Imagined wrist movement classification in single trial EEG for brain computer interface using wavelet packet

In kinaesthetic imagery, healthy subjects simulate the disability of muscle inactivity by imagining movements. EEG recordings from five healthy subjects were studied to classify the imagined left and right wrist movements performed in four directions (extension, pronation, flexion and supination). Wavelet Packet Transform (WPT) was used to extract features based on normalised energy and average amplitude for the best basis (selected by using local discriminant method). Radial Basis Function classifier was used and a four-fold cross validation was performed. The overall classification accuracy of 83% was achieved by using a vector set comprising of 10 best discriminatory features.

[1]  J. Gotman,et al.  Wavelet based automatic seizure detection in intracerebral electroencephalogram , 2003, Clinical Neurophysiology.

[2]  T. Kailath.  The Divergence and Bhattacharyya Distance Measures in Signal Selection , 1967 .

[3]  G. Pfurtscheller,et al.  Optimal spatial filtering of single trial EEG during imagined hand movement. , 2000, IEEE transactions on rehabilitation engineering : a publication of the IEEE Engineering in Medicine and Biology Society.

[4]  Lei Ding,et al.  Motor imagery classification by means of source analysis for brain–computer interface applications , 2004, Journal of neural engineering.

[5]  Aleksandar Ko.  FEATURE EXTRACTION IN DEVELOPMENT OF BRAIN-COMPUTER INTERFACE: A CASE STUDY , 1998 .

[6]  E Donchin,et al.  Brain-computer interface technology: a review of the first international meeting. , 2000, IEEE transactions on rehabilitation engineering : a publication of the IEEE Engineering in Medicine and Biology Society.

[7]  F. Sepulveda,et al.  A Comparison of Time, Frequency and ICA Based Features and Five Classifiers for Wrist Movement Classification in EEG Signals , 2005, 2005 IEEE Engineering in Medicine and Biology 27th Annual Conference.

[8]  C. A. Grimbergen,et al.  HIGH QUALITY RECORDING OF BIOELECTRIC EVENTS . I : INTERFERENCE REDUCTION , THEORY AND PRACTICE , 2009 .

[9]  H. Adeli,et al.  Analysis of EEG records in an epileptic patient using wavelet transform , 2003, Journal of Neuroscience Methods.

[10]  A. P Bradley,et al.  On wavelet analysis of auditory evoked potentials , 2004, Clinical Neurophysiology.

[11]  Jie Deng,et al.  Classification of the intention to generate a shoulder versus elbow torque by means of a time–frequency synthesized spatial patterns BCI algorithm , 2005, Journal of neural engineering.

[12]  Clemens Brunner,et al.  Mu rhythm (de)synchronization and EEG single-trial classification of different motor imagery tasks , 2006, NeuroImage.

[13]  Francisco Sepulveda,et al.  EEG Single-trial classification of four classes of imaginary wrist movements based on Gabor coefficients , 2006 .

[14]  Chng Eng Siong,et al.  High accuracy classification of EEG signal , 2004, ICPR 2004.

[15]  F. Sepulveda,et al.  Cluster separation index suggests usefulness of non-motor EEG channels in detecting wrist movement direction intention , 2004, IEEE Conference on Cybernetics and Intelligent Systems, 2004..

[16]  R Quian Quiroga,et al.  Wavelet Transform in the analysis of the frequency composition of evoked potentials. , 2001, Brain research. Brain research protocols.

[17]  Ronald R. Coifman,et al.  Local discriminant bases and their applications , 1995, Journal of Mathematical Imaging and Vision.

[18]  G Pfurtscheller,et al.  Separability of EEG signals recorded during right and left motor imagery using adaptive autoregressive parameters. , 1998, IEEE transactions on rehabilitation engineering : a publication of the IEEE Engineering in Medicine and Biology Society.