Samenvatting

Brain-computer interfaces can be used to operate
devices by detecting a person’s intention from their brain activity.
Decoding motor imagery (MI) from electroencephalogram (EEG)
signals is a commonly used approach for this purpose. To
reliably analyze and process brain activity from EEG signals,
a sufficient number of sensors is usually required. However,
a large number of sensors increases the computational cost
of decoding MI classes. Additionally, commercial devices that
measure EEG signals typically have a limited number of sensors
at their disposal. In this experimental study, we investigate the
tradeoff in accuracy and complexity when decoding MI from
a reduced number of EEG sensors. For this purpose, several
decoding pipelines were trained on EEG data that use different
subsets of electrode locations employing well-established decoding
methods. We found that there is no significant difference (p=[0.18
- 0.91]) in average decoding accuracy when using a reduced
number of sensors. The largest loss in performance for a single
individual was a reduction in mean decoding accuracy of 0.1
when using 1/8th of the available sensors. Decoding MI from a
limited number of sensors is therefore feasible, highlighting the
possibility of using commercial sensor devices for this purpose
and reducing the computational cost.
Originele taal-2English
Titel11th International IEEE EMBS Conference on Neural Engineering
Plaats van productieBaltimore, MD, USA
UitgeverijIEEE
Aantal pagina's4
ISBN van elektronische versie9781665462921
DOI's
StatusPublished - 19 mei 2023
Evenement2023 11th International IEEE/EMBS Conference on Neural Engineering - Baltimore, United States
Duur: 24 apr 202327 apr 2023
https://doi-org.myezproxy.vub.ac.be/10.1109/NER52421.2023

Publicatie series

NaamInternational IEEE/EMBS Conference on Neural Engineering, NER
Volume2023-April
ISSN van geprinte versie1948-3546
ISSN van elektronische versie1948-3554

Conference

Conference2023 11th International IEEE/EMBS Conference on Neural Engineering
Verkorte titelNER
Land/RegioUnited States
StadBaltimore
Periode24/04/2327/04/23
Internet adres

Bibliografische nota

Publisher Copyright:
© 2023 IEEE.

Copyright:
Copyright 2023 Elsevier B.V., All rights reserved.

Vingerafdruk

Duik in de onderzoeksthema's van 'Optimal sensor set for decoding motor imagery from EEG'. Samen vormen ze een unieke vingerafdruk.

Citeer dit