Pith. sign in

REVIEW 4 major objections 4 minor 54 references

SGTC: Semantic-Guided Triplet Co-training for Sparsely Annotated Semi-Supervised Medical Image Segmentation

T0 review · 4 major / 4 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 SGTC claims that three orthogonally chosen annotated slices per volume, plus CLIP-guided semantic conditioning and triplet co-training, are enough to outperform existing semi-supervised medical image segmentation methods on three public…

desk verdict A sensible three-slice annotation protocol with CLIP guidance, but the claimed SOTA edge over two-slice methods is not yet proven because the main comparison is budget-unfair. read the letter →

arxiv 2412.15526 v1 pith:QTSWM725 submitted 2024-12-20 cs.CV

Ke Yan , Qing Cai , Fan Zhang , Ziyan Cao , Zhi Liu This is my paper · ORCID
classification cs.CV
keywords semi-supervisedlearningmedicalimagesegmentationsparseannotationorthogonalslicesco-trainingCLIPpseudo-labels3D
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 a framework, SGTC, that aims to make semi-supervised medical image segmentation practical with extremely sparse annotation: only three slices per labeled volume, one from each orthogonal plane (sagittal, coronal, axial). The authors report that on LA2018, KiTS19, and LiTS, using 10% labeled cases, SGTC beats recent sparse- and dense-annotation methods on Dice, Jaccard, Hausdorff distance, and average surface distance. If the claim holds, radiologists would annotate a handful of slices instead of entire volumes while still getting state-of-the-art segmentations, especially at weak boundaries. The contribution is presented as two coupled mechanisms: semantic guidance from a frozen CLIP text encoder, and a triple-view co-training scheme that keeps three sub-networks distinct while exchanging pseudo-labels.

What carries the argument

The mechanism is a two-part architecture. First, triple-view disparity training: three V-Net sub-networks, $F_s$, $F_c$, $F_a$, are trained so that each sees two of the three orthogonal annotated slices as its own supervision, and on unlabeled volumes each receives pseudo-labels from the other two, filtered by Monte Carlo dropout uncertainty. Second, semantic-guided auxiliary learning: a frozen CLIP text encoder plus an adapter produces a text embedding $w$, which is concatenated with the global image feature and passed through an MLP to yield cross-modal parameters $\gamma$ that are element-wise added to the features before the classification layer. This mechanism carries the argument by injecting semantic context to sharpen boundaries and by preserving the spatial disparity needed for meaningful co-training.

What would settle it

Take the LA2018 or KiTS19 training set and replace the center-based slice selection with three randomly chosen orthogonal slices per volume, keeping all other training settings identical. If the Dice and surface-distance margins over BCP and Desco shrink to near zero or reverse, the advantage depends on informed slice selection rather than on the triplet co-training mechanism itself.

Watch

Extended reading notes

Core claim

The paper's central claim is that a triplet of 3D segmentation networks, each supervised by a different pair of the three orthogonal slice labels and cross-supervised on unlabeled volumes by the other two networks' uncertainty-filtered predictions, can produce segmentations that surpass previous semi-supervised methods under sparse annotation. The semantic branch takes a medical text prompt through a frozen CLIP encoder and a small adapter, concatenates the resulting text embedding with the image feature, and uses an MLP to produce per-network conditioning parameters that are added into the features before classification. The authors assert this combination yields semantic-aware, fine-granular segmentation, with the largest reported gains over the second-best method on LA2018 (6.3% Dice) and consistent improvements on the CT datasets.

Load-bearing premise

The reported gains assume a radiologist can select, for each labeled volume, three orthogonal slices that lie near the target organ's center and contain substantial foreground; the paper itself notes in the Conclusion that performance degrades when the selected slices contain limited foreground information.

Editorial extensions

If this is right

  • Radiologists could label three slices per volume instead of slice-by-slice full volumes, cutting annotation cost while reportedly improving segmentation quality on the tested organs.
  • The triplet co-training design generalizes across MR and CT modalities because it relies only on the standard three anatomical planes and a 3D backbone.
  • Pseudo-label quality is claimed to improve enough that the framework can learn effectively from about 90% unlabeled volumes per batch setting.
  • The semantic branch works with a frozen natural-image CLIP encoder, suggesting text-guided conditioning can be added to existing segmentation networks without retraining the text model.
  • The reported ablation shows both components contribute, and the dynamic loss weight $\alpha$ stabilizes training better than fixed values.

Reading between the lines

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

  • The paper's comparison with BCP at matched annotation strategy in Table 5 shows a small gap (0.915 vs 0.913 on KiTS19), which suggests the triple-view scheme may matter less than the extra third slice; a fair test would be to give BCP the same three orthogonal slices and compare.
  • Because CLIP was trained on natural images rather than medical data, the text branch may act more as a generic feature-modulation prior than as true medical semantic knowledge; replacing the CLIP embedding with a fixed random projection would test this.
  • The method assumes central slices that contain the target organ; for lesions near the periphery or for organs with high anatomical variability, the informed-slice requirement could be a practical bottleneck, and an extension could select slices automatically based on foreground likelihood.
  • The semantic conditioning is reported to improve boundary metrics, so a natural next check is to see whether the gains concentrate in Hausdorff distance and average surface distance rather than Dice.
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 / 4 minor

Summary. The paper proposes SGTC, a semi-supervised framework for medical image segmentation under a sparse annotation protocol in which each of a few labeled volumes has exactly three annotated orthogonal slices (sagittal, coronal, axial). The method combines a CLIP-based semantic-guided auxiliary learning module, which injects text-embedding-derived cross-modal parameters into three V-Net branches, with a triple-view disparity training strategy that supervises each branch with two of the three slice labels and cross-supervises unlabeled volumes with uncertainty-filtered pseudo-labels from the other two branches. Experiments on LA2018, KiTS19, and LiTS report state-of-the-art Dice and boundary metrics against seven semi-supervised baselines, together with ablations of the components, prompt templates, annotation geometry, and the loss weight alpha. Code is released.

Significance. If the reported results were robust, the paper would make a useful contribution to reducing annotation cost: the clinical protocol of annotating only three slices is plausible, the CLIP-adapter mechanism is simple to reproduce, and the triple-view co-training idea is a reasonable extension of orthogonal-slice methods such as Desco. The release of code is valuable. However, the headline comparison does not currently isolate the algorithmic contribution: Tables 1-3 give SGTC an extra annotated slice relative to every baseline, and the matched-budget comparison in Table 5 shows only a 0.2 Dice gain over BCP without error bars or significance testing. The paper's own limitation statement admits sensitivity to the choice of annotated slices, which is central to the proposed clinical protocol. I therefore see the contribution as promising but not yet established.

major comments (4)
  1. [Comparison Experiments (Tables 1-3)] The main SOTA comparison is confounded by annotation budget. In Tables 1-3 all baselines and the SGTC(Dual) ablation use CA (two orthogonal slices), while SGTC(Ours) uses SCA (three orthogonal slices). Since the text reports improvements such as +6.3% Dice over the second-best method on LA2018, the gain may simply reflect the extra slice of supervision. Internal evidence supports this concern: on LA2018, SGTC(Dual) at the same two-slice budget scores 0.739 Dice versus 0.784 for BCP, so SGTC is worse than BCP until it receives the third slice. The central claim should be re-tested with all baselines run at the SCA budget, or re-framed as a comparison of annotation protocols rather than of algorithms.
  2. [Ablation Study and Analysis (Table 5)] The controlled comparison is underpowered. Table 5, the only place where all methods use the same CAC or SCA strategy on KiTS19, reports SGTC at 0.911/0.915 Dice versus BCP at 0.909/0.913, a 0.2-point gap, with no error bars or pairwise significance test. The accompanying sentence states that the performance gains are due to the more effective triple-view disparity training strategy, which is stronger than the evidence in the table. Please provide repeated-run statistics (e.g., mean plus/minus standard deviation over multiple seeds and a paired test) or soften the claim to competitive.
  3. [Comparison Results on LiTS (Table 3)] In the LiTS comparison, the SGTC (Dual) row is reported with 80 unlabeled scans while every other row uses 90 unlabeled scans. This unequal unlabeled pool makes the row incomparable and undermines the cross-dataset consistency of the comparison; it also matters for semi-supervised methods, where the size of the unlabeled pool influences performance. Please correct the count or explain the discrepancy, and rerun the comparison with matched unlabeled data.
  4. [Implementation Details / Conclusion] The slice-selection protocol assumes access to foreground location information. The Implementation Details state that the three selected slices should contain the foreground area of the segmentation target and should be chosen as close to the center position as possible in all three planes, and the Conclusion admits performance degradation when the selected slices contain limited foreground information. Because slice selection is informed by target location, the claimed clinical scenario, in which radiologists just need to annotate three orthogonal slices, is not fully demonstrated. I ask for an experiment that varies slice positions (e.g., central versus off-center) and reports Dice, so readers can assess sensitivity to this assumption.
minor comments (4)
  1. [Methodology, Eq. (6)] Equation (6) contains a malformed normalization expression; please rewrite it cleanly and define all variables (H, W, D, k) at first use.
  2. [Implementation Details, Eq. (10)] The dynamic alpha schedule is under-specified: increased every 150 iterations does not state the increment or update rule. Since Table 7's main result depends on this schedule, please give the exact schedule (e.g., linear or step, from alpha = 0.1 to which maximum).
  3. [Ablation Study and Analysis, Figure 7] The t-SNE description is confusing: it says we trained five networks with just one single annotated slice per volume, then lists three orthogonal slices for training s, c, and a1, and three parallel slices for training a1, a2, and a3; please clarify the setup and the labels in the figure.
  4. [Comparison Experiments, first paragraph] The LiTS dataset is cited to Heller et al. 2019 in this paragraph, whereas the Datasets section cites Bilic et al. 2023; please correct the citation.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the paper's claims are empirical and not forced by construction; the unequal annotation budget in Tables 1-3 is an experimental-design concern, not circularity.

full rationale

The paper does not contain a derivation chain in which an output is defined in terms of the claimed prediction. SGTC's two components (SGAL and TVDT) are presented as architectural and loss-level contributions, and the reported results are empirical comparisons on public benchmarks. No parameter is fitted to the target metric and then renamed as a prediction, and no uniqueness theorem is imported from prior work by the same authors to forbid alternatives. The main concern of the reader's take is that Tables 1-3 give SGTC three orthogonal annotated slices (SCA) while baselines receive only two (CA), so part of the headline advantage may reflect extra supervision rather than algorithmic merit. That concern is real, but it is not circular: the paper does not define SGTC's advantage as 'has more slices'; it reports measured Dice values. Moreover, Table 5 provides a controlled comparison under identical CAC or SCA strategies, where SGTC's margin over BCP is about 0.2 Dice, and the paper acknowledges in the Conclusion that performance degrades when selected slices contain limited foreground. These are evidence-quality issues about effect size, statistical testing, and protocol validity, not cases where an equation equals its own input by construction. Self-citations appear in the reference list but are not load-bearing: no central claim rests solely on the authors' own prior results. The honest circularity finding is therefore no significant circularity.

Assumptions & free parameters 4 free parameters · 3 assumptions · 0 invented entities

The paper introduces method components (SGAL, TVDT) but no new physical or mathematical entities; all components are implemented with existing networks, adapters, and loss functions.

free parameters (4)
  • dynamic alpha schedule = 0.1 initial, increased every 150 iterations
    The schedule is chosen by the authors and shown in Table 7 to outperform fixed values; it is tuned on the evaluation datasets.
  • uncertainty threshold for pseudo-label selection = not reported
    The paper states voxels with uncertainty lower than a threshold are selected but does not give the threshold value, MC dropout settings, or how it varies across models.
  • slice selection positions = center slices containing foreground
    Implementation Details state the three slices are chosen close to the center of the target and should contain foreground; this is a data-selection choice that affects results and is acknowledged as a limitation in the Conclusion.
  • text prompt template = An image containing the [CLS], with the rest being background
    The prompt is selected by ablation and is a hand-designed input to the CLIP encoder.
assumptions (3)
  • domain assumption CLIP text embeddings, after a small adapter, provide useful semantic guidance for medical 3D segmentation despite the domain gap between natural and medical images.
    Invoked in the Semantic-Guided Auxiliary Learning mechanism (Eq. 1-3); the paper provides an ablation (Table 4) but no direct validation of the semantic alignment.
  • domain assumption Monte Carlo dropout uncertainty estimates are sufficiently well calibrated to select reliable pseudo-label voxels.
    The pseudo-label selection follows (Yu et al. 2019) with an unspecified threshold; the paper offers no analysis of calibration.
  • domain assumption Three sub-networks trained on different orthogonal slice pairs will learn complementary knowledge rather than collapse to the same solution.
    This underpins the triple-view disparity training strategy; the paper shows t-SNE visualizations but no quantitative diversity measure.

how reviews work

0 comments
Cite this review

Pith. "Pith review of SGTC: Semantic-Guided Triplet Co-training for Sparsely Annotated Semi-Supervised Medical Image Segmentation." pith.science (2026). https://pith.science/paper/QTSWM725

@misc{pith2026241215526,
  author       = {Pith},
  title        = {Pith review of: SGTC: Semantic-Guided Triplet Co-training for Sparsely Annotated Semi-Supervised Medical Image Segmentation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/QTSWM725}},
  note         = {Machine review of arXiv:2412.15526}
}
read the original abstract

Although semi-supervised learning has made significant advances in the field of medical image segmentation, fully annotating a volumetric sample slice by slice remains a costly and time-consuming task. Even worse, most of the existing approaches pay much attention to image-level information and ignore semantic features, resulting in the inability to perceive weak boundaries. To address these issues, we propose a novel Semantic-Guided Triplet Co-training (SGTC) framework, which achieves high-end medical image segmentation by only annotating three orthogonal slices of a few volumetric samples, significantly alleviating the burden of radiologists. Our method consist of two main components. Specifically, to enable semantic-aware, fine-granular segmentation and enhance the quality of pseudo-labels, a novel semantic-guided auxiliary learning mechanism is proposed based on the pretrained CLIP. In addition, focusing on a more challenging but clinically realistic scenario, a new triple-view disparity training strategy is proposed, which uses sparse annotations (i.e., only three labeled slices of a few volumes) to perform co-training between three sub-networks, significantly improving the robustness. Extensive experiments on three public medical datasets demonstrate that our method outperforms most state-of-the-art semi-supervised counterparts under sparse annotation settings. The source code is available at https://github.com/xmeimeimei/SGTC.

Figures

Figures reproduced from arXiv: 2412.15526 by the authors.

Figure 1
Figure 1. (a) The semantic guidance in visualization com [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Architecture of the proposed SGTC framework. For volumes with sparse orthogonal labels, each volume has three [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Qualitative comparisons on the LA2018 dataset with 10% labeled cases. From left to right: segmentation results [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: Qualitative comparisons on the KITS19 dataset with 10% labeled cases. [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]
Figure 5
Figure 5. Figure 5: Qualitative comparisons on the LITS dataset with 10% labeled cases. [PITH_FULL_IMAGE:figures/full_fig_p006_5.png]
Figure 7
Figure 7. Figure 7: The t-SNE visualization of feature representations [PITH_FULL_IMAGE:figures/full_fig_p007_7.png]
Figure 6
Figure 6. Figure 6: Performance comparisons of different annotations. [PITH_FULL_IMAGE:figures/full_fig_p007_6.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

54 extracted references · 49 canonical work pages

  1. [1]

    B.; Tustison, N

    Avants, B. B.; Tustison, N. J.; Song, G.; Cook, P. A.; Klein, A.; and Gee, J. C. 2011. A reproducible evaluation of ANTs similarity metric performance in brain image registration. Neuroimage, 54(3): 2033--2044

    work page 2011

  2. [2]

    M.; and Rueckert, D

    Bai, W.; Oktay, O.; Sinclair, M.; Suzuki, H.; Rajchl, M.; Tarroni, G.; Glocker, B.; King, A.; Matthews, P. M.; and Rueckert, D. 2017. Semi-supervised learning for network-based cardiac MR image segmentation. In International Conference on Medical Image Computing and Computer-Assisted Intervention., 253--260. Springer

    work page 2017

  3. [3]

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

    work page 2023

  4. [4]

    B.; Vorontsov, E.; Ben-Cohen, A.; Kaissis, G.; Szeskin, A.; Jacobs, C.; Mamani, G

    Bilic, P.; Christ, P.; Li, H. B.; Vorontsov, E.; Ben-Cohen, A.; Kaissis, G.; Szeskin, A.; Jacobs, C.; Mamani, G. E. H.; Chartrand, G.; et al. 2023. The liver tumor segmentation benchmark (LiTS). Medical Image Analysis, 84: 102680

    work page 2023

  5. [5]

    Cai, H.; Li, S.; Qi, L.; Yu, Q.; Shi, Y.; and Gao, Y. 2023. Orthogonal annotation benefits barely-supervised medical image segmentation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 3302--3311

    work page 2023

  6. [6]

    Cai, Q.; Liu, H.; Qian, Y.; Zhou, S.; Duan, X.; and Yang, Y.-H. 2019. Saliency-guided level set model for automatic object segmentation. Pattern Recognition, 93: 147--163

    work page 2019

  7. [7]

    Cai, Q.; Liu, H.; Qian, Y.; Zhou, S.; Wang, J.; and Yang, Y.-H. 2021 a . A novel hybrid level set model for non-rigid object contour tracking. IEEE Transactions on Image Processing, 31: 15--29

    work page 2021

  8. [8]

    Cai, Q.; Liu, H.; Zhou, S.; Sun, J.; and Li, J. 2018. An adaptive-scale active contour model for inhomogeneous image segmentation and bias field estimation. Pattern Recognition, 82: 79--93

    work page 2018

Show all 54 references
  1. [9]

    Cai, Q.; Qian, Y.; Zhou, S.; Li, J.; Yang, Y.-H.; Wu, F.; and Zhang, D. 2021 b . AVLSM: Adaptive variational level set model for image segmentation in the presence of severe intensity inhomogeneity and high noise. IEEE Transactions on Image Processing, 31: 43--57

    2021

  2. [10]

    Chen, D.; Bai, Y.; Shen, W.; Li, Q.; Yu, L.; and Wang, Y. 2023. Magicnet: Semi-supervised multi-organ segmentation via magic-cube partition and recovery. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 23869--23878

    2023

  3. [11]

    Chen, X.; Yuan, Y.; Zeng, G.; and Wang, J. 2021. Semi-supervised semantic segmentation with cross pseudo supervision. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2613--2622

    2021

  4. [12]

    Chen, Z.; Li, G.; and Wan, X. 2022. Align, reason and learn: Enhancing medical vision-and-language pre-training with knowledge. In Proceedings of the 30th ACM International Conference on Multimedia, 5152--5161

    2022

  5. [13]

    Cong, F.; Xu, S.; Guo, L.; and Tian, Y. 2022. Caption-aware medical VQA via semantic focusing and progressive cross-modality comprehension. In Proceedings of the 30th ACM International Conference on Multimedia, 3569--3577

    2022

  6. [14]

    Guo, S.; Cai, Q.; Qi, L.; and Dong, J. 2023. CLIP-Hand3D: Exploiting 3D Hand Pose Estimation via Context-Aware Prompting. In Proceedings of the 31st ACM International Conference on Multimedia, 4896--4907

    2023

  7. [15]

    Heller, N.; Sathianathen, N.; Kalapara, A.; Walczak, E.; Moore, K.; Kaluzniak, H.; Rosenberg, J.; Blake, P.; Rengel, Z.; Oestreich, M.; et al. 2019. The kits19 challenge data: 300 kidney tumor cases with clinical context, ct semantic segmentations, and surgical outcomes. arXiv...

    2019 arXiv

  8. [16]

    Huang, H.; Huang, Y.; Xie, S.; Lin, L.; Tong, R.; Chen, Y.-W.; Li, Y.; and Zheng, Y. 2024. Combinatorial CNN-Transformer Learning with Manifold Constraints for Semi-supervised Medical Image Segmentation. In Proceedings of the AAAI Conference on Artificial Intelligence, volume ...

    2024

  9. [17]

    P.; and Yeung, S

    Huang, S.-C.; Shen, L.; Lungren, M. P.; and Yeung, S. 2021. Gloria: A multimodal global-local representation learning framework for label-efficient medical image recognition. In Proceedings of the IEEE/CVF International Conference on Computer Vision, 3942--3951

    2021

  10. [18]

    Li, S.; Cai, H.; Qi, L.; Yu, Q.; Shi, Y.; and Gao, Y. 2022. PLN: Parasitic-like network for barely supervised medical image segmentation. IEEE Transactions on Medical Imaging, 42(3): 582--593

    2022

  11. [19]

    Li, S.; Cai, Q.; Zhang, F.; Zhang, M.; Shu, Y.; Liu, Z.; Li, H.; and Liu, L. 2024. PP-SSL: Priority-Perception Self-Supervised Learning for Fine-Grained Recognition. arXiv preprint arXiv:2412.00134

    2024 arXiv

  12. [20]

    Li, S.; Zhang, C.; and He, X. 2020. Shape-aware semi-supervised 3D semantic segmentation for medical images. In International Conference on Medical Image Computing and Computer-Assisted Intervention., 552--561. Springer

    2020

  13. [21]

    Li, X.; Yu, L.; Chen, H.; Fu, C.-W.; Xing, L.; and Heng, P.-A. 2020. Transformation-consistent self-ensembling model for semisupervised medical image segmentation. IEEE Transactions on Neural Networks and Learning Systems, 32(2): 523--534

    2020

  14. [22]

    Li, Z.; Li, Y.; Li, Q.; Wang, P.; Guo, D.; Lu, L.; Jin, D.; Zhang, Y.; and Hong, Q. 2023 a . Lvit: language meets vision transformer in medical image segmentation. IEEE Transactions on Medical Imaging

    2023

  15. [23]

    Li, Z.; Zheng, Y.; Luo, X.; Shan, D.; and Hong, Q. 2023 b . ScribbleVC: Scribble-supervised Medical Image Segmentation with Vision-Class Embedding. In Proceedings of the 31st ACM International Conference on Multimedia, 3384--3393

    2023

  16. [24]

    Liu, J.; Zhang, Y.; Chen, J.-N.; Xiao, J.; Lu, Y.; A Landman, B.; Yuan, Y.; Yuille, A.; Tang, Y.; and Zhou, Z. 2023. Clip-driven universal model for organ segmentation and tumor detection. In Proceedings of the IEEE/CVF International Conference on Computer Vision, 21152--21164

    2023

  17. [25]

    Luo, X.; Chen, J.; Song, T.; and Wang, G. 2021. Semi-supervised medical image segmentation through dual-task consistency. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 35, 8801--8809

    2021

  18. [26]

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

    2016

  19. [27]

    W.; Hallacy, C.; Ramesh, A.; Goh, G.; Agarwal, S.; Sastry, G.; Askell, A.; Mishkin, P.; Clark, J.; et al

    Radford, A.; Kim, J. W.; Hallacy, C.; Ramesh, A.; Goh, G.; Agarwal, S.; Sastry, G.; Askell, A.; Mishkin, P.; Clark, J.; et al. 2021. Learning transferable visual models from natural language supervision. In International Conference on Machine Learning, 8748--8763. PMLR

    2021

  20. [28]

    Shen, Z.; Cao, P.; Yang, H.; Liu, X.; Yang, J.; and Zaiane, O. R. 2023. Co-training with high-confidence pseudo labels for semi-supervised medical image segmentation. In Proceedings of the Thirty-Second International Joint Conference on Artificial Intelligence, 4199--4207

    2023

  21. [29]

    Shin, H.; Kim, H.; Kim, S.; Jun, Y.; Eo, T.; and Hwang, D. 2022. COSMOS: cross-modality unsupervised domain adaptation for 3D medical image segmentation based on target-aware domain translation and iterative self-training. arXiv preprint arXiv:2203.16557

    2022 arXiv

  22. [30]

    Tarvainen, A.; and Valpola, H. 2017. Mean teachers are better role models: Weight-averaged consistency targets improve semi-supervised deep learning results. Advances in Neural Information Processing Systems, 30

    2017

  23. [31]

    H.; Di Caterina, G.; and Voisey, J

    Thompson, B. H.; Di Caterina, G.; and Voisey, J. P. 2022. Pseudo-label refinement using superpixels for semi-supervised brain tumour segmentation. In 2022 IEEE 19th International Symposium on Biomedical Imaging, 1--5. IEEE

    2022

  24. [32]

    Wang, A.; Chen, H.; Lin, Z.; Ding, Z.; Liu, P.; Bao, Y.; Yan, W.; and Ding, G. 2023 a . Hierarchical Prompt Learning Using CLIP for Multi-label Classification with Single Positive Labels. In Proceedings of the 31st ACM International Conference on Multimedia, 5594--5604

    2023

  25. [33]

    Wang, H.; and Li, X. 2024. Towards generic semi-supervised framework for volumetric medical image segmentation. Advances in Neural Information Processing Systems, 36

    2024

  26. [34]

    Wang, X.; Rigall, E.; An, X.; Li, Z.; Cai, Q.; Zhang, S.; and Dong, J. 2024. A New Benchmark and Low Computational Cost Localization Method for Cephalometric Analysis. IEEE Transactions on Circuits and Systems for Video Technology

    2024

  27. [35]

    Wang, X.; Yuan, Y.; Guo, D.; Huang, X.; Cui, Y.; Xia, M.; Wang, Z.; Bai, C.; and Chen, S. 2022. SSA-Net: Spatial self-attention network for COVID-19 pneumonia infection segmentation with semi-supervised few-shot learning. Medical Image Analysis, 79: 102459

    2022

  28. [36]

    Wang, Y.; Huang, S.; Gao, Y.; Wang, Z.; Wang, R.; Sheng, K.; Zhang, B.; and Liu, S. 2023 b . Transferring CLIP's Knowledge into Zero-Shot Point Cloud Semantic Segmentation. In Proceedings of the 31st ACM International Conference on Multimedia, 3745--3754

    2023

  29. [37]

    Z.; and Wu, J

    Wu, Y.; Chen, J.; Yan, J.; Zhu, Y.; Chen, D. Z.; and Wu, J. 2023. GCL: Gradient-Guided Contrastive Learning for Medical Image Segmentation with Multi-Perspective Meta Labels. In Proceedings of the 31st ACM International Conference on Multimedia, 463--471

    2023

  30. [38]

    Xiong, Z.; Xia, Q.; Hu, Z.; Huang, N.; Bian, C.; Zheng, Y.; Vesal, S.; Ravikumar, N.; Maier, A.; Yang, X.; et al. 2021. A global benchmark of algorithms for segmenting the left atrium from late gadolinium-enhanced cardiac magnetic resonance imaging. Medical Image Analysis, 67: 101832

    2021

  31. [39]

    Xu, X.; Xiong, T.; Ding, Z.; and Tu, Z. 2023. Masqclip for open-vocabulary universal image segmentation. In Proceedings of the IEEE/CVF International Conference on Computer Vision, 887--898

    2023

  32. [40]

    Ye, Y.; Liu, Z.; Zhang, Y.; Li, J.; and Shen, H. 2022. Alleviating style sensitivity then adapting: Source-free domain adaptation for medical image segmentation. In Proceedings of the 30th ACM International Conference on Multimedia, 1935--1944

    2022

  33. [41]

    Yu, L.; Wang, S.; Li, X.; Fu, C.-W.; and Heng, P.-A. 2019. Uncertainty-aware self-ensembling model for semi-supervised 3D left atrium segmentation. In International Conference on Medical Image Computing and Computer-Assisted Intervention., 605--613. Springer

    2019

  34. [42]

    Yu, Y.; Zhan, F.; Wu, R.; Zhang, J.; Lu, S.; Cui, M.; Xie, X.; Hua, X.-S.; and Miao, C. 2022. Towards counterfactual image manipulation via clip. In Proceedings of the 30th ACM International Conference on Multimedia, 3637--3645

    2022

  35. [43]

    Yuan, Z.; Jin, Q.; Tan, C.; Zhao, Z.; Yuan, H.; Huang, F.; and Huang, S. 2023. Ramm: Retrieval-augmented biomedical visual question answering with multi-modal pre-training. In Proceedings of the 31st ACM International Conference on Multimedia, 547--556

    2023

  36. [44]

    Yun, B.; Xie, X.; Li, Q.; and Wang, Y. 2023. Uni-Dual: A Generic Unified Dual-Task Medical Self-Supervised Learning Framework. In Proceedings of the 31st ACM International Conference on Multimedia, 3887--3896

    2023

  37. [45]

    Zhang, F.; Liu, H.; Cai, Q.; Feng, C.-M.; Wang, B.; Wang, S.; Dong, J.; and Zhang, D. 2024 a . Federated Cross-Incremental Self-Supervised Learning for Medical Image Segmentation. IEEE Transactions on Neural Networks and Learning Systems

    2024

  38. [46]

    Zhang, F.; Liu, H.; Cao, C.; Cai, Q.; and Zhang, D. 2022 a . RVLSM: Robust variational level set method for image segmentation with intensity inhomogeneity and high noise. Information Sciences, 596: 439--459

    2022

  39. [47]

    Zhang, F.; Liu, H.; Duan, X.; Wang, B.; Cai, Q.; Li, H.; Dong, J.; and Zhang, D. 2024 b . DSLSM: Dual-kernel-induced statistic level set model for image segmentation. Expert Systems with Applications, 242: 122772

    2024

  40. [48]

    Zhang, F.; Liu, H.; Wang, J.; Lyu, J.; Cai, Q.; Li, H.; Dong, J.; and Zhang, D. 2024 c . Cross co-teaching for semi-supervised medical image segmentation. Pattern Recognition, 110426

    2024

  41. [49]

    Zhang, W.; Zhang, X.; Huang, S.; Lu, Y.; and Wang, K. 2022 b . Pixelseg: Pixel-by-pixel stochastic semantic segmentation for ambiguous medical images. In Proceedings of the 30th ACM International Conference on Multimedia, 4742--4750

    2022

  42. [50]

    Zhang, W.; Zhang, X.; Huang, S.; Lu, Y.; and Wang, K. 2022 c . A probabilistic model for controlling diversity and accuracy of ambiguous medical image segmentation. In Proceedings of the 30th ACM International Conference on Multimedia, 4751--4759

    2022

  43. [51]

    Zhou, H.-Y.; Lu, C.; Chen, C.; Yang, S.; and Yu, Y. 2023. A unified visual information preservation framework for self-supervised pre-training in medical image analysis. IEEE Transactions on Pattern Analysis and Machine Intelligence

    2023

  44. [52]

    Zhou, Y.; Wang, Y.; Tang, P.; Bai, S.; Shen, W.; Fishman, E.; and Yuille, A. 2019. Semi-supervised 3D abdominal multi-organ segmentation via deep multi-planar co-training. In 2019 IEEE Winter Conference on Applications of Computer Vision, 121--140. IEEE

    2019

  45. [53]

    , " * 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...

  46. [54]

    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 gl...

Pith tools

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