Attention-Enhanced U-Net for Sensor-Efficient High-Density EEG Reconstruction in Wearable Brain Monitoring Systems.

High-channel-density (HCD) electroencephalography (EEG) enables fine-grained neural sensing but is constrained by high hardware costs, spatial complexity, and limited portability. This study developed a deep learning-based method to reconstruct high-density EEG signals from low-channel-density (LCD)...

Descripción completa

Detalles Bibliográficos
Publicado en:Journal of Medical Systems Vol. 50; no. 1; pp. 1 - 15
Autores principales: Zhuang, Jyun-Rong, Guo, Pin-Cheng
Formato: equations & formulas pictorial research tables/charts tracings Journal Article
Publicado: Springer Nature 4/5/2026
Acceso en línea:Ver este registro en EBSCOhost
fields @attributes:
  recordID: 1
pdfLink:
plink: https://search.ebscohost.com/login.aspx?direct=true&db=ccm&AN=192768840&site=ehost-live
header:
  @attributes:
    shortDbName: ccm
    uiTerm: 192768840
    longDbName: CINAHL Complete
    uiTag: AN
  controlInfo:
    bkinfo:
    dissinfo:
    jinfo:
      jid:
        01485598
        4N0
      jtl: Journal of Medical Systems
      issn: 01485598
      maglogo: N
    pubinfo:
      dt: 4/5/2026
      vid: 50
      iid: 1
      pid: 237
      pub: Springer Nature
      place: New York, New York
    artinfo:
      ui:
        192768840
        192768840
        192768840
        10.1007/s10916-026-02374-5
        192768840
      ppf: 1
      ppct: 14
      formats:
      tig:
        atl: Attention-Enhanced U-Net for Sensor-Efficient High-Density EEG Reconstruction in Wearable Brain Monitoring Systems.
      aug:
        au:
          Zhuang, Jyun-Rong
          Guo, Pin-Cheng
        affil: https://ror.org/05vn3ca78 Department of Mechanical Engineering, National Chung Hsing University, 145 Xingda Rd., South Dist, 402202, Taichung City, Taiwan, R.O.C.
      sug:
        subj:
          Virtual High-Throughput Screening
          Electroencephalography
          Wearable Sensors
          Deep Learning
          Neural Networks (Computer)
          Human
          Pearson's Correlation Coefficient
          Funding Source
          Brain-Computer Interfaces
      ab: High-channel-density (HCD) electroencephalography (EEG) enables fine-grained neural sensing but is constrained by high hardware costs, spatial complexity, and limited portability. This study developed a deep learning-based method to reconstruct high-density EEG signals from low-channel-density (LCD) inputs, enabling more practical and affordable brain-monitoring systems. This study introduces VEEG-A-U-Net, a lightweight U-Net architecture enhanced with attention gates and residual learning. The model combined spherical spline interpolation with a learnable correction signal to adaptively model spatial-temporal features. The framework was trained and evaluated on the SEED dataset, using normalized mean square error (NMSE), signal-to-noise ratio (SNR), and Pearson correlation coefficient (PCC) to assess reconstruction performance. Validation was conducted through leave-one-subject-out cross-validation (LOSO-CV) and cross-dataset experiments to examine generalizability. Under the same reconstruction setting (scale factor = 2), VEEG-A-U-Net achieved competitive reconstruction performance compared with state-of-the-art methods, while requiring substantially fewer parameters and computational operations. Cross-dataset evaluations confirmed stable performance across different EEG paradigms. Inference-time analysis showed low computational latency, indicating practical feasibility for deployment in resource-constrained and edge computing environments. A preliminary clinical EEG evaluation was also conducted to explore feasibility in clinical settings.The proposed framework offers an effective and lightweight solution for reconstructing high-density EEG from sparse measurements. These findings may support the development of sensor-efficient and portable EEG systems for practical neuroengineering and brain–computer interface applications.
      pubtype: Academic Journal
      doctype:
        equations & formulas
        pictorial
        research
        tables/charts
        tracings
        Journal Article
      ougenre: Article
    language: English
    refInfo:
    holdings:
      @attributes:
        islocal: N