EEG Analysis Using ICA to Classify the Pattern

Only available on StudyMode
  • Topic: Electroencephalography, Matrices, Data
  • Pages : 9 (2082 words )
  • Download(s) : 537
  • Published : February 12, 2011
Open Document
Text Preview
EEG analysis using ICA to classify the pattern

Independent component analysis (ICA) is a method with the help of which we can find a linear representation of mixed nongaussian EEG data so that the components are statistically independent. The EEG is composed of electrical potentials arising from several sources. Eye blink artifacts and power line noise always disturb the electroencephalogram (EEG) recorded on the scalp. In this paper, Independent Component Analysis (ICA) is applied to extract eye movements and power noise of 50Hz. The effect of centering and sphering on the EEG data is studied in this paper.

Independent Component Analysis, electroencephalogram, artifacts, mixing matrix, Whitening, Sphering, uncorrelation, centering.

Electroencephalogram is the method from which the electrical activity of the brain is checked. For this electrical signals are recorded using electrodes placed on the scalp. Analysis of electroencephalogram signals is useful for diagnosis of many neurological diseases such as epilepsy , tumors , problems associated with trauma. For appropriate diagnosis of any disease analysis of the EEG signals should be perfect . For appropriate analysis we must remove the noise due to facial muscle movements , eye blinking etc., also called as ‘ARTIFACTS’. The externally recorded EEG is a highly attenuated and mixed signal since it originates from the activity of thousands of neurons, which passes through different tissue layers before reaching the recording electrodes. An important problem in EEG analysis is the mixing of the signals with various other signals , such as eye blinks , cardiac signals , line frequency noise facial muscle movements . These signals must be removed prior to any analysis. One method of achieving this is via thresholding, where any data whose amplitude exceeds a set threshold is cut off. One of the disadvantage of thresholding is the loss of valuable EEG signal. Another method for analysis and processing of the EEG signal is Independent Component Analysis (ICA) by which we are going to analyze the EEG signal.

In an EEG signal taken from sensors some artifacts like eye blinks are treated as noise or ‘unwanted signals’. Improving the ‘signal-to-noise ratio’ by filtering off the ‘noise’ could enhance the performance of the subsequent feature extraction. Independent component Analysis can be used to remove the artifacts. For application of ICA algorithm some assumptions should be considered. Some of the assumptions are as follows- (i)The recorded signals are a result of linear mixing.

(ii)The artifacts are statistically independent from the EEG signals. (iii) The number of sources are same as the number of mixtures. Independent Component Analysis in its application to EEG analysis, assume that EEG signal is composed of a finite number of components .

[pic] (1)

Here ‘t’ is a discrete time index, ‘n’ is the number of components which are mixed through unknown linear mixing process. This mixing process is described by mixing matrix ‘H’ .The elements [pic] are the weights of the mixing matrix, [pic]indicates mixing coefficient from the [pic] source to the [pic]electrode.

[pic] (2)

Here ,‘S’ consist of the signals from sources and the rows of matrix ‘x’ are the signals from sensors. X(t) are the mixed signals at the output of sensors. The above equation can be shown in the matrix form as follows:

S1h11 h12 h13…………………h1nX1
tracking img