Intra-subject class-incremental deep learning approach for EEG-based imagined speech recognition - Publication - Bridge of Knowledge

Your account

login

Search

Intra-subject class-incremental deep learning approach for EEG-based imagined speech recognition

Abstract

Brain–computer interfaces (BCIs) aim to decode brain signals and transform them into commands for device operation. The present study aimed to decode the brain activity during imagined speech. The BCI must identify imagined words within a given vocabulary and thus perform the requested action. A possible scenario when using this approach is the gradual addition of new words to the vocabulary using incremental learning methods. An issue with incremental learning methods is degradation of the decoding capacity of the original model when new classes are added. In this study, a class-incremental neural network method is proposed to increase the vocabulary of imagined speech. The results indicate a stable model that did not degenerate when a new word was integrated. The proposed method allows for the inclusion of newly imagined words without a significant loss of total accuracy for the two datasets.

Citations

  • 9

    CrossRef

  • 0

    Web of Science

  • 1 2

    Scopus

Authors (4)

Cite as

Full text

download paper
downloaded 0 times
Publication version
Accepted or Published Version
DOI:
Digital Object Identifier (open in new tab) 10.1016/j.bspc.2022.104433
License
Creative Commons: CC-BY-NC-ND open in new tab

Keywords

Details

Category:
Articles
Type:
artykuły w czasopismach
Published in:
Biomedical Signal Processing and Control no. 81,
ISSN: 1746-8094
Language:
English
Publication year:
2023
Bibliographic description:
Garcia Salinas J., Torres-García A. A., Reyes-Garćia C. A., Villaseñor-Pineda L.: Intra-subject class-incremental deep learning approach for EEG-based imagined speech recognition// Biomedical Signal Processing and Control -Vol. 81, (2023), s.104433-
DOI:
Digital Object Identifier (open in new tab) 10.1016/j.bspc.2022.104433
Sources of funding:
  • Free publication
Verified by:
Gdańsk University of Technology

seen 141 times

Recommended for you

Transfer learning in imagined speech EEG-based BCIs

2019

Tensor Decomposition for Imagined Speech Discrimination in EEG

2018

Decoding imagined speech for EEG-based BCI

  • C. A. Reyes-García,
  • A. A. Torres-García,
  • T. Hernández-del-Toro
  • + 2 authors
2024

Analysis of the Capability of Deep Learning Algorithms for EEG-based Brain-Computer Interface Implementation

2023
Meta Tags