Abstract:
We learn to map the ECG signal to emotional or physical states using a 1D-CNN followed by a Transformer. To overcome the limited number of samples, we propose a new self-...Show MoreMetadata
Abstract:
We learn to map the ECG signal to emotional or physical states using a 1D-CNN followed by a Transformer. To overcome the limited number of samples, we propose a new self-supervised learning scheme that combines latent space masking with both a temporal prediction task and a matching task, both learned via a contrastive loss. Our method outperforms the existing self-supervised approaches for ECG by a wide margin.
Published in: 2023 IEEE SENSORS
Date of Conference: 29 October 2023 - 01 November 2023
Date Added to IEEE Xplore: 28 November 2023
ISBN Information: