# Self adversarial attack as an augmentation method for immunohistochemical stainings

arXiv: https://arxiv.org/abs/2103.11362 · HTML: https://mlsanity.com/paper/2103.11362/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 8 of its 16 references that are in the map.
- First posted on arXiv 21 Mar 2021.

## Builds on (most cited first, 8 of 8)

- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [Quantifying the effects of data augmentation and stain color normalization in convolutional neural networks for computational pathology](https://mlsanity.com/paper/1902.06543/) (2019)
- [CycleGAN, a Master of Steganography](https://mlsanity.com/paper/1712.02950/) (2017)
- [PathologyGAN: Learning deep representations of cancer tissue](https://mlsanity.com/paper/1907.02644/) (2019)
- [Adversarial Self-Defense for Cycle-Consistent GANs](https://mlsanity.com/paper/1908.01517/) (2019)
- [Virtual staining for mitosis detection in Breast Histopathology](https://mlsanity.com/paper/2003.07801/) (2020)
- [Strategies for Training Stain Invariant CNNs](https://mlsanity.com/paper/1810.10338/) (2018)
- [Virtualization of tissue staining in digital pathology using an unsupervised deep learning approach](https://mlsanity.com/paper/1810.06415/) (2018)

## Built on by (most cited first, 4 of 4)

- [How to Robustify Black-Box ML Models? A Zeroth-Order Optimization Perspective](https://mlsanity.com/paper/2203.14195/) (2022)
- [Towards Measuring Domain Shift in Histopathological Stain Translation in an Unsupervised Manner](https://mlsanity.com/paper/2205.04368/) (2022)
- [HistoStarGAN: A Unified Approach to Stain Normalisation, Stain Transfer and Stain Invariant Segmentation in Renal Histopathology](https://mlsanity.com/paper/2210.09798/) (2022)
- [DSA-CycleGAN: A Domain Shift Aware CycleGAN for Robust Multi-Stain Glomeruli Segmentation](https://mlsanity.com/paper/2604.18368/) (2026)

## Abstract (arXiv)

It has been shown that unpaired image-to-image translation methods constrained by cycle-consistency hide the information necessary for accurate input reconstruction as imperceptible noise. We demonstrate that, when applied to histopathology data, this hidden noise appears to be related to stain specific features and show that this is the case with two immunohistochemical stainings during translation to Periodic acid- Schiff (PAS), a histochemical staining method commonly applied in renal pathology. Moreover, by perturbing this hidden information, the translation models produce different, plausible outputs. We demonstrate that this property can be used as an augmentation method which, in a case of supervised glomeruli segmentation, leads to improved performance.