Analysis and automatic identification of sleep stages using higher order spectra

Electroencephalogram (EEG) signals are widely used to study the activity of the brain, such as to determine sleep stages. These EEG signals are nonlinear and non-stationary in nature. It is difficult to perform sleep staging by visual interpretation and linear techniques. Thus, we use a nonlinear technique, higher order spectra (HOS), to extract hidden information in the sleep EEG signal. In this study, unique bispectrum and bicoherence plots for various sleep stages were proposed. These can be used as visual aid for various diagnostics application. A number of HOS based features were extracted from these plots during the various sleep stages (Wakefulness, Rapid Eye Movement (REM), Stage 1-4 Non-REM) and they were found to be statistically significant with p-value lower than 0.001 using ANOVA test. These features were fed to a Gaussian mixture model (GMM) classifier for automatic identification. Our results indicate that the proposed system is able to identify sleep stages with an accuracy of 88.7%.     

Evaluating True BCI Communication Rate through Mutual Information and Language Models

Brain-computer interface (BCI) systems are a promising means for restoring communication to patients suffering from “locked-in” syndrome. Research to improve system performance primarily focuses on means to overcome the low signal to noise ratio of electroencephalogric (EEG) recordings. However, the literature and methods are difficult to compare due to the array of evaluation metrics and assumptions underlying them, including that: 1) all characters are equally probable, 2) character selection is memoryless, and 3) errors occur completely at random. The standardization of evaluation metrics that more accurately reflect the amount of information contained in BCI language output is critical to make progress. We present a mutual information-based metric that incorporates prior information and a model of systematic errors. The parameters of a system used in one study were re-optimized, showing that the metric used in optimization significantly affects the parameter values chosen and the resulting system performance. The results of 11 BCI communication studies were then evaluated using different metrics, including those previously used in BCI literature and the newly advocated metric. Six studies'' results varied based on the metric used for evaluation and the proposed metric produced results that differed from those originally published in two of the studies. Standardizing metrics to accurately reflect the rate of information transmission is critical to properly evaluate and compare BCI communication systems and advance the field in an unbiased manner.     

A transition-constrained discrete hidden Markov model for automatic sleep staging

ABSTRACT: BACKGROUND: Approximately one-third of the human lifespan is spent sleeping. To diagnose sleep problems, all-night polysomnographic (PSG) recordings including electroencephalograms (EEGs), electrooculograms (EOGs) and electromyograms (EMGs), are usually acquired from the patient and scored by a well-trained expert according to Rechtschaffen & Kales (R&K) rules. Visual sleep scoring is a time-consuming and subjective process. Therefore, the development of an automatic sleep scoring method is desirable. METHOD: The EEG, EOG and EMG signals from twenty subjects were measured. In addition to selecting sleep characteristics based on the 1968 R&K rules, features utilized in other research were collected. Thirteen features were utilized including temporal and spectrum analyses of the EEG, EOG and EMG signals, and a total of 158 hours of sleep data were recorded. Ten subjects were used to train the Discrete Hidden Markov Model (DHMM), and the remaining ten were tested by the trained DHMM for recognition. Furthermore, the 2-fold cross validation was performed during this experiment. RESULTS: Overall agreement between the expert and the results presented is 85.29%. With the exception of S1, the sensitivities of each stage were more than 81%. The most accurate stage was SWS (94.9%), and the least-accurately classified stage was S1 (<34%). In the majority of cases, S1 was classified as Wake (21%), S2 (33%) or REM sleep (12%), consistent with previous studies. However, the total time of S1 in the 20 all-night sleep recordings was less than 4%. CONCLUSION: The results of the experiments demonstrate that the proposed method significantly enhances the recognition rate when compared with prior studies.     

Deterministic chaos and the first positive Lyapunov exponent: a nonlinear analysis of the human electroencephalogram during sleep

Under selected conditions, nonlinear dynamical systems, which can be described by deterministic models, are able to generate so-called deterministic chaos. In this case the dynamics show a sensitive dependence on initial conditions, which means that different states of a system, being arbitrarily close initially, will become macroscopically separated for sufficiently long times. In this sense, the unpredictability of the EEG might be a basic phenomenon of its chaotic character. Recent investigations of the dimensionality of EEG attractors in phase space have led to the assumption that the EEG can be regarded as a deterministic process which should not be mistaken for simple noise. The calculation of dimensionality estimates the degrees of freedom of a signal. Nevertheless, it is difficult to decide from this kind of analysis whether a process is quasiperiodic or chaotic. Therefore, we performed a new analysis by calculating the first positive Lyapunov exponent L ₁ from sleep EEG data. Lyapunov exponents measure the mean exponential expansion or contraction of a flow in phase space. L ₁ is zero for periodic as well as quasiperiodic processes, but positive in the case of chaotic processes expressing the sensitive dependence on initial conditions. We calculated L ₁ for sleep EEG segments of 15 healthy men corresponding to the sleep stages I, II, III, IV, and REM (according to Rechtschaffen and Kales). Our investigations support the assumption that EEG signals are neither quasiperiodic waves nor a simple noise. Moreover, we found statistically significant differences between the values of L ₁ for different sleep stages. All together, this kind of analysis yields a useful extension of the characterization of EEG signals in terms of nonlinear dynamical system theory.     

The link between cardiac autonomic activity and sleep delta power is altered in men with sleep apnea-hypopnea syndrome

We hypothesize that sleep apnea-hypopnea alters interaction between cardiac vagal modulation and sleep delta EEG. Sleep apnea-hypopnea syndrome (SAHS) is related to cardiovascular complications in men. SAHS patients show higher sympathetic activity than normal subjects. In healthy men, non-rapid eye movement (NREM) sleep is associated with cardiac vagal influence, whereas rapid eye movement (REM) sleep is linked to cardiac sympathetic activity. Interaction between cardiac autonomic modulation and delta sleep EEG is not altered across a life span nor is the delay between appearances of modifications in both signals. Healthy controls, moderate SAHS, and severe SAHS patients were compared across the first three NREM-REM cycles. Spectral analysis was applied to ECG and EEG signals. High frequency (HF) and low frequency (LF) of heart rate variability (HRV), ratio of LF/HF, and normalized (nu) delta power were obtained. A coherency analysis between HF(nu) and delta was performed, as well as a correlation analysis between obstructive apnea index (AI) or hypopnea index (HI) and gain, coherence, or phase shift. HRV components were similar between groups. In each group, HF(nu) was larger during NREM, while LF(nu) predominated across REM and wake stages. Coherence and gain between HF(nu) and delta decreased from controls to severe SAHS patients. In SAHS patients, the delay between modifications in HF(nu) and delta did not differ from zero. AI and HI correlated negatively with coherence, while HI correlated negatively with gain only. Apneas-hypopneas affect the link between cardiac sympathetic and vagal modulation and delta EEG demonstrated by the loss of cardiac autonomic activity fluctuations across shifts in sleep stages. Obstructive apneas and hypopneas alter the interaction between both signals differently.     

基于奇异值第一主成分的睡眠脑电分期方法研究

目的:脑电信号含多种噪声和伪迹,信噪比较低,特征提取前必须进行复杂的预处理,严重影响睡眠分期的速度。鉴于此,本文提出一种基于奇异值第一主成分的睡眠脑电分期方法,该方法抗噪性能较强,可省去预处理过程,减少计算量,提高睡眠分期的效率。方法:对未经过预处理的睡眠脑电进行奇异系统分析,研究奇异谱曲线,提取奇异值第一主成分,探索其随睡眠状态变化的规律。并通过支持向量机利用奇异值第一主成分对睡眠分期。结果:奇异值第一主成分不仅能表征脑电信号主体,而且可以抑制噪声、降低维数。随着睡眠的深入,奇异值第一主成分的值逐渐增大,但在REM期处于S1期和S2期之间。经MIT-BIH睡眠数据库中5例同导联位置的脑电数据测试(仅1导脑电数据),睡眠脑电分期的准确率达到86.4%。结论:在未对脑电信号进行预处理的情况下,提取的睡眠脑电的奇异值第一主成分能有效表征睡眠状态,是一种有效的睡眠分期依据。本文运用提出的方法仅采用1导脑电数据,就能得到较为满意的睡眠分期结果。该方法有较强的分类性能,且抗噪能力强,不需要对脑电作复杂的预处理,计算量小,方法简单,很大程度上提高了睡眠分期的效率。     

Feature selection for sleep/wake stages classification using data driven methods

This paper focuses on the problem of selecting relevant features extracted from human polysomnographic (PSG) signals to perform accurate sleep/wake stages classification. Extraction of various features from the electroencephalogram (EEG), the electro-oculogram (EOG) and the electromyogram (EMG) processed in the frequency and time domains was achieved using a database of 47 night sleep recordings obtained from healthy adults in laboratory settings. Multiple iterative feature selection and supervised classification methods were applied together with a systematic statistical assessment of the classification performances. Our results show that using a simple set of features such as relative EEG powers in five frequency bands yields an agreement of 71% with the whole database classification of two human experts. These performances are within the range of existing classification systems. The addition of features extracted from the EOG and EMG signals makes it possible to reach about 80% of agreement with the expert classification. The most significant improvement on classification accuracy is obtained on NREM sleep stage I, a stage of transition between sleep and wakefulness.     

Comparative analysis of EEG correlation synchronism and EEG amplitude relationships in all-night sleep

We used a new methodological approach to the evaluation of EEG synchronization based on correlation between amplitude modulation processes (EEG envelopes). We revealed: left-hemispheric dominance and dominance of frontal over occipital regions characteristic of all sleep stages; differences in synchronization in frequency bands and their patterns characteristic of a specific sleep stage; stage-dependent differences in inter-hemispheric synchrony and patterns of their changes from the frontal to occipital regions; and stage-dependent topographical distributions of high synchronization foci with respect to frequency domains. Analysis of amplitude topography also revealed left-hemispheric dominance and many significant differences in activity distribution patterns over parasagittal chains of electrodes (meridians) depending on sleep stages and frequency domains. The combination of EEG synchrony estimates with the amplitude spectral estimates made it possible to perform a reliable discriminant recognition of five sleep stages with errors in the range of 3-20%.     

Multivariate and multiorgan analysis of cardiorespiratory variability signals: the CAP sleep case.

Signals from different systems are analyzed during sleep on a beat-to-beat basis to provide a quantitative measure of synchronization with the heart rate variability (HRV) signal, oscillations of which reflect the action of the autonomic nervous system. Beat-to-beat variability signals synchronized to QRS occurrence on ECG signals were extracted from respiration, electroencephalogram (EEG) and electromyogram (EMG) traces. The analysis was restricted to sleep stage 2. Cyclic alternating pattern (CAP) periods were detected from EEG signals and the following conditions were identified: stage 2 non-CAP (2 NCAP), stage 2 CAP (2 CAP) and stage 2 CAP with myoclonus (2 CAP MC). The coupling relationships between pairs of variability signals were studied in both the time and frequency domains. Passing from 2 NCAP to 2 CAP, sympathetic activation is indicated by tachycardia and reduced respiratory arrhythmia in the heart rate signal. At the same time, we observed a marked link between EEG and HRV at the CAP frequency. During 2 CAP MC, the increased synchronization involved myoclonus and respiration. The underlying mechanism seems to be related to a global control system at the central level that involves the different systems.     

The dimensionality of human's electroencephalogram during sleep

In order to perform an analysis of nonlinear EEG-dynamics we investigated the EEG of ten male probands during sleep. According to Rechtschaffen and Kales (1968) we scored the sleep-EEG and applied an algorithm, proposed by Grassberger and Proccaccia (1983) to compute the correlation dimension of different sleep stages. The correlation dimension characterizes the dynamics of the EEG signal and estimates the degrees of freedom of the signal under study. We could demonstrate, that the EEG of slow wave sleep stages depicts a dimensionality, which is two units smaller than that of light or REM sleep.     

The prediction of EEG signals using a feedback-structured adaptive rational function filter

 In this article, we present a feedback-structured adaptive rational function filter based on a recursive modified Gram-Schmidt algorithm and apply it to the prediction of an EEG signal that has nonlinear and nonstationary characteristics. For the evaluation of the prediction performance, the proposed filter is compared with other methods, where a single-step prediction and a multi-step prediction are considered for a short-term prediction, and the prediction performance is assessed in normalized mean square error. The experimental results show that the proposed filter shows better performance than other methods considered for the short-term prediction of EEG signals. Received: 22 September 1998 / Accepted in revised form: 29 February 2000     

A model-based monitor of human sleep stages

Stochastic models are proposed for sleep and for the sleep related electroencephalogram (EEG), electrooculogram (EOG), and electromyogram (EMG). The evolution of sleep through its various stages is described as a Markov chain. The EEG is modelled using Wiener processes. The EOG and EMG are modelled as combinations of Poisson point processes and Gaussian processes, respectively. The EEG models contain a feedback structure that is based on physiological data. The maximum likelihood sleep stage monitor, that uses the sleep-related observations, has been derived and implemented. The agreement between automatic and human stage classifications of six sleep recordings was 70.6%, which was 4.5% worse than the average agreement between six human classifiers. Monitoring of simulated sleep suggests that the difficulty in separating wakefulness from stage 1 is due to poor modelling. If one ignores this difference, which, from a diagnostic point of view is fairly unimportant, the above mentioned agreement reaches 81.8%, which is 0.5% better than the corresponding average human vs human agreement.     

Evaluation of a commercially available automated sleep staging tool in healthy infants

Sleep and Biological Rhythms - The ALICE5 software package provides a commercially available automated sleep staging system designed for infants. This study aims to evaluate the accuracy of this...     

Classification des stades de sommeil par des réseaux de neurones artificiels hiérarchiques

The goal of our work is to provide an automatic analysis and decision tool for sleep stages classification based on an artificial neural networks (ANN). The first difficulty lies in choosing the physiological signals representation and in particular the electroencephalogram (EEG). Once the representation adopted, the next step is to design the optimal neural network determined by a learning and validation process of data from a set of sleep records. We studied several configurations of conventional ANN giving results varying from 62 to 71 %, then we proposed a new hierarchical configuration, which gives a rate of 74 % correct classification for six stages. These results lead us to further explore this issue at the representation and design of ANNs to improve the performance of our tool.     

Automatic EEG arousal detection for sleep apnea syndrome

Electroencephalographic (EEG) arousals are seen in EEG recordings as an awakening response of the human brain. Sleep apnea is a serious sleep disorder. Severe sleep apnea brings about EEG arousals and sleep for patients with sleep apnea syndrome (SAS) is thus frequently interrupted. The number of respiratory-related arousals during the whole night on PSG recordings is directly related to the quality of sleep. Detecting EEG arousals in the PSG record is thus a significant task for clinical diagnosis in sleep medicine. In this paper, a method for automatic detection of EEG arousals in SAS patients was proposed. To effectively detect respiratory-related arousals, threshold values were determined according to pathological events as sleep apnea and electromyogram (EMG). If resumption of ventilation (end of the apnea interval) was detected, much lower thresholds were adopted for detecting EEG arousals, including relatively doubtful arousals. Conversely, threshold was maintained high when pathological events were undetected. The proposed method was applied to polysomnographic (PSG) records of eight patients with SAS and accuracy of EEG arousal detection was verified by comparative visual inspection. Effectiveness of the proposed method in clinical diagnosis was also investigated.     

A Supplementary System for a Brain-Machine Interface Based on Jaw Artifacts for the Bidimensional Control of a Robotic Arm

Non-invasive Brain-Machine Interfaces (BMIs) are being used more and more these days to design systems focused on helping people with motor disabilities. Spontaneous BMIs translate user''s brain signals into commands to control devices. On these systems, by and large, 2 different mental tasks can be detected with enough accuracy. However, a large training time is required and the system needs to be adjusted on each session. This paper presents a supplementary system that employs BMI sensors, allowing the use of 2 systems (the BMI system and the supplementary system) with the same data acquisition device. This supplementary system is designed to control a robotic arm in two dimensions using electromyographical (EMG) signals extracted from the electroencephalographical (EEG) recordings. These signals are voluntarily produced by users clenching their jaws. EEG signals (with EMG contributions) were registered and analyzed to obtain the electrodes and the range of frequencies which provide the best classification results for 5 different clenching tasks. A training stage, based on the 2-dimensional control of a cursor, was designed and used by the volunteers to get used to this control. Afterwards, the control was extrapolated to a robotic arm in a 2-dimensional workspace. Although the training performed by volunteers requires 70 minutes, the final results suggest that in a shorter period of time (45 min), users should be able to control the robotic arm in 2 dimensions with their jaws. The designed system is compared with a similar 2-dimensional system based on spontaneous BMIs, and our system shows faster and more accurate performance. This is due to the nature of the control signals. Brain potentials are much more difficult to control than the electromyographical signals produced by jaw clenches. Additionally, the presented system also shows an improvement in the results compared with an electrooculographic system in a similar environment.     

Sepsis-induced alterations in sleep of rats

Sepsis is a systemic immune response to infection that may result in multiple organ failure and death. Polymicrobial infections remain a serious clinical problem, and in the hospital, sepsis is the number-one noncardiac killer. Although the central nervous system may be one of the first systems affected, relatively little effort has been made to determine the impact of sepsis on the brain. In this study, we used the cecal ligation and puncture (CLP) model to determine the extent to which sepsis alters sleep, the EEG, and brain temperature (Tbr) of rats. Sepsis increases the amount of time rats spend in non-rapid eye movement sleep (NREMS) during the dark period, but not during the light period. Rapid eye movements sleep (REMS) of septic rats is suppressed for about 24 h following CLP surgery, after which REMS increases during dark periods for at least three nights. The EEG is dramatically altered shortly after sepsis induction, as evidenced by reductions in slow-frequency components. Furthermore, sleep is fragmented, indicating that the quality of sleep is diminished. Effects on sleep, the EEG, and Tbr persist for at least 84 h after sepsis induction, the duration of our recording period. Immunohistochemical assays focused on brain stem mechanisms responsible for alterations in REMS, as little information is available concerning infection-induced suppression of this sleep stage. Our immunohistochemical data suggest that REMS suppression after sepsis onset may be mediated, in part, by the brain stem GABAergic system. This study demonstrates for the first time that sleep and EEG patterns are altered during CLP-induced sepsis. These data suggest that the EEG may serve as a biomarker for sepsis onset. These data also contribute to our knowledge of potential mechanisms, whereby infections alter sleep and other central nervous system functions.     

Brain oscillatory activity during sleep shows unknown dysfunctions in early encephalopathy

Electroencephalographic recordings in cirrhotic patients without overt hepatic encephalopathy (HE) have mainly been performed during wakefulness. Our aim was to quantify their alterations in nocturnal sleep electroencephalogram (EEG). In 20 patients and 20 healthy volunteers, we recorded a nocturnal digital polysomnography. Different sleep parameters were measured. Besides, we performed quantitative analysis of EEG (qEEG) as follows: spectral power in the different sleep stages was calculated in the frequency bands low δ, δ, θ, α, and σ. Also, the mean dominant frequency and Sleep Indexes were obtained. In comparison with controls, the group of patients showed (1) different alterations in both the microstructure and the macrostructure of sleep; (2) an increase in, both, θ band power and the average mean dominant frequency during rapid eye movement (REM); (3) in all sleep stages, a decrease of sleep electroencephalogram spectral power in low δ band and an increase in δ band: and (4) in stages N3 and REM, significant increases in the minimum of mean dominant frequency and in the respective sleep indexes. Therefore, in cirrhotic patients without overt HE, and likely having minimal hepatic encephalopathy, we found different alterations in both the microstructure and the macrostructure of nocturnal sleep. Also, sleep qEEG showed a brain dysfunction in slow oscillatory mechanisms intrinsic of sleep stages, with an increase in the frequency of its maximal electroencephalogram synchronization, from low δ to δ band. These alterations may reflect the onset of encephalopathy; sleep qEEG may, thus, be an adequate tool for its brain functional evaluation and follow-up.     

Automatic analysis of sleep using two parameters based on principal component analysis of electroencephalography spectral data

A computer program for the analysis of a sleep electroencephalogram (EEG) is presented. The method relies on two steps. First, a spectral analysis is performed for signals recorded from one or more electrode locations. Then, two EEG parameters are obtained by storing the spectral activity in a multidimensional space, whose dimension is reduced using principal component analysis (PCA) techniques. The main advantage of these parameters is in describing the process of sleep on a continuous scale as a function of time. Validation of the method was performed with the data collected from 16 subjects (8 young volunteers and 8 elderly insomniacs). Results snowed that the parameters correlate highly with the hypnograms established by conventional visual scoring. This signal parametrisation, however, offers more information regarding the time course of sleep, since small variations within individual sleep stages as well as smooth transitions between stages are assessed. Finally, the concurrent use of both parameters provides an original way of considering sleep as a dynamic process evolving cyclically in a single plane.