Time-frequency optimized spatial patterns for movement-related EEG decoding

Author

Bian Wu ; Yiwen Wang ; Weidong Chen ; Xiaoxiang Zheng

Author_Institution

Dept. of of Biomed. Eng., Zhejiang Univ., Hangzhou, China

fYear

2011

fDate

April 27 2011-May 1 2011

Firstpage

84

Lastpage

87

Abstract

The article presents a new method for motor-related EEG recognition which comprehensively optimizes the frequency-time-space features in a user-specific way. The method creates optimized time and frequency grids and adaptively selects channels for common spatial pattern (CSP) filters to enhance its power. The results show that the optimized features denotes a specific perspective to visualize frequency-time-spatial characteristics of motor-related EEGs, and can be used to achieve high classification accuracy.

Keywords

biomechanics; decoding; electroencephalography; medical signal processing; signal classification; time-frequency analysis; common spatial pattern filters; frequency-time-space features; high classification accuracy; motor-related EEG recognition; movement-related EEG decoding; time-frequency optimized spatial patterns; Accuracy; Classification algorithms; Electroencephalography; Fingers; Merging; Pressing; Time frequency analysis;

fLanguage

English

Publisher

ieee

Conference_Titel

Neural Engineering (NER), 2011 5th International IEEE/EMBS Conference on

Conference_Location

Cancun

ISSN

1948-3546

Print_ISBN

978-1-4244-4140-2

Type

conf

DOI

10.1109/NER.2011.5910494

Filename

5910494