Pith. sign in

REVIEW 4 major objections 4 minor 50 references

Staining normalization in histopathology: Method benchmarking using multicenter dataset

T0 review · 4 major / 4 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read A benchmark of H&E slides from 66 labs finds plain histogram matching outperforms GAN-based stain normalization.

desk verdict The 66-lab dataset is the real contribution; the benchmark is honest but its headline ranking is partly metric-aligned and should be treated as provisional. read the letter →

arxiv 2506.19106 v1 pith:ZZQ5ICL4 submitted 2025-06-23 eess.IV cs.CVq-bio.TO

Umair Khan , Jouni Härkönen , Marjukka Friman , Leena Latonen , Teijo Kuopio , Pekka Ruusuvuori This is my paper · ORCID
classification eess.IVcs.CVq-bio.TO
keywords H&Estainingstainnormalizationmulti-centerdatasetwholeslideimageshistogrammatchingCycleGANPix2pixcolorvariation
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 builds a stain-normalization benchmark from an unusual resource: sections cut from the same colon, kidney, and skin blocks and stained in 66 different laboratories across 11 countries, so that appearance differences are almost entirely due to staining. Comparing eight normalization methods on this dataset, it reports that histogram matching—a generic, non-pathology-specific technique—delivers the best quantitative color-transfer scores on all three tissue types (for example, kidney Pearson correlation 0.985), while GAN-based methods produce reasonable but not superior results and sometimes add artifacts. If the result holds, the dataset becomes a standard testbed for stain normalization, and the ranking tells multi-center pathology workflows that a simple global statistical method may be enough when tissue morphology is uniform.

What carries the argument

The central object is the dataset: whole-slide images in which 3-micron sections from the same three tissue blocks (colon, kidney, skin) were sent to 66 laboratories for routine H&E staining, then scanned and resampled to 10× magnification. The evaluation anchors on a single reference slide per tissue, selected automatically as the one whose red-to-blue mean intensity ratio is closest to one; all normalized images are compared against that reference in the lαβ color space using histogram intersection, Pearson correlation, Euclidean distance, Jensen-Shannon divergence, Fréchet Inception Distance, and SSIM. This design isolates staining variation from biological and technical factors, and it allows whole-slide normalization without tiling artifacts.

What would settle it

A blinded study in which pathologists rank the normalized images on diagnostic interpretability and a method other than histogram matching wins, or a re-run of the benchmark with a pathologist-selected reference in which the ordering changes, would falsify the paper's ranking claim.

Watch

Extended reading notes

Core claim

On a dataset that isolates inter-laboratory H&E staining variation by staining sections from identical tissue blocks in 66 laboratories, histogram matching achieves the best quantitative performance on histogram-based color-similarity metrics for skin, kidney, and colon: intersection scores of 0.891, 0.944, and 0.906, and Pearson correlation coefficients of 0.938, 0.985, and 0.935, respectively. It also gives the best Fréchet Inception Distance on skin and kidney, while Reinhard normalization is best on colon by that metric. Vahadane preserves structural similarity best (SSIM up to 0.995) but qualitatively discards the hematoxylin blue-purple component, turning sections pink. The GAN-based methods—two CycleGAN and two Pix2pix variants—do not conclusively outperform the traditional methods and produce visible hallucinations in some cases, such as context-dependent tile artifacts at tissue borders and red-transformed spindly nuclei. The authors argue that the controlled, morphologically uniform dataset and whole-slide (rather than patch-wise) application favor global statistical methods, and that one slide per laboratory is too little data for the GANs to reach their potential.

Load-bearing premise

The ranking is computed against a single reference slide per tissue, chosen by a computational rule (red-to-blue mean intensity ratio closest to one); if pathologists would pick a different target, the ranking could shift.

Editorial extensions

If this is right

  • The released dataset gives the community a public, 66-laboratory testbed for stain normalization and for grounding stain augmentation in real observed variation.
  • Global statistical methods like histogram matching may be sufficient for stain normalization in settings with uniform tissue morphology, reducing the need for GAN training.
  • GAN-based normalization should be evaluated for artifacts and may require more than one slide per laboratory to train effectively; the paper reports that a single WSI per lab was used.
  • Histogram matching makes multi-center images converge toward the reference in red/blue ratio, as shown by the post-normalization scatter plot, supporting its use for harmonizing cohorts before AI analysis.
  • Because sections come from the same blocks, the benchmark's method rankings are directly attributable to normalization performance rather than biological variation.

Reading between the lines

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

  • If the ranking carries over to downstream tasks, one would expect histogram matching to be a strong default preprocessing step for multi-center deep-learning pipelines; the paper does not test downstream tasks, so this remains an inference.
  • A reference-free or multi-reference evaluation (for example, measuring downstream segmentation accuracy or averaging over several pathologist-selected references) might produce a different ordering, since all scores are relative to one chosen slide.
  • The dataset could be used to quantify how much stain augmentation should move to cover real inter-laboratory variation; the paper suggests this use but does not demonstrate it.
  • GAN performance might improve substantially if more than one slide per laboratory were available, as the authors note; this is a testable corollary of their explanation for GAN underperformance.
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 introduces a multicenter H&E-stained tissue image dataset in which sections from the same skin, kidney, and colon blocks were stained in 66 different laboratories and scanned on a single scanner, isolating staining variation from other technical factors. Using this dataset, the authors benchmark eight stain normalization methods: four traditional methods (histogram matching, Macenko, Vahadane, Reinhard) and four GAN-based variants (two CycleGAN and two Pix2pix) evaluated with histogram-based color metrics, FID, SSIM, and a qualitative assessment on colon tissue. The central claims are that the dataset captures an unprecedented range of inter-laboratory staining variation, that histogram matching achieves the best quantitative color-fidelity scores across all three tissue types, and that GAN-based methods do not conclusively outperform traditional methods and sometimes produce artifacts.

Significance. The dataset itself is a valuable and potentially standard-setting resource for stain normalization benchmarking: the same-block, single-scanner design is careful, the scale (66 laboratories, 11 countries) is unusual, and the data are released with open-access DOIs and code. The paper also provides a useful side-by-side comparison of traditional and modern normalization methods under a controlled setup. If the ranking conclusions were robust, the work would guide method selection for multicenter pathology workflows. However, the headline result that histogram matching is 'best' is partly built into the evaluation metrics (which reward histogram transfer), is metric-dependent (as the FID results show), and is not supported by significance testing; the GAN comparison is also confounded by strongly unequal training data. These issues materially weaken the benchmarking conclusions, though they are addressable.

major comments (4)
  1. [Quantitative Evaluation; Tables 1-3 and Table 4] The four primary metrics (histogram intersection, PCC, Euclidean distance, JS divergence) are computed on channel histograms in lαβ space, which is a linear reparameterization of RGB. Histogram matching (skimage.match_histograms) is explicitly designed to transform the source histogram toward the reference histogram in RGB, so the top ranking of histogram matching on these metrics is at least partly a consequence of the evaluation measuring exactly what the method optimizes. The paper's own FID results (Table 4) demonstrate that the ranking is metric-dependent: for colon, Reinhard (92.37) and both CycleGAN variants (96.12, 96.50) beat histogram matching (99.92), and for skin, Reinhard (62.08) is close to histogram matching (61.67). The claim that histogram matching is 'the most efficient' or 'best' should be restricted to the specific histogram-similarity metrics used, and the circularity should be discussed explicitly. A non-circular validation, such as a downstream task (e.g., segmentation or classification) or a blinded pathologist preference study, would substantially strengthen the claim.
  2. [Quantitative Evaluation; Tables 1-3] No statistical significance tests are reported for any of the quantitative comparisons. Tables 1-3 list means and standard deviations only. For example, in skin tissue the histogram-matching intersection is 0.891 ± 0.034 versus 0.871 ± 0.042 for CycleGAN-ResNet with n=66; these distributions overlap considerably, and the margin may be within sampling noise. Paired tests across the 66 slides (e.g., Wilcoxon signed-rank or paired bootstrap on the per-slide scores) should be added, and the conclusions should be adjusted to state which differences are actually significant. Without this, the quantitative ranking is not statistically established.
  3. [Reference Sample Selection; Materials and Methods] For each tissue type, a single reference WSI is selected by the heuristic criterion that its red-to-blue mean intensity ratio be closest to one. This choice is not validated as representative of high-quality diagnostic H&E; a pathologist-selected reference could easily differ. Because every quantitative score in Tables 1-4 is computed relative to this one reference per tissue type, the entire method ranking depends on an unvalidated selection. The authors should either validate the computational reference against expert judgment (e.g., a small panel of pathologists rating candidate references) or show that the ranking of methods is stable under alternative reference choices, for instance by repeating the comparison with several plausible references.
  4. [GAN-based Methods Experiment Setup; Discussion] The comparison between traditional and GAN-based methods is confounded by strongly unequal training resources. CycleGAN was trained on a representative subset of 8 WSIs per tissue type, and both Pix2pix variants were trained on grayscale-RGB tile pairs from a single reference WSI. Traditional methods, in contrast, used the full WSI as the reference. The paper itself acknowledges in the Discussion that 'one WSI per laboratory is insufficient' for the deep-learning models, but this is presented as an explanation for GAN performance rather than as a limitation of the benchmark. As a result, the conclusion that GAN-based methods 'could not conclusively outperform traditional methods' is not well supported: the test is not a fair head-to-head at comparable data availability. The authors should reframe this conclusion as an observation about the current experimental setup, or include a GAN variant trained on the full dataset (or a more balanced subset), so that the method comparison is more equitable.
minor comments (4)
  1. [Abstract] The abstract contains the typo 'Pixp2pix' for Pix2pix; this should be corrected.
  2. [Quantitative Evaluation; Table 4 caption] The caption spells 'Fréchet' as 'Frachet'; please use the correct spelling throughout.
  3. [Introduction; References 40-41] The text repeatedly refers to 'De Bal et al.' but the references are by 'de Bel et al.'; the name should be spelled consistently with the cited works.
  4. [Table 5 caption] The caption says SSIM scores are 'compared against the original WSIs' but does not specify whether the reference is the original input image or the reference WSI; please clarify, because the interpretation of Vahadane's high SSIM (0.995 for skin) changes substantially depending on which comparison is meant.

Circularity Check

1 steps flagged · score 3.0 of 10

Quantitative ranking is partly circular: histogram matching is scored on histogram-similarity metrics it is designed to satisfy, so the benchmark does not independently establish it as the best normalization method.

  1. self definitional [Quantitative Evaluation (Methods); Histogram Matching (Methods); Tables 1-3]
    "First, to assess the performance on color transfer task, the normalized images were transformed to the lαβ color space and then the normalized histogram of each channel was compared with the reference image histograms using intersection, Pearson correlation coefficient (PCC), euclidean distance, and Jensen-Shannon (JS) divergence. ... It is an image processing technique that modifies pixel values in an image to transform its color distributions represented by its histogram so that it matches the color distribution of a reference image."

    The four headline quantitative metrics (intersection, PCC, Euclidean distance, JS divergence) all compare the normalized image's channel histograms against the reference histograms. The method that wins those metrics, histogram matching, is defined as the operation that makes the source histogram match the reference histogram. Thus its top ranking on those metrics is partly a restatement of how the algorithm is constructed, not an independent empirical finding. The paper's own FID results weaken the strong claim: for colon, Reinhard (92.37) and both CycleGAN variants (96.12, 96.50) beat histogram matching (99.92), and SSIM shows Vahadane best preserves structure. The dataset itself is external and no fitted parameters are called predictions, so the circularity is partial rather than total.

full rationale

This is a benchmarking paper whose primary contribution is an external, multicenter H&E dataset; the method implementations are standard off-the-shelf algorithms and the comparison is self-contained against that dataset. No fitted parameter is renamed as a prediction, and no load-bearing argument reduces to a self-citation. The one substantive circularity concern is metric-method alignment: histogram matching is explicitly designed to transform the source histogram toward the reference histogram, while the four primary evaluation metrics measure histogram similarity to the reference in lαβ space. Consequently the headline ranking of histogram matching as best on those metrics is in part guaranteed by construction rather than established by independent evidence. The paper partially mitigates this with FID, SSIM, and qualitative evaluation, and it honestly reports that on colon FID, histogram matching is not the best method. The reference-sample selection is heuristic but not circular. Overall, the central dataset contribution is not circular, but the quantitative conclusion about method superiority is only partially independent, warranting a moderate score of 3.

Assumptions & free parameters 1 free parameters · 3 assumptions · 0 invented entities

The central benchmark result rests on the reference selection rule, the histogram-based evaluation metrics, and the assumption that the same-block sections isolate staining variation. No new entities are proposed, and the main fitted choice is the GAN training subset size.

free parameters (1)
  • number of clusters for CycleGAN training subset = 8
    Selected with the within-cluster sum of squares approach and then fixed to 8 uniformly across tissue types; this choice determines which WSIs are used to train CycleGAN and therefore affects GAN performance.
assumptions (3)
  • domain assumption The sample with red-to-blue mean intensity ratio closest to 1 is a suitable reference for H&E stain normalization.
    Used to pick the target for all methods and metrics; no independent validation that this reference is diagnostically optimal.
  • domain assumption Histogram similarity in lαβ space is a valid proxy for stain normalization quality.
    Primary quantitative metrics reward exact histogram matching, which biases the ranking towards histogram matching.
  • domain assumption Sections from the same tissue block across 66 labs differ only in staining, not in tissue content.
    Enables isolation of stain variation; reasonable given the TMA punch design, but section thickness and sampling can still vary.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Staining normalization in histopathology: Method benchmarking using multicenter dataset." pith.science (2026). https://pith.science/paper/ZZQ5ICL4

@misc{pith2026250619106,
  author       = {Pith},
  title        = {Pith review of: Staining normalization in histopathology: Method benchmarking using multicenter dataset},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ZZQ5ICL4}},
  note         = {Machine review of arXiv:2506.19106}
}
read the original abstract

Hematoxylin and Eosin (H&E) has been the gold standard in tissue analysis for decades, however, tissue specimens stained in different laboratories vary, often significantly, in appearance. This variation poses a challenge for both pathologists' and AI-based downstream analysis. Minimizing stain variation computationally is an active area of research. To further investigate this problem, we collected a unique multi-center tissue image dataset, wherein tissue samples from colon, kidney, and skin tissue blocks were distributed to 66 different labs for routine H&E staining. To isolate staining variation, other factors affecting the tissue appearance were kept constant. Further, we used this tissue image dataset to compare the performance of eight different stain normalization methods, including four traditional methods, namely, histogram matching, Macenko, Vahadane, and Reinhard normalization, and two deep learning-based methods namely CycleGAN and Pixp2pix, both with two variants each. We used both quantitative and qualitative evaluation to assess the performance of these methods. The dataset's inter-laboratory staining variation could also guide strategies to improve model generalizability through varied training data

Figures

Figures reproduced from arXiv: 2506.19106 by the authors.

Figure 1
Figure 1. A collage of tissue patches representing variance in appearance. a. Image patches extracted from the dermis layer of skin tissue sections. b. Image patches extracted from kidney tissue sections. c. Image patches extracted from the smooth muscle layer of colon tissue sections*. 4/18 [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Pre-normalization red and blue channel mean intensity plot. The plot shows the spread of color variation based on the ratio of red and blue channel mean intensities of the tissue images. All three tissues, i.e., skin, kidney and colon WSIs deviate drastically from the color-balanced (dotted) line [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Skin tissue stain normalization. Stain-normalized tissue images from eight different methods, i.e., histogram matching, Macenko, Vahadane, Reinhard, CycleGAN (ResNet), CycleGAN (UNet), Pix2pix (UNet), Pixpix (DenseUNet) along with input and reference WSI. 5/18 [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (6 more)
Figure 4
Figure 4. Figure 4: Kidney tissue stain normalization. Stain-normalized tissue images from eight different methods, i.e., histogram matching, Macenko, Vahadane, Reinhard, CycleGAN (ResNet), CycleGAN (UNet), Pix2pix (UNet), Pixpix (DenseUNet) along with input and reference WSI. 6/18 [PITH…
Figure 5
Figure 5. Figure 5: Colon tissue stain normalization. Stain-normalized tissue images from eight different methods, i.e., histogram matching, Macenko, Vahadane, Reinhard, CycleGAN (ResNet), CycleGAN (UNet), Pix2pix (UNet), Pixpix (DenseUNet) along with input and reference WSI. 7/18 [PITH_…
Figure 6
Figure 6. Figure 6: Colon tissue qualitative analysis. a. Representative images of optimal as well as over- and understained H&E-slides. b. Comparisons between original, Histmatch and Reinhard normalized images from colonic mucosa-submucosa border in optimally and suboptimally stained sli…
Figure 7
Figure 7. Figure 7: Overview of the best performing methods (columns) with varying input staining quality (rows). type, representing a setup where a relatively high level of content similarity can be expected across the 66 sections. Tissue staining is one of the most critical steps in sam…
Figure 8
Figure 8. Figure 8: Post-normalization tissue patch collage showing the color uniformity. Tissue images stain normalized using histogram normalization. a. Image patches extracted from skin tissue sections*. b. Image patches extracted from kidney tissue sections. c. Image patches extracted…
Figure 9
Figure 9. Figure 9: Post-normalization red and blue channel mean intensity plot. The plot shows the convergence of red and blue channel mean intensity ratio of the tissue images after histogram matching stain normalization. All three tissue types, i.e., skin, kidney and colon WSIs are vir…

Review history (2 revisions) →

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

50 extracted references · 43 canonical work pages

  1. [1]

    W., Ciovacco, W

    Troiano, N. W., Ciovacco, W. A. & Kacena, M. A. The effects of fixation and dehydration on the histological quality of undecalcified murine bone specimens embedded in methylmethacrylate.J. histotechnology32, 27–31 (2009)

    work page 2009

  2. [2]

    A.et al.Impact of preanalytical factors during histology processing on section suitability for digital image analysis.Toxicol

    Chlipala, E. A.et al.Impact of preanalytical factors during histology processing on section suitability for digital image analysis.Toxicol. Pathol.49, 755–772 (2021)

    work page 2021

  3. [3]

    Chan, J. K. The wonderful colors of the hematoxylin–eosin stain in diagnostic surgical pathology.Int. journal surgical pathology22, 12–32 (2014). 4.Bancroft, J. D. & Gamble, M.Theory and practice of histological techniques(Elsevier health sciences, 2008)

    work page 2014

  4. [5]

    S.et al.Overview of telepathology, virtual microscopy, and whole slide imaging: prospects for the future

    Weinstein, R. S.et al.Overview of telepathology, virtual microscopy, and whole slide imaging: prospects for the future. Hum. pathology40, 1057–1069 (2009)

    work page 2009

  5. [6]

    Physical Color Calibration of Digital Pathology Scanners for Robust Artificial Intelligence Assisted Cancer Diagnosis

    Ji, X.et al.Physical color calibration of digital pathology scanners for robust artificial intelligence assisted cancer diagnosis. arXiv preprint arXiv:2307.05519(2023)

    work page Pith review arXiv 2023

  6. [7]

    Valkonen, M., Högnäs, G., Bova, G. S. & Ruusuvuori, P. Generalized fixation invariant nuclei detection through domain adaptation based deep learning.IEEE J. Biomed. Heal. Informatics25, 1747–1757 (2020)

    work page 2020

  7. [8]

    A.et al.Current and future applications of artificial intelligence in pathology: a clinical perspective.J

    Rakha, E. A.et al.Current and future applications of artificial intelligence in pathology: a clinical perspective.J. clinical pathology74, 409–414 (2021)

    work page 2021

  8. [9]

    Moscalu, M.et al.Histopathological images analysis and predictive modeling implemented in digital pathology—current affairs and perspectives.Diagnostics13, 2379 (2023)

    work page 2023

Show all 50 references
  1. [10]

    L.et al.Access to pathology and laboratory medicine services: a crucial gap.The Lancet391, 1927–1938 (2018)

    Wilson, M. L.et al.Access to pathology and laboratory medicine services: a crucial gap.The Lancet391, 1927–1938 (2018)

    2018

  2. [11]

    Ström, P.et al.Artificial intelligence for diagnosis and grading of prostate cancer in biopsies: a population-based, diagnostic study.The Lancet Oncol.21, 222–232 (2020)

    2020

  3. [12]

    C.et al.Deep learning assisted mitotic counting for breast cancer.Lab

    Balkenhol, M. C.et al.Deep learning assisted mitotic counting for breast cancer.Lab. investigation99, 1596–1606 (2019)

    2019

  4. [13]

    Zadeh Shirazi, A.et al.A deep convolutional neural network for segmentation of whole-slide pathology images identifies novel tumour cell-perivascular niche interactions that are associated with poor survival in glioblastoma.Br. J. Cancer125, 337–350 (2021)

    2021

  5. [14]

    D.et al.Methods for segmentation and classification of digital microscopy tissue images.Front

    Vu, Q. D.et al.Methods for segmentation and classification of digital microscopy tissue images.Front. bioengineering biotechnology53 (2019)

    2019

  6. [15]

    Wulczyn, E.et al.Deep learning-based survival prediction for multiple cancer types using histopathology images.PloS one15, e0233678 (2020)

    2020

  7. [16]

    Hepatology72, 2000–2013 (2020)

    Saillard, C.et al.Predicting survival after hepatocellular carcinoma resection using deep learning on histological slides. Hepatology72, 2000–2013 (2020)

    2020

  8. [17]

    cancer1, 800–810 (2020)

    Fu, Y .et al.Pan-cancer computational histopathology reveals mutations, tumor composition and prognosis.Nat. cancer1, 800–810 (2020). 18.Ahmed, A. A., Abouzid, M. & Kaczmarek, E. Deep learning approaches in histopathology.Cancers14, 5264 (2022). 19.Paige AI. Paige ai - transfo...

    2020

  9. [20]

    Pathol.19, 42 (2024)

    Dunn, C.et al.Quantitative assessment of h&e staining for pathology: development and clinical evaluation of a novel system.Diagn. Pathol.19, 42 (2024)

    2024

  10. [21]

    & Orsi, N

    Breen, J., Zucker, K., Allen, K., Ravikumar, N. & Orsi, N. M. Generative adversarial networks for stain normalisation in histopathology. InApplications of Generative AI, 227–247 (Springer, 2024)

    2024

  11. [22]

    & Doyle, S

    Therrien, R. & Doyle, S. Role of training data variability on classifier performance and generalizability. InMedical Imaging 2018: Digital Pathology, vol. 10581, 58–70 (SPIE, 2018)

    2018

  12. [23]

    image analysis58, 101544 (2019)

    Tellez, D.et al.Quantifying the effects of data augmentation and stain color normalization in convolutional neural networks for computational pathology.Med. image analysis58, 101544 (2019)

    2019

  13. [24]

    25.Goodfellow, I.et al.Generative adversarial nets.Adv

    Salvi, M.et al.Impact of stain normalization on pathologist assessment of prostate cancer: A comparative study.Cancers 15, 1503 (2023). 25.Goodfellow, I.et al.Generative adversarial nets.Adv. neural information processing systems27(2014). 16/18

    2023

  14. [26]

    & Shirley, P

    Reinhard, E., Adhikhmin, M., Gooch, B. & Shirley, P. Color transfer between images.IEEE Comput. graphics applications 21, 34–41 (2001)

    2001

  15. [27]

    In2009 IEEE international symposium on biomedical imaging: from nano to macro, 1107–1110 (IEEE, 2009)

    Macenko, M.et al.A method for normalizing histology slides for quantitative analysis. In2009 IEEE international symposium on biomedical imaging: from nano to macro, 1107–1110 (IEEE, 2009)

    2009

  16. [28]

    29.Gonzalez, R

    Vahadane, A.et al.Structure-preserving color normalization and sparse stain separation for histological images.IEEE transactions on medical imaging35, 1962–1971 (2016). 29.Gonzalez, R. C. & Woods, R. E. Digital image processing, prentice hall.Up. Saddle River, NJ(2008)

    2016

  17. [30]

    InProc Optical Tissue Image analysis in Microscopy, Histopathology and Endoscopy (MICCAI Workshop), vol

    Magee, D.et al.Colour normalisation in digital histopathology images. InProc Optical Tissue Image analysis in Microscopy, Histopathology and Endoscopy (MICCAI Workshop), vol. 100, 100–111 (Daniel Elson London, 2009)

    2009

  18. [31]

    & Thomas, N

    Niethammer, M., Borland, D., Marron, J., Woosley, J. & Thomas, N. E. Appearance normalization of histology slides. In Machine Learning in Medical Imaging: First International Workshop, MLMI 2010, Held in Conjunction with MICCAI 2010, Beijing, China, September 20, 2010. Proceed...

    2010

  19. [32]

    Fuchs, T. J. & Buhmann, J. M. Computational pathology: challenges and promises for tissue analysis.Comput. Med. Imaging Graph.35, 515–530 (2011)

    2011

  20. [33]

    & Efros, A

    Zhu, J.-Y ., Park, T., Isola, P. & Efros, A. A. Unpaired image-to-image translation using cycle-consistent adversarial networks. InProceedings of the IEEE international conference on computer vision, 2223–2232 (2017)

    2017

  21. [34]

    J., Jackson, C

    Levy, J. J., Jackson, C. R., Sriharan, A., Christensen, B. C. & Vaickus, L. J. Preliminary evaluation of the utility of deep generative histopathology image translation at a mid-sized nci cancer center.bioRxiv2020–01 (2020)

    2020

  22. [35]

    M., Boor, P

    Gadermayr, M., Appel, V ., Klinkhammer, B. M., Boor, P. & Merhof, D. Which way round? a study on the performance of stain-translation for segmenting arbitrarily dyed histological images. InMedical Image Computing and Computer Assisted Intervention–MICCAI 2018: 21st Internation...

    2018

  23. [36]

    & Latonen, L

    Koivukoski, S., Khan, U., Ruusuvuori, P. & Latonen, L. Unstained tissue imaging and virtual hematoxylin and eosin staining of histologic whole slide images.Lab. Investig.103, 100070 (2023)

    2023

  24. [37]

    & Ruusuvuori, P

    Khan, U., Koivukoski, S., Valkonen, M., Latonen, L. & Ruusuvuori, P. The effect of neural network architecture on virtual h&e staining: Systematic assessment of histological feasibility.Patterns4(2023)

    2023

  25. [38]

    & Bueno, G

    Salido, J., Vallez, N., González-López, L., Deniz, O. & Bueno, G. Comparison of deep learning models for digital h&e staining from unpaired label-free multispectral microscopy images.Comput. Methods Programs Biomed.235, 107528 (2023)

    2023

  26. [39]

    & Brox, T

    Ronneberger, O., Fischer, P. & Brox, T. U-net: Convolutional networks for biomedical image segmentation. InMedical Image Computing and Computer-Assisted Intervention–MICCAI 2015: 18th International Conference, Munich, Germany, October 5-9, 2015, Proceedings, Part III 18, 234–2...

    2015

  27. [40]

    & Litjens, G

    de Bel, T., Hermsen, M., Kers, J., van der Laak, J. & Litjens, G. Stain-transforming cycle-consistent generative adversarial networks for improved segmentation of renal histopathology. InInternational Conference on Medical Imaging with Deep Learning, 151–163 (PMLR, 2019)

    2019

  28. [41]

    & Litjens, G

    de Bel, T., Bokhorst, J.-M., van der Laak, J. & Litjens, G. Residual cyclegan for robust domain transformation of histopathological tissue slides.Med. Image Analysis70, 102004 (2021)

    2021

  29. [42]

    & Efros, A

    Isola, P., Zhu, J.-Y ., Zhou, T. & Efros, A. A. Image-to-image translation with conditional adversarial networks. In Proceedings of the IEEE conference on computer vision and pattern recognition, 1125–1134 (2017)

    2017

  30. [43]

    & Chalechale, A

    Salehi, P. & Chalechale, A. Pix2pix-based stain-to-stain translation: A solution for robust stain normalization in histopathology images analysis. In2020 International conference on machine vision and image processing (MVIP), 1–7 (IEEE, 2020)

    2020

  31. [44]

    T., Baur, C., Navab, N

    Shaban, M. T., Baur, C., Navab, N. & Albarqouni, S. Staingan: Stain style transfer for digital histological images. In2019 Ieee 16th international symposium on biomedical imaging (Isbi 2019), 953–956 (IEEE, 2019)

    2019

  32. [45]

    & Min, H

    Cho, H., Lim, S., Choi, G. & Min, H. Neural stain-style transfer learning using gan for histopathological images. arxiv 2017.arXiv preprint arXiv:1710.08543(2017])

    2017 arXiv

  33. [46]

    & Hamarneh, G

    BenTaieb, A. & Hamarneh, G. Adversarial stain transfer for histopathology image analysis.IEEE transactions on medical imaging37, 792–802 (2017). 17/18 47.Cong, C.et al.Semi-supervised adversarial learning for stain normalisation in histopathology images. InMedical Image Comput...

    2017

  34. [48]

    & Sun, J

    He, K., Zhang, X., Ren, S. & Sun, J. Deep residual learning for image recognition. InProceedings of the IEEE conference on computer vision and pattern recognition, 770–778 (2016)

    2016

  35. [49]

    C.et al.Image technology colour management-architecture, profile format, and data structure.Specif

    Consortium, I. C.et al.Image technology colour management-architecture, profile format, and data structure.Specif. ICC. 1: 2004-10 (Profile version 4.2. 0.0)(2004)

    2004

  36. [50]

    & Wojna, Z

    Szegedy, C., Vanhoucke, V ., Ioffe, S., Shlens, J. & Wojna, Z. Rethinking the inception architecture for computer vision. In Proceedings of the IEEE conference on computer vision and pattern recognition, 2818–2826 (2016)

    2016

  37. [51]

    & Hochreiter, S

    Heusel, M., Ramsauer, H., Unterthiner, T., Nessler, B. & Hochreiter, S. Gans trained by a two time-scale update rule converge to a local nash equilibrium.Adv. neural information processing systems30(2017)

    2017

  38. [52]

    & Zhang, D

    Zhang, L., Zhang, L., Mou, X. & Zhang, D. Fsim: A feature similarity index for image quality assessment.IEEE transactions on Image Process.20, 2378–2386 (2011)

    2011

  39. [53]

    & Kuopio, T

    Prezja, F., Pölönen, I., Äyrämö, S., Ruusuvuori, P. & Kuopio, T. H&e multi-laboratory staining variance exploration with machine learning.Appl. Sci.12, 7511 (2022)

    2022

  40. [54]

    T., Tessema, A

    Zewdie, E. T., Tessema, A. W. & Simegn, G. L. Classification of breast cancer types, sub-types and grade from histopathological images using deep learning technique.Heal. Technol.11, 1277–1290 (2021)

    2021

  41. [55]

    W., Zewde, E

    Habtemariam, L. W., Zewde, E. T. & Simegn, G. L. Cervix type and cervical cancer classification system using deep learning techniques.Med. devices: evidence research163–176 (2022)

    2022

  42. [56]

    InProceedings of the fifth Berkeley symposium on mathematical statistics and probability, 14, 281–297 (Oakland, CA, USA, 1967)

    MacQueen, J.et al.Some methods for classification and analysis of multivariate observations. InProceedings of the fifth Berkeley symposium on mathematical statistics and probability, 14, 281–297 (Oakland, CA, USA, 1967). 18/18

    1967

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.