Skip to content

This repository contains the official implementation of "Data-Agnostic Augmentations for Unknown Variations - Out-of-Distribution Generalisation in MRI Segmentation", under review at MIDL 2025.

License

Notifications You must be signed in to change notification settings

MIAGroupUT/augmentations-for-the-unknown

Repository files navigation

Augmentations for the Unknown

This repository contains the official implementation of "Data-Agnostic Augmentations for Unknown Variations - Out-of-Distribution Generalisation in MRI Segmentation", under review at MIDL 2025.

Our work explores novel data augmentation techniques designed to improve model generalization in medical imaging, especially in scenarios with unseen domain shifts and rare cases.

Pretrained Models and Testing Data

Pretrained models and the testing data can be found at Zenodo.

Poster

Poster

About

This repository contains the official implementation of "Data-Agnostic Augmentations for Unknown Variations - Out-of-Distribution Generalisation in MRI Segmentation", under review at MIDL 2025.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages