Pix2Pix-based Stain-to-Stain Translation: A Solution for Robust Stain Normalization in Histopathology Images Analysis

Salehi, Pegah; Chalechale, Abdolah

doi:10.48550/arxiv.2002.00647

Cited by 3 publications

(15 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Shaban et al [35] introduced CycleGANs [37] to realize the style transfer between images from two different scanners. Salehi et al [36] designed an unsupervised model based on conditional generative adversarial RestainNet networks (cGANs) [23] by coloring grayscale images to the target stain style. A multi-task network equipped with normalized branches and task-specific branches [21] is proposed to improve the generalization ability of target tasks on multiple data sets.…”

Section: B Learning-based Normalizationmentioning

confidence: 99%

“…Table I demonstrates the quantitative comparisons with two computational-based stain normalization methods [25], [27] and two GAN-based models, StainGAN [35] and STST [36].…”

Section: B Comparisons With Existing Approachesmentioning

confidence: 99%

“…The quantitative results of the segmentation under different stain normalization methods are shown in Table II. [25] 0.856 0.873 0.872 0.824 (3) Macenko [30] 0.861 0.876 0.878 0.833 (4) Vahadane [27] 0.856 0.872 0.874 0.840 (5) StainGAN [35] 0.872 0.880 0.859 0.852 (6) STST [36] 0.859 0.866 0.851 0.853 ( 7 We can easily observe that all the stain normalization methods can reduce the stain inconsistency of the data, thereby help improve the segmentation task, shown in (2) to (7). ( 4) uses the non-negative matrix factorization method to obtain the dye concentration matrix to generate a more detailed normalized image, so obtains a better segmentation effect than (2) and (3).…”

Section: Stain Normalization In the Segmentation Taskmentioning

confidence: 99%

“…On the contrary, (4) also consumes a longer time in stain normalization. In this task, the segmentation [25] 0.903 0.930 0.906 (3) Macenko [30] 0.895 0.923 0.896 (4) StainGAN [35] 0.905 0.931 0.911 (5) STST [36] 0.879 0.912 0.886 (6) Ours 0.910 0.935 0.919 model maps pixels into differentiated representations based on the significant color and morphological differences between tissues to achieve the purpose of gland segmentation. Therefore, to ensure the efficiency of the segmentation model, the normalized image should have a broader color distribution to maintain the distinguishability of the histological tissue clusters.…”

Section: Stain Normalization In the Segmentation Taskmentioning

confidence: 99%

“…Moreover, optimization-based approaches are easily trapped in local optimum [21]. Recently, GANbased models [35], [36] have been proposed and achieved promising results. Due to the data-driven nature, they require a large amount of paired training samples, which are hard to collect.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

RestainNet: a self-supervised digital re-stainer for stain normalization

Zhao¹,

Lin²,

Chen³

et al. 2022

Preprint

View full text Add to dashboard Cite

Color inconsistency is an inevitable challenge in computational pathology, which generally happens because of stain intensity variations or sections scanned by different scanners. It harms the pathological image analysis methods, especially the learning-based models. A series of approaches have been proposed for stain normalization. However, most of them are lack of flexibility in practice. In this paper, we formulated stain normalization as a digital re-staining process and proposed a self-supervised learning model, which is called RestainNet. Our network is regarded as a digital re-stainer which learns how to re-stain an unstained (grayscale) image. Two digital stains, Hematoxylin (H) and Eosin (E), were extracted from the original image by Beer-Lambert's Law. We proposed a staining loss to maintain the correctness of stain intensity during the restaining process. Thanks to the self-supervised nature, paired training samples are no longer necessary, which demonstrates great flexibility in practical usage. Our RestainNet outperforms existing approaches and achieves state-of-the-art performance with regard to color correctness and structure preservation. We further conducted experiments on the segmentation and classification tasks and the proposed RestainNet achieved outstanding performance compared with SOTA methods. The self-supervised design allows the network learn any staining style with no extra effort.

show abstract

Section: B Learning-based Normalizationmentioning

confidence: 99%

“…Table I demonstrates the quantitative comparisons with two computational-based stain normalization methods [25], [27] and two GAN-based models, StainGAN [35] and STST [36].…”

Section: B Comparisons With Existing Approachesmentioning

confidence: 99%

Section: Stain Normalization In the Segmentation Taskmentioning

confidence: 99%

Section: Stain Normalization In the Segmentation Taskmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

RestainNet: a self-supervised digital re-stainer for stain normalization

Zhao¹,

Lin²,

Chen³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

Assessment of Color Reproducibility and Mitigation of Color Variation in Whole Slide Image Scanners for Toxicologic Pathology

Chu,

Ge,

Eastham

et al. 2023

Toxicol Pathol

View full text Add to dashboard Cite

Digital pathology workflows in toxicologic pathology rely on whole slide images (WSIs) from histopathology slides. Inconsistent color reproduction by WSI scanners of different models and from different manufacturers can result in different color representations and inter-scanner color variation in the WSIs. Although pathologists can accommodate a range of color variation during their evaluation of WSIs, color variability can degrade the performance of computational applications in digital pathology. In particular, color variability can compromise the generalization of artificial intelligence applications to large volumes of data from diverse sources. To address these challenges, we developed a process that includes two modules: (1) assessing the color reproducibility of our scanners and the color variation among them and (2) applying color correction to WSIs to minimize the color deviation and variation. Our process ensures consistent color reproduction across WSI scanners and enhances color homogeneity in WSIs, and its flexibility enables easy integration as a post-processing step following scanning by WSI scanners of different models and from different manufacturers.

show abstract

A New Method of Artificial-Intelligence-Based Automatic Identification of Lymphovascular Invasion in Urothelial Carcinomas

Ceachi,

Cioplea,

Mustatea

et al. 2024

Diagnostics

View full text Add to dashboard Cite

The presence of lymphovascular invasion (LVI) in urothelial carcinoma (UC) is a poor prognostic finding. This is difficult to identify on routine hematoxylin–eosin (H&E)-stained slides, but considering the costs and time required for examination, immunohistochemical stains for the endothelium are not the recommended diagnostic protocol. We developed an AI-based automated method for LVI identification on H&E-stained slides. We selected two separate groups of UC patients with transurethral resection specimens. Group A had 105 patients (100 with UC; 5 with cystitis); group B had 55 patients (all with high-grade UC; D2-40 and CD34 immunohistochemical stains performed on each block). All the group A slides and 52 H&E cases from group B showing LVI using immunohistochemistry were scanned using an Aperio GT450 automatic scanner. We performed a pixel-per-pixel semantic segmentation of selected areas, and we trained InternImage to identify several classes. The DiceCoefficient and Intersection-over-Union scores for LVI detection using our method were 0.77 and 0.52, respectively. The pathologists’ H&E-based evaluation in group B revealed 89.65% specificity, 42.30% sensitivity, 67.27% accuracy, and an F1 score of 0.55, which is much lower than the algorithm’s DCC of 0.77. Our model outlines LVI on H&E-stained-slides more effectively than human examiners; thus, it proves a valuable tool for pathologists.

show abstract

Pix2Pix-based Stain-to-Stain Translation: A Solution for Robust Stain Normalization in Histopathology Images Analysis

Cited by 3 publications

References 0 publications

RestainNet: a self-supervised digital re-stainer for stain normalization

RestainNet: a self-supervised digital re-stainer for stain normalization

Assessment of Color Reproducibility and Mitigation of Color Variation in Whole Slide Image Scanners for Toxicologic Pathology

A New Method of Artificial-Intelligence-Based Automatic Identification of Lymphovascular Invasion in Urothelial Carcinomas

Contact Info

Product

Resources

About