Juliette Spinnato – J. Spinnato (I2M) : Finding EEG Space-time-scale localized features using Matrix-based penalized discriminant analysis

Carte non disponible

Date/heure
Date(s) - 13 juin 2014

Catégories Pas de Catégories


Finding EEG Space-time-scale localized features using Matrix-based penalized discriminant analysis\nBy Juliette Spinnato, I2M\n\nIn this talk we will explain a new method for constructing and selecting of discriminant space-time-scale features for electroencephalogram\n(EEG) signal classification, suitable for Error Related Potentials (ErrP) detection in brain-computer interface (BCI). The method rests\non a new variant of matrix-variate Linear Discriminant Analysis (LDA), and differs from previously proposed approaches in mainly\nthree ways.\nFirst, a discrete wavelet expansion is introduced for mapping time-courses to time-scale coefficients, yielding time-scale localized features.\nSecond, the matrix-variate LDA is modified in such a way that it yields an interesting duality property, that makes interpretation easier.\nThird, a space penalization is introduced using a surface Laplacian, so as to enforce spatial smoothness.\nThe proposed approaches, termed D-MLDA and D-MPDA are tested on EEG signals, with the goal of detecting ErrP. Numerical results\nshow that D-MPDA outperforms D-MLDA and other matrix-variate LDA techniques. In addition this method produces relevant features\nfor interpretation in ErrP signals.[

Juliette Spinnato – J. Spinnato (I2M) : Finding EEG Space-time-scale localized features using Matrix-based penalized discriminant analysis

Carte non disponible

Date/heure
Date(s) - 13 juin 2014

Catégories Pas de Catégories


Finding EEG Space-time-scale localized features using Matrix-based penalized discriminant analysis\nBy Juliette Spinnato, I2M\n\nIn this talk we will explain a new method for constructing and selecting of discriminant space-time-scale features for electroencephalogram\n(EEG) signal classification, suitable for Error Related Potentials (ErrP) detection in brain-computer interface (BCI). The method rests\non a new variant of matrix-variate Linear Discriminant Analysis (LDA), and differs from previously proposed approaches in mainly\nthree ways.\nFirst, a discrete wavelet expansion is introduced for mapping time-courses to time-scale coefficients, yielding time-scale localized features.\nSecond, the matrix-variate LDA is modified in such a way that it yields an interesting duality property, that makes interpretation easier.\nThird, a space penalization is introduced using a surface Laplacian, so as to enforce spatial smoothness.\nThe proposed approaches, termed D-MLDA and D-MPDA are tested on EEG signals, with the goal of detecting ErrP. Numerical results\nshow that D-MPDA outperforms D-MLDA and other matrix-variate LDA techniques. In addition this method produces relevant features\nfor interpretation in ErrP signals.[