Academic Journal
Synergistic EEG signal processing for brain-computer interfaces using hybrid MothCray optimization and deep learning.
| Τίτλος: | Synergistic EEG signal processing for brain-computer interfaces using hybrid MothCray optimization and deep learning. |
|---|---|
| Συγγραφείς: | Kumari A; Department of Computer Science and Engineering, National Institute of Technology Goa, Cuncolim, Goa, 403703, India. Electronic address: annupriya@nitgoa.ac.in., Edla DR; Department of Computer Science and Engineering, National Institute of Technology Goa, Cuncolim, Goa, 403703, India. Electronic address: dr.reddy@nitgoa.ac.in., Ramesh D; Department of Computer Science and Engineering, Indian Institute of Technology (Indian School of Mines) Dhanbad, 826004, India; Department of Computer Science, Vizja University, Poland. Electronic address: drramesh@iitism.ac.in., Shirodkar VR; Department of Information Technology Goa College of Engineering, Farmagudi, 403401, Goa, India. Electronic address: vaishali@nitgoa.ac.in. |
| Πηγή: | Neuroscience [Neuroscience] 2026 Oct 15; Vol. 614, pp. 47-58. Date of Electronic Publication: 2026 Aug 04. |
| Τύπος έκδοσης: | Journal Article |
| Γλώσσα: | English |
| Στοιχεία περιοδικού: | Publisher: Elsevier Science Country of Publication: United States NLM ID: 7605074 Publication Model: Print-Electronic Cited Medium: Internet ISSN: 1873-7544 (Electronic) Linking ISSN: 03064522 NLM ISO Abbreviation: Neuroscience Subsets: MEDLINE |
| Imprint Name(s): | Publication: [New York?] : Elsevier Science Original Publication: Oxford, Elmsford, N. Y., Pergamon Press |
| Ιατρικοί όροι (MeSH): | Electroencephalography*/methods , Brain*/physiology , Brain-Computer Interfaces* , Deep Learning* , Signal Processing, Computer-Assisted*, Humans ; Algorithms |
| Περίληψη: | Electroencephalography (EEG) based brain-computer interface (BCI) systems require optimal channel selection to achieve high signal quality, reduced setup complexity, and robust usability. This study introduces a comprehensive signal-processing framework to improve the efficiency and accuracy of EEG-driven BCIs. Signal pre-processing incorporates standard EEG denoising techniques, including notch filtering, independent component analysis (ICA), and segmentation into temporal windows to mitigate artifacts and enhance data quality. The channel selection phase employs the novel mothcray optimization (MCO) algorithm, a hybrid approach that integrates moth flame optimization and crayfish optimization to identify the most informative channels. Feature extraction encompasses time-domain statistics, frequency-domain attributes, and connectivity measures, enabling a rich representation of underlying neural dynamics. Classification is performed using an advanced deep neural network architecture tailored for the spatial-temporal characteristics of EEG data. Validated on the BCI competition IV dataset IIa, the proposed model achieves an accuracy rate of 93.92%, outperforming established methods and underscoring its effectiveness. The proposed MCO-based channel selection, multi-domain feature fusion, and enhanced EEGNet classifier demonstrate a significant step forward for practical BCI deployment that requires fast setup, robust operation, and efficient use of resources. It is especially advantageous in applications that rely on a reduced number of EEG channels to enable more comfortable and portable headsets, must sustain stable performance under varying recording conditions in home- or clinic-based neurorehabilitation and assistive communication settings. (Copyright © 2026 International Brain Research Organization (IBRO). Published by Elsevier Inc. All rights reserved.) |
| Competing Interests: | Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. |
| Contributed Indexing: | Keywords: Brain-computer interface; Channel selection; Deep neural network; Electroencephalography; Mothcray optimization |
| Entry Date(s): | Date Created: 20260804 Date Completed: 20260907 Latest Revision: 20260907 |
| Update Code: | 20260908 |
| DOI: | 10.1016/j.neuroscience.2026.07.060 |
| PMID: | 42551709 |
| Βάση Δεδομένων: | MEDLINE |
| ISSN: | 1873-7544 |
|---|---|
| DOI: | 10.1016/j.neuroscience.2026.07.060 |