Pith. sign in

REVIEW 5 major objections 5 minor 59 references

Efficient Prototype Consistency Learning in Medical Image Segmentation via Joint Uncertainty and Data Augmentation

T0 review · 5 major / 5 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read This paper claims that enriching class prototypes with CutMix-augmented data and JUQ-weighted pseudo-labels yields state-of-the-art semi-supervised medical image segmentation.

desk verdict A reprint of the authors' own BIBM 2024 paper with an undefined label-leaking mask in the unlabeled prototype equations; the memory trick is the only real novelty. read the letter →

arxiv 2505.16283 v1 pith:OBWH6DEQ submitted 2025-05-22 cs.CV

Lijian Li , Yuanpeng He , Chi-Man Pun This is my paper
classification cs.CV
keywords semi-supervisedlearningmedicalimagesegmentationprototypeuncertaintyquantificationdataaugmentationCutMixMeanTeachermemory-efficient3D
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

This paper tries to establish that semi-supervised medical image segmentation can be improved by enlarging the pool of samples used to build class prototypes, adding CutMix augmentation and a joint uncertainty measure to filter pseudo-labels. The method, EPCL-JUDA, builds on the Mean-Teacher framework and fuses prototypes from labeled, original unlabeled, and augmented unlabeled data into global prototypes. The authors report Dice 92.24, Jaccard 85.57, 95HD 4.63, and ASD 1.47 on the Left Atrium dataset at 20% labeled ratio, with leading numbers on Pancreas-NIH and Type B Aortic Dissection. They also claim a prototype network cuts GPU memory substantially, from 18.65 GB to 4.21 GB when generating features from the first decoder layer. If true, this would ease both the annotation burden and the hardware demands of 3D medical segmentation.

What carries the argument

The central object is the global class prototype, formed by fusing labeled and unlabeled prototypes. Labeled prototypes are computed by masked average pooling on features from original plus CutMix-augmented labeled data; unlabeled prototypes are pooled separately from original and augmented unlabeled features, masked by pseudo-labels weighted by JUQ. JUQ multiplies normalized distributional uncertainty (exponential of negative variance) with normalized complement entropy to produce per-voxel reliability weights. A prototype network of three 3D convolutional layers compresses feature maps to the number of classes before upsampling, which is what reduces memory. Cosine similarity maps between global prototypes and student features then drive the consistency losses.

What would settle it

A controlled experiment on the same framework comparing JUQ against plain entropy as the uncertainty measure, evaluated on a dataset with added label noise or a domain shift: if entropy matches or beats JUQ there, the claimed benefit of the joint measure would be falsified. More directly, computing the correlation between JUQ weights and voxel-wise prediction correctness on a held-out set would test whether the reliability map ranks voxels correctly.

Watch

Extended reading notes

Core claim

The central claim is that EPCL-JUDA outperforms previous state-of-the-art semi-supervised segmentation methods by combining four ingredients: CutMix-augmented labeled data for prototype generation, joint uncertainty quantification (JUQ) to weight pseudo-labels, fusion of labeled and unlabeled prototypes into global prototypes, and a lightweight prototype network for memory-efficient feature compression. On the LA dataset at 20% labeled ratio, the method achieves Dice 92.24, Jaccard 85.57, 95HD 4.63, and ASD 1.47; on Pancreas-NIH and TBAD it reports similarly leading scores. The ablation studies attribute the largest gain to the data augmentation, with JUQ and the prototype network contributing smaller but consistent improvements.

Load-bearing premise

The joint uncertainty formula (JUQ) is assumed to rank voxel reliability correctly for both original and augmented unlabeled data, but the paper does not validate this ranking against ground truth; if the weights mis-rank voxels, the pseudo-labels and unlabeled prototypes built from them are corrupted.

Editorial extensions

If this is right

  • If the reported gains hold, prototype-based semi-supervised segmentation can benefit directly from data augmentation on the labeled branch, not just on the unlabeled branch.
  • The JUQ weighting scheme could be transferred to other pseudo-label pipelines as a generic per-voxel confidence filter.
  • The prototype network shows that generating prototypes from low-dimensional class-feature maps is a viable memory-saving alternative to using high-dimensional decoder features.
  • The strong results on the aortic dissection dataset suggest the method handles thin, complex anatomical structures such as vessel walls.

Reading between the lines

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

  • The JUQ formula is hand-designed and unvalidated against ground-truth voxel correctness; its advantage over plain entropy may be dataset-dependent and should be tested under label noise or domain shift.
  • The memory savings from the prototype network could enable larger batch sizes or higher-resolution inputs in 3D segmentation, which might further improve accuracy beyond what the paper reports.
  • Since CutMix on labeled data enriches prototype diversity, other augmentation strategies that preserve label alignment, such as Mixup or controlled synthetic deformations, could be plugged into the same fusion framework.
  • Equation 6 applies a soft reliability weighting to pseudo-labels; a hard threshold variant might behave differently under distribution shift and is worth exploring.
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.

Referee Report

5 major / 5 minor

Summary. This paper proposes EPCL-JUDA, a semi-supervised 3D medical image segmentation method built on a Mean-Teacher architecture. Labeled images are augmented with CutMix and concatenated with originals to produce labeled prototypes by masked average pooling. Original and augmented unlabeled data are processed separately, with a joint uncertainty quantification (JUQ) combining normalized prediction variance and entropy to weight pseudo-labels. Labeled and unlabeled prototypes are fused into global prototypes, and prototype-to-feature cosine similarity maps are supervised with labels and weighted pseudo-labels through consistency losses. A small prototype network is introduced to lower GPU memory relative to using decoder features. The paper reports Dice, Jaccard, 95HD, and ASD on Left Atrium, Pancreas-NIH, and Type B Aortic Dissection at 10% and 20% labeled ratios, claiming state-of-the-art results, plus ablations of JUQ, data augmentation, the prototype network, and unlabeled-data combination patterns.

Significance. The claimed contribution is incremental but potentially useful: augmenting the prototype pool with CutMix and weighting pseudo-labels by a product of variance- and entropy-based reliability are reasonable design choices, and the memory reduction from the prototype network is a concrete engineering benefit. The reported Dice improvements over strong baselines (e.g., 92.24 vs. 91.69 for UPCoL on LA at 20%) are consistent across three datasets and four metrics, and the ablations isolate the main components. However, the contributions as written are not verifiable: the unlabeled prototype equations appear to use ground-truth labels, the self-citation in [23] describes the identical method, and the quantitative comparisons lack error bars or significance tests. These issues must be resolved before the empirical superiority claim can be accepted.

major comments (5)
  1. [Section III-D, Eqs. (8)-(9)] In Section III-A, labels y_a are defined only for the labeled subset D_l, while D_u contains only images. Nevertheless, Eqs. (8) and (9) compute the unlabeled prototypes p_c^{u1} and p_c^{u2} using the ground-truth indicator 1[y_a^p = c], the same mask as the labeled prototype in Eq. (7). Taken literally, this leaks label information into the semi-supervised pipeline and would invalidate the reported results; if the intended mask is the reliable pseudo-label \hat{pl}^p_{u,a} from Eq. (6), the paper never specifies how a probability vector is converted to the binary mask 1[\cdot] (argmax, threshold, or soft weighting), and the denominator \sum_p 1[y_a^p = c] can be zero for classes absent from a sample. The unlabeled prototype construction must be rewritten with an explicit pseudo-label mask, a defined threshold/soft-weighting rule, and a non-zero denominator handling, so that the SOTA claims in Tables I-III can be verified without label leakage.
  2. [Section II-B and References] Section II-B and reference [23] present EPCL-JUDA [23] as prior work ("Building upon these concepts, EPCL-JUDA [23] combines..."), yet [23] has the identical title, the same authors, and reports the same core method. At the same time, the paper's contributions in Section I introduce EPCL-JUDA as a new proposal. The manuscript needs a clear statement of what this version adds over the BIBM 2024 paper (e.g., extended experiments, new ablations, memory analysis), or the novelty claim must be revised; otherwise the contribution is self-referential and cannot be evaluated.
  3. [Section III-C, Eq. (6)] Section III-C, Eq. (6) is not a valid reliability weighting as typeset. It reads \hat{pl}^p_{u,a} = (1/(HWD))(1 - JUQ \sum_p JUQ) \otimes pl^p_{u,a}; the term "1 - JUQ \cdot \sum_p JUQ" is dimensionally mixed and can be negative or unbounded, and the factor 1/(HWD) is unexplained. This pseudo-label formula is the input to the unlabeled prototypes (Eqs. (8)-(9)) and the consistency losses (Eqs. (12)-(13)), so it must be corrected to the intended normalization (e.g., (1 - JUQ/\sum_p JUQ)) and its range stated.
  4. [Section IV-B and Tables I-III] Section IV-B states that 5-fold cross-validation is used for LA and Pancreas-NIH and 3-fold for TBAD, yet Tables I-III report only single point estimates for every metric. Without means, standard deviations, or significance tests, the claimed superiority over previous state of the art (e.g., Table I: EPCL-JUQ 92.24 Dice vs. UPCoL 91.69) may be within fold-to-fold variability. Please report mean \pm std over folds and, where possible, paired significance tests.
  5. [Section III-D, Eq. (10)] Section III-D, Eq. (10) defines p_c^u = \lambda_1 p_c^l + \lambda_2 p_c^m, but the surrounding text says that original and augmented unlabeled prototypes are fused to form p_c^u, and p_c^m is never defined. Additionally, \lambda_1 and \lambda_2 are described as fusion coefficients for the two unlabeled prototypes, which does not match the equation. This equation must be corrected to show how p_c^{u1} and p_c^{u2} are combined before the global fusion in Eq. (11).
minor comments (5)
  1. [Throughout] The acronym is inconsistent throughout: EPCL-JUDA in the title and abstract, EPCL-JUQ in the result tables and discussion, and EPCL-JUA in Section V and Fig. 3; standardize one name.
  2. [Table IV] Table IV lists the setting as "Labeled 8(20%) Unlabled 0", which contradicts Table I's 16(20%) labeled / 64(80%) unlabeled for LA; correct the scan counts and the "Unlabled" typo.
  3. [Table II] Table II repeats the same values for UA-MT, SASSNet, and DTC under both 10% and 20% labeled ratios (e.g., UA-MT 77.26/63.82/11.90/3.06), which looks like a copy-paste error; the 20% rows should be actual measurements.
  4. [Fig. 3] Fig. 3's caption contains garbled path strings instead of a readable description of the reliability maps.
  5. [Abstract and Implementation Details] The code is only promised ("will be released soon") and no repository link is given; since the equations are not fully reproducible as written, providing the code would materially help verification.

Circularity Check

2 steps flagged · score 6.0 of 10

The paper's central unlabeled-prototype derivation is self-definitional as written (Eqs. 8-9 use ground-truth labels), and the proposed method is simultaneously attributed to the authors' own prior paper [23], making the novelty claim self-referential.

  1. self definitional [Section III-A and Section III-D, Eqs. (8)-(9)]
    "the unlabeled subset only contains M images denoted as Du = {(xa)}2B a=B+1 ... The original unlabeled prototype for class c is formulated as follows: pc u1 = 1 B BX a=1 (1 − JUQ PH×W ×D p=1 JUQ) H × W × D PH×W ×D p=1 f u1,p a 1 [ya p = c] PH×W ×D p=1 1 [yap = c]"

    For a in the unlabeled subset Du, ya is not defined; Du is declared to contain only images xa. Eq. (8) (and identically Eq. (9) for augmented unlabeled data) nevertheless masks the pooling with the ground-truth indicator 1[ya^p=c], the same indicator used for labeled prototypes in Eq. (7). Literally, the 'unlabeled' prototype is constructed from ground-truth labels, so the unsupervised prototype claim reduces by definition to a labeled computation and would leak labels into the semi-supervised pipeline. If pseudo-labels were intended, the paper never specifies how the probability vector becomes the hard mask 1[.], and the denominator can be zero for absent classes. The SOTA claim in Tables I-III is therefore not supported by the written equations.

  2. self citation load bearing [Section II-B (Related Works) and Reference [23]; see also Introduction]
    "Building upon these concepts, EPCL-JUDA [23] combines CutMix augmentation and prototype consistency learning with uncertainty modeling, resulting in effective global prototypes and refined pseudo-labels. [23] Li, L., He, Y., Pun, C.M.: Efficient prototype consistency learning in semi-supervised medical image segmentation via joint uncertainty and data augmentation. In: 2024 IEEE International Conference on Bioinformatics and Biomedicine (BIBM). pp. 2114–2121. IEEE (2024)"

    The abstract and introduction present EPCL-JUDA as newly proposed, while the related-work section attributes the very same method, under the same name and with the same three authors and title, to reference [23]. The novelty/identity of the method is thus grounded in the authors' own prior publication rather than in an independent source. This self-citation is not the basis of the empirical benchmark numbers, which are compared with external baselines, but the 'proposal' claim reduces to a self-reference by construction.

full rationale

The strongest non-circular part is the empirical evaluation: Tables I-III measure Dice/Jaccard/95HD/ASD on public benchmarks against external baselines (UA-MT, BCP, Co-BioNet, UPCoL, etc.), and the ablations compare variants of the method itself. Those comparisons are not circular. However, the method's core derivation of unlabeled prototypes is not self-contained as written. Section III-A defines Du = {(xa)} without labels, yet Eqs. (8)-(9) pool unlabeled features using the ground-truth mask 1[ya^p=c], identical to the labeled-prototype mask in Eq. (7). On a literal reading this leaks labels into the 'unlabeled' branch and makes the unlabeled prototype a labeled computation by construction; on the intended pseudo-label reading the paper omits the threshold/argmax rule, so the equation is undefined. Either way the central SOTA claim is not derivable from the written formulas. Separately, the paper cites its own identical prior publication [23] as the basis of EPCL-JUDA, making the novelty claim self-referential; this is a self-citation load, though it does not by itself invalidate the external benchmarks. I therefore assign 6: partial circularity/self-definition in the central prototype construction, plus a load-bearing self-citation, while the benchmark comparisons themselves retain independent content. The JUQ formula (Eq. 5) is an unvalidated heuristic rather than a circular step; it is better treated as a correctness/reproducibility risk.

Assumptions & free parameters 5 free parameters · 5 assumptions · 1 invented entities

The framework is an engineering combination of known building blocks. The main free parameters are fusion coefficients and a schedule; the ad hoc JUQ reliability formula is the most invention-like component. The prototype network is the only newly introduced architectural entity, with partial supporting measurements.

free parameters (5)
  • λ1 (labeled prototype fusion coefficient) = 1
    Set to 1 in Section IV-B implementation details; hand-chosen without sensitivity analysis.
  • λ2 (unlabeled prototype fusion coefficient) = 1
    Same as λ1, set to 1 in Section IV-B; hand-chosen.
  • λcon (Gaussian warming-up schedule) = 0 to 1 ramp
    Standard Mean-Teacher schedule from ref [37]; used to control labeled vs unlabeled prototype fusion in Eq. 11 and consistency weighting in Eq. 14.
  • Prototype network depth = 3 layers
    Described as 'the optimal number of prototype network is 3' in Section IV-B; chosen by ablation rather than derived.
  • CutMix mixing ratio = unspecified
    No value reported anywhere; required to reproduce the augmentation and the prototype calculations.
assumptions (5)
  • domain assumption Teacher EMA pseudo-labels are accurate enough to supervise consistency training.
    Standard Mean-Teacher assumption; invoked throughout Section III, especially in the JUQ weighting and prototype generation.
  • domain assumption Cosine similarity between features and prototypes approximates class membership likelihood.
    Borrowed from SG-One and UPCoL, references [58] and [26]; used to build similarity maps for the consistency losses in Eqs. 12-13.
  • domain assumption Masked average pooling with (mixed) labels produces valid class prototypes.
    From PANet [39]; used in Eqs. 7-9 for labeled and unlabeled prototypes.
  • ad hoc to paper The JUQ product of normalized variance and inverted entropy measures voxel reliability.
    No derivation or standalone validation; introduced in Section III-C, Eq. 5, and only qualitatively illustrated in Fig. 3.
  • domain assumption Reported performance numbers from prior papers are directly comparable without re-running those baselines.
    Tables I-III compare against numbers taken from other publications; no shared implementation, seeds, or hardware are described.
invented entities (1)
  • Prototype network independent evidence
    purpose: Replaces high-memory decoder-feature upsampling with a small 3-layer convolutional network that produces prototype features at class dimension, reducing VRAM and slightly improving segmentation accuracy.
    Supported by Table V (memory drops from 18.65 GB to 4.21 GB for decoder layer 1) and the w/o PRO ablation in Table IV; however, no code is released and architecture details are minimal.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Efficient Prototype Consistency Learning in Medical Image Segmentation via Joint Uncertainty and Data Augmentation." pith.science (2026). https://pith.science/paper/OBWH6DEQ

@misc{pith2026250516283,
  author       = {Pith},
  title        = {Pith review of: Efficient Prototype Consistency Learning in Medical Image Segmentation via Joint Uncertainty and Data Augmentation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/OBWH6DEQ}},
  note         = {Machine review of arXiv:2505.16283}
}
read the original abstract

Recently, prototype learning has emerged in semi-supervised medical image segmentation and achieved remarkable performance. However, the scarcity of labeled data limits the expressiveness of prototypes in previous methods, potentially hindering the complete representation of prototypes for class embedding. To overcome this issue, we propose an efficient prototype consistency learning via joint uncertainty quantification and data augmentation (EPCL-JUDA) to enhance the semantic expression of prototypes based on the framework of Mean-Teacher. The concatenation of original and augmented labeled data is fed into student network to generate expressive prototypes. Then, a joint uncertainty quantification method is devised to optimize pseudo-labels and generate reliable prototypes for original and augmented unlabeled data separately. High-quality global prototypes for each class are formed by fusing labeled and unlabeled prototypes, which are utilized to generate prototype-to-features to conduct consistency learning. Notably, a prototype network is proposed to reduce high memory requirements brought by the introduction of augmented data. Extensive experiments on Left Atrium, Pancreas-NIH, Type B Aortic Dissection datasets demonstrate EPCL-JUDA's superiority over previous state-of-the-art approaches, confirming the effectiveness of our framework. The code will be released soon.

Figures

Figures reproduced from arXiv: 2505.16283 by the authors.

Figure 1
Figure 1. The overview of the proposed EPCL-JUDA framework. For training process, labeled data is augmented using techniques like CutMix and combined [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. The visualizations of experimental results on Left Atrium dataset. [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. The reliability map generated by Entropy and JUQ [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

59 extracted references · 47 canonical work pages

  1. [23]

    In: 2024 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)

    Li, L., He, Y ., Pun, C.M.: Efficient prototype consistency learning in semi-supervised medical image segmentation via joint uncertainty and data augmentation. In: 2024 IEEE International Conference on Bioinformatics and Biomedicine (BIBM). pp. 2114–2121. IEEE (2024)

    work page 2024

  2. [1]

    In: Ghahramani, Z., Welling, M., Cortes, C., Lawrence, N.D., Wein- berger, K.Q

    Bachman, P., Alsharif, O., Precup, D.: Learning with pseudo-ensembles. In: Ghahramani, Z., Welling, M., Cortes, C., Lawrence, N.D., Wein- berger, K.Q. (eds.) Advances in Neural Information Processing Sys- tems 27: Annual Conference on Neural Information Processing Systems 2014, December 8-13 2014, Montreal, Quebec, Canada. pp. 3365–3373 (2014), https://pr...

    work page 2014

  3. [2]

    In: Descoteaux, M., Maier-Hein, L., Franz, A.M., Jannin, P., Collins, D.L., Duchesne, S

    Bai, W., Oktay, O., Sinclair, M., Suzuki, H., Rajchl, M., Tarroni, G., Glocker, B., King, A.P., Matthews, P.M., Rueckert, D.: Semi- supervised learning for network-based cardiac MR image segmentation. In: Descoteaux, M., Maier-Hein, L., Franz, A.M., Jannin, P., Collins, D.L., Duchesne, S. (eds.) Medical Image Computing and Computer Assisted Intervention -...

    2017

  4. [3]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

    Bai, Y ., Chen, D., Li, Q., Shen, W., Wang, Y .: Bidirectional copy-paste for semi-supervised medical image segmentation. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 11514–11524 (2023)

    work page 2023

  5. [4]

    Advances in neural information processing systems 32 (2019)

    Berthelot, D., Carlini, N., Goodfellow, I., Papernot, N., Oliver, A., Raf- fel, C.A.: Mixmatch: A holistic approach to semi-supervised learning. Advances in neural information processing systems 32 (2019)

    work page 2019

  6. [5]

    In: ICASSP 2025- 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

    Bi, Y ., Che, E., Chen, Y ., He, Y ., Qu, J.: Multi-prototype-based em- bedding refinement for medical image segmentation. In: ICASSP 2025- 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). pp. 1–5. IEEE (2025)

    work page 2025

  7. [6]

    In: Shen, D., Liu, T., Peters, T.M., Staib, L.H., Es- sert, C., Zhou, S., Yap, P., Khan, A.R

    Bortsova, G., Dubost, F., Hogeweg, L., Katramados, I., de Bruijne, M.: Semi-supervised medical image segmentation via learning consistency under transformations. In: Shen, D., Liu, T., Peters, T.M., Staib, L.H., Es- sert, C., Zhou, S., Yap, P., Khan, A.R. (eds.) Medical Image Computing and Computer Assisted Intervention - MICCAI 2019 - 22nd International ...

    work page 2019

  8. [7]

    In: IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2021, virtual, June 19- 25, 2021

    Chen, X., Yuan, Y ., Zeng, G., Wang, J.: Semi-supervised semantic segmentation with cross pseudo supervision. In: IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2021, virtual, June 19- 25, 2021. pp. 2613–2622. Computer Vision Foundation / IEEE (2021). https://doi.org/10.1109/CVPR46437.2021.00264

    arXiv 2021

Show all 59 references
  1. [8]

    In: Proceedings of the European Conference on Computer Vision (ECCV)

    Dvornik, N., Mairal, J., Schmid, C.: Modeling visual context is key to augmenting object detection datasets. In: Proceedings of the European Conference on Computer Vision (ECCV). pp. 364–380 (2018)

    2018

  2. [9]

    In: 31st British Machine Vision Conference 2020, BMVC 2020, Virtual Event, UK, September 7-10, 2020

    French, G., Laine, S., Aila, T., Mackiewicz, M., Finlayson, G.D.: Semi- supervised semantic segmentation needs strong, varied perturbations. In: 31st British Machine Vision Conference 2020, BMVC 2020, Virtual Event, UK, September 7-10, 2020. BMV A Press (2020), https://www. bm...

    2020

  3. [10]

    Iranian Journal of Fuzzy Systems 19(3), 171–186 (2022)

    He, Y ., Deng, Y .: Ordinal fuzzy entropy. Iranian Journal of Fuzzy Systems 19(3), 171–186 (2022)

    2022

  4. [11]

    arXiv preprint arXiv:2404.06181 (2024)

    He, Y .: Epl: Evidential prototype learning for semi-supervised medical image segmentation. arXiv preprint arXiv:2404.06181 (2024)

    2024 arXiv

  5. [12]

    In: 2024 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)

    He, Y ., Bi, Y ., Li, L., Pun, C.M., Jiao, W., Jin, Z.: Mutual evidential deep learning for semi-supervised medical image segmentation. In: 2024 IEEE International Conference on Bioinformatics and Biomedicine (BIBM). pp. 2010–2017. IEEE (2024)

    2024

  6. [13]

    Soft Computing 27(11), 6973– 6981 (2023)

    He, Y ., Deng, Y .: Ordinal belief entropy. Soft Computing 27(11), 6973– 6981 (2023)

    2023

  7. [14]

    Information Sciences 621, 749–765 (2023)

    He, Y ., Deng, Y .: Tdqmf: Two-dimensional quantum mass function. Information Sciences 621, 749–765 (2023)

    2023

  8. [15]

    arXiv preprint arXiv:2404.06177 (2024)

    He, Y ., Li, L.: Uncertainty-aware evidential fusion-based learn- ing for semi-supervised medical image segmentation. arXiv preprint arXiv:2404.06177 (2024)

    2024 arXiv

  9. [16]

    In: ICASSP 2024-2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

    He, Y ., Li, L., Zhan, T., Jiao, W., Pun, C.M.: Generalized uncertainty- based evidential fusion with hybrid multi-head attention for weak- supervised temporal action localization. In: ICASSP 2024-2024 IEEE International Conference on Acoustics, Speech and Signal Processing (I...

    2024

  10. [17]

    Pattern Recognition 166, 111639 (2025)

    He, Y ., Li, L., Zhan, T., Pun, C.M., Jiao, W., Jin, Z.: Co-evidential fusion with information volume for semi-supervised medical image segmentation. Pattern Recognition 166, 111639 (2025)

    2025

  11. [18]

    International Journal of Intelligent Systems 36(5), 1914–1942 (2021)

    He, Y ., Xiao, F.: Conflicting management of evidence combination from the point of improvement of basic probability assignment. International Journal of Intelligent Systems 36(5), 1914–1942 (2021)

    2021

  12. [19]

    Applied Intelligence 52(4), 4473–4487 (2022)

    He, Y ., Xiao, F.: A new base function in basic probability assignment for conflict management. Applied Intelligence 52(4), 4473–4487 (2022)

    2022

  13. [20]

    arXiv preprint arXiv:2501.11388 (2025)

    Huang, C.j., He, Y ., Han, X., Jiao, W., Jin, Z., Wang, L.: Unitrans: A unified vertical federated knowledge transfer framework for enhancing cross-hospital collaboration. arXiv preprint arXiv:2501.11388 (2025)

    2025 arXiv

  14. [21]

    CoRR abs/2001.04647 (2020), https: //arxiv.org/abs/2001.04647

    Kim, J., Jang, J., Park, H.: Structured consistency loss for semi- supervised semantic segmentation. CoRR abs/2001.04647 (2020), https: //arxiv.org/abs/2001.04647

    2020 arXiv

  15. [22]

    In: IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2022, New Orleans, LA, USA, June 18-24, 2022

    Kwon, D., Kwak, S.: Semi-supervised semantic segmentation with error localization network. In: IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2022, New Orleans, LA, USA, June 18-24, 2022. pp. 9947–9957. IEEE (2022)

    2022

  16. [24]

    In: ICASSP 2025-2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) (2025)

    Li, L., He, Y ., Pun, C.M.: An adaptive framework for multi-view clustering leveraging conditional entropy optimization. In: ICASSP 2025-2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) (2025)

    2025

  17. [25]

    In: Martel, A.L., Abolmaesumi, P., Stoyanov, D., Mateus, D., Zuluaga, M.A., Zhou, S.K., Racoceanu, D., Joskowicz, L

    Li, S., Zhang, C., He, X.: Shape-aware semi-supervised 3d semantic segmentation for medical images. In: Martel, A.L., Abolmaesumi, P., Stoyanov, D., Mateus, D., Zuluaga, M.A., Zhou, S.K., Racoceanu, D., Joskowicz, L. (eds.) Medical Image Computing and Computer Assisted Interve...

    2020

  18. [26]

    In: International Conference on Medical Image Computing and Computer-Assisted Intervention

    Lu, W., Lei, J., Qiu, P., Sheng, R., Zhou, J., Lu, X., Yang, Y .: Upcol: Uncertainty-informed prototype consistency learning for semi- supervised medical image segmentation. In: International Conference on Medical Image Computing and Computer-Assisted Intervention. pp. 662–672...

    2023

  19. [27]

    Luo, X., Chen, J., Song, T., Wang, G.: Semi-supervised medical image segmentation through dual-task consistency. In: Thirty-Fifth AAAI Con- ference on Artificial Intelligence, AAAI 2021, Thirty-Third Conference on Innovative Applications of Artificial Intelligence, IAAI 2021, ...

    2021

  20. [28]

    In: Medical Image Computing and Computer Assisted Intervention–MICCAI 2021: 24th International Conference, Strasbourg, France, September 27–October 1, 2021, Proceedings, Part II 24

    Luo, X., Liao, W., Chen, J., Song, T., Chen, Y ., Zhang, S., Chen, N., Wang, G., Zhang, S.: Efficient semi-supervised gross target volume of nasopharyngeal carcinoma segmentation via uncertainty rectified pyra- mid consistency. In: Medical Image Computing and Computer Assisted...

    2021

  21. [29]

    Medical Image Analysis 80, 102517 (2022)

    Luo, X., Wang, G., Liao, W., Chen, J., Song, T., Chen, Y ., Zhang, S., Metaxas, D.N., Zhang, S.: Semi-supervised medical image segmentation via uncertainty rectified pyramid consistency. Medical Image Analysis 80, 102517 (2022)

    2022

  22. [30]

    In: 2016 fourth international conference on 3D vision (3DV)

    Milletari, F., Navab, N., Ahmadi, S.A.: V-net: Fully convolutional neural networks for volumetric medical image segmentation. In: 2016 fourth international conference on 3D vision (3DV). pp. 565–571. Ieee (2016)

    2016

  23. [31]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

    Ouali, Y ., Hudelot, C., Tami, M.: Semi-supervised semantic segmenta- tion with cross-consistency training. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 12674– 12684 (2020)

    2020

  24. [32]

    Peiris, H., Chen, Z., Egan, G., Harandi, M.: Duo-segnet: adversarial dual-views for semi-supervised medical image segmentation. In: Med- ical Image Computing and Computer Assisted Intervention–MICCAI 2021: 24th International Conference, Strasbourg, France, September 27– Octobe...

    2021

  25. [33]

    Nature Machine Intelligence 5(7), 724–738 (2023)

    Peiris, H., Hayat, M., Chen, Z., Egan, G., Harandi, M.: Uncertainty- guided dual-views for semi-supervised volumetric medical image seg- mentation. Nature Machine Intelligence 5(7), 724–738 (2023)

    2023

  26. [34]

    In: Medical Image Computing and Computer-Assisted Intervention–MICCAI 2015: 18th International Conference, Munich, Germany, October 5-9, 2015, Proceedings, Part I

    Roth, H.R., Lu, L., Farag, A., Shin, H.C., Liu, J., Turkbey, E.B., Summers, R.M.: Deeporgan: Multi-level deep convolutional networks for automated pancreas segmentation. In: Medical Image Computing and Computer-Assisted Intervention–MICCAI 2015: 18th International Conference, ...

    2015

  27. [35]

    pp. 556–564. Springer (2015)

    2015

  28. [36]

    In: Lee, D.D., Sugiyama, M., von Luxburg, U., Guyon, I., Garnett, R

    Sajjadi, M., Javanmardi, M., Tasdizen, T.: Regularization with stochas- tic transformations and perturbations for deep semi-supervised learn- ing. In: Lee, D.D., Sugiyama, M., von Luxburg, U., Guyon, I., Garnett, R. (eds.) Advances in Neural Information Process- ing Systems 29...

    2016

  29. [37]

    IEEE transactions on medical imaging 41(3), 608– 620 (2021)

    Shi, Y ., Zhang, J., Ling, T., Lu, J., Zheng, Y ., Yu, Q., Qi, L., Gao, Y .: Inconsistency-aware uncertainty estimation for semi-supervised medical image segmentation. IEEE transactions on medical imaging 41(3), 608– 620 (2021)

    2021

  30. [38]

    Advances in neural information processing systems 30 (2017)

    Tarvainen, A., Valpola, H.: Mean teachers are better role models: Weight-averaged consistency targets improve semi-supervised deep learning results. Advances in neural information processing systems 30 (2017)

    2017

  31. [39]

    In: Proceedings of the AAAI Conference on Artificial Intelligence

    Tu, P., Huang, Y ., Zheng, F., He, Z., Cao, L., Shao, L.: Guidedmix- net: Semi-supervised semantic segmentation by using labeled images as reference. In: Proceedings of the AAAI Conference on Artificial Intelligence. vol. 36, pp. 2379–2387 (2022)

    2022

  32. [40]

    In: proceedings of the IEEE/CVF international conference on computer vision

    Wang, K., Liew, J.H., Zou, Y ., Zhou, D., Feng, J.: Panet: Few-shot image semantic segmentation with prototype alignment. In: proceedings of the IEEE/CVF international conference on computer vision. pp. 9197–9206 (2019)

    2019

  33. [41]

    IEEE Transactions on Medical Imaging (2023)

    Wang, P., Peng, J., Pedersoli, M., Zhou, Y ., Zhang, C., Desrosiers, C.: Cat: Constrained adversarial training for anatomically-plausible semi- supervised segmentation. IEEE Transactions on Medical Imaging (2023)

    2023

  34. [42]

    In: IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2022, New Orleans, LA, USA, June 18-24, 2022

    Wang, Y ., Wang, H., Shen, Y ., Fei, J., Li, W., Jin, G., Wu, L., Zhao, R., Le, X.: Semi-supervised semantic segmen- tation using unreliable pseudo-labels. In: IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2022, New Orleans, LA, USA, June 18-24, 2022. pp...

    2022

  35. [43]

    IEEE Trans

    Wu, L., Fang, L., He, X., He, M., Ma, J., Zhong, Z.: Querying labeled for unlabeled: Cross-image semantic consistency guided semi-supervised semantic segmentation. IEEE Trans. Pattern Anal. Mach. Intell. 45(7), 8827–8844 (2023)

    2023

  36. [44]

    In: International Conference on Medical Image Computing and Computer- Assisted Intervention

    Wu, Y ., Wu, Z., Wu, Q., Ge, Z., Cai, J.: Exploring smoothness and class-separation for semi-supervised medical image segmentation. In: International Conference on Medical Image Computing and Computer- Assisted Intervention. pp. 34–43. Springer (2022)

    2022

  37. [45]

    In: de Brui- jne, M., Cattin, P.C., Cotin, S., Padoy, N., Speidel, S., Zheng, Y ., Essert, C

    Wu, Y ., Xu, M., Ge, Z., Cai, J., Zhang, L.: Semi-supervised left atrium segmentation with mutual consistency training. In: de Brui- jne, M., Cattin, P.C., Cotin, S., Padoy, N., Speidel, S., Zheng, Y ., Essert, C. (eds.) Medical Image Computing and Computer Assisted Interventi...

    2021

  38. [46]

    Medical Image Anal

    Xia, Y ., Yang, D., Yu, Z., Liu, F., Cai, J., Yu, L., Zhu, Z., Xu, D., Yuille, A.L., Roth, H.: Uncertainty-aware multi- view co-training for semi-supervised medical image segmenta- tion and domain adaptation. Medical Image Anal. 65, 101766 (2020). https://doi.org/10.1016/j.med...

    2020

  39. [47]

    In: International Con- ference on Medical Image Computing and Computer-Assisted Interven- tion

    Xiang, J., Qiu, P., Yang, Y .: Fussnet: Fusing two sources of uncertainty for semi-supervised medical image segmentation. In: International Con- ference on Medical Image Computing and Computer-Assisted Interven- tion. pp. 481–491. Springer (2022)

    2022

  40. [48]

    Medical Image Anal

    Xiong, Z., Xia, Q., Hu, Z., Huang, N., Bian, C., Zheng, Y ., Vesal, S., Ravikumar, N., Maier, A.K., Yang, X., Heng, P., Ni, D., Li, C., Tong, Q., Si, W., Puybareau, ´E., Khoudli, Y ., G´eraud, T., Zhao, J.: A global bench- mark of algorithms for segmenting the left atrium from...

    2021

  41. [49]

    In: Meila, M., Zhang, T

    Xu, Y ., Shang, L., Ye, J., Qian, Q., Li, Y ., Sun, B., Li, H., Jin, R.: Dash: Semi-supervised learning with dynamic thresholding. In: Meila, M., Zhang, T. (eds.) Proceedings of the 38th International Conference on Machine Learning, ICML 2021, 18-24 July 2021, Virtual Event. P...

    2021

  42. [50]

    Xu, Z., Wang, Y ., Lu, D., Yu, L., Yan, J., Luo, J., Ma, K., Zheng, Y ., Tong, R.K.: All-around real label supervision: Cyclic pro- totype consistency learning for semi-supervised medical image seg- mentation. IEEE J. Biomed. Health Informatics 26(7), 3174–3184 (2022). https:/...

    2022

  43. [51]

    Frontiers in Physiology 12, 732711 (2021)

    Yao, Z., Xie, W., Zhang, J., Dong, Y ., Qiu, H., Yuan, H., Jia, Q., Wang, T., Shi, Y ., Zhuang, J., et al.: Imagetbad: A 3d computed tomography angiography image dataset for automatic segmentation of type-b aortic dissection. Frontiers in Physiology 12, 732711 (2021)

    2021

  44. [52]

    In: Medical Image Computing and Computer Assisted Intervention– MICCAI 2019: 22nd International Conference, Shenzhen, China, Octo- ber 13–17, 2019, Proceedings, Part II 22

    Yu, L., Wang, S., Li, X., Fu, C.W., Heng, P.A.: Uncertainty-aware self-ensembling model for semi-supervised 3d left atrium segmentation. In: Medical Image Computing and Computer Assisted Intervention– MICCAI 2019: 22nd International Conference, Shenzhen, China, Octo- ber 13–17...

    2019

  45. [53]

    In: Shen, D., Liu, T., Peters, T.M., Staib, L.H., Essert, C., Zhou, S., Yap, P., Khan, A.R

    Yu, L., Wang, S., Li, X., Fu, C., Heng, P.: Uncertainty-aware self- ensembling model for semi-supervised 3d left atrium segmentation. In: Shen, D., Liu, T., Peters, T.M., Staib, L.H., Essert, C., Zhou, S., Yap, P., Khan, A.R. (eds.) Medical Image Computing and Com- puter Assis...

    2019

  46. [54]

    In: Proceedings of the IEEE/CVF international conference on computer vision

    Yun, S., Han, D., Oh, S.J., Chun, S., Choe, J., Yoo, Y .: Cutmix: Reg- ularization strategy to train strong classifiers with localizable features. In: Proceedings of the IEEE/CVF international conference on computer vision. pp. 6023–6032 (2019)

    2019

  47. [55]

    Neuroimage 31(3), 1116–1128 (2006)

    Yushkevich, P.A., Piven, J., Cody Hazlett, H., Gimpel Smith, R., Ho, S., Gee, J.C., Gerig, G.: User-guided 3D active contour segmentation of anatomical structures: Significantly improved efficiency and reliability. Neuroimage 31(3), 1116–1128 (2006)

    2006

  48. [56]

    In: Ranzato, M., Beygelzimer, A., Dauphin, Y .N., Liang, P., Vaughan, J.W

    Zhang, B., Wang, Y ., Hou, W., Wu, H., Wang, J., Okumura, M., Shinozaki, T.: Flexmatch: Boosting semi-supervised learning with cur- riculum pseudo labeling. In: Ranzato, M., Beygelzimer, A., Dauphin, Y .N., Liang, P., Vaughan, J.W. (eds.) Advances in Neural Information Process...

    2021

  49. [57]

    arXiv preprint arXiv:1710.09412 (2017)

    Zhang, H., Cisse, M., Dauphin, Y .N., Lopez-Paz, D.: mixup: Beyond empirical risk minimization. arXiv preprint arXiv:1710.09412 (2017)

    2017 arXiv

  50. [58]

    In: IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2021, virtual, June 19-25, 2021

    Zhang, P., Zhang, B., Zhang, T., Chen, D., Wang, Y ., Wen, F.: Prototyp- ical pseudo label denoising and target structure learning for domain adaptive semantic segmentation. In: IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2021, virtual, June 19-25, 2021. p...

    2021

  51. [59]

    IEEE transactions on cybernetics 50(9), 3855–3865 (2020)

    Zhang, X., Wei, Y ., Yang, Y ., Huang, T.S.: Sg-one: Similarity guidance network for one-shot semantic segmentation. IEEE transactions on cybernetics 50(9), 3855–3865 (2020)

    2020

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.