Decomposition and classification of electroencephalography data

1 In this study, we aim to automatically identify multiple artifact types in EEG. 2 We used multinomial regression to classify independent components of EEG data, select3 ing from 65 spatial, spectral, and temporal features of independent components using forward 4 selection. The classifier identified neural and five non-neural types of components. 5 Between subjects within studies, high classification performances were obtained. Between 6 studies, however, classification was more difficult. For neural vs. non-neural classifications, 7 performance was on par with previous results obtained by others. 8 We found that automatic separation of multiple artifact classes is possible with a small 9 feature set. 10 Our method can reduce manual workload and allow for the selective removal of artifact 11 classes. Identifying artifacts during EEG recording may be used to instruct subjects to refrain 12 from activity causing them. 13