Title :
Geometric subspace methods and time-delay embedding for EEG artifact removal and classification
Author :
Anderson, Charles W. ; Knight, James N. ; O´Connor, Tim ; Kirby, Michael J. ; Sokolov, Artem
Author_Institution :
Dept. of Comput. Sci., Colorado State Univ., Fort Collins, CO, USA
fDate :
6/1/2006 12:00:00 AM
Abstract :
Generalized singular-value decomposition is used to separate multichannel electroencephalogram (EEG) into components found by optimizing a signal-to-noise quotient. These components are used to filter out artifacts. Short-time principal components analysis of time-delay embedded EEG is used to represent windowed EEG data to classify EEG according to which mental task is being performed. Examples are presented of the filtering of various artifacts and results are shown of classification of EEG from five mental tasks using committees of decision trees.
Keywords :
decision trees; delays; electroencephalography; filtering theory; medical signal processing; principal component analysis; signal classification; singular value decomposition; EEG artifact removal; decision trees; filtering; generalized singular-value decomposition; geometric subspace methods; mental task; multichannel electroencephalogram; short-time principal components analysis; signal classification; time-delay embedding; Classification tree analysis; Data mining; Decision trees; Electrodes; Electroencephalography; Filtering; Filters; Optimization methods; Principal component analysis; Signal analysis; Artifact; brain–computer interface (BCI); classification; electroencephalogram (EEG); principal components analysis; time-delay embedding; Algorithms; Artifacts; Brain; Communication Aids for Disabled; Electroencephalography; Evoked Potentials; Humans; Man-Machine Systems; Pattern Recognition, Automated; User-Computer Interface;
Journal_Title :
Neural Systems and Rehabilitation Engineering, IEEE Transactions on
DOI :
10.1109/TNSRE.2006.875527
