Graph Neural Networks on SPD Manifolds for Motor Imagery Classification: A Perspective from the Time-Frequency Analysis
arXiv:2211.02641 · doi:10.1109/TNNLS.2023.3307470
Abstract
The motor imagery (MI) classification has been a prominent research topic in brain-computer interfaces based on electroencephalography (EEG). Over the past few decades, the performance of MI-EEG classifiers has seen gradual enhancement. In this study, we amplify the geometric deep learning-based MI-EEG classifiers from the perspective of time-frequency analysis, introducing a new architecture called Graph-CSPNet. We refer to this category of classifiers as Geometric Classifiers, highlighting their foundation in differential geometry stemming from EEG spatial covariance matrices. Graph-CSPNet utilizes novel manifold-valued graph convolutional techniques to capture the EEG features in the time-frequency domain, offering heightened flexibility in signal segmentation for capturing localized fluctuations. To evaluate the effectiveness of Graph-CSPNet, we employ five commonly-used publicly available MI-EEG datasets, achieving near-optimal classification accuracies in nine out of eleven scenarios. The Python repository can be found at https://github.com/GeometricBCI/Tensor-CSPNet-and-Graph-CSPNet.
arXiv comment15 pages, 5 figures, 6 Tables; This work has been accepted by the IEEE Transactions on Neural Networks and Learning Systems, 2023. Copyright will be transferred without notice, after which this version may no longer be accessible
References in corpus (14)
- A Comprehensive Survey on Graph Neural Networks
- EEGNet: A Compact Convolutional Network for EEG-based Brain-Computer Interfaces
- Deep learning with convolutional neural networks for EEG decoding and visualization
- Geometric deep learning: going beyond Euclidean data
- Hyperbolic Graph Convolutional Neural Networks
- Riemannian Adaptive Optimization Methods
- FBCNet: A Multi-view Convolutional Neural Network for Brain-Computer Interface
- Federated Transfer Learning for EEG Signal Classification
- Tensor-CSPNet: A Novel Geometric Deep Learning Framework for Motor Imagery Classification
- Riemannian batch normalization for SPD neural networks
- Geoopt: Riemannian Optimization in PyTorch
- MAtt: A Manifold Attention Network for EEG Decoding
- SPD domain-specific batch normalization to crack interpretable unsupervised domain adaptation in EEG
- Score-Based Data Generation for EEG Spatial Covariance Matrices: Towards Boosting BCI Performance