Riemannian classification of EEG signals with missing values
This work addresses missing data handling in EEG classification for brain-computer interfaces, representing an incremental improvement over existing methods.
The paper tackled the problem of classifying EEG signals with missing values by proposing a strategy using observed-data likelihood within an EM algorithm, which generally performed better than existing methods on real EEG data for brain-computer interface tasks.
This paper proposes a strategy to handle missing data for the classification of electroencephalograms using covariance matrices. It relies on the observed-data likelihood within an expectation-maximization algorithm. This approach is compared to two existing state-of-the-art methods: (i) covariance matrices computed with imputed data; (ii) Riemannian averages of partially observed covariance matrix. All approaches are combined with the minimum distance to Riemannian mean classifier and applied to a classification task of two widely known paradigms of brain-computer interfaces. In addition to be applicable for a wider range of missing data scenarios, the proposed strategy generally performs better than other methods on the considered real EEG data.