Benchmarking brain–computer interface algorithms: Riemannian approaches vs convolutional neural networks

To date, a comprehensive comparison of Riemannian decoding methods with deep convolutional neural networks for EEG-based brain-computer interfaces remains absent from published work. We address this research gap by using MOABB, The Mother Of All BCI Benchmarks, to compare novel convolutional neural...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Journal of neural engineering Jg. 21; H. 4
Hauptverfasser: Eder, Manuel, Xu, Jiachen, Grosse-Wentrup, Moritz
Format: Journal Article
Sprache:Englisch
Veröffentlicht: England IOP Publishing 01.08.2024
Schlagworte:
Algorithms
benchmarking
Benchmarking - methods
brain-computer interface
Brain-Computer Interfaces
classification
convolutional neural network
electroencephalography
Electroencephalography - methods
Evoked Potentials, Visual - physiology
Humans
Imagination - physiology
machine learning
Neural Networks, Computer
Riemannian geometry
Riemannian geometry
electroencephalography
benchmarking
brain-computer interface
convolutional neural network
classification
machine learning
ISSN:1741-2560, 1741-2552, 1741-2552
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:To date, a comprehensive comparison of Riemannian decoding methods with deep convolutional neural networks for EEG-based brain-computer interfaces remains absent from published work. We address this research gap by using MOABB, The Mother Of All BCI Benchmarks, to compare novel convolutional neural networks to state-of-the-art Riemannian approaches across a broad range of EEG datasets, including motor imagery, P300, and steady-state visual evoked potentials paradigms. We systematically evaluated the performance of convolutional neural networks, specifically EEGNet, shallow ConvNet, and deep ConvNet, against well-established Riemannian decoding methods using MOABB processing pipelines. This evaluation included within-session, cross-session, and cross-subject methods, to provide a practical analysis of model effectiveness and to find an overall solution that performs well across different experimental settings. We find no significant differences in decoding performance between convolutional neural networks and Riemannian methods for within-session, cross-session, and cross-subject analyses. The results show that, when using traditional Brain-Computer Interface paradigms, the choice between CNNs and Riemannian methods may not heavily impact decoding performances in many experimental settings. These findings provide researchers with flexibility in choosing decoding approaches based on factors such as ease of implementation, computational efficiency or individual preferences.
Bibliographie:JNE-107434.R1
ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISSN:1741-2560
1741-2552
1741-2552
DOI:10.1088/1741-2552/ad6793