SDREAMER: Self-distilled Mixture-of-Modality-Experts Transformer for Automatic Sleep Staging
Research output: Contribution to conference › Paper › Research › peer-review
Standard
SDREAMER : Self-distilled Mixture-of-Modality-Experts Transformer for Automatic Sleep Staging. / Chen, Jingyuan; Yao, Yuan; Anderson, Mie; Hauglund, Natalie; Kjaerby, Celia; Untiet, Verena; Nedergaard, Maiken; Luo, Jiebo.
2023. 131-142 Paper presented at 2023 IEEE International Conference on Digital Health, ICDH 2023, Hybrid, Chicago, United States.Research output: Contribution to conference › Paper › Research › peer-review
Harvard
APA
Vancouver
Author
Bibtex
}
RIS
TY - CONF
T1 - SDREAMER
T2 - 2023 IEEE International Conference on Digital Health, ICDH 2023
AU - Chen, Jingyuan
AU - Yao, Yuan
AU - Anderson, Mie
AU - Hauglund, Natalie
AU - Kjaerby, Celia
AU - Untiet, Verena
AU - Nedergaard, Maiken
AU - Luo, Jiebo
N1 - Funding Information: ACKNOWLEDGMENTS Research reported in this publication was supported by the National Institutes of Health under Award Number U19NS128613. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. Publisher Copyright: © 2023 IEEE.
PY - 2023
Y1 - 2023
N2 - Automatic sleep staging based on electroencephalography (EEG) and electromyography (EMG) signals is an important aspect of sleep-related research. Current sleep staging methods suffer from two major drawbacks. First, there are limited information interactions between modalities in the existing methods. Second, current methods do not develop unified models that can handle different sources of input. To address these issues, we propose a novel sleep stage scoring model sDREAMER, which emphasizes cross-modality interaction and per-channel performance. Specifically, we develop a mixture-of-modality-expert (MoME) model with three pathways for EEG, EMG, and mixed signals with partially shared weights. We further propose a self-distillation training scheme for further information interaction across modalities. Our model is trained with multi-channel inputs and can make classifications on either single-channel or multi-channel inputs. Experiments demonstrate that our model outperforms the existing transformer-based sleep scoring methods for multi-channel inference. For single-channel inference, our model also outperforms the transformer-based models trained with single-channel signals.
AB - Automatic sleep staging based on electroencephalography (EEG) and electromyography (EMG) signals is an important aspect of sleep-related research. Current sleep staging methods suffer from two major drawbacks. First, there are limited information interactions between modalities in the existing methods. Second, current methods do not develop unified models that can handle different sources of input. To address these issues, we propose a novel sleep stage scoring model sDREAMER, which emphasizes cross-modality interaction and per-channel performance. Specifically, we develop a mixture-of-modality-expert (MoME) model with three pathways for EEG, EMG, and mixed signals with partially shared weights. We further propose a self-distillation training scheme for further information interaction across modalities. Our model is trained with multi-channel inputs and can make classifications on either single-channel or multi-channel inputs. Experiments demonstrate that our model outperforms the existing transformer-based sleep scoring methods for multi-channel inference. For single-channel inference, our model also outperforms the transformer-based models trained with single-channel signals.
KW - distillation
KW - mixture-of-modality experts
KW - sleep scoring
KW - transformer
U2 - 10.1109/ICDH60066.2023.00028
DO - 10.1109/ICDH60066.2023.00028
M3 - Paper
AN - SCOPUS:85172375046
SP - 131
EP - 142
Y2 - 2 July 2023 through 8 July 2023
ER -
ID: 373667363