Electroencephalogram
215
spectrum are characteristic of problem solving and complex tasks. Lower frequencies such as delta and theta waves are prevalent in children and during rest. There are
several standard ways of providing calibrated tests on the functioning of the brain.
Many tests involve sensory stimuli and specific restrictive assignments to generate
EPs. Typical diagnostic applications of EEG analysis are as follows: diagnosis of
epilepsy, monitoring of brain activity of patients under anesthesia, analyzing sleep
disorders, and initial tests for identifying suspected brain tumor.
PROBLEMS
10.1 Import the EEG signal in the file “p_10_1.xls” and plot the 10 channels.* The
sampling rate was 256 samples per second with 3.9 ms period.
a. D etermine the dominant frequency of channel 0 and compare this to the
dominant frequency of channel 8.
b. In channel 0, using the power spectra of the signal, locate the motion artifacts that are characterized by a gradual change in the trend (DC offset)
of the signal. This artifact is resulting from either electrode movement or
muscle action.
10.2 U se the MATLAB ® to read the file “p_10_2.xls” and plot the four graphs.
The sampling rate of this data file is 100 Hz, and the file spans 10 s worth
of data. The data are of a suspected petit mal assault in a 13-year-old
patient. †
a. D etect the spike-and-sharp-wave complexes with wavelet analysis and
determine the duration of these events.
b. Using the duration of these events, and by considering the typical frequency or duration of petit mal patterns, verify that these assaults fall in
the petit mal category.
10.3 Import the data in file “p_10_3.xls” and plot the EEGs. The sampling rate of
the data was 173.61 Hz. The EEG pattern of “p_10_3.xls” represents a grand
mal assault. ‡
a. Determine the normal EEG frequency spectrum.
b. Determine the onset of the epileptic EEG pattern.
c. Plot the power spectrum of the signal.
d. Ca lculate the fractal dimension, signal complexity, and signal mobility of this signal and compare it with the same values for the signal in
Problem 10.4.
10.4 Import the data in file “p_10_4.xls” and plot the single EEG. This file shows
tonic–clonic or grand mal assault seizures. The data were recorded with a C4
electrode on the right central portion of the scalp with the earlobe as reference.
* The data file “p_10_1.xls” is a portion of 64 recordings of a study by Dr. Henri Begleiter at the
Neurodynamics Laboratory at the State University of New York Health Center at Brooklyn in the
effects of alcoholism on the brain activity and on the EEG.
† The data file is from Alpo Vaerri and a research team at the Tampere University of Technology,
Tampere, Finland.
‡ Courtesy of Ralph G. Andrzejak, Department while at the Department of Epileptology of the
University of Bonn, Bonn, Germany.
215
spectrum are characteristic of problem solving and complex tasks. Lower frequencies such as delta and theta waves are prevalent in children and during rest. There are
several standard ways of providing calibrated tests on the functioning of the brain.
Many tests involve sensory stimuli and specific restrictive assignments to generate
EPs. Typical diagnostic applications of EEG analysis are as follows: diagnosis of
epilepsy, monitoring of brain activity of patients under anesthesia, analyzing sleep
disorders, and initial tests for identifying suspected brain tumor.
PROBLEMS
10.1 Import the EEG signal in the file “p_10_1.xls” and plot the 10 channels.* The
sampling rate was 256 samples per second with 3.9 ms period.
a. D etermine the dominant frequency of channel 0 and compare this to the
dominant frequency of channel 8.
b. In channel 0, using the power spectra of the signal, locate the motion artifacts that are characterized by a gradual change in the trend (DC offset)
of the signal. This artifact is resulting from either electrode movement or
muscle action.
10.2 U se the MATLAB ® to read the file “p_10_2.xls” and plot the four graphs.
The sampling rate of this data file is 100 Hz, and the file spans 10 s worth
of data. The data are of a suspected petit mal assault in a 13-year-old
patient. †
a. D etect the spike-and-sharp-wave complexes with wavelet analysis and
determine the duration of these events.
b. Using the duration of these events, and by considering the typical frequency or duration of petit mal patterns, verify that these assaults fall in
the petit mal category.
10.3 Import the data in file “p_10_3.xls” and plot the EEGs. The sampling rate of
the data was 173.61 Hz. The EEG pattern of “p_10_3.xls” represents a grand
mal assault. ‡
a. Determine the normal EEG frequency spectrum.
b. Determine the onset of the epileptic EEG pattern.
c. Plot the power spectrum of the signal.
d. Ca lculate the fractal dimension, signal complexity, and signal mobility of this signal and compare it with the same values for the signal in
Problem 10.4.
10.4 Import the data in file “p_10_4.xls” and plot the single EEG. This file shows
tonic–clonic or grand mal assault seizures. The data were recorded with a C4
electrode on the right central portion of the scalp with the earlobe as reference.
* The data file “p_10_1.xls” is a portion of 64 recordings of a study by Dr. Henri Begleiter at the
Neurodynamics Laboratory at the State University of New York Health Center at Brooklyn in the
effects of alcoholism on the brain activity and on the EEG.
† The data file is from Alpo Vaerri and a research team at the Tampere University of Technology,
Tampere, Finland.
‡ Courtesy of Ralph G. Andrzejak, Department while at the Department of Epileptology of the
University of Bonn, Bonn, Germany.