clinical Image-Based Procedures, and Skin Image Analysis: First International Workshop, OR 2....

    2018

  39. [47]

    H.; Tudosiu, P.-D.; Dafflon, J.; Da Costa, P

    Pinaya, W. H.; Tudosiu, P.-D.; Dafflon, J.; Da Costa, P. F.; Fernandez, V.; Nachev, P.; Ourselin, S.; and Cardoso, M. J. 2022. Brain imaging generation with latent diffusion models. In MICCAI Workshop on Deep Generative Models, 117--126. Springer

    2022

  40. [48]

    Qin, T.; Wang, Z.; He, K.; Shi, Y.; Gao, Y.; and Shen, D. 2020. Automatic data augmentation via deep reinforcement learning for effective kidney tumor segmentation. In IEEE International Conference on Acoustics, Speech and Signal Processing, 1419--1423. IEEE

    2020

  41. [49]

    I.; Etmann, C.; McCague, C.; Beer, L.; et al

    Roberts, M.; Driggs, D.; Thorpe, M.; Gilbey, J.; Yeung, M.; Ursprung, S.; Aviles-Rivero, A. I.; Etmann, C.; McCague, C.; Beer, L.; et al. 2021. Common pitfalls and recommendations for using machine learning to detect and prognosticate for COVID-19 using chest radiographs and C...

    2021

  42. [50]

    Rombach, R.; Blattmann, A.; Lorenz, D.; Esser, P.; and Ommer, B. 2022. High-resolution image synthesis with latent diffusion models. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 10684--10695

    2022

  43. [51]

    Ronneberger, O.; Fischer, P.; and Brox, T. 2015. U-Net : Convolutional networks for biomedical image segmentation. In Medical Image Computing and Computer-Assisted Intervention, 234--241. Springer

    2015

  44. [52]

    Russakovsky, O.; Deng, J.; Su, H.; Krause, J.; Satheesh, S.; Ma, S.; Huang, Z.; Karpathy, A.; Khosla, A.; Bernstein, M.; et al. 2015. ImageNet large scale visual recognition challenge. International Journal of Computer Vision, 115: 211--252

    2015

  45. [53]

    Sanderson, E.; and Matuszewski, B. J. 2022. FCN-transformer feature fusion for polyp segmentation. In Annual Conference on Medical Image Understanding and Analysis, 892--907. Springer

    2022

  46. [54]

    Schlemper, J.; Oktay, O.; Schaap, M.; Heinrich, M.; Kainz, B.; Glocker, B.; and Rueckert, D. 2019. Attention gated networks: Learning to leverage salient regions in medical images. Medical Image Analysis, 53: 197--207

    2019

  47. [55]

    B.; Chen, I

    Seyyed-Kalantari, L.; Zhang, H.; McDermott, M. B.; Chen, I. Y.; and Ghassemi, M. 2021. Underdiagnosis bias of artificial intelligence algorithms applied to chest radiographs in under-served patient populations. Nature Medicine, 27(12): 2176--2182

    2021

  48. [56]

    D.; Johansen, D.; Riegler, M

    Srivastava, A.; Jha, D.; Chanda, S.; Pal, U.; Johansen, H. D.; Johansen, D.; Riegler, M. A.; Ali, S.; and Halvorsen, P. 2021. MSRF-Net : a multi-scale residual fusion network for biomedical image segmentation. IEEE Journal of Biomedical and Health Informatics, 26(5): 2252--2263

    2021

  49. [57]

    Su, Z.; Yao, K.; Yang, X.; Huang, K.; Wang, Q.; and Sun, J. 2023. Rethinking data augmentation for single-source domain generalization in medical image segmentation. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 37(2), 2366--2374

    2023

  50. [58]

    Tang, F.; Ding, J.; Wang, L.; Xian, M.; and Ning, C. 2023. Multi-Level Global Context Cross Consistency Model for Semi-Supervised Ultrasound Image Segmentation with Diffusion Model. arXiv preprint arXiv:2305.09447

    2023 arXiv

  51. [59]

    Wang, J.; Lan, C.; Liu, C.; Ouyang, Y.; Qin, T.; Lu, W.; Chen, Y.; Zeng, W.; and Yu, P. 2022. Generalizing to unseen domains: A survey on domain generalization. IEEE Transactions on Knowledge and Data Engineering

    2022

  52. [60]

    Wightman, R.; Touvron, H.; and J \'e gou, H. 2021. ResNet strikes back: An improved training procedure in timm. arXiv preprint arXiv:2110.00476

    2021 arXiv

  53. [61]

    S.; and Xie, S

    Woo, S.; Debnath, S.; Hu, R.; Chen, X.; Liu, Z.; Kweon, I. S.; and Xie, S. 2023. ConvNeXt V2 : Co-designing and scaling convnets with masked autoencoders. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 16133--16142

    2023

  54. [62]

    Xu, J.; Li, M.; and Zhu, Z. 2020. Automatic data augmentation for 3D medical image segmentation. In Medical Image Computing and Computer-Assisted Intervention, 378--387. Springer

    2020

  55. [63]

    Xu, Z.; Liu, D.; Yang, J.; Raffel, C.; and Niethammer, M. 2021. Robust and Generalizable Visual Representation Learning via Random Convolutions. In International Conference on Learning Representations

    2021

  56. [64]

    Yang, D.; Roth, H.; Xu, Z.; Milletari, F.; Zhang, L.; and Xu, D. 2019. Searching learning strategy with reinforcement learning for 3D medical image segmentation. In Medical Image Computing and Computer-Assisted Intervention, 3--11. Springer

    2019

  57. [65]

    X.; and Xue, Y

    Ye, J.; Ni, H.; Jin, P.; Huang, S. X.; and Xue, Y. 2023. Synthetic augmentation with large-scale unconditional pre-training. In International Conference on Medical Image Computing and Computer-Assisted Intervention, 754--764. Springer

    2023

  58. [66]

    Yi, X.; Walia, E.; and Babyn, P. 2019. Generative adversarial network in medical imaging: A review. Medical Image Analysis, 58: 101552

    2019

  59. [67]

    J.; Chun, S.; Choe, J.; and Yoo, Y

    Yun, S.; Han, D.; Oh, S. J.; Chun, S.; Choe, J.; and Yoo, Y. 2019. CutMix : Regularization strategy to train strong classifiers with localizable features. In Proceedings of the IEEE/CVF International Conference on Computer Vision, 6023--6032

    2019

  60. [68]

    Zhang, L.; Rao, A.; and Agrawala, M. 2023. Adding conditional control to text-to-image diffusion models. In Proceedings of the IEEE/CVF International Conference on Computer Vision, 3836--3847

    2023

  61. [69]

    Zhou, K.; Liu, Z.; Qiao, Y.; Xiang, T.; and Loy, C. C. 2022 a . Domain generalization: A survey. IEEE Transactions on Pattern Analysis and Machine Intelligence

    2022

  62. [70]

    Zhou, K.; Yang, Y.; Qiao, Y.; and Xiang, T. 2021. Domain Generalization with MixStyle . In International Conference on Learning Representations

    2021

  63. [71]

    Zhou, Z.; Qi, L.; Yang, X.; Ni, D.; and Shi, Y. 2022 b . Generalizable cross-modality medical image segmentation via style augmentation and dual normalization. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 20856--20865

    2022

  64. [72]

    M.; Lekadir, K.; Vesal, S.; RaviKumar, N.; et al

    Zhuang, X.; Xu, J.; Luo, X.; Chen, C.; Ouyang, C.; Rueckert, D.; Campello, V. M.; Lekadir, K.; Vesal, S.; RaviKumar, N.; et al. 2022. Cardiac segmentation on late gadolinium enhancement MRI : a benchmark study from multi-sequence cardiac MR segmentation challenge. Medical Imag...

    2022

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.