What is the best data augmentation for 3D brain tumor segmentation?
This work addresses data scarcity in medical imaging for brain tumor segmentation, but it is incremental as it applies existing augmentation methods to a specific domain.
The study tackled the problem of limited annotated data for 3D brain tumor segmentation by testing various data augmentation techniques, finding that brightness adjustment and elastic deformation improved a standard 3D U-Net's performance, though combinations did not provide additional gains.
Training segmentation networks requires large annotated datasets, which in medical imaging can be hard to obtain. Despite this fact, data augmentation has in our opinion not been fully explored for brain tumor segmentation. In this project we apply different types of data augmentation (flipping, rotation, scaling, brightness adjustment, elastic deformation) when training a standard 3D U-Net, and demonstrate that augmentation significantly improves the network's performance in many cases. Our conclusion is that brightness augmentation and elastic deformation work best, and that combinations of different augmentation techniques do not provide further improvement compared to only using one augmentation technique. Our code is available at https://github.com/mdciri/3D-augmentation-techniques.