Data-Agnostic Augmentations for Unknown Variations: Out-of-Distribution Generalisation in MRI Segmentation
By: Puru Vaish , Felix Meister , Tobias Heimann and more
Potential Business Impact:
Makes medical scans more accurate for doctors.
Medical image segmentation models are often trained on curated datasets, leading to performance degradation when deployed in real-world clinical settings due to mismatches between training and test distributions. While data augmentation techniques are widely used to address these challenges, traditional visually consistent augmentation strategies lack the robustness needed for diverse real-world scenarios. In this work, we systematically evaluate alternative augmentation strategies, focusing on MixUp and Auxiliary Fourier Augmentation. These methods mitigate the effects of multiple variations without explicitly targeting specific sources of distribution shifts. We demonstrate how these techniques significantly improve out-of-distribution generalization and robustness to imaging variations across a wide range of transformations in cardiac cine MRI and prostate MRI segmentation. We quantitatively find that these augmentation methods enhance learned feature representations by promoting separability and compactness. Additionally, we highlight how their integration into nnU-Net training pipelines provides an easy-to-implement, effective solution for enhancing the reliability of medical segmentation models in real-world applications.
Similar Papers
MediAug: Exploring Visual Augmentation in Medical Imaging
CV and Pattern Recognition
Improves medical scans to find diseases better.
When Dynamic Data Selection Meets Data Augmentation
Machine Learning (CS)
Trains computers faster with less data.
SGD-Mix: Enhancing Domain-Specific Image Classification with Label-Preserving Data Augmentation
CV and Pattern Recognition
Makes computer pictures more real for learning.