Howard Civil Service International House, Taipei, Taiwan, April 9-11, 2015
[2]
Noise Reduction and Brain Mapping based Robust
Principal Component Analysis
Arjon Turnip
Abstract—Analysis of EEG activity usually raises the problem value(there has been an ample amount of research toward this
of differentiating between genuine EEG activity and that which is goal)[8-13]. Many of these newer approaches involve
introduced through a variety of external influence. These techniques including independent component analysis, neural
artifacts may affect the outcome of the EEG recording. In this networks, Kohonen maps, principal component analysis
paper, wavelet denoising and band pass filter for preprocessing (PCA),and other methods which were either unavailable or
and a robustprincipal component analysis algorithm for much less well known during the early days of EEG signal
extraction are proposed to remove the artifacts. The algorithm is processing. Techniques for online cancelation of EEG noise
designed to adaptively derive a relatively small number of without rejecting other data are remains difficult in EEG
decorrelated linear combinations of a set of random zero-mean
processing software but necessary one for BCI designers.In
variables while retaining as much of the information from the
this paper, we introduce a novel method (i.e., robust principal
original variables as possible. The proposed method was tested in
real EEG records acquired from eight subjects. The experimental
component analysis (RPCA)) for automatic noise cancelation
result show that the proposed method can effectively remove the which can significantly enhance prominence of the spike in
artifacts from all subjects. the clean EEG signals. This method is based on separation of
the raw multichanel EEG recordings into a relatively small
Keywords—Noise, Artifacts, Principal Component Analysis, number of decorrelated linear combinations of a set of
EEG, BCI random zero-mean variables while retaining as much of the
information from the original variables as possible.
I. Introduction
When brain cells (neurons) are activated, the synaptic
II. DataAcquisition
currents are produced within the dendrites. This current In the experiment, eight healthy adult subjects (all men,
generates a magnetic field measurable by electromyogram ranging in age between 20-22 old years) participated in this
(EMG) machines and a secondary electrical field over the study, which are 5 subjects have slight hair and the rest are
scalp measurable by Electroencephalogram (EEG)systems thick hair. All of the subjects do same experiment with three
[1].EEG is the noninvasive measurement of the electrical stimuluswhich are baseline, closed eye, and blink eye
activity on the scalp over multiple areas of the brain. The condition. It has been made sure that every participant was not
measured of currents that flow during synaptic excitations of in a stress mental condition while the experiment was being
the dendrites of many pyramidal neurons in the cerebral cortex conducted. All participant picture show in Fig.1.During the
is called EEG signal. EEG signal, which is important in experiment, allparticipant were sitting in a comfortable chair
clinical application(i.e., for diagnosing, monitoring, and in front of 14” monitor at a distance of about 1 m. To build the
[2] latar managing neurological disorders) and in research field (i.e., experiment setting used in the EEG sample collecting process,
belakang brain computer interface (BCI) application), is widely affected we utilize the OpenVibe software to perform the data
prgf 1 by a variety of large signal contaminations or noise. Artifacts acquisition, stimulus visualization, and EEG recording with
can dramatically alter the signal recorded at all scalp sites, the function block that already built in. This experiment
especially those closest to the source of the noise [2-7]. Hence, consists of two sessions, first session had a period of 130
a necessary stage in EEG processing is artifact or noise seconds with 2stimulus(i.e., baseline (eye normal) and closed
cancelation. Determining what is considered artifact, how eye condition). Second sessions had a period of 66 seconds
much artifact is excessive, and removing artifacts from real with 2 others stimulus (i.e., baseline (eye normal) and blink
data require a skilled human or a sophisticated artificial eye condition).The EEG signals are recorded continuously
system [7]. using six electrodes (channels) at F7, F8, T7, T8, O1 and O2
that represent the visual of human brain and digitized at a 128
In a BCI application that removed all data containing Hz sampling rate.The six electrodes configuration is shown in
artifact might be left with too little clean data to be of practical Fig. 2.
use. Therefore,the use of noise cancelation method to the EEG
records and leaving clean data would be of tremendous
1
978-1-4799-8069-7/15/$31.00 ©2015 IEEE 550
∑ , (i=1,2,...,n) (2)
where , , .The
minimization of these cost function ns by the gradient descent
technique leads to an adaptive learn
ning algorithm
(3)
(4)
or in a discrete-time version as
1 , (5)
with 0 0, 0.
2
551
Since the artifacts can randomly occur and are unexpected,
F8 they are difficult to identify. Thus, instead of detecting and
removing artifacts, our approach is to extract the event-related
T8
components based on a global pattern that encapsulates
C hannels
T8
O2 V. Conclusion
Channels
4
Channels
Acknowledgment
3
This research was supported by the thematic program
2 through the Bandung Technical Management Unit for
Instrumentation Development (Deputy for Scientific Services)
1
and the excellent program through the Research Center for
Physics (Deputy for Engineering Sciences) funded by
0 1 μ volt 20 40 60
Time [s]
80 100 120
Indonesian Institute of Sciences, Indonesia.
Fig. 4 Extracted EEG signals using robust PCA based moving average. References
Ch1 Ch2 Ch3 [1] S. Sanei, J.A. Chambers, EEG Signal Processing, USA: John Wiley and
Sons, 2007.
[2] A. Jafarifarmand, M.A Badamchizadeh, “Artifacts removal in EEG
signal using a new neural network enhanced adaptive filter”,
Neurocomputing, vol. 103, pp. 222–23, 2013.
[3] G. Geetha and S. N. Geethalakshmi,“Artifact Removal from EEG using
Spatially Constrained Independent Component Analysis and Wavelet
Denoising with Otsu’s Thresholding Technique”, Procedia Engineering,
vol. 30, pp. 1064 – 1071, 2012.
Ch4 Ch5 Ch6
+ [4] S. O’Regan, S. Faul, and W. Marnane,“Automatic detection of EEG
artefacts arising from head movements using EEG and gyroscope
signals”, Medical Engineering & Physics, vol. 35, pp. 867– 874, 2013.
[5] K. J. Mullinger, J. Havenhand, and R. Bowtell,“Identifying the sources
of the pulse artefact in EEG recordings made inside an MR scanner”,
NeuroImage, vol. 71,pp. 75–83, 2013.
- [6] M. E. H. Chowdhury, K. J. Mullinger, P. Glover, and R.
Bowtell,“Reference layer artefact subtraction (RLAS): A novel method
Fig. 5 Coresponding spike of waveforms of extracted EEG signals (in Fig. 5).
3
552
of minimizing EEG artefacts during simultaneous fMRI”, NeuroImage, [11] A. Turnip & K.-S. Hong, “Classifying mental activities from EEG-P300
vol. 84,pp. 307–319, 2014. signals using adaptive neural network”, Int. J. Innov. Comp. Inf.
[7] B. Allison, P3 or not P3: Toward a better P300 BCI. University of Control, 8(7), 2012.
California, San Diego, 2003. [12] A. Turnip, K.-S. Hong & M.-Y. Jeong, “Real-time feature extraction of
[8] C. Babiloni, F. Vecchio, F. Infarinato, P. Buffo, N. Marzano, D. Spada, P300 component using adaptive nonlinear principal component
S. Rossi, I. Bruni, P. M. Rossini, & D. Perani, “Simultaneous recording analysis”, BioMedical Engineering OnLine, 10(83), 2011.
of electroencephalographic data in musicians playing in ensemble”, [13] K. I. Diamantaras and S.Y. Kung, Principal Component Neural
Cortex, 47(9), pp. 1082-1090, 2011. Networks. Theory and Applications.Adaptive and Learning Systems for
[9] J. Escudero, R. Hornero, D. Abasolo, A. Fernandez, “Quantitative Signal Processing, Communications and Control.John Wiley & Sons
evaluation of artifact removal in real magnetoencephalogram signals Inc., New York, 1996.
with blind source separation”, Annals of Biomedical Engineering, 39(8), [14] A. Cichocki and R. Unbehauen, “Robust estimation of principal
pp. 2274-2286, 2011. components in real time”, Electronics Letters, 29(21): pp. 1869–1870,
[10] W. O. Tatum, B. A. Dworetzky & D. L. Schomer, “ Artifact and 1993.
recording concepts in EEG”, Journal of Clinical Neurophysiology, [15] M. Falkenstein, J. Hoormann, and J. Hohnsbein,“ERP components in
28(3), pp. 252-263, 2011. Go/Nogo tasks and their relation to inhibition”, Acta Psychologica, vol.
101, No. 2-3, pp. 267-291, 1999.
4
553