Single trial EEG classification of observed wrist movements

• Published in: 2009 4th International IEEE/EMBS Conference on Neural Engineering pp. 359 - 362
• Main Authors: Lakany, H., Valsan, G., Conway, B.A.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01-04-2009
• Subjects: Brain computer interface; Brain computer interfaces; Data mining; Decoding; EEG; Electrodes; Electroencephalography; Humans; Neurons; Scalp; single trial classification; Time frequency analysis; Wrist; wrist movements
• ISBN: 1424420725; 9781424420728
• ISSN: 1948-3546; 1948-3554
• DOI: 10.1109/NER.2009.5109307

In this paper, we present the results of single trial EEG classification of observed wrist movements. This study is part of our endeavour to develop brain computer interfaces as an assistive device for people with severe motor disabilities. Our methods rely on a simple but robust algorithm that requires no subject training to modulate brain activity. We adopt a method based on extraction and selection of statistically significant time-frequency features using ANOVA and principal component analysis. Classification results achieved ~ 80% (plusmn12 %).