MI-EEGNET: A novel convolutional neural network for motor imagery classification.
Convolutional neural networks
Deep learning
Electroencephalography
Motor imagery
Journal
Journal of neuroscience methods
ISSN: 1872-678X
Titre abrégé: J Neurosci Methods
Pays: Netherlands
ID NLM: 7905558
Informations de publication
Date de publication:
01 04 2021
01 04 2021
Historique:
received:
11
04
2020
revised:
08
12
2020
accepted:
09
12
2020
pubmed:
19
12
2020
medline:
1
7
2021
entrez:
18
12
2020
Statut:
ppublish
Résumé
Brain-computer interfaces (BCI) permits humans to interact with machines by decoding brainwaves to command for a variety of purposes. Convolutional neural networks (ConvNet) have improved the state-of-the-art of motor imagery decoding in an end-to-end approach. However, shallow ConvNets usually perform better than their deep counterparts. Thus, we aim to design a novel ConvNet that is deeper than the existing models, with an increase in terms of performances, and with optimal complexity. We develop a ConvNet based on Inception and Xception architectures that uses convolutional layers to extract temporal and spatial features. We adopt separable convolutions and depthwise convolutions to enable faster and efficient ConvNet. Then, we introduce a new block that is inspired by Inception to learn more rich features to improve the classification performances. The obtained results are comparable with other state-of-the-art techniques. Also, the weights of the convolutional layers give us some insights onto the learned features and reveal the most relevant ones. We show that our model significantly outperforms Filter Bank Common Spatial Pattern (FBCSP), Riemannian Geometry (RG) approaches, and ShallowConvNet (p < 0.05). The obtained results prove that motor imagery decoding is possible without handcrafted features.
Sections du résumé
BACKGROUND
Brain-computer interfaces (BCI) permits humans to interact with machines by decoding brainwaves to command for a variety of purposes. Convolutional neural networks (ConvNet) have improved the state-of-the-art of motor imagery decoding in an end-to-end approach. However, shallow ConvNets usually perform better than their deep counterparts. Thus, we aim to design a novel ConvNet that is deeper than the existing models, with an increase in terms of performances, and with optimal complexity.
NEW METHOD
We develop a ConvNet based on Inception and Xception architectures that uses convolutional layers to extract temporal and spatial features. We adopt separable convolutions and depthwise convolutions to enable faster and efficient ConvNet. Then, we introduce a new block that is inspired by Inception to learn more rich features to improve the classification performances.
RESULTS
The obtained results are comparable with other state-of-the-art techniques. Also, the weights of the convolutional layers give us some insights onto the learned features and reveal the most relevant ones.
COMPARISON WITH EXISTING METHOD(S)
We show that our model significantly outperforms Filter Bank Common Spatial Pattern (FBCSP), Riemannian Geometry (RG) approaches, and ShallowConvNet (p < 0.05).
CONCLUSIONS
The obtained results prove that motor imagery decoding is possible without handcrafted features.
Identifiants
pubmed: 33338542
pii: S0165-0270(20)30460-X
doi: 10.1016/j.jneumeth.2020.109037
pii:
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
109037Informations de copyright
Copyright © 2020. Published by Elsevier B.V.