Spectral analysis and data classification in magnetoencephalography
A. V. Derguzov, Sergey Aleksandrovich Makhortykh · Pattern Recognition and Image Analysis · 2006
A method for classifying types of brain activity in magnetoencephalographic (MEG) signals is proposed. Sources of abnormal cortical activity are localized by performing a generalized spectral analysis in the space of Fourier coefficients of the expansions of recorded signals in spherical harmonics. The basic principles of the method are discussed, and the results of its application to actual MEG records are presented in the case of Parkinson’s disease.