Skip to main content
. Author manuscript; available in PMC: 2023 Aug 31.
Published in final edited form as: Proc IEEE Comput Soc Conf Comput Vis Pattern Recognit. 2023 Aug 22;2023:19798–19808. doi: 10.1109/cvpr52729.2023.01896

Figure 5. Hierarchical self-supervised discriminative learning without strong data augmentations.

Figure 5.

Randomly sampled patch representations are shown after SimCLR and HiDisc-Patient pretraining without the use of strong, domain-agnostic data augmentations on SRH dataset. HiDisc achieves high-quality representation learning, while SimCLR is unable to learn semantically meaningful features via instance discrimination alone. HGG, high grade glioma; LGG, low grade glioma; Normal, normal brain tissue