Pith. sign in

REVIEW 4 major objections 6 minor 51 references

HisynSeg: Weakly-Supervised Histopathological Image Segmentation via Image-Mixing Synthesis and Consistency Regularization

T0 review · 4 major / 6 minor · reviewed 2026-08-10 · deepseek-v4-flash

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

Pith's one-line read HisynSeg transforms weakly supervised histopathology segmentation into fully supervised training by synthesizing multi-tissue images with exact pixel masks from single-tissue sources, achieving state-of-the-art mIoU on three datasets.

desk verdict A solid, thoroughly evaluated synthesis-based WSSS system whose main claims are slightly oversold and whose single-tissue purity assumption needs an audit. read the letter →

arxiv 2412.20924 v1 pith:W6NZHYRI submitted 2024-12-30 cs.CV cs.AI

Zijie Fang , Yifeng Wang , Peizhang Xie , Zhi Wang , Yongbing Zhang This is my paper · ORCID
classification cs.CVcs.AI
keywords weakly-supervisedsemanticsegmentationhistopathologyimage-mixingsynthesisBéziermaskgenerationMosaictransformationconsistencyregularizationclassactivationmapcomputationalpathology
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

HisynSeg tries to make weakly supervised tissue segmentation in histopathology match fully supervised quality while using only image-level labels. Its key move is to exploit the homogeneity of histopathology images: an image whose label names exactly one tissue type can be treated as having a perfect pixel mask, so mixing such images produces synthetic multi-tissue images with exact masks for free. Two recipes, a Mosaic-style splice and a Bézier-curve blend, create realistic tissue boundaries; a discriminator filters out composites that look fake; and a consistency regularizer lets real unlabeled images participate in training. The paper reports state-of-the-art mIoU of 76.66%, 70.97%, and 76.19% on WSSS4LUAD, BCSS-WSSS, and LUAD-HistoSeg, and shows the weakly trained model lands within about one point of fully supervised models on the BCSS benchmark. The authors acknowledge that the construction rests on the purity of single-label images and that MIL-style labels would cause over-segmentation.

What carries the argument

The load-bearing mechanism is the exact-mask synthesis from single-tissue images: because histopathology backgrounds are white and tissues of one category cluster together, every non-background pixel of a single-label image can be assigned that label. Two generators build multi-tissue composites from these images: Mosaic transformation selects a random anchor point and fills the four quadrants with crops of gridded single-tissue images, and Bézier mask generation forms a closed smooth shape from end-to-end cubic Bézier curves and blends two images inside and outside the mask. The compositions are filtered by a ResNet-18 image discriminator, trained with real multi-tissue images as positive examples, which rejects composites scored as fake. Training then combines a Dice segmentation loss on the synthetic masks, an L1 consistency regularization between the downsampled segmentation probability map and the softmaxed activation map from a 1x1 convolution on the shared encoder, and a multi-label soft-margin classification loss; an optional iterative round replaces the consistency loss with pseudo-masks from the model itself.

What would settle it

Take a random sample of the single-label images used for synthesis, run a fully supervised reference segmenter (or ask a pathologist) to count pixels of other tissue classes, and retrain HisynSeg after removing any image with a nontrivial secondary-tissue fraction; if the exact-mask premise holds, removal should not hurt performance, and if it does not, the reported gains should shrink accordingly.

Watch

Extended reading notes

Core claim

The central claim is that the homogeneity of histopathological images lets a weakly supervised segmentation problem be converted into a fully supervised one: a single-label image's whole foreground can be assigned its image-level class, and compositing two or four such images yields diverse multi-tissue training images with exact ground-truth masks. Mosaic transformation splices four gridded single-tissue images at a random anchor point, while Bézier mask generation blends two images inside and outside a closed smooth curve formed by cubic Bézier segments, matching the curved boundaries of real tissue. A ResNet-18 discriminator, trained separately for each synthesis strategy, discards composites it classifies as fake, and a self-supervised consistency loss aligns a downsampled segmentation probability map with the softmax of a shared-encoder activation map, so real images contribute without needing masks. On WSSS4LUAD, BCSS-WSSS, and LUAD-HistoSeg the method reports mIoU of 76.66%, 70.97%, and 76.19%, respectively, and on the BCSS benchmark its weakly supervised result lands within 0.9 mIoU of the best fully supervised model tested.

Load-bearing premise

The framework assumes that every training image whose image-level label contains exactly one tissue class really is pure, so all non-background pixels belong to that class; if single-label images contain traces of other tissues, the synthetic masks are silently wrong.

Editorial extensions

If this is right

  • On BCSS, the weakly supervised model's 70.97% mIoU lands within 0.9 points of the best fully supervised DeepLabV3+ with EfficientNet-b6 (71.83%), and it beats U-Net and U-Net++ with ResNet-50 backbones, so the gap to full supervision is small.
  • Because the masks are exact by construction, the framework sidesteps the under-activation and over-activation that limit CAM-based pseudo-masks.
  • The two synthesis strategies also act as data augmentation: adding synthesized images (without using their masks) raises SC-CAM from 72.26% to 76.20% mIoU on LUAD-HistoSeg and improves most other weakly supervised frameworks.
  • With only 10 single-label images per tissue class for synthesis, mIoU stays above 75% on LUAD-HistoSeg, still beating the previous best baseline (WSSS-Tissue, 74.38%).
  • An iterative round that replaces the consistency loss with the model's own pseudo-masks for real images yields further gains on BCSS-WSSS and LUAD-HistoSeg.

Reading between the lines

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

  • The purity assumption can be audited directly: a fully supervised reference model run over single-label images would reveal how often secondary tissues appear, and a purity filter could replace the hard "single tissue" criterion.
  • The same recipe should transfer to other imaging domains with homogeneous contents and smooth boundaries, such as bright-field microscopy of cultured cells, where one-label images are cheap and synthetic exact masks could support weakly supervised segmentation.
  • The paper's stated limitation that the discriminator only filters suggests a closed-loop variant: use the discriminator's confidence as a training signal for a learnable compositor, which could reduce the compute wasted on rejected synthetic images.
  • On MIL-labeled datasets (a patch is 'tumor' if it contains any tumor), the method would over-segment; a testable extension is to estimate patch-level purity and restrict synthesis to high-purity single-label images.
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

4 major / 6 minor

Summary. The paper proposes HisynSeg, a weakly-supervised histopathology image segmentation framework that avoids CAM-based pseudo-mask generation by synthesizing training images with pixel-level masks. Two synthesis strategies are introduced: Mosaic transformation and Bézier-mask-based image mixing, both operating on histopathology images that contain a single tissue type, so that the composite images inherit pixel-exact masks. A ResNet-18 discriminator filters synthesized images for authenticity, and a segmentation model is trained with a Dice loss on the surviving synthetic images. Real images without pixel masks are additionally used through an auxiliary classification head and a consistency regularization that aligns the decoder's softmax probability map with the encoder's activation map. An iterative self-training variant is also described. Experiments on WSSS4LUAD, BCSS-WSSS, and LUAD-HistoSeg report state-of-the-art mIoU values (76.66%, 70.97%, 76.19%) with permutation tests over five seeds, along with extensive ablations of synthesis strategies, filtering, loss combinations, loss weights, backbones, and limited single-label data.

Significance. If the reported results hold, the paper introduces a genuinely different paradigm for weakly supervised histopathology segmentation: instead of refining imperfect CAMs, it uses the homogeneity of histological images to obtain pixel-level masks for synthesized images. The strengths of the manuscript are substantial: code is released, experiments are run over five seeds, permutation tests accompany pairwise comparisons, ablations cover most design choices, and the method is compared against both CAM-based WSSS methods and semi-supervised methods. The synthesis+filtering pipeline is also evaluated with FID/KID, which is a useful addition. The central conceptual claim, however, is broader than what the method actually delivers: real images are still supervised through CAM consistency and through pseudo-masks produced by the model itself, so the paper does not literally transform WSSS into a fully supervised problem. The key premise that a single-tissue image-level label implies a pixel-pure mask is not audited, and this premise is load-bearing for the entire synthesis module.

major comments (4)
  1. [§III-A, first paragraph] The load-bearing premise that 'we can obtain accurate semantic segmentation masks for histopathological images with a single tissue type' is not audited. A single-label image guarantees only that the annotator chose not to mark a second tissue; it does not guarantee that every non-background pixel belongs to the labeled class, nor that white background is not included in the tissue region. Since background pixels are excluded from evaluation (Section IV-B), a synthesized mask that paints white regions as tumor or stroma is not penalized at test time, but it teaches the model to emit tissue in background areas. The risk is concrete for WSSS4LUAD, whose image-level labels were produced by a ResNet-38 with pathologist review rather than derived from pixel masks (Section IV-A, dataset a). I request a quantitative purity audit for each dataset: per-class statistics of non-background pixel fractions in the selected single-label images, and ideally a manual or pixel-mask-based verification on a sample. Without this, the reported SOTA numbers rest on an unverified premise exactly at the point where weak supervision is converted to strong supervision.
  2. [Abstract and Section I, contributions] The claim that HisynSeg 'successfully transforms the weakly-supervised semantic segmentation problem into a fully-supervised one' is overstated. Real images without masks are still trained through Eq. (9), which aligns the decoder's probability map to the encoder's own activation map Fc, and Section III-D retrains the model on pseudo-masks generated by the model itself for real images. The synthesis module supplies fully supervised samples, but the overall framework is a hybrid of fully supervised training on synthetic data, CAM-consistency regularization on real data, and self-training. Please either soften the wording (e.g., 'largely avoids CAM-based pseudo-mask generation' or 'converts the training signal into fully supervised form for synthetic images') or add an experiment that isolates the contribution of synthetic masks while ablating Eq. (9) and Section III-D, showing the performance contribution of each component.
  3. [Eq. (9) and Section III-C] There is a potential circularity issue in the consistency regularization: Lreg aligns the segmentation probability map to the encoder's own tissue activation map. If the encoder produces under-activated or over-activated CAMs, as the paper itself argues in the introduction and Figure 1, then this loss can reinforce those errors instead of correcting them. The paper does not analyze this failure mode, and the loss-ablation table (Table XIV) shows only that removing Lreg hurts performance, not that the consistency term is robust to CAM errors. Please add a discussion and, ideally, an experiment that measures how the quality of Fc (e.g., its IoU against the synthesized masks or against ground-truth masks on a validation subset) relates to the contribution of Lreg.
  4. [Section III-D and Table II] The iterative training strategy is applied only to BCSS-WSSS and LUAD-HistoSeg, not to WSSS4LUAD, with the statement that no performance improvement was empirically found. Since Section III-D retrains on self-generated pseudo-masks, it is a form of self-training that can introduce confirmation bias. The paper does not report what happens on the two datasets when iterative training is removed, nor does it analyze the quality of the pseudo-masks used in the retraining stage. Please report the ablation of the iterative strategy on the two datasets where it is used, and show the IoU of the pseudo-masks against ground truth (where available) to demonstrate that self-training is not merely consolidating the model's own errors.
minor comments (6)
  1. [Eq. (2)] The word 'bionormial' should be 'binomial', and the definition of the binomial coefficient should be given with a conventional factorial expression.
  2. [Eq. (1) and Figure 3] The notation in Eq. (1) for the Mosaic image layout is not fully self-explanatory; please clarify the correspondence between the four quadrants and the four cropped images, and check that Figure 3's arrow labels are legible when printed.
  3. [Section IV-C and Table V] There is a discrepancy in the number of synthesized images used for discriminator evaluation: the text says 10,000 images, but the footnote to Table V says 50,000 images are synthesized for the WSSS4LUAD Mosaic row. Please correct or reconcile this.
  4. [Tables II and III] The abbreviation 'Ours (Conf.)' is not defined; please spell out that it refers to the conference version [18].
  5. [Section IV-D] The permutation test procedure is described briefly; please specify how many permutations were used, whether the test is on paired samples, and how the p-value is computed, so that the statistical claims are reproducible.
  6. [Section V] The limitation paragraph correctly notes that the filtering module does not feed back into the synthesis module; this limitation is also visible in the numbers (Table V), where only a small fraction of synthesized images are classified as real for some datasets. Please quantify how many images are discarded by the filtering module for each dataset and strategy, since this directly affects the effective training set size.

Circularity Check

0 steps flagged · score 2.0 of 10

No significant circularity: the synthetic pixel masks are constructed from weak labels and final metrics are computed on held-out pixel ground truth; the self-citation to the authors' conference version is not load-bearing.

full rationale

The central derivation is self-contained against external benchmarks. Single-label histopathology images are used to compose synthetic images whose pixel masks are known by construction from the constituent image-level labels (Eqs. 1 and 4), and the segmentation model is trained on those masks via Dice loss (Eq. 6). All reported mIoU/fwIoU values are evaluated on held-out pixel-level ground truth of WSSS4LUAD, BCSS-WSSS, and LUAD-HistoSeg, so the final prediction does not reduce to the training labels. The only self-citation is to the authors' conference version [18], used as a baseline and as a starting point; the paper re-describes the synthesis and adds the Bézier strategy, filtering, and consistency regularization, so the argument is not carried by an unverified self-citation. Equations (9) and Section III-D do create self-distillation loops: real images are supervised by the encoder's own activation map, and after training the model is retrained on its own pseudo-masks. These loops provide no new independent pixel evidence, but they are auxiliary to the synthesis-based core and the final test evaluation remains on independent masks. The Section V limitation that the framework 'relies on accurate image-level labels' is a correctness risk rather than circularity: impure single-label images would corrupt the synthetic masks and degrade, not tautologically produce, the reported test accuracy. No equation or fitted parameter is equivalent to the target result by construction; the score only reflects the minor, non-load-bearing self-citation.

Assumptions & free parameters 6 free parameters · 6 assumptions · 0 invented entities

The central claim rests on domain assumptions about label purity and the validity of synthetic training data, plus several hand-set hyperparameters. No new physical or conceptual entities are introduced; the masks, discriminators, and regularizers are algorithmic components.

free parameters (6)
  • Anchor range for Mosaic transformation = alpha=0.2, beta=0.8
    Chosen by hand in Section III-A.1; controls how central the splice point can be; not ablated.
  • Mosaic grid size m = m=2
    Chosen in Section III-A.1; controls number of tiles; not ablated.
  • Number of filtered synthesized images per strategy = 3,600 for WSSS4LUAD and LUAD-HistoSeg; 7,200 for BCSS-WSSS
    Set in Section IV-C; not ablated; affects training set size.
  • Discriminator acceptance threshold = 0.5
    Section III-B keeps synthetic images with probability greater than 0.5; threshold not ablated.
  • Bézier control-point count N = not reported
    Section III-A.2 says N random points are taken but no value or range is given; affects mask smoothness and is not ablated.
  • Loss weights for Lseg, Lreg, Lcls = 1, 1, 1
    Default weights in Eq. 12; robustness is tested in Table XV, so this is a lower concern.
assumptions (6)
  • domain assumption A histopathological image with a single tissue-type label has a pixel-level mask equal to all non-background pixels.
    Section III-A uses this to justify exact synthetic masks; no purity audit is provided.
  • domain assumption Image-level labels in WSSS4LUAD, BCSS-WSSS, and LUAD-HistoSeg are accurate enough to identify single-tissue images.
    Section IV-A describes label creation but does not quantify label noise for single-label images.
  • domain assumption Synthesized images after discriminator filtering are sufficiently representative of real tumor microenvironments to train a generalizable segmentation model.
    Defended via FID/KID and qualitative examples, but no direct proof that the filtered synthetic distribution matches the real test distribution.
  • domain assumption The real/fake classification score from a ResNet-18 discriminator is a valid proxy for segmentation usefulness of a synthesized image.
    Section III-B; validated indirectly by Table XIII rather than by a direct usefulness metric.
  • domain assumption Self-training with model-generated pseudo-masks improves rather than amplifies errors.
    Section III-D retrains on pseudo-masks for real images without an ablation isolating this step.
  • standard math Bézier curve construction and derivative continuity condition are standard and correctly applied.
    Equations 2 and 3 give the curve formula and smoothness constraint; these are uncontroversial background.

how reviews work

0 comments
Cite this review

Pith. "Pith review of HisynSeg: Weakly-Supervised Histopathological Image Segmentation via Image-Mixing Synthesis and Consistency Regularization." pith.science (2026). https://pith.science/paper/W6NZHYRI

@misc{pith2026241220924,
  author       = {Pith},
  title        = {Pith review of: HisynSeg: Weakly-Supervised Histopathological Image Segmentation via Image-Mixing Synthesis and Consistency Regularization},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/W6NZHYRI}},
  note         = {Machine review of arXiv:2412.20924}
}
read the original abstract

Tissue semantic segmentation is one of the key tasks in computational pathology. To avoid the expensive and laborious acquisition of pixel-level annotations, a wide range of studies attempt to adopt the class activation map (CAM), a weakly-supervised learning scheme, to achieve pixel-level tissue segmentation. However, CAM-based methods are prone to suffer from under-activation and over-activation issues, leading to poor segmentation performance. To address this problem, we propose a novel weakly-supervised semantic segmentation framework for histopathological images based on image-mixing synthesis and consistency regularization, dubbed HisynSeg. Specifically, synthesized histopathological images with pixel-level masks are generated for fully-supervised model training, where two synthesis strategies are proposed based on Mosaic transformation and B\'ezier mask generation. Besides, an image filtering module is developed to guarantee the authenticity of the synthesized images. In order to further avoid the model overfitting to the occasional synthesis artifacts, we additionally propose a novel self-supervised consistency regularization, which enables the real images without segmentation masks to supervise the training of the segmentation model. By integrating the proposed techniques, the HisynSeg framework successfully transforms the weakly-supervised semantic segmentation problem into a fully-supervised one, greatly improving the segmentation accuracy. Experimental results on three datasets prove that the proposed method achieves a state-of-the-art performance. Code is available at https://github.com/Vison307/HisynSeg.

Figures

Figures reproduced from arXiv: 2412.20924 by the authors.

Figure 1
Figure 1. Examples of histopathological images and pseudo-masks generated by CAM. For comparison, ground-truth masks are also pro￾vided. Nearest and bilinear represent the utilized interpolation method. Blue and red circles highlight the under-activated and over-activated regions, respectively. Black arrows indicate the introduced noise caused by interpolation. Red pixels represent tumor epithelial and green pixels stand for … view at source ↗
Figure 3
Figure 3. An overview of the HisynSeg framework. The framework is composed of three modules. In the image-mixing synthesis module, synthesized images and masks are generated by two proposed strategies, namely Mosaic transformation and Bezier mask generation. Next, the synthesized ´ image filtering module selects authentic images from the synthesized images. Finally, a segmentation model is trained in the histopathological ima… view at source ↗
Figure 4
Figure 4. The segmentation boundaries between different types of tissues for (a) a real image, (b) a synthesized image by Mosaic transformation, and (c) a synthesized image by Bezier mask generation. The red ´ lines/curves in the images represent the segmentation boundaries. Bezier curve can be uniquely determined by ´ n + 1 control points, denoted as P0, P1, · · · , Pn. The curve equation C(t) with parameter t can be calcula… view at source ↗
Figures from the paper (7 more)
Figure 5
Figure 5. Figure 5: The architecture of the discriminator, which is based on the backbone of ResNet-18. The last fully connected layer (i.e., 1 × 1 convolution) of our discriminator is modified to output a 2D vector for real and fake image classification. The kernel shapes and strides of …
Figure 6
Figure 6. Figure 6: The qualitative illustrations of the segmentation masks of different methods over the WSSS4LUAD dataset. Blue, green, and orange pixels represent tumor, stroma, and normal tissues. The segmentation masks with red boundaries in the first column are the ground truths. Th…
Figure 7
Figure 7. Figure 7: Examples of synthesized images and masks with different strate￾gies. Red, green, blue, and orange pixels represent tumor epithelial, necrosis, lymphocyte, and tumor-associated stroma tissues. TABLE VII COMPARISON OF SYNTHESIS STRATEGIES OVER THE LUAD-HISTOSEG DATASET. …
Figure 8
Figure 8. Figure 8: Segmentation results of different synthesis strategies over the LUAD-HistoSeg dataset. Arrows show mispredictions associated with the artifacts in the synthesized images. Red, green, blue, and orange pixels represent tumor epithelial, necrosis, lymphocyte, and tumor-as…
Figure 9
Figure 9. Figure 9: The comparison of real images and synthesized images with Mosaic transformation or Bezier mask generation before and after filtering ´ over WSSS4LUAD, BCSS-WSSS, and LUAD-HistoSeg datasets. TABLE XII ABLATION STUDIES USING SEPARATE SYNTHESIS STRATEGIES OVER THE LUAD-HI…
Figure 10
Figure 10. Figure 10: Visualization of predicted masks with separate synthesis strategies over the LUAD-HistoSeg dataset. The black arrows highlight the disadvantages of each strategy. Red, green, and orange pixels rep￾resent tumor epithelial, necrosis, and tumor-associated stroma tissues.…
Figure 11
Figure 11. Figure 11: The training loss curves of HisynSeg for each sub-loss over the three datasets. The solid lines represent the mean loss curves over different runs, and the shadows stand for standard deviations. TABLE XIV ABLATION STUDIES ON LOSS COMBINATIONS OVER THE LUAD-HISTOSEG DA…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

51 extracted references · 40 canonical work pages

  1. [1]

    Cancer statistics, 2023.,

    R. L. Siegel, K. D. Miller, N. S. Wagle, and A. Jemal, “Cancer statistics, 2023.,” CA: A Cancer Journal for Clinicians , vol. 73, no. 1, 2023

    work page 2023

  2. [2]

    Us food and drug administration approval of whole slide imaging for primary diagnosis: a key milestone is reached and new questions are raised,

    A. J. Evans, T. W. Bauer, M. M. Bui, T. C. Cornish, H. Duncan, E. F. Glassy, J. Hipp, R. S. McGee, D. Murphy, C. Myers, et al. , “Us food and drug administration approval of whole slide imaging for primary diagnosis: a key milestone is reached and new questions are raised,” Archives of Pathology & Laboratory Medicine, vol. 142, no. 11, pp. 1383–1387, 2018

    work page 2018

  3. [3]

    Deep learning in cancer pathology: a new generation of clinical biomarkers,

    A. Echle, N. T. Rindtorff, T. J. Brinker, T. Luedde, A. T. Pearson, and J. N. Kather, “Deep learning in cancer pathology: a new generation of clinical biomarkers,” British Journal of Cancer, vol. 124, no. 4, pp. 686– 696, 2021

    work page 2021

  4. [4]

    Geospatial immune variability illuminates differential evolution of lung adenocarcinoma,

    K. AbdulJabbar, S. E. A. Raza, R. Rosenthal, M. Jamal-Hanjani, S. Veeriah, A. Akarca, T. Lund, D. A. Moore, R. Salgado, M. Al Bakir, et al., “Geospatial immune variability illuminates differential evolution of lung adenocarcinoma,” Nature Medicine , vol. 26, no. 7, pp. 1054– 1062, 2020

    work page 2020

  5. [5]

    U-net: Convolutional net- works for biomedical image segmentation,

    O. Ronneberger, P. Fischer, and T. Brox, “U-net: Convolutional net- works for biomedical image segmentation,” in International Conference on Medical Image Computing and Computer-Assisted Intervention , pp. 234–241, Springer, 2015

    2015

  6. [6]

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

    Z. Zhou, M. M. Rahman Siddiquee, N. Tajbakhsh, and J. Liang, “Unet++: A nested u-net architecture for medical image segmenta- tion,” in Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support: 4th International Workshop, DLMIA 2018, and 8th International Workshop, ML-CDS 2018, Held in Conjunction with MICCAI 2018, Pro...

    work page 2018

  7. [7]

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

    H. Cao, Y . Wang, J. Chen, D. Jiang, X. Zhang, Q. Tian, and M. Wang, “Swin-unet: Unet-like pure transformer for medical image segmenta- tion,” in Proceedings of the European Conference on Computer Vision (ECCV), pp. 205–218, Springer, 2022

    work page 2022

  8. [8]

    Multi-layer pseudo-supervision for histopathology tissue semantic segmentation using patch-level classification labels,

    C. Han, J. Lin, J. Mai, Y . Wang, Q. Zhang, B. Zhao, X. Chen, X. Pan, Z. Shi, Z. Xu, et al., “Multi-layer pseudo-supervision for histopathology tissue semantic segmentation using patch-level classification labels,” Medical Image Analysis , p. 102487, 2022

    work page 2022

Show all 51 references
  1. [9]

    Structured crowdsourcing enables convolutional segmentation of histology images,

    M. Amgad, H. Elfandy, H. Hussein, L. A. Atteya, M. A. Elsebaie, L. S. Abo Elnasr, R. A. Sakr, H. S. Salem, A. F. Ismail, A. M. Saad, et al., “Structured crowdsourcing enables convolutional segmentation of histology images,” Bioinformatics, vol. 35, no. 18, pp. 3461–3467, 2019

    2019

  2. [10]

    Histosegnet: Semantic segmentation of histological tissue type in whole slide images,

    L. Chan, M. S. Hosseini, C. Rowsell, K. N. Plataniotis, and S. Damask- inos, “Histosegnet: Semantic segmentation of histological tissue type in whole slide images,” in Proceedings of the IEEE/CVF International Conference on Computer Vision , pp. 10662–10671, 2019. 18 IEEE TRAN...

    2019

  3. [11]

    What’s the point: Semantic segmentation with point supervision,

    A. Bearman, O. Russakovsky, V . Ferrari, and L. Fei-Fei, “What’s the point: Semantic segmentation with point supervision,” in European Conference on Computer Cision , pp. 549–565, Springer, 2016

    2016

  4. [12]

    Scribblesup: Scribble- supervised convolutional networks for semantic segmentation,

    D. Lin, J. Dai, J. Jia, K. He, and J. Sun, “Scribblesup: Scribble- supervised convolutional networks for semantic segmentation,” in Pro- ceedings of the IEEE conference on Computer Vision and Pattern Recognition, pp. 3159–3167, 2016

    2016

  5. [13]

    Learning deep features for discriminative localization,

    B. Zhou, A. Khosla, A. Lapedriza, A. Oliva, and A. Torralba, “Learning deep features for discriminative localization,” inProceedings of the IEEE conference on computer vision and pattern recognition , pp. 2921–2929, 2016

    2016

  6. [14]

    Self-supervised equivariant attention mechanism for weakly supervised semantic seg- mentation,

    Y . Wang, J. Zhang, M. Kan, S. Shan, and X. Chen, “Self-supervised equivariant attention mechanism for weakly supervised semantic seg- mentation,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pp. 12275–12284, 2020

    2020

  7. [15]

    Hamil: High-resolution activation maps and interleaved learning for weakly supervised seg- mentation of histopathological images,

    L. Zhong, G. Wang, X. Liao, and S. Zhang, “Hamil: High-resolution activation maps and interleaved learning for weakly supervised seg- mentation of histopathological images,” IEEE Transactions on Medical Imaging, vol. 42, no. 10, pp. 2912–2923, 2023

    2023

  8. [16]

    Ima- genet: A large-scale hierarchical image database,

    J. Deng, W. Dong, R. Socher, L.-J. Li, K. Li, and L. Fei-Fei, “Ima- genet: A large-scale hierarchical image database,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pp. 248–255, Ieee, 2009

    2009

  9. [17]

    Wsss4luad: Grand challenge on weakly-supervised tissue semantic segmentation for lung adenocarcinoma,

    C. Han, X. Pan, L. Yan, H. Lin, B. Li, S. Yao, S. Lv, Z. Shi, J. Mai, J. Lin, et al., “Wsss4luad: Grand challenge on weakly-supervised tissue semantic segmentation for lung adenocarcinoma,” arXiv preprint arXiv:2204.06455, 2022

    2022 arXiv

  10. [18]

    Weakly- supervised semantic segmentation for histopathology images based on dataset synthesis and feature consistency constraint,

    Z. Fang, Y . Chen, Y . Wang, Z. Wang, X. Ji, and Y . Zhang, “Weakly- supervised semantic segmentation for histopathology images based on dataset synthesis and feature consistency constraint,” in Proceedings of the AAAI Conference on Artificial Intelligence , vol. 37, pp. 606–613, 2023

    2023

  11. [19]

    Weakly-supervised semantic segmentation via sub-category exploration,

    Y .-T. Chang, Q. Wang, W.-C. Hung, R. Piramuthu, Y .-H. Tsai, and M.- H. Yang, “Weakly-supervised semantic segmentation via sub-category exploration,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pp. 8991–9000, 2020

    2020

  12. [20]

    Grad-cam: Visual explanations from deep networks via gradient-based localization,

    R. R. Selvaraju, M. Cogswell, A. Das, R. Vedantam, D. Parikh, and D. Batra, “Grad-cam: Visual explanations from deep networks via gradient-based localization,” in 2017 IEEE International Conference on Computer Vision (ICCV) , pp. 618–626, 2017

    2017

  13. [21]

    Online easy example mining for weakly-supervised gland segmentation from histology images,

    Y . Li, Y . Yu, Y . Zou, T. Xiang, and X. Li, “Online easy example mining for weakly-supervised gland segmentation from histology images,” in International Conference on Medical Image Computing and Computer- Assisted Intervention, pp. 578–587, Springer, 2022

    2022

  14. [22]

    Semi-supervised pathological image segmentation via cross distillation of multiple attentions and seg-cam consistency,

    L. Zhong, X. Luo, X. Liao, S. Zhang, and G. Wang, “Semi-supervised pathological image segmentation via cross distillation of multiple attentions and seg-cam consistency,” Pattern Recognition , vol. 152, p. 110492, 2024

    2024

  15. [23]

    Camel: A weakly supervised learning framework for histopathology image segmentation,

    G. Xu, Z. Song, Z. Sun, C. Ku, Z. Yang, C. Liu, S. Wang, J. Ma, and W. Xu, “Camel: A weakly supervised learning framework for histopathology image segmentation,” in Proceedings of the IEEE/CVF International Conference on Computer Vision , pp. 10682–10691, 2019

    2019

  16. [24]

    Constrained deep weak supervision for histopathology image segmentation,

    Z. Jia, X. Huang, I. Eric, C. Chang, and Y . Xu, “Constrained deep weak supervision for histopathology image segmentation,” IEEE Transactions on Medical Imaging , vol. 36, no. 11, pp. 2376–2388, 2017

    2017

  17. [25]

    Generative adversarial nets,

    I. J. Goodfellow, J. Pouget-Abadie, M. Mirza, B. Xu, D. Warde-Farley, S. Ozair, A. Courville, and Y . Bengio, “Generative adversarial nets,” in Proceedings of the 27th International Conference on Neural Information Processing Systems, vol. 2, p. 2672–2680, 2014

    2014

  18. [26]

    Image-to-image translation with conditional adversarial networks,

    P. Isola, J.-Y . Zhu, T. Zhou, and A. A. Efros, “Image-to-image translation with conditional adversarial networks,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition , pp. 1125– 1134, 2017

    2017

  19. [27]

    Unpaired image-to-image translation using cycle-consistent adversarial networks,

    J.-Y . Zhu, T. Park, P. Isola, and A. A. Efros, “Unpaired image-to-image translation using cycle-consistent adversarial networks,” in Proceedings of the IEEE International Conference on Computer Vision , pp. 2223– 2232, 2017

    2017

  20. [28]

    Unsupervised multi- domain progressive stain transfer guided by style encoding dictionary,

    X. Guan, Y . Wang, Y . Lin, X. Li, and Y . Zhang, “Unsupervised multi- domain progressive stain transfer guided by style encoding dictionary,” IEEE Transactions on Image Processing , vol. 33, pp. 767–779, 2024

    2024

  21. [29]

    Denoising diffusion probabilistic mod- els,

    J. Ho, A. Jain, and P. Abbeel, “Denoising diffusion probabilistic mod- els,” in Proceedings of the 34th International Conference on Neural Information Processing Systems , vol. 33, pp. 6840–6851, 2020

    2020

  22. [30]

    Ilvr: Conditioning method for denoising diffusion probabilistic models,

    J. Choi, S. Kim, Y . Jeong, Y . Gwon, and S. Yoon, “Ilvr: Conditioning method for denoising diffusion probabilistic models,” in Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV) , pp. 14347–14356, 2021

    2021

  23. [31]

    High- resolution image synthesis with latent diffusion models,

    R. Rombach, A. Blattmann, D. Lorenz, P. Esser, and B. Ommer, “High- resolution image synthesis with latent diffusion models,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recogni- tion, pp. 10684–10695, 2022

    2022

  24. [32]

    A morphology focused diffusion proba- bilistic model for synthesis of histopathology images,

    P. A. Moghadam, S. Van Dalen, K. C. Martin, J. Lennerz, S. Yip, H. Farahani, and A. Bashashati, “A morphology focused diffusion proba- bilistic model for synthesis of histopathology images,” in Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision , ...

    2000

  25. [33]

    Diffmix: Diffusion model-based data syn- thesis for nuclei segmentation and classification in imbalanced pathology image datasets,

    H.-J. Oh and W.-K. Jeong, “Diffmix: Diffusion model-based data syn- thesis for nuclei segmentation and classification in imbalanced pathology image datasets,” in International Conference on Medical Image Comput- ing and Computer-Assisted Intervention , pp. 337–345, Springer, 2023

    2023

  26. [34]

    Cutmix: Reg- ularization strategy to train strong classifiers with localizable features,

    S. Yun, D. Han, S. J. Oh, S. Chun, J. Choe, and Y . Yoo, “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

  27. [35]

    Yolov4: Op- timal speed and accuracy of object detection,

    A. Bochkovskiy, C.-Y . Wang, and H.-Y . M. Liao, “Yolov4: Op- timal speed and accuracy of object detection,” arXiv preprint arXiv:2004.10934, 2020

    2004 arXiv

  28. [36]

    Interpolation using b ´ezier curves,

    G. Elber, “Interpolation using b ´ezier curves,” in Graphics Gems III (IBM Version), pp. 133–136, Elsevier, 1992

    1992

  29. [37]

    Deep residual learning for image recognition,

    K. He, X. Zhang, S. Ren, and J. Sun, “Deep residual learning for image recognition,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 770–778, 2016

    2016

  30. [38]

    Adam: A method for stochastic optimization,

    D. P. Kingma and J. Ba, “Adam: A method for stochastic optimization,” arXiv preprint arXiv:1412.6980 , 2014

    2014 arXiv

  31. [39]

    Encoder- decoder with atrous separable convolution for semantic image segmen- tation,

    L.-C. Chen, Y . Zhu, G. Papandreou, F. Schroff, and H. Adam, “Encoder- decoder with atrous separable convolution for semantic image segmen- tation,” in Proceedings of the European Conference on Computer Vision (ECCV), pp. 801–818, 2018

    2018

  32. [40]

    Efficientnet: Rethinking model scaling for con- volutional neural networks,

    M. Tan and Q. Le, “Efficientnet: Rethinking model scaling for con- volutional neural networks,” in International Conference on Machine Learning, pp. 6105–6114, PMLR, 2019

    2019

  33. [41]

    Decoupled weight decay regularization,

    I. Loshchilov and F. Hutter, “Decoupled weight decay regularization,” in International Conference on Learning Representations , 2019

    2019

  34. [42]

    Towards a general-purpose foundation model for computational pathology,

    R. J. Chen, T. Ding, M. Y . Lu, D. F. Williamson, G. Jaume, A. H. Song, B. Chen, A. Zhang, D. Shao, M. Shaban, et al. , “Towards a general-purpose foundation model for computational pathology,” Nature Medicine, vol. 30, no. 3, pp. 850–862, 2024

    2024

  35. [43]

    A visual-language foundation model for computational pathology,

    M. Y . Lu, B. Chen, D. F. Williamson, R. J. Chen, I. Liang, T. Ding, G. Jaume, I. Odintsov, L. P. Le, G. Gerber, et al. , “A visual-language foundation model for computational pathology,” Nature Medicine , vol. 30, no. 3, pp. 863–874, 2024

    2024

  36. [44]

    Semi-supervised semantic segmen- tation with cross-consistency training,

    Y . Ouali, C. Hudelot, and M. Tami, “Semi-supervised semantic segmen- tation with cross-consistency training,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pp. 12674– 12684, 2020

    2020

  37. [45]

    Con- sistency regularisation in varying contexts and feature perturbations for semi-supervised semantic segmentation of histology images,

    R. M. S. Bashir, T. Qaiser, S. E. A. Raza, and N. M. Rajpoot, “Con- sistency regularisation in varying contexts and feature perturbations for semi-supervised semantic segmentation of histology images,” Medical Image Analysis, vol. 91, p. 102997, 2024

    2024

  38. [46]

    Semi-supervised semantic segmentation with cross pseudo supervision,

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

    2021

  39. [47]

    Revisiting weak- to-strong consistency in semi-supervised semantic segmentation,

    L. Yang, L. Qi, L. Feng, W. Zhang, and Y . Shi, “Revisiting weak- to-strong consistency in semi-supervised semantic segmentation,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 7236–7246, 2023

    2023

  40. [48]

    Fast ff-to-ffpe whole slide image translation via laplacian pyramid and contrastive learn- ing,

    L. Fan, A. Sowmya, E. Meijering, and Y . Song, “Fast ff-to-ffpe whole slide image translation via laplacian pyramid and contrastive learn- ing,” in International Conference on Medical Image Computing and Computer-Assisted Intervention, pp. 409–419, Springer, 2022

    2022

  41. [49]

    A deep- learning model for transforming the style of tissue images from cryosec- tioned to formalin-fixed and paraffin-embedded,

    K. B. Ozyoruk, S. Can, B. Darbaz, K. Bas ¸ak, D. Demir, G. I. Gokceler, G. Serin, U. P. Hacisalihoglu, E. Kurtulus ¸, M. Y . Lu, et al. , “A deep- learning model for transforming the style of tissue images from cryosec- tioned to formalin-fixed and paraffin-embedded,” Nature B...

    2022

  42. [50]

    Digestpath: A benchmark dataset with challenge review for the pathological detection and segmentation of digestive-system,

    Q. Da, X. Huang, Z. Li, Y . Zuo, C. Zhang, J. Liu, W. Chen, J. Li, D. Xu, Z. Hu, et al., “Digestpath: A benchmark dataset with challenge review for the pathological detection and segmentation of digestive-system,” Medical Image Analysis , vol. 80, p. 102485, 2022

    2022

  43. [51]

    Re-labeling imagenet: from single to multi-labels, from global to localized labels,

    S. Yun, S. J. Oh, B. Heo, D. Han, J. Choe, and S. Chun, “Re-labeling imagenet: from single to multi-labels, from global to localized labels,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 2340–2350, 2021

    2021

Pith tools

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