Automated detection of EEG artifacts during sleep: preprocessing for all-night spectral analysis.

This paper describes a simple artifact detection algorithm which can be used when large amounts of EEG data are to be automatically processed via spectral analysis techniques in a general purpose digital computer, and visual inspection of each EEG epoch becomes an impossible task. The technique is based on a chi-square (chi(2)) goodness-of-fit test to a Gaussian distribution (CSQ), and it was applied to EEG epochs each 30 sec long. This test proved to be very sensitive to non-stationarities in the EEG amplitude distribution for a particular epoch, and it produced a large value for the chi(2) coefficient when an artifact was present. EEG epochs that gave rise to chi(2) coefficients of value larger than a heuristically determined minimum were discarded from further analysis. The above technique enabled efficient data reduction and reliable automatic off-line processing of 50 nights of sleep EEG via spectral techniques.     

EEG analysis of healthy adults in the highlands of Xining

An EEG analysis of 400 healthy adults in Xining, Qinghai Province, about 2260 meters above the sea level was made. A comparison of the data with those obtained from a healthy group of adults in the Beijing flatlands was carried out. It was found that (1) the incidence of abnormal EEG of the examinees in Xining was higher (32.25%) than that in Beijing, (2) the index in Xining was lower than that in Beijing, (3) the slow waves occurred more frequently in the Xining group, (4) the spikes were seen in 9.25% of the examined subjects and the sharp waves in 3.0% of the examined subjects in Xining, and (5) the physiological reactions were not so sensitive in the Xining group.     

Spectral analysis of sleep EEG in patients with restless legs syndrome.

OBJECTIVE: Conventional analyses of sleep EEG recordings according to standard criteria indicate severe sleep disturbances in patients with restless legs syndrome (RLS). Spectral analysis of sleep EEG may be a sensitive tool to detect functional alterations of sleep mechanisms beyond the visual scoring of polysomnographic records. We analysed sleep EEG spectral power differences between RLS patients and healthy subjects. Furthermore, we studied the relationship of sleep EEG spectral power to the occurrence of periodic leg movements in sleep (PLMS) and arousal events. METHODS: Sleep EEGs from 20 patients with idiopathic RLS and of 20 age and sex matched healthy subjects were investigated. The spectral analysis was carried out on the same 30s epochs for which sleep stages had been determined. As a first step, whole-night spectral power excluding epochs with an arousal or with a PLMS was compared separately for REM and NREM sleep between RLS and healthy subjects. In a second step, we evaluated the spectral effects of PLMS, PLMS with associated arousals and isolated arousals relative to epochs of sleep without such events in both groups. In this part of the analysis, we only included the epochs of sleep stage 2 (the main and most stable non-REM sleep stage) and of REM sleep. RESULTS: Spectral power of all sleep epochs (excluding arousals and PLMS) did not differ between patients with RLS and healthy subjects. As expected, arousals and PLMS-associated arousals resulted in a significant increase in higher-frequency activity (alpha, beta1, beta2 and gamma bands) in both groups. Spectral power in epochs with PLMS alone did not significantly differ from spectral power in epochs without PLMS and without arousal in any of the groups. CONCLUSIONS: We found no evidence for an altered cortical activity in sleep stage 2 and REM sleep epochs in RLS patients compared to that in healthy controls if epochs with arousals were not considered. Furthermore, while PLMS associated with an arousal have a high impact on EEG spectra, the effect of a PLMS without arousal seems to be minor and transient. SIGNIFICANCE: Our data suggest that RLS related symptoms may intermittently disrupt sleep but do not appear to involve a persistent disturbance of the basic sleep generating patterns.     

Spectral characteristics of sleep EEG in chronic insomnia

To determine whether the spectral characteristics of the sleep electroencephalogram (EEG) of insomniacs differ from that of healthy subjects, we compared in each of the first four non-rapid eye movement (NREM) and rapid eye movement (REM) episodes: (a) the time courses of absolute power, averaged over the subjects in each group, for the delta, theta, alpha, sigma and beta frequency bands; (b) the relationship between these time courses; and (c) the overnight trend of integrated power in each frequency band. The results show that NREM power, for all frequencies below the beta range, has slower rise rates and reaches lower levels in the insomniac group, whereas beta power is significantly increased. In REM, insomniacs show lower levels in the delta and theta bands, whereas power in the faster frequency bands is significantly increased. Thus, the pathophysiology of insomnia is characterized not only by the generally acknowledged slow wave deficiency, but also by an excessive hyperarousal of the central nervous system throughout the night, affecting both REM and NREM sleep. This hyperarousal is interpreted in terms of the neuronal group theory of sleep which provides a possible explanation for the discrepancies observed between subjective impressions and objective measures of sleep. Also, it is suggested that the progressive hyperpolarization of the thalamocortical neurons as sleep deepens is slower in the patient population and that this may explain the observed slow wave deficiency. The homeostatic control of slow wave activity, on the other hand, would appear to be intact in the patient population.     

The time course of high-frequency bands (15-45 Hz) in all-night spectral analysis of sleep EEG.

OBJECTIVE: The EEG spectral content of all-night sleep recordings obtained in 7 healthy young subjects, aged 18-20 years, including frequencies up to 45 Hz, was studied in order to detect eventual changes in the high-frequency range similar to those reported by magnetic field recording during REM sleep at 40 Hz. METHODS: For this purpose, power spectra were calculated with a fast Fourier transform and the power of the bands ranging 0.75-4.50 Hz (Delta), 4.75-7.75 (Theta), 8.00-12.25 (Alpha), 12.50-15.00 (Sigma), 15.25-24.75 (Beta), 25.00-34.75 (Gamma1), and 35.00-44.75 (Gamma 2) was calculated for-the whole period of analysis (7 h). Also two additional time series: the ratio between Beta and Gamma2, and between Gamma1 and Gamma2 were calculated (Beta and Gamma ratios). RESULTS: Beta and Gamma1 showed small changes with a tendency to increase during REM sleep; Gamma2, on the contrary, showed small changes with a tendency to decrease during REM sleep. Beta and Gamma ratio peaks were clearly correlated with the occurrence of REM sleep. The small changes shown by Beta, Gamma1 and Gamma2 were not statistically significant; on the contrary, Beta ratio and Gamma ratio showed the most important statistical significance values being highest during REM sleep and lowest during slow-wave sleep. Finally, the calculation of the linear correlation coefficient and of the cross-correlation between the different bands showed a clear reciprocity between Delta and Beta and Gamma ratios. CONCLUSIONS: Our study shows a new method for the analysis of high frequencies (up to 45 Hz) in the scalp-recorded sleep EEG which allowed us to better define, as compared to previous studies on the same topic, the changes in power characteristically associated with REM sleep and correlated with the REM/non-REM ultradian rhythm, and to propose it as a tool for future studies.     

EEG alpha band coherence analysis in healthy adults: preliminary results

We studied the occipital inter-hemispheric coherence of Electroencephalogram (electrodes O1-O2) for alpha band (alpha1--8.0 to 10.0 Hz and alpha2--10.1 to 12.5 Hz) in two groups of healthy individuals (young adults and subjects older than 50 years-old), to assess if there is significant difference between this two age groups. No significant difference in alpha band coherences was found between these two age groups.     

Diurnal variations of EEG power in healthy adults

EEG power variations were evaluated in 9 healthy young adults from 8.45 a.m. to 9 p.m. and at 7 a.m. the next day. EEG signals were obtained from 16 electrodes in closed eyes and open eyes situations. Diurnal power variations were calculated for each frequency component, according to the recording situation (RS) and to the scalp site. Regarding values in the early morning on the first day, the power of almost all the frequency components showed an important diurnal increase. It came back close to initial values at 7 a.m. on the second day, which is in agreement with the existence of EEG circadian variations. Diurnal evolutions were dependent on the frequency components: the higher the frequency, the later was its diumal maximum. For many frequency components, the diurnal variation was dependent on RS and the scalp topography. All these characteristics could be used to split the classical EEG bands, especially the delta and alpha bands and be useful for physiological and pharmacological research.     

Spectral analysis of burst periods in EEG from healthy and post-asphyctic full-term neonates.

OBJECTIVE: To investigate whether the periodic EEG patterns seen in healthy and sick full term neonates (trace alternant and burst suppression, respectively) have different frequency characteristics. METHODS: Burst episodes were selected from the EEGs of 9 healthy and 9 post-asphyctic full-term neonates and subjected to power spectrum analysis. Powers in two bands were estimated; 0-4 and 4-30 Hz, designated low- and high-frequency activity, respectively (LFA, HFA). The spectral edge frequency (SEF) was also assessed. RESULTS: In bursts, the LFA power was lower in periods of burst suppression as compared to those of trace alternant. The parameter that best discriminated between the groups was the relative amount of low- and high-frequency activity. The SEF parameter had a low sensitivity to the group differences. In healthy neonates, the LFA power was higher over the posterior right as compared to the posterior left region. CONCLUSIONS: Spectral power of low frequencies differs significantly between the burst episodes of healthy and sick neonates. SIGNIFICANCE: These results can be used when monitoring cerebral function in neonates.     

Laterality of delta waves during all-night sleep

In order to clarify the functional asymmetry of the brain function during sleep, period-amplitude analysis of delta electroencephalogram activity was performed on polysomnograms in 12 right-handed healthy males. Electroencephalograms were recorded from disc electrodes placed at C3, C4, O1 and O2 (10-20 electrode system), using A1 +A2 for reference. Although there were no significant differences in delta counts between O1 and O2, delta counts of C3 were significantly larger than those of C4. These results suggest that there exists distinct laterality in the number of delta waves in the central region, reflecting the functional asymmetry of the brain during sleep.     

Effects of hypnotics on the sleep EEG of healthy young adults: new data and psychopharmacologic implications

Benzodiazepine hypnotics increase NREM sleep and alter its EEG by reducing delta (0.3–3 Hz) and increasing sigma (12–15 Hz) and beta (15–23 Hz) activity. We tested whether the nonbenzodiazepine hypnotic, zolpidem (10 mg), produced the same pattern of sleep and EEG changes as two “classical” benzodiazepines, triazolam (0.25 mg) and temazepam (30 mg). Sleep EEG of 16 subjects was analyzed with period amplitude analysis for 3 nights during drug administration or placebo. The effects of zolpidem were in the same direction but generally of smaller magnitude than those of the classical benzodiazepines. These differences are more likely the result of non-equivalent dosages than different pharmacologic actions. Period amplitude analysis showed that the decreased delta activity resulted mainly from a decrease in wave amplitude. In contrast, the increased sigma and beta activity were produced by increased wave incidence. Delta suppression increased with repeated drug administration but sigma and beta stimulation did not. While these findings have little relevance for the clinical choice of hypnotics they may hold important implications for the brain mechanisms involved in hypnotic tolerance and withdrawal delirium.     

Physiologic all-night sleep in patients with severe forms of epilepsy

100 all night sleep records in 90 patients with various forms of epilepsy and 10 patients with syncope were analyzed. There were 10 patients with generalized epilepsy, 41-with partial epilepsy with complex symptomatology and temporal foci, 23--with mixed seizures and frontal focal changes and 16 patients with partial epilepsy with simple seizures and various location of EEG foci. Normal sleep pattern was present in 21% of cases. The most frequent changes of sleep pattern were: prolongation of sleep onset and the latency of the first episode of REM, instability of sleep stages and absence of sleep spindles. Interictal discharges appeared mostly in all sleep stages of NREM. 50% of epileptic patients showed focal spikes in REM. Nocturnal seizures occurred in 18 patients, in several of them very frequently.     

Spectral EEG analysis following hemispheric stroke

Quantitative EEG frequency analysis was performed within the acute stage and after the recovery in 40 patients with hemispheric stroke in order to analyze ipsi- and contralateral alpha peak frequency (APF) and band power changes. Localization of hemispheric lesion was determined by computer tomography. Changes of clinical scores were compared with the alpha asymmetries. In the cases of small subcortical infarcts good improvement of alpha activity was observed over the affected hemisphere; contralateral APF was relatively preserved. Bilateral symmetric reduction of APF was found in territorial middle cerebral artery infarcts, with poor tendency of recovery of alpha power and neurologic status. These findings suggest transitory derangement of alpha generators in the contralateral hemisphere evidenced by APF and power asymmetries. EEG signs of contralateral alpha reduction may be due to the remote effect of primary ischemic lesion indicating an electrical diaschisis phenomenon in the acute phase of stroke. EEG signs of diaschisis may anticipate a poor recovery of alpha activity and clinical status in the post-stroke period.     

Non-linear analysis of the sleep EEG

A sleep electroencephalogram was analyzed by non-linear analysis. The polysomnography of a healthy male subject was analyzed and the correlation dimensions calculated. The correlation dimensions decreased from the 'awake' stage to sleep stages 1-3 and increased during rapid eye movement (REM) sleep. These results were seen during each sleep cycle. In each sleep cycle, the correlation dimensions decrease for slow wave sleep, and increase for REM sleep.