Pith. sign in

REVIEW 4 major objections 5 minor 85 references

MATCH: Multi-faceted Adaptive Topo-Consistency for Semi-Supervised Histopathology Segmentation

T0 review · 4 major / 5 minor · reviewed 2026-08-04 · deepseek-v4-flash

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

Pith's one-line read A self-consistency signal from perturbed predictions—not a hand-picked threshold—tells a semi-supervised histopathology segmenter which connected components are real, reducing topological errors at 10% and 20% labels.

desk verdict Solid incremental extension of TopoSemiSeg with a new matching scheme and dual-level consistency; the stability-as-correctness assumption is unvalidated but the empirical case holds together. read the letter →

arxiv 2510.01532 v2 pith:B47J64AP submitted 2025-10-02 cs.CV

Meilong Xu , Xiaoling Hu , Shahira Abousamra , Chen Li , Chao Chen This is my paper · ORCID
classification cs.CV
keywords semi-supervisedsegmentationhistopathologypersistenthomologytopologicalconsistencyMCdropouttemporalglandnuclei
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's central claim is that in semi-supervised histopathology segmentation, biologically meaningful gland and nucleus structures can be identified without ground truth by asking which connected components survive random perturbations of the model. It proposes a training signal that pushes those surviving components to be sharply defined while suppressing unstable ones, replacing the fixed persistence threshold used by earlier topological consistency methods. Across three histopathology datasets at 10% and 20% labeled data, the method reports lower topological error metrics while matching or slightly improving object-level Dice, and it produces cell counts closer to ground truth. If the paper is right, topological correctness in dense medical images can be extracted from unlabeled data through consistency alone.

What carries the argument

The load-bearing object is the persistence diagram: each dot records when a connected component of the super-level set is born and dies as the likelihood threshold sweeps down. MATCH-Pair computes a similarity S_ij = w_i w_j * IoU(M_i, M_j) * (1 - d_ij / d_max) between components from two prediction maps, where w is normalized persistence, M is the spatial mask of the component, and d is birth-point distance, then solves a one-to-one assignment. MATCH-Global applies this pairwise assignment between consecutive facets and takes connected components of the resulting graph to give each structure a stable identity across all dropout runs and snapshots. The dual-level losses L_match and L_diag th

What would settle it

Take unlabeled histopathology patches in which a known stain artifact makes the network merge two adjacent glands in every dropout realization and every snapshot pair. If the merged component is still marked 'matched' and pulled toward persistence while the correct split remains unmatched, and the Betti error does not improve, the consensus premise is falsified. A more direct test: on a small set with expert masks, compute the precision of matched 0-D components against real structures; if persistence across perturbations does not predict correspondence to real structures, the central claim fa

Watch

Extended reading notes

Core claim

The central claim is that stable topological components can be identified without labels from the model's own perturbations, and that using them as self-supervision lowers topological error. The method extracts connected components from likelihood maps via persistent homology with a super-level set filtration, gathers several maps via Monte Carlo dropout and from consecutive training snapshots, and matches components between maps using a score built from normalized persistence, spatial overlap, and birth-point distance; a global pass chains pairwise matches into identities across all maps. Matched components are pulled toward high birth probability and low death probability, unmatched compon

Load-bearing premise

The load-bearing premise is that a connected component which repeats across MC-dropout runs and consecutive training snapshots is biologically real and its unmatched counterpart is noise; if the model is stably wrong in the same way, the self-supervision reinforces the error.

Editorial extensions

If this is right

  • At 10% and 20% labeled data, topological errors (Betti Error, Betti Matching Error, DIU) drop relative to the previous threshold-based topological SSL baseline on CRAG, GlaS, and MoNuSeg, while object-level Dice stays comparable or slightly higher.
  • On MoNuSeg, the predicted total cell count is closer to the ground-truth count (7,511 vs 7,877 for the topological baseline; per-image absolute error 106.21 vs 132.36), so fixing a few topological pixels changes a biological readout.
  • The mechanism transfers to 1-D structures (road segmentation) and to multi-class nuclei segmentation, where it improves topological metrics class by class.
  • The four dropout predictions yield a pixel-wise variance map that correlates with segmentation errors (Pearson 0.73-0.77), giving uncertainty estimation as a by-product with no extra training objective.

Reading between the lines

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

  • If 'survives perturbation' is accepted as a label-free marker of real structure, the same matching-consistency recipe could transfer to other dense-object domains such as electron microscopy or aerial imagery, and to higher-dimensional topological features such as loops, at the price of the extra forward passes the paper acknowledges.
  • Because the method already produces error-correlated uncertainty maps, a natural next step is to make the consensus signal drive active-learning queries, asking a pathologist to label only the patches where matched and unmatched components conflict.
  • The deepest unresolved point is circularity: stability across perturbations is treated as truth. An external calibration—comparing matched components against human annotations on a small held-out set, or injecting known consistent artifacts to see whether they get reinforced—would show how far the assumption holds.
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 / 5 minor

Summary. The paper proposes MATCH, a semi-supervised segmentation framework for histopathology that enforces topological consistency across multiple perturbed predictions. It introduces MATCH-Pair and MATCH-Global to match 0-dimensional persistent-homology features across MC-dropout realizations and temporal training snapshots without ground truth, then uses two consistency losses (Lmatch for matched components, Ldiag for unmatched components) to push matched structures toward high persistence and unmatched structures toward zero persistence. The method is evaluated on CRAG, GlaS, and MoNuSeg at 10% and 20% labeled-data settings, reporting lower Betti Error, Betti Matching Error, and DIU than TopoSemiSeg and other SSL baselines, with comparable pixel-wise Dice. Additional ablations cover matching algorithms, loss components, hyperparameters, backbones, perturbation types, density, 1-D topology, and downstream cell counting.

Significance. If the reported gains hold, MATCH offers a practical recipe for topology-aware semi-supervised segmentation in dense histopathology, where false merges/splits are clinically important. The paper ships code, evaluates on three public datasets plus Roads and MoNuSAC, includes extensive ablations, reports an uncertainty analysis, and includes a downstream counting experiment. These are genuine strengths. The central idea—using multi-facet stability to identify reliable topological structure without hand-picked thresholds—is plausible and the GT-based evaluation prevents the consistency losses from being purely circular. However, several load-bearing validation gaps remain, especially concerning statistical rigor and whether stability across perturbations is a valid proxy for biological correctness.

major comments (4)
  1. [§4.2, Table 7] The density-aware claim is not supported by the presented comparison. The Sparse row reports only Ours (≤30 cells), while the Crowded rows compare TopoSemiSeg and Ours on ≥100 cells. There is no TopoSemiSeg baseline on the same sparse subset, so the statement that MATCH 'excels in sparse fields' cannot be concluded. Please report both methods on identical sparse/crowded test subsets (and on the same number of patches) with significance testing.
  2. [Table 1 caption] The caption states 'statistically significant best results' but no significance tests or number of runs are reported anywhere. With values such as CRAG 10% BE 0.197±0.012 vs 0.227±0.014, the reader cannot tell whether the gains are significant. Report the number of independent runs and paired significance tests (or confidence intervals) for the main comparisons, or soften the 'statistically significant' wording.
  3. [§3.3, Table 13] The matched/unmatched dichotomy is load-bearing: the method assumes that components stable across MC-dropout and temporal snapshots are biologically meaningful, while unmatched components are noise. The losses then reinforce matched components (P_birth→1, P_death→0) and suppress unmatched ones. If the student is stably but systematically wrong (e.g., consistent oversegmentation or a stable false merge), those errors will be classified as 'matched' and reinforced. Table 13 directly shows that the source of self-consistency matters: switching from student-only to teacher–student consistency degrades Dice_Obj from 0.909 to 0.885. Please add an oracle or precision/recall analysis of matched components against ground-truth components, or an experiment with a deliberately biased student, to bound this risk and support the generalization claim.
  4. [§3.1, Table 2] The matching algorithm is a core contribution, but the comparison omits the closest spatial-aware matching baseline, Wen et al. [63]. Figure 2 and Table 2 compare only Wasserstein and Betti Matching. Including [63] (or explaining why it cannot be run) is important to substantiate the claim that MATCH-Pair achieves accurate matching without ground truth.
minor comments (5)
  1. [§10 (Supplementary)] Bintra and Btemp are not explicitly stated in the implementation details; Table 4 implies the optimal value is 4, but the text should state the default values used in the main experiments.
  2. [Figure 2] The claim that MATCH-Pair achieves 'almost completely accurate matching' is based on a visual example. Quantifying matching accuracy against ground-truth correspondences would strengthen the claim.
  3. [Supplementary Table 18] The references in the cell-counting table appear inconsistent: PMT is cited as [8] and TopoSemiSeg as [57], but in the main reference list [8] is Fleming et al. and [57] is Tarvainen and Valpola. Please correct the citation numbering.
  4. [§12 (Supplementary)] Typo: 'varing' should be 'varying' in the first sentence of the λ_intra/λ_temp ablation.
  5. [References] Reference [6] contains 'F oundations' and reference [42] contains 'V olodymyr' with stray spaces; please fix.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: the main reported gains are judged by ground-truth topological metrics, and the self-citations are not load-bearing.

full rationale

The paper's central empirical claim—lower BE/BME/DIU on CRAG, GlaS, and MoNuSeg—is evaluated with ground-truth-based topological metrics on held-out test sets (Table 1, Sec. 4.1), so the improvements are not produced by construction from the loss definitions. The dual-level losses (Lintra, Ltemp; Sec. 3.3) directly encourage matched components to become persistent and unmatched components to become less persistent, but this only shows that the method enforces its own consistency objective; it does not by itself establish the externally measured topological fidelity, which is the actual claim. The load-bearing assumption that persistence across MC-dropout and temporal snapshots corresponds to biological correctness is unvalidated (no oracle or precision/recall analysis of matched components), and Table 13 shows stability across a chosen perturbation set does not guarantee correctness. That is a correctness/limitation concern, not a circular derivation; the Limitations section (Sec. 13) and negative broader-impact statement (Sec. 14) acknowledge the risk of propagating inaccuracies from poorly matched structures. Self-citations (TopoSemiSeg [69], temporal-view works [35,36]) are used as baselines or design inspiration rather than as load-bearing mathematical justifications; no uniqueness theorem or fitted parameter is renamed as a prediction. The matching algorithm is a new heuristic with external evaluation against Wasserstein and Betti matching. Overall, no circular step of the kind defined by the rubric is present, and the paper is self-contained enough for its empirical claims.

Assumptions & free parameters 8 free parameters · 6 assumptions · 0 invented entities

The method introduces no new physical or architectural entities, but relies on several hand-set hyperparameters and two domain assumptions: (i) persistent features in likelihood maps correspond to objects, and (ii) consensus across the student's own perturbed predictions identifies truth. The second is the main epistemic risk.

free parameters (8)
  • τ_primary = 0.1
    Matching-score threshold; hand-set and ablated in Table 15; low value chosen to be inclusive.
  • B_intra = 4
    Number of MC-dropout facets; tuned via sensitivity Table 4; 4 chosen as optimal.
  • B_temp = 4
    Number of temporal snapshots; tuned via Table 4.
  • λ_intra = 0.001
    Weight of intra-topological consistency loss; tuned in Table 10.
  • λ_temp = 0.001
    Weight of temporal-topological consistency loss; tuned in Table 10.
  • dropout_rate = 0.2
    MC-dropout rate; chosen via Table 16; 20-30% plateau.
  • EMA α = 0.999
    Teacher EMA decay; tuned via Table 11.
  • λ_cons scale k = 0.1
    Gaussian ramp-up scale for pixel-wise consistency; chosen without a dedicated ablation.
assumptions (6)
  • domain assumption 0-D superlevel persistent homology features of a likelihood map correspond to biological objects (glands/nuclei) and their merge/split events.
    Used throughout Sections 3.1-3.3; no proof that PH components align with instances beyond the application context.
  • ad hoc to paper Components that match across MC-dropout and temporal snapshots are reliable; unmatched components are noise.
    Central inductive bias, Sec. 1 and Fig. 1; no ground-truth validation; can fail for stable systematic errors.
  • ad hoc to paper The Hungarian assignment with similarity S_ij = w1w2·IoU·(1-d/d_max) recovers correct feature correspondences between noisy persistence diagrams.
    Heuristic cost in Sec. 3.1-3.2; no correctness guarantee; evaluation is only indirect via segmentation metrics.
  • domain assumption Predictions from consecutive training snapshots provide a useful perturbation view for SSL.
    Borrowed from temporal ensembling/PMT; assumed in Sec. 3.3; the storage/use of Btemp snapshots is not fully specified.
  • standard math Standard persistent homology stability and discrete flood-fill region extraction hold for the discretized likelihood maps.
    Background TDA (Sec. 7); no custom mathematics introduced.
  • domain assumption The mean-teacher EMA framework provides stable pseudo-targets for SSL.
    Standard SSL assumption from [57], used in Preliminaries; not re-derived.

how reviews work

0 comments
Cite this review

Pith. "Pith review of MATCH: Multi-faceted Adaptive Topo-Consistency for Semi-Supervised Histopathology Segmentation." pith.science (2026). https://pith.science/paper/B47J64AP

@misc{pith2026251001532,
  author       = {Pith},
  title        = {Pith review of: MATCH: Multi-faceted Adaptive Topo-Consistency for Semi-Supervised Histopathology Segmentation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/B47J64AP}},
  note         = {Machine review of arXiv:2510.01532}
}
read the original abstract

In semi-supervised segmentation, capturing meaningful semantic structures from unlabeled data is essential. This is particularly challenging in histopathology image analysis, where objects are densely distributed. To address this issue, we propose a semi-supervised segmentation framework designed to robustly identify and preserve relevant topological features. Our method leverages multiple perturbed predictions obtained through stochastic dropouts and temporal training snapshots, enforcing topological consistency across these varied outputs. This consistency mechanism helps distinguish biologically meaningful structures from transient and noisy artifacts. A key challenge in this process is to accurately match the corresponding topological features across the predictions in the absence of ground truth. To overcome this, we introduce a novel matching strategy that integrates spatial overlap with global structural alignment, minimizing discrepancies among predictions. Extensive experiments demonstrate that our approach effectively reduces topological errors, resulting in more robust and accurate segmentations essential for reliable downstream analysis. Code is available at https://github.com/Melon-Xu/MATCH.

Figures

Figures reproduced from arXiv: 2510.01532 by the authors.

Figure 1
Figure 1. Intuition of the proposed framework. (a) Colored likelihood maps are coming from the MC [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Comparison of our matching with Betti Matching [ [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Overview of the proposed MATCH framework with dual-level topological consistency. [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: Pipeline of the MATCH-Pair algorithm between two persistence diagrams. [PITH_FULL_IMAGE:figures/full_fig_p005_4.png]
Figure 5
Figure 5. Figure 5: Qualitative illustration of MC dropout predictions (after the model convergence). [PITH_FULL_IMAGE:figures/full_fig_p008_5.png]
Figure 6
Figure 6. Figure 6: Qualitative results for semi-supervised methods on [PITH_FULL_IMAGE:figures/full_fig_p009_6.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

85 extracted references · 4 linked inside Pith

  1. [69]

    Semi-supervised segmentation of histopathology images with noise-aware topological consistency

    Meilong Xu, Xiaoling Hu, Saumya Gupta, Shahira Abousamra, and Chao Chen. Semi-supervised segmentation of histopathology images with noise-aware topological consistency. InECCV, 2024

    2024

  2. [63]

    Topology-preserving image segmentation with spatial-aware persistent feature matching.arXiv preprint arXiv:2412.02076, 2024

    Bo Wen, Haochen Zhang, Dirk-Uwe G Bartsch, William R Freeman, Truong Q Nguyen, and Cheolhong An. Topology-preserving image segmentation with spatial-aware persistent feature matching.arXiv preprint arXiv:2412.02076, 2024

    arXiv 2024

  3. [1]

    Pseudo-label guided contrastive learning for semi-supervised medical image segmentation

    Hritam Basak and Zhaozheng Yin. Pseudo-label guided contrastive learning for semi-supervised medical image segmentation. InCVPR, 2023

    2023

  4. [2]

    Topologically faithful multi-class segmentation in medical images

    Alexander H Berger, Laurin Lux, Nico Stucki, Vincent Bürgin, Suprosanna Shit, Anna Banaszak, Daniel Rueckert, Ulrich Bauer, and Johannes C Paetzold. Topologically faithful multi-class segmentation in medical images. InMICCAI, 2024

    2024

  5. [3]

    Mixmatch: A holistic approach to semi-supervised learning

    David Berthelot, Nicholas Carlini, Ian Goodfellow, Nicolas Papernot, Avital Oliver, and Colin A Raffel. Mixmatch: A holistic approach to semi-supervised learning. InNeurIPS, 2019

    2019

  6. [4]

    Encoder-decoder with atrous separable convolution for semantic image segmentation

    Liang-Chieh Chen, Yukun Zhu, George Papandreou, Florian Schroff, and Hartwig Adam. Encoder-decoder with atrous separable convolution for semantic image segmentation. InECCV, 2018

    2018

  7. [5]

    A topological loss function for deep-learning based image segmentation using persistent homology

    James R Clough, Nicholas Byrne, Ilkay Oksuz, Veronika A Zimmer, Julia A Schnabel, and Andrew P King. A topological loss function for deep-learning based image segmentation using persistent homology. TPAMI, 2020

    2020

  8. [6]

    Lipschitz functions have l p-stable persistence.F oundations of Computational Mathematics, 2010

    David Cohen-Steiner, Herbert Edelsbrunner, John Harer, and Yuriy Mileyko. Lipschitz functions have l p-stable persistence.F oundations of Computational Mathematics, 2010

    2010

Show all 85 references
  1. [7]

    American Mathematical Soc., 2010

    Herbert Edelsbrunner and John Harer.Computational topology: an introduction. American Mathematical Soc., 2010

    2010

  2. [8]

    Colorectal carcinoma: Pathologic aspects.Journal of gastrointestinal oncology, 2012

    Matthew Fleming, Sreelakshmi Ravula, Sergei F Tatishchev, and Hanlin L Wang. Colorectal carcinoma: Pathologic aspects.Journal of gastrointestinal oncology, 2012

    2012

  3. [9]

    Dropout as a bayesian approximation: Representing model uncertainty in deep learning

    Yarin Gal and Zoubin Ghahramani. Dropout as a bayesian approximation: Representing model uncertainty in deep learning. InICML, 2016

    2016

  4. [10]

    Pmt: Progressive mean teacher via exploring temporal consistency for semi-supervised medical image segmentation

    Ning Gao, Sanping Zhou, Le Wang, and Nanning Zheng. Pmt: Progressive mean teacher via exploring temporal consistency for semi-supervised medical image segmentation. InECCV, 2024

    2024

  5. [11]

    Mild-net: Minimal information loss dilated network for gland instance segmentation in colon histology images.MedIA, 2019

    Simon Graham, Hao Chen, Jevgenij Gamper, Qi Dou, Pheng-Ann Heng, David Snead, Yee Wah Tsang, and Nasir Rajpoot. Mild-net: Minimal information loss dilated network for gland instance segmentation in colon histology images.MedIA, 2019

    2019

  6. [12]

    Hover-net: Simultaneous segmentation and classification of nuclei in multi-tissue histology images.MedIA, 2019

    Simon Graham, Quoc Dang Vu, Shan E Ahmed Raza, Ayesha Azam, Yee Wah Tsang, Jin Tae Kwak, and Nasir Rajpoot. Hover-net: Simultaneous segmentation and classification of nuclei in multi-tissue histology images.MedIA, 2019

    2019

  7. [13]

    Semi-supervised learning by entropy minimization

    Yves Grandvalet and Yoshua Bengio. Semi-supervised learning by entropy minimization. InNeurIPS, 2004

    2004

  8. [14]

    On calibration of modern neural networks

    Chuan Guo, Geoff Pleiss, Yu Sun, and Kilian Q Weinberger. On calibration of modern neural networks. In ICML, 2017

    2017

  9. [15]

    Learning topological interactions for multi-class medical image segmentation

    Saumya Gupta, Xiaoling Hu, James Kaan, Michael Jin, Mutshipay Mpoy, Katherine Chung, Gagandeep Singh, Mary Saltz, Tahsin Kurc, Joel Saltz, et al. Learning topological interactions for multi-class medical image segmentation. InECCV, 2022

    2022

  10. [16]

    Topology-aware uncertainty for image segmentation

    Saumya Gupta, Yikai Zhang, Xiaoling Hu, Prateek Prasanna, and Chao Chen. Topology-aware uncertainty for image segmentation. InNeurIPS, 2023

    2023

  11. [17]

    Toposeg: Topology-aware nuclear instance segmentation

    Hongliang He, Jun Wang, Pengxu Wei, Fan Xu, Xiangyang Ji, Chang Liu, and Jie Chen. Toposeg: Topology-aware nuclear instance segmentation. InICCV, 2023

    2023

  12. [18]

    Cellvit: Vision transformers for precise cell segmentation and classification.MedIA, 2024

    Fabian Hörst, Moritz Rempe, Lukas Heine, Constantin Seibold, Julius Keyl, Giulia Baldini, Selma Ugurel, Jens Siveke, Barbara Grünwald, Jan Egger, et al. Cellvit: Vision transformers for precise cell segmentation and classification.MedIA, 2024

    2024

  13. [19]

    Lora: Low-rank adaptation of large language models

    Edward J Hu, Yelong Shen, Phillip Wallis, Zeyuan Allen-Zhu, Yuanzhi Li, Shean Wang, Lu Wang, Weizhu Chen, et al. Lora: Low-rank adaptation of large language models. InICLR, 2022

    2022

  14. [20]

    Structure-aware image segmentation with homotopy warping

    Xiaoling Hu. Structure-aware image segmentation with homotopy warping. InNeurIPS, 2022. 11

    2022

  15. [21]

    Topology-preserving deep image segmentation

    Xiaoling Hu, Fuxin Li, Dimitris Samaras, and Chao Chen. Topology-preserving deep image segmentation. InNeurIPS, 2019

    2019

  16. [22]

    Learning probabilistic topological representations using discrete morse theory

    Xiaoling Hu, Dimitris Samaras, and Chao Chen. Learning probabilistic topological representations using discrete morse theory. InICLR, 2023

    2023

  17. [23]

    Topology-aware segmentation using discrete morse theory

    Xiaoling Hu, Yusu Wang, Li Fuxin, Dimitris Samaras, and Chao Chen. Topology-aware segmentation using discrete morse theory. InICLR, 2021

    2021

  18. [24]

    Adversarial learning for semi-supervised semantic segmentation.arXiv preprint arXiv:1802.07934, 2018

    Wei-Chih Hung, Yi-Hsuan Tsai, Yan-Ting Liou, Yen-Yu Lin, and Ming-Hsuan Yang. Adversarial learning for semi-supervised semantic segmentation.arXiv preprint arXiv:1802.07934, 2018

    2018 arXiv

  19. [25]

    nnu-net: a self-configuring method for deep learning-based biomedical image segmentation.Nature methods, 2021

    Fabian Isensee, Paul F Jaeger, Simon AA Kohl, Jens Petersen, and Klaus H Maier-Hein. nnu-net: a self-configuring method for deep learning-based biomedical image segmentation.Nature methods, 2021

    2021

  20. [26]

    An introduction to variational methods for graphical models.Machine learning, 1999

    Michael I Jordan, Zoubin Ghahramani, Tommi S Jaakkola, and Lawrence K Saul. An introduction to variational methods for graphical models.Machine learning, 1999

    1999

  21. [27]

    Evolutionary characterization of lung adenocarcinoma morphology in tracerx.Nature medicine, 2023

    Takahiro Karasaki, David A Moore, Selvaraju Veeriah, Cristina Naceur-Lombardelli, Antonia Toncheva, Neil Magno, Sophia Ward, Maise Al Bakir, Thomas BK Watkins, Kristiana Grigoriadis, et al. Evolutionary characterization of lung adenocarcinoma morphology in tracerx.Nature medic...

    2023

  22. [28]

    Adam: A method for stochastic optimization.arXiv preprint arXiv:1412.6980, 2014

    Diederik P Kingma. Adam: A method for stochastic optimization.arXiv preprint arXiv:1412.6980, 2014

    2014 arXiv

  23. [29]

    Segment anything

    Alexander Kirillov, Eric Mintun, Nikhila Ravi, Hanzi Mao, Chloe Rolland, Laura Gustafson, Tete Xiao, Spencer Whitehead, Alexander C Berg, Wan-Yen Lo, et al. Segment anything. InICCV, 2023

    2023

  24. [30]

    Enhancing sam with efficient prompting and preference optimization for semi-supervised medical image segmentation

    Aishik Konwer, Zhijian Yang, Erhan Bas, Cao Xiao, Prateek Prasanna, Parminder Bhatia, and Taha Kass-Hout. Enhancing sam with efficient prompting and preference optimization for semi-supervised medical image segmentation. InCVPR, 2025

    2025

  25. [31]

    The hungarian method for the assignment problem.Naval research logistics quarterly, 1955

    Harold W Kuhn. The hungarian method for the assignment problem.Naval research logistics quarterly, 1955

    1955

  26. [32]

    A multi-organ nucleus segmentation challenge

    Neeraj Kumar, Ruchika Verma, Deepak Anand, Yanning Zhou, Omer Fahri Onder, Efstratios Tsougenis, Hao Chen, Pheng-Ann Heng, Jiahui Li, Zhiqiang Hu, et al. A multi-organ nucleus segmentation challenge. TMI, 2019

    2019

  27. [33]

    Temporal ensembling for semi-supervised learning.arXiv preprint arXiv:1610.02242, 2016

    Samuli Laine and Timo Aila. Temporal ensembling for semi-supervised learning.arXiv preprint arXiv:1610.02242, 2016

    2016 arXiv

  28. [34]

    Semi-supervised medical image segmentation using adversarial consistency learning and dynamic convolution network

    Tao Lei, Dong Zhang, Xiaogang Du, Xuan Wang, Yong Wan, and Asoke K Nandi. Semi-supervised medical image segmentation using adversarial consistency learning and dynamic convolution network. TMI, 2022

    2022

  29. [35]

    Calibrating uncertainty for semi-supervised crowd counting

    Chen Li, Xiaoling Hu, Shahira Abousamra, and Chao Chen. Calibrating uncertainty for semi-supervised crowd counting. InICCV, 2023

    2023

  30. [36]

    Confidence estimation using unlabeled data

    Chen Li, Xiaoling Hu, and Chao Chen. Confidence estimation using unlabeled data. InICLR, 2023

    2023

  31. [37]

    Transformation- consistent self-ensembling model for semi-supervised medical image segmentation.TNNLS, 2020

    Xiaomeng Li, Lequan Yu, Hao Chen, Chi-Wing Fu, Lei Xing, and Pheng-Ann Heng. Transformation- consistent self-ensembling model for semi-supervised medical image segmentation.TNNLS, 2020

    2020

  32. [38]

    Semi-supervised medical image segmentation through dual-task consistency

    Xiangde Luo, Jieneng Chen, Tao Song, and Guotai Wang. Semi-supervised medical image segmentation through dual-task consistency. InAAAI, 2021

    2021

  33. [39]

    Semi-supervised medical image segmentation via uncertainty rectified pyramid consistency.MedIA, 2022

    Xiangde Luo, Guotai Wang, Wenjun Liao, Jieneng Chen, Tao Song, Yinan Chen, Shichuan Zhang, Dimitris N Metaxas, and Shaoting Zhang. Semi-supervised medical image segmentation via uncertainty rectified pyramid consistency.MedIA, 2022

    2022

  34. [40]

    Topograph: An efficient graph-based framework for strictly topology preserving image segmentation

    Laurin Lux, Alexander H Berger, Alexander Weers, Nico Stucki, Daniel Rueckert, Ulrich Bauer, and Johannes C Paetzold. Topograph: An efficient graph-based framework for strictly topology preserving image segmentation. InICLR, 2025

    2025

  35. [41]

    Segment anything in medical images

    Jun Ma, Yuting He, Feifei Li, Lin Han, Chenyu You, and Bo Wang. Segment anything in medical images. Nature Communications, 2024

    2024

  36. [42]

    University of Toronto (Canada), 2013

    V olodymyr Mnih.Machine learning for aerial image labeling. University of Toronto (Canada), 2013. 12

    2013

  37. [43]

    Rodolfo Montironi, Roberta Mazzuccheli, Marina Scarpelli, Antonio Lopez-Beltran, Giovanni Fellegara, and Ferran Algaba. Gleason grading of prostate cancer in needle biopsies or radical prostatectomy specimens: contemporary approach, current clinical significance and sources of...

    2005

  38. [44]

    Exploring uncertainty measures in deep networks for multiple sclerosis lesion detection and segmentation.MedIA, 2020

    Tanya Nair, Doina Precup, Douglas L Arnold, and Tal Arbel. Exploring uncertainty measures in deep networks for multiple sclerosis lesion detection and segmentation.MedIA, 2020

    2020

  39. [45]

    Semi-supervised histopathology image segmentation with feature diversified collaborative learning

    Thanh-Huy Nguyen, Nguyen Lan Vi Vu, Hoang-Thien Nguyen, Quang-Vinh Dinh, Xingjian Li, and Min Xu. Semi-supervised histopathology image segmentation with feature diversified collaborative learning. In AAAI Bridge Program on AI for Medicine and Healthcare, 2025

    2025

  40. [46]

    Semi-supervised semantic segmentation with cross- consistency training

    Yassine Ouali, Céline Hudelot, and Myriam Tami. Semi-supervised semantic segmentation with cross- consistency training. InCVPR, 2020

    2020

  41. [47]

    Pytorch: An imperative style, high-performance deep learning library.arXiv preprint arXiv:1912.01703, 2019

    A Paszke. Pytorch: An imperative style, high-performance deep learning library.arXiv preprint arXiv:1912.01703, 2019

    1912 arXiv

  42. [48]

    Karl Pearson. Vii. note on regression and inheritance in the case of two parents.proceedings of the royal society of London, 1895

  43. [49]

    U-net: Convolutional networks for biomedical image segmentation

    Olaf Ronneberger, Philipp Fischer, and Thomas Brox. U-net: Convolutional networks for biomedical image segmentation. InMICCAI, 2015

    2015

  44. [50]

    Reference-guided pseudo- label generation for medical semantic segmentation

    Constantin Marc Seibold, Simon Reiß, Jens Kleesiek, and Rainer Stiefelhagen. Reference-guided pseudo- label generation for medical semantic segmentation. InAAAI, 2022

    2022

  45. [51]

    Revisiting and maximizing temporal knowledge in semi-supervised semantic segmentation.arXiv preprint arXiv:2405.20610, 2024

    Wooseok Shin, Hyun Joon Park, Jin Sob Kim, and Sung Won Han. Revisiting and maximizing temporal knowledge in semi-supervised semantic segmentation.arXiv preprint arXiv:2405.20610, 2024

    2024

  46. [52]

    cldice-a novel topology-preserving loss function for tubular structure segmentation

    Suprosanna Shit, Johannes C Paetzold, Anjany Sekuboyina, Ivan Ezhov, Alexander Unger, Andrey Zhylka, Josien PW Pluim, Ulrich Bauer, and Bjoern H Menze. cldice-a novel topology-preserving loss function for tubular structure segmentation. InCVPR, 2021

    2021

  47. [53]

    Gland segmentation in colon histology images: The glas challenge contest.MedIA, 2017

    Korsuk Sirinukunwattana, Josien PW Pluim, Hao Chen, Xiaojuan Qi, Pheng-Ann Heng, Yun Bo Guo, Li Yang Wang, Bogdan J Matuszewski, Elia Bruni, Urko Sanchez, et al. Gland segmentation in colon histology images: The glas challenge contest.MedIA, 2017

    2017

  48. [54]

    Tint fill

    Alvy Ray Smith. Tint fill. InProceedings of the 6th annual conference on Computer graphics and interactive techniques, 1979

    1979

  49. [55]

    Fixmatch: Simplifying semi-supervised learning with consistency and confidence

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

    2020

  50. [56]

    Topologically faithful image segmentation via induced matching of persistence barcodes

    Nico Stucki, Johannes C Paetzold, Suprosanna Shit, Bjoern Menze, and Ulrich Bauer. Topologically faithful image segmentation via induced matching of persistence barcodes. InICML, 2023

    2023

  51. [57]

    Mean teachers are better role models: Weight-averaged consistency targets improve semi-supervised deep learning results

    Antti Tarvainen and Harri Valpola. Mean teachers are better role models: Weight-averaged consistency targets improve semi-supervised deep learning results. InNeurIPS, 2017

    2017

  52. [58]

    Monusac2020: A multi-organ nuclei segmentation and classification challenge.TMI, 2021

    Ruchika Verma, Neeraj Kumar, Abhijeet Patil, Nikhil Cherian Kurian, Swapnil Rane, Simon Graham, Quoc Dang Vu, Mieke Zwager, Shan E Ahmed Raza, Nasir Rajpoot, et al. Monusac2020: A multi-organ nuclei segmentation and classification challenge.TMI, 2021

    2021

  53. [59]

    Advent: Adversarial entropy minimization for domain adaptation in semantic segmentation

    Tuan-Hung Vu, Himalaya Jain, Maxime Bucher, Matthieu Cord, and Patrick Pérez. Advent: Adversarial entropy minimization for domain adaptation in semantic segmentation. InCVPR, 2019

    2019

  54. [60]

    Topogan: A topology-aware generative adversarial network

    Fan Wang, Huidong Liu, Dimitris Samaras, and Chao Chen. Topogan: A topology-aware generative adversarial network. InECCV, 2020

    2020

  55. [61]

    Ta-net: Topology-aware network for gland segmenta- tion

    Haotian Wang, Min Xian, and Aleksandar Vakanski. Ta-net: Topology-aware network for gland segmenta- tion. InWACV, 2022

    2022

  56. [62]

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

    Kaiping Wang, Bo Zhan, Chen Zu, Xi Wu, Jiliu Zhou, Luping Zhou, and Yan Wang. Semi-supervised medical image segmentation via a tripled-uncertainty guided mean teacher model with contrastive learning. MedIA, 2022. 13

    2022

  57. [64]

    Cross-patch dense contrastive learning for semi-supervised segmentation of cellular nuclei in histopathologic images

    Huisi Wu, Zhaoze Wang, Youyi Song, Lin Yang, and Jing Qin. Cross-patch dense contrastive learning for semi-supervised segmentation of cellular nuclei in histopathologic images. InCVPR, 2022

    2022

  58. [65]

    Otoo, and Kenji Suzuki

    Kesheng Wu, Ekow J. Otoo, and Kenji Suzuki. Optimizing two-pass connected-component labeling algorithms.Pattern Analysis and Applications, 2009

    2009

  59. [66]

    Entropy-guided contrastive learning for semi-supervised medical image segmentation.IET Image Processing, 2024

    Junsong Xie, Qian Wu, and Renju Zhu. Entropy-guided contrastive learning for semi-supervised medical image segmentation.IET Image Processing, 2024

    2024

  60. [67]

    Deep segmentation-emendation model for gland instance segmentation

    Yutong Xie, Hao Lu, Jianpeng Zhang, Chunhua Shen, and Yong Xia. Deep segmentation-emendation model for gland instance segmentation. InMICCAI, 2019

    2019

  61. [68]

    Topocellgen: Generating histopathology cell topology with a diffusion model

    Meilong Xu, Saumya Gupta, Xiaoling Hu, Chen Li, Shahira Abousamra, Dimitris Samaras, Prateek Prasanna, and Chao Chen. Topocellgen: Generating histopathology cell topology with a diffusion model. InCVPR, 2025

    2025

  62. [70]

    Ambiguity-selective consistency regularization for mean-teacher semi-supervised medical image segmentation.MedIA, 2023

    Zhe Xu, Yixin Wang, Donghuan Lu, Xiangde Luo, Jiangpeng Yan, Yefeng Zheng, and Raymond Kai-yu Tong. Ambiguity-selective consistency regularization for mean-teacher semi-supervised medical image segmentation.MedIA, 2023

    2023

  63. [71]

    3d topology-preserving segmentation with compound multi-slice representation

    Jiaqi Yang, Xiaoling Hu, Chao Chen, and Chialing Tsai. 3d topology-preserving segmentation with compound multi-slice representation. InISBI, 2021

    2021

  64. [72]

    A topological-attention convlstm network and its application to em images

    Jiaqi Yang, Xiaoling Hu, Chao Chen, and Chialing Tsai. A topological-attention convlstm network and its application to em images. InMICCAI, 2021

    2021

  65. [73]

    Anomaly-guided weakly supervised lesion segmentation on retinal oct images

    Jiaqi Yang, Nitish Mehta, Gozde Demirci, Xiaoling Hu, Meera S Ramakrishnan, Mina Naguib, Chao Chen, and Chia-Ling Tsai. Anomaly-guided weakly supervised lesion segmentation on retinal oct images. MedIA, 2024

    2024

  66. [74]

    A multimodal approach combining structural and cross-domain textual guidance for weakly supervised oct segmentation.arXiv preprint arXiv:2411.12615, 2024

    Jiaqi Yang, Nitish Mehta, Xiaoling Hu, Chao Chen, and Chia-Ling Tsai. A multimodal approach combining structural and cross-domain textual guidance for weakly supervised oct segmentation.arXiv preprint arXiv:2411.12615, 2024

    2024

  67. [75]

    Enhancing pseudo label quality for semi-supervised domain-generalized medical image segmentation

    Huifeng Yao, Xiaowei Hu, and Xiaomeng Li. Enhancing pseudo label quality for semi-supervised domain-generalized medical image segmentation. InAAAI, 2022

    2022

  68. [76]

    Mine your own anatomy: Revisiting medical image segmentation with extremely limited labels.TPAMI, 2024

    Chenyu You, Weicheng Dai, Fenglin Liu, Yifei Min, Nicha C Dvornek, Xiaoxiao Li, David A Clifton, Lawrence Staib, and James S Duncan. Mine your own anatomy: Revisiting medical image segmentation with extremely limited labels.TPAMI, 2024

    2024

  69. [77]

    Rethinking semi-supervised medical image segmentation: A variance-reduction perspective

    Chenyu You, Weicheng Dai, Yifei Min, Fenglin Liu, David Clifton, S Kevin Zhou, Lawrence Staib, and James Duncan. Rethinking semi-supervised medical image segmentation: A variance-reduction perspective. InNeurIPS, 2023

    2023

  70. [78]

    Simcvd: Simple contrastive voxel-wise representation distillation for semi-supervised medical image segmentation.TMI, 2022

    Chenyu You, Yuan Zhou, Ruihan Zhao, Lawrence Staib, and James S Duncan. Simcvd: Simple contrastive voxel-wise representation distillation for semi-supervised medical image segmentation.TMI, 2022

    2022

  71. [79]

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

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

    2019

  72. [80]

    Topology-preserving hard pixel mining for tubular structure segmentation

    Guoqing Zhang, Caixia Dong, and Yang Li. Topology-preserving hard pixel mining for tubular structure segmentation. InBMVC, 2023

    2023

  73. [81]

    Boostmis: Boosting medical image semi-supervised learning with adaptive pseudo labeling and informative active annotation

    Wenqiao Zhang, Lei Zhu, James Hallinan, Shengyu Zhang, Andrew Makmur, Qingpeng Cai, and Beng Chin Ooi. Boostmis: Boosting medical image semi-supervised learning with adaptive pseudo labeling and informative active annotation. InCVPR, 2022

    2022

  74. [82]

    Discriminative error prediction network for semi-supervised colon gland segmentation.MedIA, 2022

    Zhenxi Zhang, Chunna Tian, Harrison X Bai, Zhicheng Jiao, and Xilan Tian. Discriminative error prediction network for semi-supervised colon gland segmentation.MedIA, 2022. 14

    2022

  75. [83]

    Xnet: Wavelet-based low and high frequency fusion networks for fully-and semi-supervised semantic segmentation of biomedical images

    Yanfeng Zhou, Jiaxing Huang, Chenlong Wang, Le Song, and Ge Yang. Xnet: Wavelet-based low and high frequency fusion networks for fully-and semi-supervised semantic segmentation of biomedical images. In ICCV, 2023

    2023

  76. [84]

    Xnet v2: Fewer limitations, better results and greater universality

    Yanfeng Zhou, Lingrui Li, Zichen Wang, Guole Liu, Ziwen Liu, and Ge Yang. Xnet v2: Fewer limitations, better results and greater universality. InBIBM, 2024

    2024

  77. [85]

    Unet++: A nested u-net architecture for medical image segmentation

    Zongwei Zhou, Md Mahfuzur Rahman Siddiquee, Nima Tajbakhsh, and Jianming Liang. Unet++: A nested u-net architecture for medical image segmentation. InDeep learning in medical image analysis and multimodal learning for clinical decision support: 4th international workshop, DLMI...

    2018

Pith tools

Reviewed August 4, 2026 · model on record in the stance chip above.