Although significant advances have been made in classification of electroencephalograms (EEGs) in biomedical signal analysis, supervised approaches require high numbers of labeled data which are expensive and challenging to collect in medical environments. Self-supervised learning has become an exciting alternative, as it can be used to obtain useful representation learning from recordings that are not labeled. In this paper, a self-supervised geometric learning framework is proposed, which combines the Symmetric Positive Definite (SPD) manifold representation and contrastive learning objective for robust EEG classification. Multi-channel EEG-based covariance structures are projected onto Riemannian manifolds, where the nonlinear relationships between neural signals are preserved. The objective, which is a combination of the geometric and self-supervised objectives, significantly decreases the dependency on annotation by 65% and achieves a classification accuracy of 98.2%, outperforming CNN, Transformer and fully supervised baselines. The proposed framework shows high generalizability and is applicable to motor imagery, mental arithmetic and clinical brain-computer interface (BCI) tasks.
Rear wings, front wings and floor edges are examples of or more distinctive parts of the car's struc...
In F1 vehicles, Aero-elastic wing flutter is a serious concern on the integrity of the wing structur...
The classification methods based on EEG have been shown to be valuable for EEG diagnosis of neurolog...
In this paper, we have come to know that the epilepsy is a common neural disorder in children which ...
