Official implementation of Transformer-based Models for Unsupervised Anomaly Segmentation in Brain MR Images.
Paper accepted in the International MICCAI Brainlesion 2022 Workshop
@InProceedings{10.1007/978-3-031-33842-7_3,
author="Ghorbel, Ahmed
and Aldahdooh, Ahmed
and Albarqouni, Shadi
and Hamidouche, Wassim",
editor="Bakas, Spyridon
and Crimi, Alessandro
and Baid, Ujjwal
and Malec, Sylwia
and Pytlarz, Monika
and Baheti, Bhakti
and Zenk, Maximilian
and Dorent, Reuben",
title="Transformer Based Models for Unsupervised Anomaly Segmentation in Brain MR Images",
booktitle="Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries",
year="2023",
publisher="Springer Nature Switzerland",
address="Cham",
pages="25--44",
isbn="978-3-031-33842-7"
}
MICCAI BrainLes 2022 Workshop, Transformer, Autoencoder, TensorFlow, Keras, Anomaly Segmentation, Unsupervised, Neuroimaging, Deeplearning
Python >= 3.6
All packages used in this repository are listed in requirements.txt. To install those, run:
pip3 install -r requirements.txt
Transformers_Unsupervised_Anomaly_Segmentation/
│
├── Transformers_Unsupervised_Anomaly_Segmentation.ipynb - Jupyter notebook to work on Google Colab
│
├── data/
│ └── data.txt - datasets descriptions and download link
│
├── models/ - Models defining, training and evaluating
│ ├── Autoencoders/
│ ├── DCAE.py
│ └── ...
│ ├── Latent Variable models/
│ ├── VAE.py
│ └── ...
│ └── Transformer based models/
│ ├── B_TAE.py
│ └── ...
│
├── saved/ - saving folder
│
└── scripts/ - small utility scripts
├── utils.py
└── ...
Every model can be trained and tested individually using the scripts which are provided in the models/* folders.
Training can be started by importing Transformers_Unsupervised_Anomaly_Segmentation.ipynb in Google Colab.
This github repository is linked and can directly loaded into the notebook. However, the datasets have to be stored so that Google Colab can access them.
Either uploading by a zip-file or uploading it to Google Drive and mounting the drive.
Please do not hesitate to open an issue to inform of any problem you may find within this repository.
This project is inspired by the comparative study paper on Autoencoders for Unsupervised Anomaly Segmentation in Brain MR Images: A Comparative Study.
@article{baur2021autoencoders,
title={Autoencoders for unsupervised anomaly segmentation in brain mr images: A comparative study},
author={Baur, Christoph and Denner, Stefan and Wiestler, Benedikt and Navab, Nassir and Albarqouni, Shadi},
journal={Medical Image Analysis},
pages={101952},
year={2021},
publisher={Elsevier}
}
This project is licensed under the GNU General Public License v3.0. See LICENSE for more details