Learning to segment with limited annotations: Self-supervised pretraining with regression and contrastive loss in MRI
This work addresses the problem of reducing annotation costs for medical image segmentation, particularly in MRI applications like liver and prostate segmentation, but it is incremental as it builds on existing self-supervised methods.
The paper tackles the challenge of limited manual annotations for deep learning segmentation models by proposing self-supervised pretraining with regression and contrastive losses on MRI data, showing that models pretrained this way achieve comparable performance with fewer labeled datasets and that contrastive loss outperforms regression loss.
Obtaining manual annotations for large datasets for supervised training of deep learning (DL) models is challenging. The availability of large unlabeled datasets compared to labeled ones motivate the use of self-supervised pretraining to initialize DL models for subsequent segmentation tasks. In this work, we consider two pre-training approaches for driving a DL model to learn different representations using: a) regression loss that exploits spatial dependencies within an image and b) contrastive loss that exploits semantic similarity between pairs of images. The effect of pretraining techniques is evaluated in two downstream segmentation applications using Magnetic Resonance (MR) images: a) liver segmentation in abdominal T2-weighted MR images and b) prostate segmentation in T2-weighted MR images of the prostate. We observed that DL models pretrained using self-supervision can be finetuned for comparable performance with fewer labeled datasets. Additionally, we also observed that initializing the DL model using contrastive loss based pretraining performed better than the regression loss.