Linking stages of non-rapid-eye-movement sleep to the spectral EEG markers of the drives for sleep and wake

The conventional staging classification reduces all patterns of sleep polysomnogram signals to a small number of yes-or-no variables labeled wake or a stage of sleep (e.g., W, N1, N2, N3, and R for wake, the 1st, 2nd and 3rd stages of non-rapid-eye-movement sleep, and rapid-eye-movement sleep, respectively). However, the neurobiological underpinnings of such stages remained to be elucidated. We tried to evaluate their link to scores on the 1st and 2nd principal components of the EEG spectrum (1PCS and 2PCS), the markers of two major groups of promoters/inhibitors of sleep/wakefulness delineated as the drives for sleep and wake, respectively. On two occasions, polysomnographic records were obtained from 69 university students during 50-min afternoon naps, and 30-s stage epochs were assigned to 1PCS and 2PCS. Results suggested two-dimensionality of the structure of individual differences in amounts of stages. Amount of N1 loaded exclusively on one of two dimensions associated with 1PCS, amounts of W and N2 loaded exclusively on another dimension associated with 2PCS, and amount of N3 equally loaded on both dimensions. Scores demonstrated stability within each stage, but a drastic change in just one of two scores occurred during transitions from one stage to another on the way from wakefulness to deeper sleep (e.g., 2PCS changed from >0 to <0 during transition W→N1, 1PCS changed from <0 to >0 during transition N1→N2). Therefore, the transitions between stages observed during short naps might be linked to rapid changes in the reciprocal interactions between the promoters/inhibitors of sleep/wakefulness.

Comparison of resting-state EEG between adults with Down syndrome and typically developing controls

Background Down syndrome (DS) is the most common genetic cause of intellectual disability (ID) worldwide. Understanding electrophysiological characteristics associated with DS provides potential mechanistic insights into ID, helping inform biomarkers and targets for intervention. Currently, electrophysiological characteristics associated with DS remain unclear due to methodological differences between studies and inadequate controls for cognitive decline as a potential cofounder. Methods Eyes-closed resting-state EEG measures (specifically delta, theta, alpha, and beta absolute and relative powers, and alpha peak amplitude, frequency and frequency variance) in occipital and frontal regions were compared between adults with DS (with no diagnosis of dementia or evidence of cognitive decline) and typically developing (TD) matched controls (n = 25 per group). Results We report an overall ‘slower’ EEG spectrum, characterised by higher delta and theta power, and lower alpha and beta power, for both regions in people with DS. Alpha activity in particular showed strong group differences, including lower power, lower peak amplitude and greater peak frequency variance in people with DS. Conclusions Such EEG ‘slowing’ has previously been associated with cognitive decline in both DS and TD populations. These findings indicate the potential existence of a universal EEG signature of cognitive impairment, regardless of origin (neurodevelopmental or neurodegenerative), warranting further exploration.

Features importance in seizure classification using scalp EEG reduced to single timeseries

Seizure detection and seizure-type classification are best performed using intra-cranial or full-scalp electroencephalogram (EEG). In embedded wearable systems however, recordings from only a few electrodes are available, reducing the spatial resolution of the signals to a handful of timeseries at most. Taking this constraint into account, we tested the performance of multiple classifiers using a subset of the EEG recordings by selecting a single trace from the montage or performing a dimensionality reduction over each hemispherical space. Our results support that Random Forest (RF) classifiers lead most efficient and stable classification performances over Support Vector Machines (SVM). Interestingly, tracking the feature importances using permutation tests reveals that classical EEG spectrum power bands display different rankings across the classifiers: low frequencies (delta, theta) are most important for SVMs while higher frequencies (alpha, gamma) are more relevant for RF and Decision Trees. We reach up to 94.3% +/- 5.3% accuracy in classifying absence from tonic-clonic seizures using state-of-art sampling methods for unbalanced datasets and leave-patients-out 3-fold cross-validation policy.

Is So Called “Split Alpha” in EEG Spectral Analysis a Result of Methodological and Interpretation Errors?

This paper attempts to explain some methodological issues regarding EEG signal analysis which might lead to misinterpretation and therefore to unsubstantiated conclusions. The so called “split-alpha,” a “new phenomenon” in EEG spectral analysis described lately in few papers is such a case. We have shown that spectrum feature presented as a “split alpha” can be the result of applying improper means of analysis of the spectrum of the EEG signal that did not take into account the significant properties of the applied Fast Fourier Transform (FFT) method. Analysis of the shortcomings of the FFT method applied to EEG signal such as limited duration of analyzed signal, dependence of frequency resolution on time window duration, influence of window duration and shape, overlapping and spectral leakage was performed. Our analyses of EEG data as well as simulations indicate that double alpha spectra called as “split alpha” can appear, as spurious peaks, for short signal window when the EEG signal being studied shows multiple frequencies and frequency bands. These peaks have no relation to any frequencies of the signal and are an effect of spectrum leakage. Our paper is intended to explain the reasons underlying a spectrum pattern called as a “split alpha” and give some practical indications for using spectral analysis of EEG signal that might be useful for readers and allow to avoid EEG spectrum misinterpretation in further studies and publications as well as in clinical practice.

EEG power spectral density in locked-in and completely locked-in state patients: a longitudinal study

Persons with their eye closed and without any means of communication is said to be in a completely locked-in state (CLIS) while when they could still open their eyes actively or passively and have some means of communication are said to be in locked-in state (LIS). Two patients in CLIS without any means of communication, and one patient in the transition from LIS to CLIS with means of communication, who have Amyotrophic Lateral Sclerosis were followed at a regular interval for more than 1 year. During each visit, resting-state EEG was recorded before the brain–computer interface (BCI) based communication sessions. The resting-state EEG of the patients was analyzed to elucidate the evolution of their EEG spectrum over time with the disease’s progression to provide future BCI-research with the relevant information to classify changes in EEG evolution. Comparison of power spectral density (PSD) of these patients revealed a significant difference in the PSD’s of patients in CLIS without any means of communication and the patient in the transition from LIS to CLIS with means of communication. The EEG of patients without any means of communication is devoid of alpha, beta, and higher frequencies than the patient in transition who still had means of communication. The results show that the change in the EEG frequency spectrum may serve as an indicator of the communication ability of such patients.

Automatic Wake-Sleep Stages Classification using Electroencephalogram Instantaneous Frequency and Envelope Tracking

BackgroundThe study of cerebral activity during sleep using the electroencephalograph (EEG) is a major research field in neuroscience. Despite the rich literature in this field, the automatic and accurate categorization of wake-sleep stages remains an open problem.New MethodA robust model-based Kalman filtering scheme is proposed for tracking the poles of a second order time-varying autoregressive model fitted over the EEG acquired during different wake/sleep stages. The pole angle/phase is regarded as the dominant frequency of the EEG spectrum (known as the instantaneous frequency in literature). The frequency resolution is improved by splitting the wide frequency band to subbands corresponding to well-known brain rhythms. Using recent findings in field of EEG phase/frequency tracking, the instantaneous envelope of the narrow-band signal’s analytic form is also tracked as a complementary feature.ResultsThe minimal set of instantaneous frequency and envelope features is employed in three classification schemes, using training labels from R&k and AASM sleep scoring standards. The LDA classifier resulted in the highest performance using the proposed feature set.Comparison with Existing MethodsThe proposed method resulted in a higher mean decoding accuracy and a lower standard deviation on the entire dataset, as compared with state-of-the-art techniques.ConclusionsThe accurate tracking of the instantaneous frequency and envelope are highly informative for sleep stage scoring. The proposed method is shown to have additional applications, including the prediction of wake-sleep transition, which can be used for drowsiness detection from the EEG.

The influence of cholinergic drugs on electroencephalogram of the gonadectomized female rabbits

The results of investigation has shown, that after the gonadectomy of the rabbits-females the general power of EEG spectrum (GPS of EEG) changed under influence of cholinergic drugs. The sterilizathion of the animals leads to disorders of interaction between M- and N-cholinergic mechanisms in the brain. In particular, blockade of M-cholinoreceptors by metamizyl in intact and ovariectomized rabbits increased the GPS of EEG. The simultaneous administration of metamizyl with galantamine to intact females led to even greater increase of GPS of EEG, whereas the sterilized rabbits demonstrated its reduction. On the contrary, the application of the N-cholinoreceptors inhibitor gangleron with inhibitor of acetylcholinesterase (AChE) galantamine reduced the GPS of EEG in intact animals and increased it in gonadectomized rabbits in comparison with gangleron alone. Thus, the effect of M,N-cholinoblockators in combination with AChE inhibitor in sterilized rabbits changed the EEG spectrum to opposite in comparison with intact females.

Exploring EEG Spectral Patterns in Episodic and Chronic Migraine During the Interictal State: Determining Frequencies of Interest in the Resting State

Objective The analysis of particular (electroencephalographic) EEG frequency bands has revealed new insights relative to the neural dynamics that, when studying the EEG spectrum as a whole, would have remained hidden. This study is aimed at characterizing spectral resting state EEG patterns for assessing possible differences of episodic and chronic migraine during the interictal period. For that purpose, a novel methodology for analyzing specific frequencies of interest was performed. Methods Eighty-seven patients with migraine (45 with episodic and 42 with chronic migraine) and 39 age- and sex-matched controls performed a resting-state EEG recording. Spectral measures were computed using conventional frequency bands. Additionally, particular frequency bands were determined to distinguish between controls and migraine patients, as well as between migraine subgroups. Results Frequencies ranging from 11.6 Hz to 12.8 Hz characterized migraine as a whole, with differences evident in the central and left parietal regions (controlling for false discovery rate). An additional band between 24.1 Hz and 29.8 Hz was used to discriminate between migraine subgroups. Interestingly, the power in this band was positively correlated with time from onset in episodic migraine, but no correlation was found for chronic migraine. Conclusions Specific frequency bands were proposed to identify the spectral characteristics of the electrical brain activity in migraine during the interictal stage. Our findings support the importance of discriminating between migraine subgroups to avoid hiding relevant features in migraine.