EEG-Based Feature Classification Combining 3D-Convolutional Neural Networks with Generative Adversarial Networks for Motor Imagery

Aug 29, 2024Journal of integrative neuroscience

Classifying Brain Signals of Imagined Movement Using 3D Neural Networks and Data Generation Models

AI simplified

Circadian Biology on OpenScience ↗PubMed ↗DOI ↗OA ↗

Abstract

The 3D-CNN-GAN models achieved within-session motor imagery accuracies of 77.03% on the GigaDB dataset.

The introduction of a novel 3D-convolutional neural network-generative adversarial network architecture improves feature decoding from EEG signals.
Using a sliding window technique, temporal, frequency, and phase features were extracted to create a three-dimensional feature map.
Generative adversarial networks synthesized artificial data, enhancing the dataset and improving the detection of brain connectivity patterns.
On the GigaDB dataset, the 3D-CNN-GAN model showed a 0.54% increase in accuracy compared to the 3D-CNN model.
Cross-session motor imagery accuracies reached 63.04% with the 3D-CNN-GAN on the SHU dataset, indicating improved generalizability.

AI simplified

BACKGROUND: The adoption of convolutional neural networks (CNNs) for decoding electroencephalogram (EEG)-based motor imagery (MI) in brain-computer interfaces has significantly increased recently. The effective extraction of motor imagery features is vital due to the variability among individuals and temporal states.

METHODS: This study introduces a novel network architecture, 3D-convolutional neural network-generative adversarial network (3D-CNN-GAN), for decoding both within-session and cross-session motor imagery. Initially, EEG signals were extracted over various time intervals using a sliding window technique, capturing temporal, frequency, and phase features to construct a temporal-frequency-phase feature (TFPF) three-dimensional feature map. Generative adversarial networks (GANs) were then employed to synthesize artificial data, which, when combined with the original datasets, expanded the data capacity and enhanced functional connectivity. Moreover, GANs proved capable of learning and amplifying the brain connectivity patterns present in the existing data, generating more distinctive brain network features. A compact, two-layer 3D-CNN model was subsequently developed to efficiently decode these TFPF features.

RESULTS: Taking into account session and individual differences in EEG data, tests were conducted on both the public GigaDB dataset and the SHU laboratory dataset. On the GigaDB dataset, our 3D-CNN and 3D-CNN-GAN models achieved two-class within-session motor imagery accuracies of 76.49% and 77.03%, respectively, demonstrating the algorithm's effectiveness and the improvement provided by data augmentation. Furthermore, on the SHU dataset, the 3D-CNN and 3D-CNN-GAN models yielded two-class within-session motor imagery accuracies of 67.64% and 71.63%, and cross-session motor imagery accuracies of 58.06% and 63.04%, respectively.

CONCLUSIONS: The 3D-CNN-GAN algorithm significantly enhances the generalizability of EEG-based motor imagery brain-computer interfaces (BCIs). Additionally, this research offers valuable insights into the potential applications of motor imagery BCIs.

Full Text

Full text is available at the source.

View full text ↗

EEG-Based Feature Classification Combining 3D-Convolutional Neural Networks with Generative Adversarial Networks for Motor Imagery

Abstract

Full Text

You found one interesting study. We’ll send the next 7.

what lands in your inbox each week:

Recent issues from the circadian biology brief

Abstract

Full Text

Related papers

You found one interesting study. We’ll send the next 7.

what lands in your inbox each week:

Recent issues from the circadian biology brief