Pith. sign in

REVIEW 4 major objections 5 minor 49 references

DiM: $f$-Divergence Minimization Guided Sharpness-Aware Optimization for Semi-supervised Medical Image Segmentation

T0 review · 4 major / 5 minor · reviewed 2026-08-12 · deepseek-v4-flash

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

Pith's one-line read DiM adds f-divergence alignment to sharpness-aware optimization and reports higher Dice, Jaccard, and boundary metrics on semi-supervised Fundus and Prostate segmentation than prior methods.

desk verdict The paper's central f-divergence alignment loss, as written, is not the f-divergence defined in Eq. 1 and is not computable without a density-ratio estimator that is never described. read the letter →

arxiv 2411.12350 v1 pith:IAA25VFW submitted 2024-11-19 cs.CV cs.AI

Bingli Wang , Houcheng Su , Nan Yin , Mengzhu Wang , Li Shen This is my paper · ORCID
classification cs.CVcs.AI
keywords f-divergencesharpness-awareminimizationsemi-supervisedlearningmedicalimagesegmentationdomainshiftfeaturealignmententropymixed-domain
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 DiM, a training method for semi-supervised medical image segmentation that combines sharpness-aware minimization (SAM) with an f-divergence alignment term. The central claim is that minimizing the f-divergence between the logit distributions of labeled and unlabeled images keeps the unlabeled feature distribution close to the labeled one, counteracting feature drift when training and test data come from different domains. On Fundus and Prostate benchmarks with very few labeled samples, DiM reports higher Dice and Jaccard scores and lower boundary distance metrics than previous semi-supervised and unsupervised domain adaptation methods, approaching an upper bound trained on all labels. The intended payoff is a model that performs evenly across source and target domains instead of overfitting to the labeled domain.

What carries the argument

The load-bearing object is the f-divergence alignment loss $L_{\mathrm{align}} = D_f(p_{\mathrm{label}}\|p_{\mathrm{unlabel}}) = \mathbb{E}_{x\sim p_{\mathrm{unlabel}}}[f(p_{\mathrm{label}}(x)/p_{\mathrm{unlabel}}(x))]$, evaluated for Jeffrey, Jensen-Shannon, or Pearson divergences via Monte Carlo samples from the unlabeled logits. This term is added to a SAM-based objective that also uses selective entropy minimization, $S(x)E(x;\theta)$ with an entropy threshold, and a perturbation step $\hat{\epsilon}(\theta) = \rho\,\mathrm{sign}(\nabla_\theta E)|\nabla_\theta E|/\|\nabla_\theta E\|_2$ to seek flat minima. The f-divergence term is what carries the cross-domain alignment claim: it penalizes mismatch between the labeled and unlabeled prediction distributions, while the SAM term stabilizes the model under that penalty.

What would settle it

Run DiM and a SAM-only baseline on the same labeled/unlabeled split while logging the Monte Carlo f-divergence estimate on held-out unlabeled data; if the estimate is high-variance across seeds or stops decreasing while Dice is still improving, or if lowering $L_{\mathrm{align}}$ does not improve target-domain Dice relative to the SAM-only baseline, the central claim fails. A cleaner control is to replace $L_{\mathrm{align}}$ with the same loss computed on randomly permuted unlabeled logits: if target-domain performance does not drop, the alignment term is not doing the claimed work.

Watch

Extended reading notes

Core claim

The paper's discovery claim is that f-divergence minimization is the missing ingredient in sharpness-aware semi-supervised segmentation: SAM flattens the loss landscape, and f-divergence aligns the labeled and unlabeled logit distributions, so the model converges to a flat minimum that is also domain-aligned. The total objective is $L_{\mathrm{total}} = L_s + \lambda(L_{\mathrm{in}} + L_{\mathrm{out}} + \lambda L_{\mathrm{sym}}) + L_{\mathrm{align}}$, with $L_{\mathrm{align}} = D_f(p_{\mathrm{label}}\|p_{\mathrm{unlabel}})$ computed by Monte Carlo estimation from minibatches. In the reported experiments, DiM achieves an average Dice of 87.28 on the Fundus optic cup/disc task with 20 labeled samples, compared with 85.85 for MiDSS, and 87.45 on the Prostate task with 40 labeled samples, compared with 85.37 for MiDSS, while also reducing Hausdorff distance and average surface distance. The ablation tables show the best results when SAM and f-divergence are used together, and the t-SNE visualizations show tighter cross-domain clusters than MiDSS.

Load-bearing premise

The load-bearing premise is that the f-divergence between labeled and unlabeled logit distributions can be reliably estimated by Monte Carlo from minibatches of four labeled and four unlabeled samples in high-dimensional logit space, and that minimizing this proxy meaningfully aligns the underlying features.

Editorial extensions

If this is right

  • If DiM's claim is correct, semi-supervised medical image segmentation can be made robust to domain shift without any extra annotations, by adding a single divergence term to an existing SAM-based pipeline.
  • The reported gains on Fundus (Dice 85.85 to 87.28 with 20 labels) and Prostate (85.37 to 87.45 with 40 labels) suggest that f-divergence alignment recovers a large part of the gap toward fully supervised upper bounds.
  • Because the method matches or beats dedicated unsupervised domain adaptation baselines without using target labels, it offers a simpler alternative to UDA for mixed-domain clinical data.
  • The choice of divergence function matters: Jensen-Shannon and Jeffrey variants appear more reliable than Pearson, which gives practitioners a concrete selection rule.

Reading between the lines

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

  • One implicit consequence is that the same alignment loss could be applied to other semi-supervised dense prediction tasks, such as natural-image semantic segmentation or MRI/CT multi-organ segmentation, wherever labeled and unlabeled data come from different acquisition sites.
  • The paper never specifies how the density ratio $p_{\mathrm{label}}/p_{\mathrm{unlabel}}$ is formed from a minibatch of four labeled and four unlabeled logits; a natural extension is to replace the Monte Carlo ratio with a learned density-ratio estimator or a dual f-divergence estimator and test whether small-batch training becomes more stable.
  • If the benefit of $L_{\mathrm{align}}$ comes mostly from regularizing the labeled logits rather than from true distribution alignment, then a simpler feature-consistency or entropy-regularization baseline might reproduce the gains; comparing DiM against such baselines with the divergence term ablated would settle that.
  • The reported convergence curves show stable validation Dice across four fundus domains, which suggests the method may also serve as a diagnostic probe: monitoring the f-divergence estimate during training could indicate when domain shift is being corrected.
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

4 major / 5 minor

Summary. The paper proposes DiM, a method for semi-supervised medical image segmentation under domain shift. DiM combines sharpness-aware minimization (SAM) with selective entropy minimization and an alignment loss Lalign that is claimed to minimize the f-divergence between the logit distributions of labeled and unlabeled data. Experiments on Fundus (optic cup/disc) and Prostate datasets report improved Dice/Jaccard and lower surface-distance metrics relative to MiDSS and several other semi-supervised and domain-adaptation baselines, with ablations over SAM and different f-divergence variants. The core mechanism of the paper is the f-divergence alignment term, but its definition and implementation are not fully specified, which affects the reproducibility and attribution of the empirical gains.

Significance. The problem addressed is important: semi-supervised medical image segmentation with data coming from multiple clinical sites is a realistic setting, and improving robustness to domain shift is valuable. If the f-divergence alignment term were properly defined and reliably estimated, the paper would offer a useful extension to SAM-based semi-supervised segmentation. The paper also provides a broad comparison across four fundus domains and six prostate sites, with ablations and visualizations. The main weakness is that the central f-divergence term is not computed as written, no density-ratio estimator is described, and the reported improvements lack variance estimates. These issues are load-bearing for the paper's central claim, so the contribution is currently conditional on substantial revision.

major comments (4)
  1. [Section 3.1, Eq. (1)-(2)] Equation (2) defines Lalign as E_x~p_unlabel[f(p_label(x)/p_unlabel(x))], silently omitting the boundary term f'(∞)p_label(p_unlabel=0) that is explicitly present in Eq. (1). For the Jensen-Shannon and Pearson divergences in Table 1, f'(∞) is nonzero, so whenever the supports of the labeled and unlabeled logit distributions differ — exactly the regime the paper says it handles — Lalign is not equal to D_f. The authors should either retain the boundary term in Lalign or justify why it can be ignored; as written, the central identity Lalign = D_f is false in the intended setting.
  2. [Section 3.1, Monte Carlo estimation] The paper states that the f-divergence is 'computed via Monte Carlo estimation based on samples from p_unlabel' but never specifies how the density ratio p_label(x)/p_unlabel(x) is estimated. With a batch of 4 labeled and 4 unlabeled samples and high-dimensional logits, the empirical supports of the two distributions are almost surely disjoint; a naive plug-in estimate makes Lalign constant for JS divergence and infinite for forward KL divergence, with zero or undefined gradients. A concrete estimator (e.g., a variational critic, a kernel-based ratio estimate, or a low-dimensional projection with explicit density estimation) is required for the method to be reproducible and for the f-divergence rows in Table 4 to be attributable to f-divergence minimization.
  3. [Section 4.3, Tables 2-3] Tables 2 and 3 report all segmentation metrics as single numbers with no standard deviations, numbers of seeds, or significance tests. The reported gains over MiDSS are modest in places (Fundus average DC 87.28 vs. 85.85; Prostate average DC 87.45 vs. 85.37) and are not consistent across all domains, so the claimed state-of-the-art performance is not statistically supported. Please report mean and standard deviation over at least three independent runs and state the model-selection and seed protocol.
  4. [Section 4.2, upper-bound comparison] The 'Upper bound' rows in Tables 2 and 3 are said to be taken from the MiDSS paper, which used UCP within the FixMatch framework with all training data labeled, rather than being computed with the authors' own pipeline, U-Net backbone, SAM optimizer, and data splits. Such an external result cannot serve as a directly comparable fully supervised upper bound for DiM. The authors should either compute their own fully supervised upper bound under identical settings or clearly label the row as an external result and avoid comparing against it as an upper bound for their method.
minor comments (5)
  1. [Section 3.3, Eq. (10)] The term Lsym in Eq. (10) is never defined, and the symbol λ appears twice in the same expression; please define Lsym and clarify whether both occurrences denote the same time-dependent weight λ(t).
  2. [Section 4.4, Table 4] The 'baseline' row in Table 4 reports DC 88.27 on Domain 1, whereas the U-Net result with 20 labeled samples in Table 2 is 59.54/73.89; please clarify what the ablation baseline is and how it relates to the main experimental setting.
  3. [Section 4.2, comparison methods] UA-MT, FixMatch, CPS, CoraNet, SS-Net, BCP, CauSSL, and MiDSS are semi-supervised methods, not 'supervised techniques' as written in the method list; please correct this description.
  4. [Section 3.2] The sentence beginning 'Directly using gradient norms to filter out unreliable test samples is challenging...' appears twice in consecutive paragraphs; please remove the duplicate.
  5. [Throughout] Please proofread for typos and formatting artifacts, including 'utilizef-divergence' at the start of Section 3.1, 'Sencondly' in Section 4.4, and the figure cross-references in Section 4.5, where the text refers to 'Figure 5' for Fundus curves that appear in Figure 4.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the f-divergence loss is an empirically evaluated regularizer, not a prediction fitted to its own target.

full rationale

The paper's claimed derivation is an empirical method rather than a formal derivation with a forced conclusion. The f-divergence alignment loss (Eq. 2) is introduced as an additional regularizer; the final objective Ltotal (Eq. 10) combines it with supervised loss, intermediate consistency losses, and SAM-based entropy minimization. The experimental sections then compare the full method against external SOTA baselines on two benchmarks. No parameter is fitted to the benchmark and then reported as a prediction; no result is defined in terms of the quantity it is supposed to establish. The ablation study attributes gains to SAM and f-divergence, but those gains are measured, not derived from the definition of the loss. The paper cites prior work (SAM, MiDSS, FixMatch, etc.), but all are external baselines or standard building blocks, and none of the load-bearing claims depends on a citation to the present authors' own prior results. The most significant weakness is a mathematical gap: Eq. 1 includes an f'(∞)·plabel(punlabel=0) term that Eq. 2 drops, and the text never specifies how plabel/punlabel is estimated from 4+4 logits; however, an incorrect or underspecified loss is a correctness/reproducibility problem, not circularity, because the loss does not reduce to its own input by construction. No circular pattern of the seven enumerated kinds is present.

Assumptions & free parameters 4 free parameters · 4 assumptions · 0 invented entities

The method's core loss terms rely on several unstated hyperparameters (rho, E0, lambda schedule) and on the assumption that Monte Carlo f-divergence estimation from small batches is a reliable alignment signal. No new physical or mathematical entities are introduced.

free parameters (4)
  • SAM perturbation radius rho = not reported
    Controls the size of weight perturbation in Eq (4)-(5); never specified in the text.
  • Entropy threshold E0 = not reported
    Used in the selective entropy mask S(x) in Eq (3); threshold is never defined or derived.
  • Loss weight lambda = lambda(t)=e^{-5(1-t/t_total)}
    Time-dependent coefficient for unsupervised terms in Eq (10); schedule is fixed but rationale is not given.
  • Choice of f-divergence = JS used for main results
    The paper tests JS, Jeffrey, and Pearson and picks JS for the main experiments, making the choice a free parameter tuned on the validation set.
assumptions (4)
  • standard math f-divergence is a valid distance between distributions and Monte Carlo estimation is unbiased
    The method relies on standard properties of f-divergence, invoked in Section 3.1 without proof.
  • domain assumption Logits of labeled and unlabeled data are comparable distributions over the same discrete set X
    Section 3.1 assumes this comparability; if supports differ, the f-divergence may be undefined or infinite.
  • ad hoc to paper Minimizing f-divergence between labeled and unlabeled logits improves cross-domain generalization
    This is the central empirical premise, stated in Section 3.1 but not derived from first principles.
  • domain assumption The SAM gradient approximation Eq (5) is accurate enough for the entropy objective
    The paper adopts the standard SAM perturbation formula from Foret et al. without re-derivation.

how reviews work

0 comments
Cite this review

Pith. "Pith review of DiM: $f$-Divergence Minimization Guided Sharpness-Aware Optimization for Semi-supervised Medical Image Segmentation." pith.science (2026). https://pith.science/paper/IAA25VFW

@misc{pith2026241112350,
  author       = {Pith},
  title        = {Pith review of: DiM: $f$-Divergence Minimization Guided Sharpness-Aware Optimization for Semi-supervised Medical Image Segmentation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/IAA25VFW}},
  note         = {Machine review of arXiv:2411.12350}
}
abstract

As a technique to alleviate the pressure of data annotation, semi-supervised learning (SSL) has attracted widespread attention. In the specific domain of medical image segmentation, semi-supervised methods (SSMIS) have become a research hotspot due to their ability to reduce the need for large amounts of precisely annotated data. SSMIS focuses on enhancing the model's generalization performance by leveraging a small number of labeled samples and a large number of unlabeled samples. The latest sharpness-aware optimization (SAM) technique, which optimizes the model by reducing the sharpness of the loss function, has shown significant success in SSMIS. However, SAM and its variants may not fully account for the distribution differences between different datasets. To address this issue, we propose a sharpness-aware optimization method based on $f$-divergence minimization (DiM) for semi-supervised medical image segmentation. This method enhances the model's stability by fine-tuning the sensitivity of model parameters and improves the model's adaptability to different datasets through the introduction of $f$-divergence. By reducing $f$-divergence, the DiM method not only improves the performance balance between the source and target datasets but also prevents performance degradation due to overfitting on the source dataset.

Figures

Figures reproduced from arXiv: 2411.12350 by the authors.

Figure 1
Figure 1. A T-sne visualization analysis was performed on the [PITH_FULL_IMAGE:figures/full_fig_p006_1.png] view at source ↗
Figure 2
Figure 2. Visual comparison of segmentation results on Fundus dataset across different models. Red and green represent the Optical Cup [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗
Figure 3
Figure 3. Visual results on Prostate dataset. 0 10 20 30 40 50 Epoch 0.0 0.2 0.4 0.6 0.8 1.0 Loss (a) Domain 1 Loss 0 10 20 30 40 50 Epoch 0.0 0.2 0.4 0.6 0.8 1.0 Loss (b) Domain 2 Loss 0 10 20 30 40 50 Epoch 0.0 0.2 0.4 0.6 0.8 1.0 Loss (c) Domain 3 Loss 0 10 20 30 40 50 Epoch 0.0 0.2 0.4 0.6 0.8 1.0 Loss (d) Domain 4 Loss 0 10 20 30 40 50 Epoch 0.0 0.2 0.4 0.6 0.8 1.0 Cup Dice Disc Dice (e) Domain 1 Dice 0 10 20 30 40 50 Ep… view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Validation Loss and Dice across four domains on Fundus dataset. [PITH_FULL_IMAGE:figures/full_fig_p008_4.png]
Figure 5
Figure 5. Figure 5: Validation Loss and Dice for Domain 1 on the Prostate [PITH_FULL_IMAGE:figures/full_fig_p008_5.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

49 extracted references · 41 canonical work pages

  1. [1]

    Towards understanding sharpness-aware minimization

    Maksym Andriushchenko and Nicolas Flammarion. Towards understanding sharpness-aware minimization. In ICML, pages 639–668. PMLR, 2022. 3

    work page 2022

  2. [2]

    Bidirectional copy-paste for semi-supervised medical image segmentation

    Yunhao Bai, Duowen Chen, Qingli Li, Wei Shen, and Yan Wang. Bidirectional copy-paste for semi-supervised medical image segmentation. In CVPR, pages 11514–11524, 2023. 5

    work page 2023

  3. [3]

    Semi-supervised medical image segmentation via learning consistency un- der transformations

    Gerda Bortsova, Florian Dubost, Laurens Hogeweg, Ioan- nis Katramados, and Marleen De Bruijne. Semi-supervised medical image segmentation via learning consistency un- der transformations. In MICCAI, pages 810–818. Springer,

    work page

  4. [4]

    Unsupervised bidirectional cross-modality adaptation via deeply synergistic image and feature alignment for med- ical image segmentation

    Cheng Chen, Qi Dou, Hao Chen, Jing Qin, and Pheng Ann Heng. Unsupervised bidirectional cross-modality adaptation via deeply synergistic image and feature alignment for med- ical image segmentation. TMI, 39(7):2494–2505, 2020. 5

    work page 2020

  5. [5]

    Semi-supervised semantic segmentation with cross pseudo supervision

    Xiaokang Chen, Yuhui Yuan, Gang Zeng, and Jingdong Wang. Semi-supervised semantic segmentation with cross pseudo supervision. In CVPR, pages 2613–2622, 2021. 5

    work page 2021

  6. [6]

    Image segmentation based on multi-region multi-scale local binary fitting and kullback–leibler divergence

    Dansong Cheng, Feng Tian, Lin Liu, Xiaofang Liu, and Ye Jin. Image segmentation based on multi-region multi-scale local binary fitting and kullback–leibler divergence. Signal, Image and Video Processing, 12:895–903, 2018. 2

    work page 2018

  7. [7]

    Big data in healthcare: management, analysis and future prospects

    Sabyasachi Dash, Sushil Kumar Shakyawar, Mohit Sharma, and Sandeep Kaushik. Big data in healthcare: management, analysis and future prospects. Journal of Big Data, 6(1):1– 25, 2019. 1

    work page 2019

  8. [8]

    Sharpness-aware minimization for efficiently improving generalization

    Pierre Foret, Ariel Kleiner, Hossein Mobahi, and Behnam Neyshabur. Sharpness-aware minimization for efficiently improving generalization. arXiv preprint arXiv:2010.01412,

    arXiv 2010

Show all 49 references
  1. [9]

    Unsupervised domain adaptation by backpropagation

    Yaroslav Ganin and Victor Lempitsky. Unsupervised domain adaptation by backpropagation. In ICML, pages 1180–1189. PMLR, 2015. 2

    2015

  2. [10]

    Bias learning, knowl- edge sharing

    Joumana Ghosn and Yoshua Bengio. Bias learning, knowl- edge sharing. TNN, 14(4):748–765, 2003. 2

    2003

  3. [11]

    Domain adaptation for medical image analysis: a survey

    Hao Guan and Mingxia Liu. Domain adaptation for medical image analysis: a survey. IEEE Transactions on Biomedical Engineering, 69(3):1173–1185, 2021. 1

    2021

  4. [12]

    Deep residual learning for image recognition

    Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. Deep residual learning for image recognition. In CVPR, pages 770–778, 2016. 5

    2016

  5. [13]

    H-denseunet: hybrid densely con- nected unet for liver and tumor segmentation from ct vol- umes

    Xiaomeng Li, Hao Chen, Xiaojuan Qi, Qi Dou, Chi-Wing Fu, and Pheng-Ann Heng. H-denseunet: hybrid densely con- nected unet for liver and tumor segmentation from ct vol- umes. TMI, 37(12):2663–2674, 2018. 1

    2018

  6. [14]

    Feddg: Federated domain generalization on medical image segmentation via episodic learning in continuous fre- quency space

    Quande Liu, Cheng Chen, Jing Qin, Qi Dou, and Pheng-Ann Heng. Feddg: Federated domain generalization on medical image segmentation via episodic learning in continuous fre- quency space. In CVPR, pages 1013–1023, 2021. 1

    2021

  7. [15]

    Deep unsupervised domain adaptation: A review of recent ad- vances and perspectives.APSIPA Transactions on Signal and Information Processing, 11(1), 2022

    Xiaofeng Liu, Chaehwa Yoo, Fangxu Xing, Hyejin Oh, Georges El Fakhri, Je-Won Kang, Jonghye Woo, et al. Deep unsupervised domain adaptation: A review of recent ad- vances and perspectives.APSIPA Transactions on Signal and Information Processing, 11(1), 2022. 2

    2022

  8. [16]

    Conditional adversarial domain adapta- tion

    Mingsheng Long, Zhangjie Cao, Jianmin Wang, and Michael I Jordan. Conditional adversarial domain adapta- tion. NeurIPS, 31, 2018. 2

    2018

  9. [17]

    Unsuper- vised domain adaptation for cardiac segmentation: Towards structure mutual information maximization

    Changjie Lu, Shen Zheng, and Gaurav Gupta. Unsuper- vised domain adaptation for cardiac segmentation: Towards structure mutual information maximization. In CVPR, pages 2588–2597, 2022. 5

    2022

  10. [18]

    Uncertainty-aware pseudo-label and consistency for semi- supervised medical image segmentation

    Liyun Lu, Mengxiao Yin, Liyao Fu, and Feng Yang. Uncertainty-aware pseudo-label and consistency for semi- supervised medical image segmentation. Biomedical Signal Processing and Control, 79:104203, 2023. 2

    2023

  11. [19]

    Semi-supervised medical image segmentation through dual- task consistency

    Xiangde Luo, Jieneng Chen, Tao Song, and Guotai Wang. Semi-supervised medical image segmentation through dual- task consistency. In AAAI, pages 8801–8809, 2021. 2

    2021

  12. [20]

    Semi-supervised medical image segmen- tation via cross teaching between cnn and transformer

    Xiangde Luo, Minhao Hu, Tao Song, Guotai Wang, and Shaoting Zhang. Semi-supervised medical image segmen- tation via cross teaching between cnn and transformer. InIn- ternational conference on medical imaging with deep learn- ing, pages 820–833. PMLR, 2022. 2

    2022

  13. [21]

    Constructing and exploring intermediate domains in mixed domain semi-supervised medical image segmenta- tion

    Qinghe Ma, Jian Zhang, Lei Qi, Qian Yu, Yinghuan Shi, and Yang Gao. Constructing and exploring intermediate domains in mixed domain semi-supervised medical image segmenta- tion. In CVPR, pages 11642–11651, 2024. 1, 5

    2024

  14. [22]

    Constructing and exploring intermediate domains in mixed domain semi-supervised medical image segmenta- tion

    Qinghe Ma, Jian Zhang, Lei Qi, Qian Yu, Yinghuan Shi, and Yang Gao. Constructing and exploring intermediate domains in mixed domain semi-supervised medical image segmenta- tion. In CVPR, pages 11642–11651, 2024. 2

    2024

  15. [23]

    Caussl: Causality-inspired semi-supervised learning for medical image segmentation

    Juzheng Miao, Cheng Chen, Furui Liu, Hao Wei, and Pheng- Ann Heng. Caussl: Causality-inspired semi-supervised learning for medical image segmentation. In ICCV, pages 21426–21437, 2023. 1, 2, 5

    2023

  16. [24]

    U-net: Convolutional networks for biomedical image segmentation

    Olaf Ronneberger, Philipp Fischer, and Thomas Brox. U-net: Convolutional networks for biomedical image segmentation. In MICCAI, pages 234–241. Springer, 2015. 1, 5

    2015

  17. [25]

    Opportunities and challenges in using real-world data for health care.The Jour- nal of Clinical Investigation, 130(2):565–574, 2020

    Vivek A Rudrapatna, Atul J Butte, et al. Opportunities and challenges in using real-world data for health care.The Jour- nal of Clinical Investigation, 130(2):565–574, 2020. 1

    2020

  18. [26]

    Semi-supervised domain adaptation via minimax entropy

    Kuniaki Saito, Donghyun Kim, Stan Sclaroff, Trevor Darrell, and Kate Saenko. Semi-supervised domain adaptation via minimax entropy. In ICCV, pages 8050–8058, 2019. 2

    2019

  19. [27]

    Strong-weak distribution alignment for adaptive ob- ject detection

    Kuniaki Saito, Yoshitaka Ushiku, Tatsuya Harada, and Kate Saenko. Strong-weak distribution alignment for adaptive ob- ject detection. In CVPR, pages 6956–6965, 2019. 2

    2019

  20. [28]

    Learning from synthetic data: Addressing domain shift for semantic segmentation

    Swami Sankaranarayanan, Yogesh Balaji, Arpit Jain, Ser Nam Lim, and Rama Chellappa. Learning from synthetic data: Addressing domain shift for semantic segmentation. In CVPR, pages 3752–3761, 2018. 2

    2018

  21. [29]

    Inconsistency-aware uncertainty estimation for semi-supervised medical image segmentation

    Yinghuan Shi, Jian Zhang, Tong Ling, Jiwen Lu, Yefeng Zheng, Qian Yu, Lei Qi, and Yang Gao. Inconsistency-aware uncertainty estimation for semi-supervised medical image segmentation. TMI, 41(3):608–620, 2021. 5

    2021

  22. [30]

    A dirt-t approach to unsupervised domain adaptation

    Rui Shu, Hung H Bui, Hirokazu Narui, and Stefano Ermon. A dirt-t approach to unsupervised domain adaptation. arXiv preprint arXiv:1802.08735, 2018. 2

    2018 arXiv

  23. [31]

    Fixmatch: Simplifying semi-supervised learning with consistency and confidence

    Kihyuk Sohn, David Berthelot, Nicholas Carlini, Zizhao Zhang, Han Zhang, Colin A Raffel, Ekin Dogus Cubuk, Alexey Kurakin, and Chun-Liang Li. Fixmatch: Simplifying semi-supervised learning with consistency and confidence. NeurIPS, 33:596–608, 2020. 5

    2020

  24. [32]

    R ´enyi divergence and kullback-leibler divergence

    Tim Van Erven and Peter Harremos. R ´enyi divergence and kullback-leibler divergence. IEEE Transactions on Informa- tion Theory, 60(7):3797–3820, 2014. 2

    2014

  25. [33]

    Semi-supervised medical image segmentation via a tripled-uncertainty guided mean teacher model with contrastive learning

    Kaiping Wang, Bo Zhan, Chen Zu, Xi Wu, Jiliu Zhou, Lup- ing Zhou, and Yan Wang. Semi-supervised medical image segmentation via a tripled-uncertainty guided mean teacher model with contrastive learning. Medical Image Analysis , 79:102447, 2022. 1

    2022

  26. [34]

    Deep visual domain adapta- tion: A survey

    Mei Wang and Weihong Deng. Deep visual domain adapta- tion: A survey. Neurocomputing, 312:135–153, 2018. 2

    2018

  27. [35]

    Semi-supervised learning by augmented distribution alignment

    Qin Wang, Wen Li, and Luc Van Gool. Semi-supervised learning by augmented distribution alignment. In ICCV, pages 1466–1475, 2019. 1

    2019

  28. [36]

    U-medsam: Uncertainty-aware medsam for medical image segmentation

    Xin Wang, Xiaoyu Liu, Peng Huang, Pu Huang, Shu Hu, and Hongtu Zhu. U-medsam: Uncertainty-aware medsam for medical image segmentation. arXiv preprint arXiv:2408.08881, 2024. 3

    2024 arXiv

  29. [37]

    Swinmm: masked multi-view with swin trans- formers for 3d medical image segmentation

    Yiqing Wang, Zihan Li, Jieru Mei, Zihao Wei, Li Liu, Chen Wang, Shengtian Sang, Alan L Yuille, Cihang Xie, and Yuyin Zhou. Swinmm: masked multi-view with swin trans- formers for 3d medical image segmentation. In MICCAI, pages 486–496, 2023. 2

    2023

  30. [38]

    A survey of transfer learning

    Karl Weiss, Taghi M Khoshgoftaar, and DingDing Wang. A survey of transfer learning. Journal of Big Data , 3:1–40,

  31. [39]

    Mutual consistency learning for semi-supervised medical image segmentation

    Yicheng Wu, Zongyuan Ge, Donghao Zhang, Minfeng Xu, Lei Zhang, Yong Xia, and Jianfei Cai. Mutual consistency learning for semi-supervised medical image segmentation. Medical Image Analysis, 81:102530, 2022. 2

    2022

  32. [40]

    Exploring smoothness and class-separation for semi-supervised medical image segmentation

    Yicheng Wu, Zhonghua Wu, Qianyi Wu, Zongyuan Ge, and Jianfei Cai. Exploring smoothness and class-separation for semi-supervised medical image segmentation. In MICCAI, pages 34–43. Springer, 2022. 5

    2022

  33. [41]

    Fda: Fourier domain adaptation for semantic segmentation

    Yanchao Yang and Stefano Soatto. Fda: Fourier domain adaptation for semantic segmentation. In CVPR, pages 4085–4095, 2020. 5

    2020

  34. [42]

    Uncertainty-aware self-ensembling model for semi-supervised 3d left atrium segmentation

    Lequan Yu, Shujun Wang, Xiaomeng Li, Chi-Wing Fu, and Pheng-Ann Heng. Uncertainty-aware self-ensembling model for semi-supervised 3d left atrium segmentation. InMICCAI, pages 605–613. Springer, 2019. 5

    2019

  35. [43]

    Collab- orative unsupervised domain adaptation for medical image diagnosis

    Yifan Zhang, Ying Wei, Qingyao Wu, Peilin Zhao, Shuaicheng Niu, Junzhou Huang, and Mingkui Tan. Collab- orative unsupervised domain adaptation for medical image diagnosis. TIP, 29:7834–7844, 2020. 1

    2020

  36. [44]

    Le-uda: Label-efficient unsu- pervised domain adaptation for medical image segmentation

    Ziyuan Zhao, Fangcheng Zhou, Kaixin Xu, Zeng Zeng, Cun- tai Guan, and S Kevin Zhou. Le-uda: Label-efficient unsu- pervised domain adaptation for medical image segmentation. TMI, 42(3):633–646, 2022. 1

    2022

  37. [45]

    Imb- sam: A closer look at sharpness-aware minimization in class-imbalanced recognition

    Yixuan Zhou, Yi Qu, Xing Xu, and Hengtao Shen. Imb- sam: A closer look at sharpness-aware minimization in class-imbalanced recognition. In ICCV, pages 11345–11355,

  38. [46]

    Unet++: Redesigning skip connections to exploit multiscale features in image segmen- tation

    Zongwei Zhou, Md Mahfuzur Rahman Siddiquee, Nima Tajbakhsh, and Jianming Liang. Unet++: Redesigning skip connections to exploit multiscale features in image segmen- tation. TMI, 39(6):1856–1867, 2019. 1

    2019

  39. [47]

    A brief introduction to weakly supervised learning

    Zhi-Hua Zhou. A brief introduction to weakly supervised learning. National Science Review, 5(1):44–53, 2018. 1

    2018

  40. [48]

    Surrogate gap minimization improves sharpness-aware training

    Juntang Zhuang, Boqing Gong, Liangzhe Yuan, Yin Cui, Hartwig Adam, Nicha Dvornek, Sekhar Tatikonda, James Duncan, and Ting Liu. Surrogate gap minimization improves sharpness-aware training. arXiv preprint arXiv:2203.08065,

    arXiv

  41. [49]

    Challenges and methodologies of fully automatic whole heart segmentation: a review

    Xiahai Zhuang et al. Challenges and methodologies of fully automatic whole heart segmentation: a review. Journal of Healthcare Engineering, 4:371–407, 2013. 2

    2013

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.