Seigo Toi

Quantification of EEG irregularity by use of the entropy of the power spectrum

A new method for quantifying irregularity of EEGs is proposed in this study. The entropy, an information measure, determines the uniformity of proportion distribution. The peakedness or flatness of the distribution of the EEG power spectrum, representing EEG rhythmicity, can be measured by the entropy, because the power spectrum consists of proportions of power at each frequency. The irregularity of the EEG was measured by the entropy of the power spectrum, called an irregularity index (II). The II was obtained from the power spectrum at F3, F4, C3, C4, P3, P4, O1 and O2 during rest and mental arithmetic in 10 normal subjects. Relative band powers of delta, theta, alpha and beta bands and alpha peak frequency were also obtained. EEGs during rest were significantly more irregular anteriorly than in the occipital areas. Alpha activity was also more irregular in the anterior region. A greater degree of EEG desynchronization during mental arithmetic was found over the left hemisphere and the right occipital area. The II was more sensitive to such desynchronization than alpha band power and alpha peak frequency. The differences in spectral structures between rest and mental arithmetic conditions, mainly over the left hemisphere, were also confirmed by the Kullback-Leibler information.

Analysis of rapidly changing EEGs before generalized spike and wave complexes

Ten epileptic patients who had absence and tonic-clonic seizures were examined. They showed almost generalized 2-5 Hz spike and wave complexes (SWCs) with maxima at the frontal or central locations. Their EEGs were recorded from F3, F4, C3, C4, P3, P4, O1 and O2. One 15 sec EEG epoch, which included the background activity followed by SWCs, was divided into 20 segments. On the basis of Akaikes information criterion obtained from an AR model fitted to each segment by the least squares method, a distance measure between segments was obtained. The 20 segments were classified according to the distance measure with the combined use of both cluster analysis and multidimensional scaling. The power spectrum was also obtained for each segment. There were 2-6 clusters at some locations, always including the largest amplitude locations of the SWCs (the frontal or central locations). The clusters could be grouped into 3 major types, each forming a period: (1) one large cluster distant from the SWCs (the period of background activity); (2) small clusters just before the SWCs (the period just before the SWCs); (3) small clusters during the SWCs (the period of the SWCs). The period just before the SWCs, which occurred earliest at the largest amplitude location, may be a transition state of the background activity to the SWCs. A few segments remote from the SWCs belonged to the small cluster just before or during the SWCs in some patients, thus suggesting that an EEG event similar to EEG changes just before or during the SWCs occurred. The EEG event buried within the background can be considered as a poorly developed epileptiform discharge. A gradual increase or decrease in alpha frequency before the SWCs was found in most patients; this suggests that changes in the level of vigilance occur before SWCs.

Potential flow of frontal midline theta activity during a mental task in the human electroencephalogram

The movement of potential field (potential flow) of frontal midline theta activity (Fm theta) as well as its potential distribution was examined in 7 subjects by using optical flow detection techniques in image processing. Electroencephalograms (EEGs) over the fronto-central region were recorded from 13 electrodes near the frontal midline (Fz) while the subjects were performing a mental task. The potential flow of Fm theta was estimated on a frame consisting of a square grid with Fz at its center. In regions anterior to Fz, the direction of potential flow was from lateral to medial, whereas it was from medial to lateral in regions posterior to Fz. The peak-to-trough amplitude distribution was round or oval with a maximum just anterior to Fz. The source density distribution showed the greatest potential along the midline in the frontal region and bilaterally symmetric smaller maximum areas mostly in the prefrontal regions. Our findings suggested the presence of 2 different source areas of opposite direction in each hemisphere in spite of a round or oval amplitude distribution.

A new segmentation method of electroencephalograms by use of Akaike's information criterion

Non-stationary EEGs, whose statistical properties change with time, were segmented into stationary segments to closely track the behavior of EEG characteristics. We have developed a new segmentation method of optimizing segmentation parameters by using AIC (Akaikes information criterion) as an objective criterion. We applied the segmentation method to EEGs. The instantaneous power spectra of EEGs estimated with wavelet transform were compared with the segmented EEGs. EEGs were recorded from F3, F4, C3, C4, P3, P4, 01 and 02 in 13 normal subjects. Artifact-free 15-s epochs were taken at each electrode location. Each epoch was divided into stationary segments, consisting of several fixed intervals, by optimizing 2 segmentation parameters (interval length and starting point) so that the sum of AICs for several sequences of segments could be the smallest. The EEG segmentation could represent differences in the power spectra between segments. The average length of segments during relaxed wakefulness was 6.0 +/- 3.8 s. The EEG segmentation during mental arithmetic could detect the start of mental arithmetic.

INTER-SITE EEG RELATIONSHIPS BEFORE WIDESPREAD EPILEPTIFORM DISCHARGES

EEG interactions of the background among locations can start to change just before epileptiform discharges. Such interactions were investigated with relative power contribution analysis based on a multivariate autoregressive model, which permits determination of multiple causal relations of EEGs among locations. EEGs from F3, F4, P3, P4, T3 and T4 were examined in 10 epileptic patients with asymmetric spike and wave complexes (SWCs). A 12.5-s epoch just before SWCs was divided into stationary segments throughout 6 locations with a segmentation method. In segments long before SWCs, most power at each location was generated from its own location. In segments immediately preceding SWCs, contributions from other locations, particularly from the hemisphere with smaller SWCs, increased. Overall EEG relationships among 6 locations were examined by an entropy which measures the uniformness of the spatial distribution of power contribution. The entropy significantly increased gradually toward SWCs. Our findings demonstrated stronger interactions among locations just before epileptiform discharges, suggesting a transitional state from background EEG to epileptiform discharges.

Increases in the power spectral slope of background electroencephalogram just prior to asymmetric spike and wave complexes in epileptic patients

Moment-to-moment electroencephalogram (EEG) changes just before spike and wave complexes (SWCs) were investigated by using a non-stationary power spectral analysis in 10 epileptic patients with asymmetric SWCs. The instantaneous power spectra of background EEG were estimated for a 10-s period just before SWCs in 10 locations. The spectral shape, showing a l/f amplitude characteristic, was assessed by the negative slope (spectral slope) of a straight line on a plot of log power versus frequency. Spectral slope significantly increased toward SWCs at locations with greater SWCs. This finding suggested structural changes in EEG frequency composition just before SWCs because of an epileptogenic process.

Abnormality of background EEG determined by the entropy of power spectra in epileptic patients

Relationships between epileptiform discharges and background activity were examined by power spectral entropy (PSE), measuring a degree of EEG irregularity. The EEGs were recorded from 10 electrodes placed at F3, F4, C3, C4, T3, T4, P3, P4, O1 and O2 in 11 epileptic patients with widespread lateralized spike and wave complexes (SWCs). Bipolar records were also made from the antero-posterior derivations. The locations of the maximum PSE coincided with those of the maximum amplitude of SWC in most of the patients. Bipolar derivations with the maximum PSE always included the locations with the maximum PSE obtained from a linked ears reference. Pearsons correlation coefficient between PSE and SWC amplitude was 0.62 +/- 0.14 (mean +/- S.D.) in 11 patients, thus indicating that the scalp distribution of PSE was closely related to that of the amplitude of SWC. These findings suggest that the background EEG is disorganized in or near the epileptogenic focus. A focal background abnormality can therefore be estimated by PSE.

Changes in the fractal dimension of alpha envelope from wakefulness to drowsiness in the human electroencephalogram

Changes in the fractal dimension of the waxing and waning of alpha activity from wakefulness to drowsiness were investigated in 10 normal subjects. The envelope of filtered alpha activity was estimated with the Hilbert transform. A subset time series was constructed from the alpha envelope data taken at an interval (k). The points of the logarithm of interval k vs. the logarithm of the curve length of the subset time series (log L(k)) were fitted by a two-segment curve, consisting of small and long line segments (segments 1 and 2). The fractal dimension was estimated from the slope of straight line for each segment by using Higuchis method. The mean fractal dimension was 1.16 for segment 1 and 1.95 for segment 2 during wakefulness and 1.15 for segment 1 and 1.89 for segment 2 during drowsiness. The fractal dimension significantly decreased in both segments during drowsiness, thus, indicating decreases in the complexity of alpha envelope.

Moving potential field of frontal midline theta activity during a mental task

The movement of potential field (potential flow) of frontal midline theta activity (Fm theta) was examined by using optical flow detection techniques in image processing in 6 normal subjects. Electroencephalograms (EEGs) over the fronto-central region were recorded from 13 electrodes around Fz. The potential flow of Fm theta was estimated on a frame which consisted of a square grid of 19 x 19 points in which each of the 13 electrodes was positioned with Fz at the center of grid. The peak-to-trough amplitude distribution was rounded or oval with a maximum just anterior to Fz. In each hemisphere, potential field moved medially in a region anterior to Fz, posteriorly near or at the midline and laterally in a region posterior to Fz. In the temporal sequence of potential field, at each point velocity directions tended to change progressively clockwise or counterclockwise. In the whole frame, a rotary movement of potential flow was found with a center at the middle on each side. Our findings suggested that a potential maximum at the midline may be due to overlapping of fields of opposite direction over the frontal scalp, arising from sources of similar orientation in both hemispheres. It can be assumed that Fm theta propagates in a clockwise direction in the left hemisphere and in a counterclockwise direction in the right hemisphere.

Potential flow of a activity in the human electroencephalogram

Abstract The movement of potential field of a activity was investigated with a potential flow technique. a activity was divided into slow, medium and fast subbands. The peak-to-peak amplitude and source density distribution of each subband was also constructed. The potential flow of medium a was from medial to lateral in the anterior region, and from lateral to medial in the posterior region. The flow patterns of slow and fast a were more complicated. Similar flow directions formed a group. Different groups met or diverged at discontinuous grid points around the local maxima of potential distribution. The discontinuous points of medium a were centered in the parieto-occipital region, and those of slow and fast a were more equally distributed over the scalp. Against potential gradients, several groups of similar flow directions participated in the formation of potential field. The generation of potential field was assumed to be controlled widely for medium a and locally for slow and fast a.