https://github.com/xtarx/StainGAN

https://github.com/xtarx/StainGAN

Last synced: about 1 month ago
JSON representation

• Host: GitHub
• URL: https://github.com/xtarx/StainGAN
• Owner: xtarx
• Created: 2018-08-21T20:24:15.000Z (over 6 years ago)
• Default Branch: master
• Last Pushed: 2021-02-04T21:38:30.000Z (almost 4 years ago)
• Last Synced: 2024-10-22T09:49:27.148Z (2 months ago)
• Language: Python
• Size: 9.5 MB
• Stars: 62
• Watchers: 3
• Forks: 30
• Open Issues: 6
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

• awesome-pathology - StainGAN - Stain style transfer for digital histological images. (Software / Model)

README

        
# StainGAN

StainGAN implementation based on Cycle-Consistency Concept

For more information visit website.

## Structure

 * Stain-Transfer Model

 * Pre-processing.

 * Post-processing.

 * Evaluation 

## Datasets

The evaluation was done using the Camelyon16 challenge (https://camelyon16.grand-challenge.org/) consisting of 400 whole-slide images collected

in two different labs in Radboud University Medical Center (lab 1) and University

Medical Center Utrecht (lab 2). Otsu thresholding was used to remove the

background, Afterwards, 40, 000 256 × 256 patches were generated on the x40

magnification level, 30, 000 were used for training and 10, 000 used for validation

from lab 1 and 10, 000 patches were generated for testing from lab 2.

Patches can be found here: https://campowncloud.in.tum.de/index.php/s/iGgQ9vdHiMZsFJB?path=%2FStainGAN_camelyon16 

**Any use of the dataset or anypart of the code should be cited**

## Citation

If you use this code for your research, please cite our papers.

```

@inproceedings{shaban2019staingan,

  title={Staingan: Stain style transfer for digital histological images},

  author={Shaban, M Tarek and Baur, Christoph and Navab, Nassir and Albarqouni, Shadi},

  booktitle={2019 Ieee 16th international symposium on biomedical imaging (Isbi 2019)},

  pages={953--956},

  year={2019},

  organization={IEEE}

}

```

## Todo

- [x] Submit Matlab Image Similarity code

- [ ] Submit Trained Model and sample images

- [ ] Update Readme with more examples and explanation

## Acknowledgments

Code is inspired by [pytorch-DCGAN](https://github.com/pytorch/examples/tree/master/dcgan) and [CycleGAN](https://github.com/junyanz/CycleGAN).

Contact Shadi Albarqouni