Scholarlink

<p>Sleep staging is essential for assessing sleep quality and diagnosing sleep disorders, yet current deep learning approaches face two major challenges: conventional fusion methods fail to account for the varying contributions of different modalities across sleep stages, and raw sleep data are often contaminated by artifacts that can interfere with model performance. We propose SleepGMUformer, a gated multimodal temporal neural network that processes multidomain sleep data including EEG (Fpz-Cz, Pz-Oz), EOG, and wearable biosignals (heart rate, motion, steps). Our architecture comprises: (1) a preprocessing module for signal alignment, artifact removal, and EEG detrending; (2) single-channel temporal feature extraction using transformer encoders; and (3) a novel Gated Multimodal Unit (GMU) for dynamic, instance-level modality weighting. Extensive experiments demonstrate that our model achieves state-of-the-art performance with accuracies of 85.7% on SleepEDF-78 and 94.5% on the WristHR-Motion-Sleep dataset, outperforming existing methods by 1-4%. The GMU module provides interpretable insights into modality contributions across different sleep stages. SleepGMUformer effectively addresses key limitations in multimodal sleep staging by adaptively weighting modality contributions and robustly processing heterogeneous data sources. Our work enables more accurate and interpretable sleep staging, bridging the gap between clinical PSG and consumer wearable devices.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Gated adaptive feature weighting for multi-modal sleep staging using polysomnography and wearable biosignals

  • 期刊论文

  • 发表日期：

    2026年7月1日

查看全文
  • Chenjun Zhao,
  • Xuesen Niu,
  • Xinglin Yu,
  • Long Chen,
  • Na Lv,
  • Huiyu Zhou,
  • Aite Zhao

摘要

Sleep staging is essential for assessing sleep quality and diagnosing sleep disorders, yet current deep learning approaches face two major challenges: conventional fusion methods fail to account for the varying contributions of different modalities across sleep stages, and raw sleep data are often contaminated by artifacts that can interfere with model performance. We propose SleepGMUformer, a gated multimodal temporal neural network that processes multidomain sleep data including EEG (Fpz-Cz, Pz-Oz), EOG, and wearable biosignals (heart rate, motion, steps). Our architecture comprises: (1) a preprocessing module for signal alignment, artifact removal, and EEG detrending; (2) single-channel temporal feature extraction using transformer encoders; and (3) a novel Gated Multimodal Unit (GMU) for dynamic, instance-level modality weighting. Extensive experiments demonstrate that our model achieves state-of-the-art performance with accuracies of 85.7% on SleepEDF-78 and 94.5% on the WristHR-Motion-Sleep dataset, outperforming existing methods by 1-4%. The GMU module provides interpretable insights into modality contributions across different sleep stages. SleepGMUformer effectively addresses key limitations in multimodal sleep staging by adaptively weighting modality contributions and robustly processing heterogeneous data sources. Our work enables more accurate and interpretable sleep staging, bridging the gap between clinical PSG and consumer wearable devices.