Efficient EEG Imagined Speech Decoding Using Channel Selection and Residual Attention-Based Lightweight Model
IEEE Sensors Journal · 2026-11 · 26 (21) · 33078-33091
DOI: 10.1109/jsen.2026.3734776 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Sensors Journal, 2026-11. Verfasst von Anand Mohan und 1 weiteren Autor:innen. Die Arbeit nennt 38 Literaturangaben.
Abstract (Originalsprache)
Imagined speech is the internal articulation of words without vocalization. It offers a path for developing silent communication systems through non-invasive brain-computer interfaces (BCIs). EEG signals are high-dimensional, noisy, and non-stationary, making classification challenging. The primary issues arise from processing data from multiple EEG channels, many of which may contain irrelevant information. Various channel selection strategies have been investigated. ERP-based, statistical, entropy-based, spectral, and connectivity-driven approaches reduce computational overhead and enhance signal relevance. A hybrid strategy combining insights from multiple domains was also explored to select the most discriminative channels. Complementing this, a comprehensive investigation of multiple phase-based feature combinations was studied, such as those integrating phase with spectral power, entropy, Hjorth parameters, statistical descriptors, coherence, and time-frequency representations. These features capture dynamic phase interactions and neural synchrony patterns essential for imagined speech. To unify these contributions, a novel decoding framework, ERP-DWPS-ResNet (Event-Related Potential-guided Dynamic Wavelet Phase-Statistical Residual Network), is introduced. This architecture combines ERP-guided channel selection, dynamic wavelet-based signal augmentation, phase-statistical feature fusion, and a lightweight residual attention-based deep learning model for classification. For classification, a novel lightweight deep learning model named Residual Hierarchically Converging Spatial and Channel Attention Network (Res-HCSATNet) is introduced. The proposed framework achieves classification accuracies of 85.0%, 83.0%, 84.5%, and 94.1% for long words, short words, vowels, and short–long word pairs, respectively. Compared with Transformer and BiLSTM baselines, the proposed Res-HCSATNet achieves comparable or higher accuracy with only 161,191 trainable parameters and an inference latency of 0.78–0.84 ms.
Zitieren
APA
Mohan, A., & Anand, R. S. (2026). Efficient EEG Imagined Speech Decoding Using Channel Selection and Residual Attention-Based Lightweight Model. IEEE Sensors Journal, 26(21), 33078-33091. https://doi.org/10.1109/jsen.2026.3734776
Vancouver
Mohan A, Anand RS. Efficient EEG Imagined Speech Decoding Using Channel Selection and Residual Attention-Based Lightweight Model. IEEE Sensors Journal. 2026;26(21):33078-33091. doi:10.1109/jsen.2026.3734776
BibTeX
@article{Mohan2026,
author = {Mohan, Anand and Anand, R. S.},
title = {Efficient EEG Imagined Speech Decoding Using Channel Selection and Residual Attention-Based Lightweight Model},
journal = {IEEE Sensors Journal},
year = {2026},
volume = {26},
number = {21},
pages = {33078--33091},
doi = {10.1109/jsen.2026.3734776}
}
EUVIMED