Structure-Preserving Multi-Domain Stain Color Augmentation using Style-Transfer with Disentangled Representations

@inproceedings{Wagner2021StructurePreservingMS,
  title={Structure-Preserving Multi-Domain Stain Color Augmentation using Style-Transfer with Disentangled Representations},
  author={Sophia J. Wagner and Nadieh Khalili and Raghav Sharma and Melanie Boxberg and Carsten Marr and Walter de Back and Tingying Peng},
  booktitle={MICCAI},
  year={2021}
}
In digital pathology, different staining procedures and scanners cause substantial color variations in whole-slide images (WSIs), especially across different laboratories. These color shifts result in a poor generalization of deep learning-based methods from the training domain to external pathology data. To increase test performance, stain normalization techniques are used to reduce the variance between training and test domain. Alternatively, color augmentation can be applied during training… Expand

Figures and Tables from this paper

References

SHOWING 1-10 OF 19 REFERENCES
Stain Style Transfer of Histopathology Images Via Structure-Preserved Generative Learning
TLDR
This study proposes two stain style transfer models, SSIM-GAN and DSCSI-GAN, based on the generative adversarial networks, which outperform prior arts in terms of generating more stain-consistent images, better preserving histological information in images, and obtaining significantly higher learning efficiency. Expand
Residual cyclegan for robust domain transformation of histopathological tissue slides
TLDR
This paper proposes to reduce performance variability by using -consistent generative adversarial (CycleGAN) networks to remove staining variation by improving upon the regular CycleGAN by incorporating residual learning. Expand
Quantifying the effects of data augmentation and stain color normalization in convolutional neural networks for computational pathology
TLDR
A novel unsupervised method to perform stain colornormalization using a neural network is proposed, providing practical guidelines on how to use stain color augmentation and stain color normalization in future computational pathology applications. Expand
Staingan: Stain Style Transfer for Digital Histological Images
TLDR
This work proposes a deep-learning solution inspired by CycleGANs that is trained end-to-end, eliminating the need for an expert to pick a representative reference slide, and showed superior results quantitatively and qualitatively against the state of the art methods. Expand
Generative Adversarial Networks in Digital Pathology: A Survey on Trends and Future Potential
TLDR
The recent state-of-the-art developments in a generalizing notation are summarized, the main applications of GANs are presented, and an outlook of some chosen promising approaches and their possible future applications are given. Expand
Multimarginal Wasserstein Barycenter for Stain Normalization and Augmentation
TLDR
A new approach based on the multimarginal Wasserstein barycenter to normalize and augment H&E stained images given one or more references to drive stain normalization and augmentation simultaneously is presented. Expand
Structure-Preserving Color Normalization and Sparse Stain Separation for Histological Images
TLDR
Stain density correlation with ground truth and preference by pathologists were higher for images normalized using the method when compared to other alternatives, and a computationally faster extension of this technique is proposed for large whole-slide images that selects an appropriate patch sample instead of using the entire image to compute the stain color basis. Expand
Stain Specific Standardization of Whole-Slide Histopathological Images
TLDR
The results of the empirical evaluations collectively demonstrate the potential contribution of the proposed standardization algorithm to improved diagnostic accuracy and consistency in computer-aided diagnosis for histopathology data. Expand
DRIT++: Diverse Image-to-Image Translation via Disentangled Representations
TLDR
This work presents an approach based on disentangled representation for generating diverse outputs without paired training images that can generate diverse and realistic images on a wide range of tasks without pairedTraining data. Expand
Distribution Matching Losses Can Hallucinate Features in Medical Image Translation
This paper discusses how distribution matching losses, such as those used in CycleGAN, when used to synthesize medical images can lead to mis-diagnosis of medical conditions. It seems appealing toExpand
...
1
2
...