Preservational Learning Improves Self-supervised Medical Image Models by Reconstructing Diverse Contexts
This work addresses the challenge of improving self-supervised representation learning for medical image analysis, offering a domain-specific incremental advance.
The paper tackles the problem of preserving maximal information in self-supervised learning for medical images by introducing Preservational Learning, which reconstructs diverse contexts to complement contrastive methods, resulting in PCRL outperforming self-supervised and supervised baselines in 5 classification/segmentation tasks.
Preserving maximal information is one of principles of designing self-supervised learning methodologies. To reach this goal, contrastive learning adopts an implicit way which is contrasting image pairs. However, we believe it is not fully optimal to simply use the contrastive estimation for preservation. Moreover, it is necessary and complemental to introduce an explicit solution to preserve more information. From this perspective, we introduce Preservational Learning to reconstruct diverse image contexts in order to preserve more information in learned representations. Together with the contrastive loss, we present Preservational Contrastive Representation Learning (PCRL) for learning self-supervised medical representations. PCRL provides very competitive results under the pretraining-finetuning protocol, outperforming both self-supervised and supervised counterparts in 5 classification/segmentation tasks substantially.