A Novel Deep Learning Approach With Data Augmentation to Classify Motor Imagery Signals

TLDR

Brain‑computer interfaces rely on accurate classification of non‑invasive EEG signals, and deep learning has become popular for feature extraction and classification, yet its performance depends heavily on large training datasets. This study proposes a novel approach that combines deep learning with data augmentation to improve EEG classification for motor imagery. The authors generate synthetic EEG frames by mixing intrinsic mode functions from empirical mode decomposition and converting all data into tensor representations via complex Morlet wavelets, then train two models—a convolutional neural network and a wavelet neural network that replaces convolutional layers with wavelets—to classify motor‑imagery signals. Experimental results demonstrate that the augmented EEG data significantly enhance neural‑network training, yielding higher classification accuracies than existing methods for motor‑imagery signals, and the wavelet neural network also performs well on steady‑state visual evoked potential classification.

Abstract

Brain-computer interface provides a new communication bridge between the human mind and devices, depending largely on the accurate classification and identification of non-invasive EEG signals. Recently, the deep learning approaches have been widely used in many fields to extract features and classify various types of data successfully. However, the deep learning approach requires massive data to train its neural networks, and the amount of data impacts greatly on the quality of the classifiers. This paper proposes a novel approach that combines deep learning and data augmentation for EEG classification. We applied the empirical mode decomposition on the EEG frames and mixed their intrinsic mode functions to create new artificial EEG frames, followed by transforming all EEG data into tensors as inputs of the neural network by complex Morlet wavelets. We proposed two neural networks-convolutional neural network and wavelet neural network-to train the weights and classify two classes of motor imagery signals. The wavelet neural network is a new type of neural network using wavelets to replace the convolutional layers. The experimental results show that the artificial EEG frames substantially improve the training of neural networks, and both two networks yield relatively higher classification accuracies compared to prevailing approaches. Meanwhile, we also verified the performance of our new proposed wavelet neural network model in the classification of steady-state visual evoked potentials.