CLOCS, a contrastive learning method for cardiac signals, consistently outperformed state-of-the-art methods BYOL and SimCLR on downstream tasks and achieved strong generalization with 25% of labelled data.
CLOCS is a novel contrastive learning method for cardiac signals that outperforms state-of-the-art methods and achieves strong generalization with limited labelled data.
The healthcare industry generates troves of unlabelled physiological data. This data can be exploited via contrastive learning, a self-supervised pre-training method that encourages representations of instances to be similar to one another. We propose a family of contrastive learning methods, CLOCS, that encourages representations across space, time, and patients to be similar to one another. We show that CLOCS consistently outperforms the state-of-the-art methods, BYOL and SimCLR, when performing a linear evaluation of, and fine-tuning on, downstream tasks. We also show that CLOCS achieves strong generalization performance with only 25\% of labelled training data. Furthermore, our training procedure naturally generates patient-specific representations that can be used to quantify patient-similarity.
Kiyasseh et al. (Wed,) conducted a other in Cardiac signals. CLOCS (Contrastive Learning of Cardiac Signals) vs. BYOL and SimCLR was evaluated on Performance on downstream tasks (linear evaluation and fine-tuning). CLOCS, a contrastive learning method for cardiac signals, consistently outperformed state-of-the-art methods BYOL and SimCLR on downstream tasks and achieved strong generalization with 25% of labelled data.