Skip to content

H&E slides color normalisation using generative adversarial networks

Notifications You must be signed in to change notification settings

clemsgrs/stain-normalization

Folders and files

NameName
Last commit message
Last commit date

Latest commit

3ae055e · Dec 1, 2021

History

51 Commits
Feb 7, 2021
Feb 27, 2021
Mar 1, 2021
Feb 27, 2021
Feb 27, 2021
Dec 1, 2021
Feb 8, 2021
Feb 27, 2021
Feb 27, 2021
Feb 6, 2021
Feb 6, 2021
Feb 8, 2021
Feb 27, 2021
Feb 27, 2021

Repository files navigation

Stain Normalization

H&E slides color normalisation using generative adversarial networks
Academic project
Code based on StainGAN rep: https://github.com/xtarx/StainGAN

Dataset

I'm using the public MITOS-ATYPIE141 dataset (see more at: https://mitos-atypia-14.grand-challenge.org/).
The original dataset consists of 284 frames at x20 magnification selected in breast cancer biopsy slides by the team of Professor Frédérique Capron, head of the Pathology Department at Pitié-Salpêtrière Hospital in Paris, France. Each slide was stained with standard hematoxylin and eosin (H&E) dyes and was scanned by 2 scanners: Aperio Scanscope XT and Hamamatsu Nanozoomer 2.0-HT.

How to use the code?

To run training and testing, just download the train_inference.ipynb notebook and open it with Google Colab.
In the notebook, we first download the data, then run training and/or inference using the respective config files.

Results

rand10_results

About

H&E slides color normalisation using generative adversarial networks

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published