Fig. 3: Illustration of various data augmentation techniques available in GaNDLF showing the image and overlaid segmentation.
From: GaNDLF: the generally nuanced deep learning framework for scalable end-to-end clinical workflows
a Original image, b affine augmentation, c elastic augmentation, d flip augmentation, e bias augmentation, f blur augmentation, g ghosting augmentation, h noise augmentation, i spike augmentation, and j motion augmentation.
