EEG-based motor imagery classification using convolutional neural networks with local reparameterization trick

作者：

Highlights：

• Proposed a novel convolutional neural network approach for classification of EEG signals.

• Using local reparameterize trick to obtain an efficient estimator.

• Classification accuracy of greater than 92% was achieved by a global classifier.

• The model can be used in handling individual variability issue.

摘要

•Proposed a novel convolutional neural network approach for classification of EEG signals.•Using local reparameterize trick to obtain an efficient estimator.•Classification accuracy of greater than 92% was achieved by a global classifier.•The model can be used in handling individual variability issue.

论文关键词：Electroencephalogram (EEG),Motor Imagery (MI),Deep learning (DL),Convolutional neural networks (CNNs),Local Reparameterization Trick,Classification

论文评审过程：Received 31 January 2021, Revised 16 July 2021, Accepted 21 September 2021, Available online 27 September 2021, Version of Record 7 October 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.115968