Key result
Self-supervised pretraining outperforms learning from scratch for cardiac ultrasound segmentation and EF estimation.
Why the study?
Does self-supervised learning combined with supervised contrastive learning improve cardiac ultrasound segmentation compared to learning from scratch?
Does self-supervised learning combined with supervised contrastive learning improve cardiac ultrasound segmentation compared to learning from scratch?
Answers are grounded in this paper and cite their sources.
Self-supervised learning improves cardiac ultrasound segmentation and ejection fraction evaluation, reducing the need for large amounts of dense pixel-level annotations.
Deep learning models trained with an insufficient volume of data can often fail to generalize between different equipment, clinics, and clinicians or fail to achieve acceptable performance. We improve cardiac ultrasound segmentation models using unlabeled data to learn recurrent anatomical representations via self-supervision. In addition, we leverage supervised local contrastive learning on sparse labels to improve the segmentation and reduce the need for large amounts of dense pixel-level supervisory annotations. Then, we implement supervised fine-tuning to segment key temporal anatomical features to estimate the cardiac Ejection Fraction (EF). We show that pretraining the network weights using self-supervised learning for subsequent supervised contrastive learning outperforms learning from scratch, validated using two state-of-the-art segmentation models, the DeepLabv3+ and Attention U-Net.Clinical relevance-This work has clinical relevance for assisting physicians when conducting cardiac function evaluations. We improve cardiac ejection fraction evaluation compared to previous methods, helping to alleviate the burden associated with acquiring labeled images.
No takes yet. Share an insight, caveat, or question.
Lamoureux et al. (2023) studied Cardiac ultrasound segmentation. Self-supervised learning and supervised local contrastive learning vs. Learning from scratch was evaluated on Segmentation performance and cardiac Ejection Fraction estimation. Pretraining network weights using self-supervised learning for subsequent supervised contrastive learning outperformed learning from scratch for cardiac ultrasound segmentation and EF estimation.
Synapse has enriched 10 related papers. Consider them for comparative context: