Translation Consistent Semi-supervised Segmentation for 3D Medical Images. (arXiv:2203.14523v2 [cs.CV] UPDATED)

3D medical image segmentation methods have been successful, but their
dependence on large amounts of voxel-level annotated data is a disadvantage
that needs to be addressed given the high cost to obtain such annotation.
Semi-supervised learning (SSL) solve this issue by training models with a large
unlabelled and a small labelled dataset. The most successful SSL approaches are
based on consistency learning that minimises the distance between model
responses obtained from perturbed views of the unlabelled data. These
perturbations usually …

annotated data annotation arxiv consistent context cost data dataset image images issue medical responses segmentation semi-supervised semi-supervised learning small ssl supervised learning training translation voxel