Classification of Three Anesthesia Stages Based on Near-Infrared Spectroscopy Signals.

Proper monitoring of anesthesia stages can guarantee the safe performance of clinical surgeries. In this study, different anesthesia stages were classified using near-infrared spectroscopy (NIRS) signals with machine learning. The cerebral hemodynamic variables of right proximal oxyhemoglobin (HbO2) in maintenance (MNT), emergence (EM) and the consciousness (CON) stage were collected and then the differences between the three stages were compared by phase-amplitude coupling (PAC). Then combined with time-domain including linear (mean, standard deviation, max, min and range), nonlinear (sample entropy) and power in frequency-domain signal features, feature selection was performed and finally classification was performed by support vector machine (SVM) classifier. The results show that the PAC of the NIRS signal was gradually enhanced with the deepening of anesthesia level. A good three-classification accuracy of 69.27% was obtained, which exceeded the result of classification of any single category feature. These results indicate the fesibility of NIRS signals in performing three or even more anesthesia stage classifications, providing insight into the development of new anesthesia monitoring modalities.

Monitoring the Depth of Anesthesia Based on Phase-Amplitude Coupling of Near-Infrared Spectroscopy Signals.

Accurate monitoring of the depth of anesthesia (DOA) is essential to ensure the safety of the operation. In this study, a new index using near-infrared spectroscopy (NIRS) signal was proposed to assess the relationship between the DOA and cerebral hemodynamic variables. METHODS: Four cerebral hemodynamic variables of 15 patients were collected, including left, right, proximal, distal, oxygenated (HbO 2) and deoxygenated (Hb) hemoglobin concentration changes. The Phase-Amplitude coupling (PAC), an adaptation of cross-frequency coupling to reflect the modulation of the amplitude of high-frequency signals by the phase of low-frequency signals, was measured and the modulation index (MI) was obtained to monitor the DOA afterwards. Meanwhile, the BIS value based on electroencephalogram is also measured and compared. RESULTS: Compared with awake period, in anesthesia maintenance period, the PAC was strengthened. The analysis of receiver operating characteristic (ROC) curve showed that the MI, especially the MI of rp-HbO2, could effectively discriminate these two periods. Additionally, during the whole anesthesia process, the BIS value was statistically consistent with the MI of cerebral hemodynamic variables, and cerebral hemodynamic variables were immune from interference by clinical electric devices. CONCLUSION: The MI of cerebral hemodynamic variables was appropriate to be used as a new index to monitor the DOA. SIGNIFICANCE: This study is of great significance to the development of new modes of anesthesia monitoring and new decoding methods, and is expected to develop a high-performance anesthesia monitoring system.

Decoding Multi-Class EEG Signals of Hand Movement Using Multivariate Empirical Mode Decomposition and Convolutional Neural Network.

Brain-computer interface (BCI) is a technology that connects the human brain and external devices. Many studies have shown the possibility of using it to restore motor control in stroke patients. One specific challenge of such BCI is that the classification accuracy is not high enough for multi-class movements. In this study, by using Multivariate Empirical Mode Decomposition (MEMD) and Convolutional Neural Network (CNN), a novel algorithm (MECN) was proposed to decode EEG signals for four kinds of hand movements. Firstly, the MEMD was used to decompose the movement-related electroencephalogram (EEG) signals to obtain the multivariate intrinsic empirical functions (MIMFs). Then, the optimal MIMFs fusion was performed based on sequential forward selection algorithm. Finally, the selected MIMFs were input to the CNN model for discriminating four kinds of hand movements. The average classification accuracy of thirteen subjects over the six-fold cross-validation reached 81.14% for 2s-data before the movement onset and 81.08% for 2s-data after the movement onset. The MECN method achieved statistically significant improvement on the state-of-the-art methods. The results showed that the algorithm proposed in this study can effectively decode four kinds of hand movements based on EEG signals.

Hybrid motion artifact detection and correction approach for functional near-infrared spectroscopy measurements.

SIGNIFICANCE: Functional near-infrared spectroscopy (fNIRS) is a promising optical neuroimaging technique, measuring the hemodynamic signals from the cortex. However, improving signal quality and reducing artifacts arising from oscillation and baseline shift (BS) are still challenging up to now for fNIRS applications. AIM: Considering the advantages and weaknesses of the different algorithms to reduce the artifact effect in fNIRS signals, we propose a hybrid artifact detection and correction approach. APPROACH: First, distinct artifact detection was realized through an fNIRS detection strategy. Then the artifacts were divided into three categories: BS, slight oscillation, and severe oscillation. A comprehensive correction was applied through three main steps: severe artifact correction by cubic spline interpolation, BS removal by spline interpolation, and slight oscillation reduction by dual-threshold wavelet-based method. RESULTS: Using fNIRS data acquired during whole night sleep monitoring, we compared the performance of our approach with existing algorithms in signal-to-noise ratio (SNR) and Pearson's correlation coefficient (R). We found that the proposed method showed improvements in performance in SNR and R with strong stability. CONCLUSIONS: These results suggest that the new hybrid artifact detection and correction method enhances the viability of fNIRS as a functional neuroimaging modality.

Improvement in EEG Source Imaging Accuracy by Means of Wavelet Packet Transform and Subspace Component Selection.

The electroencephalograph (EEG) source imaging (ESI) method is a non-invasive method that provides high temporal resolution imaging of brain electrical activity on the cortex. However, because the accuracy of EEG source imaging is often affected by unwanted signals such as noise or other source-irrelevant signals, the results of ESI are often incongruous with the real sources of brain activities. This study presents a novel ESI method (WPESI) that is based on wavelet packet transform (WPT) and subspace component selection to image the cerebral activities of EEG signals on the cortex. First, the original EEG signals are decomposed into several subspace components by WPT. Second, the subspaces associated with brain sources are selected and the relevant signals are reconstructed by WPT. Finally, the current density distribution in the cerebral cortex is obtained by establishing a boundary element model (BEM) from head MRI and applying the appropriate inverse calculation. In this study, the localization results obtained by this proposed approach were better than those of the original sLORETA approach (OESI) in the computer simulations and visual evoked potential (VEP) experiments. For epilepsy patients, the activity sources estimated by this proposed algorithm conformed to the seizure onset zones. The WPESI approach is easy to implement achieved favorable accuracy in terms of EEG source imaging. This demonstrates the potential for use of the WPESI algorithm to localize epileptogenic foci from scalp EEG signals.

Reliability of EEG microstate analysis at different electrode densities during propofol-induced transitions of brain states.

Electroencephalogram (EEG) microstate analysis is a promising and effective spatio-temporal method that can segment signals into several quasi-stable classes, providing a great opportunity to investigate short-range and long-range neural dynamics. However, there are still many controversies in terms of reproducibility and reliability when selecting different parameters or datatypes. In this study, five electrode configurations (91, 64, 32, 19, and 8 channels) were used to measure the reliability of microstate analysis at different electrode densities during propofol-induced sedation. First, the microstate topography and parameters at five different electrode densities were compared in the baseline (BS) condition and the moderate sedation (MD) condition, respectively. The intraclass correlation coefficient (ICC) and coefficient of variation (CV) were introduced to quantify the consistency of the microstate parameters. Second, statistical analysis and classification between BS and MD were performed to determine whether the microstate differences between different conditions remained stable at different electrode densities, and ICC was also calculated between the different conditions to measure the consistency of the results in a single condition. The results showed that in both the BS or MD condition, respectively, there were few significant differences in the microstate parameters among the 91-, 64-, and 32-channel configurations, with most of the differences observed between the 19- or 8-channel configurations and the other configurations. The ICC and CV data also showed that the consistency among the 91-, 64-, and 32-channel configurations was better than that among all five electrode configurations after including the 19- and 8-channel configurations. Furthermore, the significant differences between the conditions in the 91-channel configuration remained stable at the 64- and 32-channel resolutions, but disappeared at the 19- and 8-channel resolutions. In addition, the classification and ICC results showed that the microstate analysis became unreliable with fewer than 20 electrodes. The findings of this study support the hypothesis that microstate analysis of different brain states is more reliable with higher electrode densities; the use of a small number of channels is not recommended.

Effective Brain State Estimation During Propofol-Induced Sedation Using Advanced EEG Microstate Spectral Analysis.

Brain states are patterns of neuronal synchrony, and the electroencephalogram (EEG) microstate provides a promising tool to characterize and analyze the synchronous neural firing. However, the topographical spectral information for each predominate microstate is still unclear during the switch of consciousness, such as sedation, and the practical usage of the EEG microstate is worth probing. Also, the mechanism behind the anesthetic-induced alternations of brain states remains poorly understood. In this study, an advanced EEG microstate spectral analysis was utilized using multivariate empirical mode decomposition in Hilbert-Huang transform. The practicability was further investigated in scalp EEG recordings during the propofol-induced transition of consciousness. The process of transition from the awake baseline to moderate sedation was accompanied by apparent increases in microstate (A, B, and F) energy, especially in the whole-brain delta band, frontal alpha band and beta band. In comparison to other effective EEG-based parameters that commonly used to measure anesthetic depth, using the selected spectral features reached better performance (80% sensitivity, 90% accuracy) to estimate the brain states during sedation. The changes in microstate energy also exhibited high correlations with individual behavioral data during sedation. In a nutshell, the EEG microstate spectral analysis is an effective method to estimate brain states during propofol-induced sedation, giving great insights into the underlying mechanism. The generated spectral features can be promising markers to dynamically assess the consciousness level.

Seizure Prediction Using Directed Transfer Function and Convolution Neural Network on Intracranial EEG.

Automatic seizure prediction promotes the development of closed-loop treatment system on intractable epilepsy. In this study, by considering the specific information exchange between EEG channels from the perspective of whole brain activities, the convolution neural network (CNN) and the directed transfer function (DTF) were merged to present a novel method for patient-specific seizure prediction. Firstly, the intracranial electroencephalogram (iEEG) signals were segmented and the information flow features of iEEG signals were calculated by using the DTF algorithm. Then, these features were reconstructed as the channel-frequency maps according to channel pairs and the frequency of information flow. Finally, these maps were fed into the CNN model and the outputs were post-processed by the moving average approach to predict the epileptic seizures. By the evaluation of cross-validation method, the proposed algorithm achieved the averaged sensitivity of 90.8%, the averaged false prediction rate of 0.08 per hour. Compared to the random predictor and other existing algorithms tested on the Freiburg EEG dataset, our proposed method achieved better performance for seizure prediction in all patients. These results demonstrated that the proposed algorithm could provide an robust seizure prediction solution by using deep learning to capture the brain network changes of iEEG signals from epileptic patients.

Non-Canonical Microstate Becomes Salient in High Density EEG During Propofol-Induced Altered States of Consciousness.

Dynamically assessing the level of consciousness is still challenging during anesthesia. With the help of Electroencephalography (EEG), the human brain electric activity can be noninvasively measured at high temporal resolution. Several typical quasi-stable states are introduced to represent the oscillation of the global scalp electric field. These so-called microstates reflect spatiotemporal dynamics of coherent neural activities and capture the switch of brain states within the millisecond range. In this study, the microstates of high-density EEG were extracted and investigated during propofol-induced transition of consciousness. To analyze microstates on the frequency domain, a novel microstate-wise spectral analysis was proposed by the means of multivariate empirical mode decomposition and Hilbert-Huang transform. During the transition of consciousness, a map with a posterior central maximum denoted as microstate F appeared and became salient. The current results indicated that the coverage, occurrence, and power of microstate F significantly increased in moderate sedation. The results also demonstrated that the transition of brain state from rest to sedation was accompanied by significant increase in mean energy of all frequency bands in microstate F. Combined with studies on the possible cortical sources of microstates, the findings reveal that non-canonical microstate F is highly associated with propofol-induced altered states of consciousness. The results may also support the inference that this distinct topography can be derived from canonical microstate C (anterior-posterior orientation). Finally, this study further develops pertinent methodology and extends possible applications of the EEG microstate during propofol-induced anesthesia.

Monitoring the Depth of Anesthesia Through the Use of Cerebral Hemodynamic Measurements Based on Sample Entropy Algorithm.

OBJECTIVE: The aim of this study is to explore the relationship between the depth of anesthesia and the cerebral hemodynamic variables during the complete anesthesia process. METHODS: In this study, near-infrared spectroscopy signals were used to record eight kinds of cerebral hemodynamic variables, including left, right, proximal, distal deoxygenated (Hb) and oxygenated (HbO2) hemoglobin concentration changes. Then, by measuring the complexity information of cerebral hemodynamic variables, the sample entropy was calculated as a new index of monitoring the depth of anesthesia. RESULTS: By means of receiver operating characteristic curve analysis, the sample entropy approach was proved to effectively discriminate anesthesia maintenance and waking phases. The discriminatory ability of HbO2 signals was stronger than that of Hb signals and the distal signals had weaker discrimination capability when compared with the proximal signals. In addition, there was statistical consistency between the bispectral index and sample entropy of cerebral hemodynamic variables during the complete anesthesia process. Moreover, the cerebral hemodynamic signals could not be interfered by clinical electrical devices. CONCLUSION: The sample entropy of cerebral hemodynamic variables could be suitable as a new index for monitoring the depth of anesthesia. SIGNIFICANCE: This study is very meaningful for developing new modality and decoding methods in perspective of anesthesia surveillance and may result in the anesthesia monitoring system with high performance.

An improved method to evaluate heart rate variability based on time-variant cardiorespiratory relation.

Frequency domain analysis of heart rate variability (HRV) is a noninvasive method to evaluate the autonomic nervous system (ANS), but the traditional parameters of HRV, i.e., the power spectra of the high-frequency (HF) and low-frequency bands (LF), cannot estimate the activity of the parasympathetic (PNS) and sympathetic nervous systems (SNS) well. The aim of our study was to provide a corrected method to better distinguish the contributions of the PNS and SNS in the HRV spectrum. Respiration has a gating effect on cardiac vagal efferent activity, which induces respiration-locked heart rate (HR) changes because of the fast effect of the PNS. So the respiration-related heart rate (HRr) is closely related to PNS activity. In this study, HR was decomposed into HRr and the respiration-unrelated component (HRru) based on empirical mode decomposition (EMD) and the relationship between HR and respiration. Time-frequency analysis of HRr and HRru was defined as HFr and LFru, respectively, with specific adaptive bands for every signal. Two experimental data sets, representing SNS and PNS activation, respectively, were used for efficiency analysis of our method. Our results show that the corrected HRV predicted ANS activity well. HFr could be an index of PNS activity, LFru mainly reflected SNS activity, and LFru/HFr could be more accurate in representing the sympathovagal balance.NEW & NOTEWORTHY This study includes the time-varying relationship between respiration and heart rate in the analysis of heart rate variability. Correction for low-frequency and high-frequency components based on respiration significantly improved evaluation of the sympathetic and parasympathetic nervous systems.

[Hidden Markov models for human interactive behavior].

In order to meet the requirements in the cooperation and competition experiments for an individual patient in clinical application, two human interactive behavior key-press models based on hidden Markov model (HMM) were proposed. To validate the cooperative and competitive models, a verification experimental task was designed and the data were collected. The correlation of the score and subjects' participation level has been used to analyze the reasonability verification. Behavior verification was conducted by comparing the statistical difference in response time for subjects between human-human and human-computer experiment. In order to verify the physiological validity of the models, we have utilized the coherence analysis to analyze the deep information of prefrontal brain area. Reasonability verification shows that the correlation coefficient for the training data and the testing data is 0.883 1 and 0.578 6 respectively based on cooperation model, and 0.813 1 and 0.617 8 respectively based on the competition model. The behavioral verification result shows that the cooperation and competition models have an accuracy of 71.43% respectively. The results of physiological validity show that the deep information of prefrontal brain area could been extracted based on the cooperation and competition models, and reveal the consistency of coherence between the double key-press cooperative and competitive experiments, respectively. Above all, the high consistency is obtained between the cooperatio/competition model and the double key-press experiment by the behavioral and physiological evaluation results. Consequently, the cooperation and competition models could be applied to clinical trials.

Probing prefrontal cortex hemodynamic alterations during facial emotion recognition for major depression disorder through functional near-infrared spectroscopy.

OBJECTIVE: A serious issue in psychiatric practice is a lack of specific, objective biomarker to assist clinicians in establishing differential diagnosis and improving individualized treatment. Major depression disorder (MDD) is characterized by poorer ability in processing of facial emotional expressions. APPROACH: Applying a portable neuroimaging system using near-infrared spectroscopy, we investigated the prefrontal cortex hemodynamic activation changes during facial emotion recognition and rest periods for 27 MDD patients compared with 24 healthy controls (HC). MAIN RESULTS: The hemodynamic changes in the left prefrontal cortex for the MDD group showed significant differences in the median values and the Mayer wave power ratios of the oxygenated hemoglobin (oxy-Hb) and deoxygenated hemoglobin (deoxy-Hb) during the emotional face recognition compared with the HC subjects, indicating the abnormal oxidative metabolism and weaker local hemodynamic oscillations for the MDD. The mean cross wavelet coefficients and the average wavelet coherence coefficient between oxy-Hb and deoxy-Hb over the left prefrontal cortex, and also between the bilateral oxy-Hb in the MDD patients were significantly lower than the HC group, demonstrating abnormal locally functional connectivity over the left prefrontal cortex, and the inter-hemispheric connection between the bilateral prefrontal cortices. SIGNIFICANCE: These results suggested that the hemodynamic changes over the left prefrontal cortex and between the bilateral prefrontal cortices detected by fNIRS could provide reliable predictors for the diagnosis of the depression in clinic, and also supported the rationale for use of transcranial magnetic stimulation over the left dorsolateral prefrontal cortex to restore excitability of prefrontal cortex that exhibits diminished regulation of emotion-generative systems in the MDD patients.

Technology Development for Simultaneous Wearable Monitoring of Cerebral Hemodynamics and Blood Pressure.

For many cerebrovascular diseases both blood pressure (BP) and hemodynamic changes are important clinical variables. In this paper, we describe the development of a novel approach to noninvasively and simultaneously monitor cerebral hemodynamics, BP, and other important parameters at high temporal resolution (250 Hz sampling rate). In this approach, cerebral hemodynamics are acquired using near infrared spectroscopy based sensors and algorithms, whereas continuous BP is acquired by superficial temporal artery tonometry with pulse transit time based drift correction. The sensors, monitoring system, and data analysis algorithms used in the prototype for this approach are reported in detail in this paper. Preliminary performance tests demonstrated that we were able to simultaneously and noninvasively record and reveal cerebral hemodynamics and BP during people's daily activity. As examples, we report dynamic cerebral hemodynamic and BP fluctuations during postural changes and micturition. These preliminary results demonstrate the feasibility of our approach, and its unique power in catching hemodynamics and BP fluctuations during transient symptoms (such as syncope) and revealing the dynamic features of related events.

Epileptic Seizure Detection in Long-Term EEG Recordings by Using Wavelet-Based Directed Transfer Function.

GOAL: The accurate automatic detection of epileptic seizures is very important in long-term electroencephalogram (EEG) recordings. In this study, the wavelet decomposition and the directed transfer function (DTF) algorithm were combined to present a novel wavelet-based directed transfer function (WDTF) method for the patient-specific seizure detection. METHODS: First, five subbands were extracted from 19-channel EEG signals by using wavelet decomposition in a sliding window. Second, the information flow characteristics of five subbands and full frequency band of EEG signals were calculated by the DTF method. The intensity of the outflow information was then used to reduce the feature dimensionality. Finally, all features were combined to identify interictal and ictal EEG segments by the support vector machine classifier. RESULTS: By using fivefold cross validation, the proposed method had achieved excellent performance with the average accuracy of 99.4%, the average selectivity of 91.1%, the average sensitivity of 92.1%, the average specificity of 99.5%, and the average detection rate of 95.8%. CONCLUSION: The WDTF method is able to enhance seizure detection results in long-term EEG recordings of focal epilepsy patients. SIGNIFICANCE: This study may lead to the development of seizure detection system with high performance, thus reducing the workload of epileptologists and facilitating to take corresponding steps promptly after the seizure onset. The high-frequency activity in the epilepsy brain may be of great importance for investigating the pathological mechanism and treatment of seizure.

Epileptic Seizure Detection Based on Partial Directed Coherence Analysis.

Long-term video EEG epilepsy monitoring can help doctors diagnose and cure epilepsy. The workload of doctors to read the EEG signals of epilepsy patients can be effectively reduced by automatic seizure detection. The application of partial directed coherence (PDC) analysis as mechanism for feature extraction in the scalp EEG recordings for seizure detection could reflect the physiological changes of brain activity before and after seizure onsets. In this study, a new approach on the basis of PDC was proposed to detect the seizure intervals of epilepsy patients. First of all, the multivariate autoregressive model was established for a moving window and the direction and intensity of information flow based on PDC analysis was calculated. Then, the outflow information related to certain EEG channel could be obtained by summing up the intensity of information flow propagated to other EEG channels in order to reduce the feature dimensionality. At last, according to the pathological features of epileptic seizures, the outflow information was regarded as the input vectors to a support vector machine classifier for discriminating interictal periods and ictal periods of EEG signals. The proposed method had achieved a good performance with the correct rate of 98.3%, the selectivity rate of 67.88%, the sensitivity rate of 91.44%, the specificity rate of 99.34%, and the average detection rate of 95.39%, which demonstrated that this method was suitable for detecting the seizure intervals of epilepsy patients. By comparing with other existing techniques, the proposed method based on PDC analysis achieved significant improvement in terms of seizure detection.

The Removal of EOG Artifacts From EEG Signals Using Independent Component Analysis and Multivariate Empirical Mode Decomposition.

The recorded electroencephalography (EEG) signals are usually contaminated by electrooculography (EOG) artifacts. In this paper, by using independent component analysis (ICA) and multivariate empirical mode decomposition (MEMD), the ICA-based MEMD method was proposed to remove EOG artifacts (EOAs) from multichannel EEG signals. First, the EEG signals were decomposed by the MEMD into multiple multivariate intrinsic mode functions (MIMFs). The EOG-related components were then extracted by reconstructing the MIMFs corresponding to EOAs. After performing the ICA of EOG-related signals, the EOG-linked independent components were distinguished and rejected. Finally, the clean EEG signals were reconstructed by implementing the inverse transform of ICA and MEMD. The results of simulated and real data suggested that the proposed method could successfully eliminate EOAs from EEG signals and preserve useful EEG information with little loss. By comparing with other existing techniques, the proposed method achieved much improvement in terms of the increase of signal-to-noise and the decrease of mean square error after removing EOAs.

Development of motion resistant instrumentation for ambulatory near-infrared spectroscopy.

Ambulatory near-infrared spectroscopy (aNIRS) enables recording of systemic or tissue-specific hemodynamics and oxygenation during a person's normal activities. It has particular potential for the diagnosis and management of health problems with unpredictable and transient hemodynamic symptoms, or medical conditions requiring continuous, long-duration monitoring. aNIRS is also needed in conditions where regular monitoring or imaging cannot be applied, including remote environments such as during spaceflight or at high altitude. One key to the successful application of aNIRS is reducing the impact of motion artifacts in aNIRS recordings. In this paper, we describe the development of a novel prototype aNIRS monitor, called NINscan, and our efforts to reduce motion artifacts in aNIRS monitoring. Powered by 2 AA size batteries and weighting 350 g, NINscan records NIRS, ECG, respiration, and acceleration for up to 14 h at a 250 Hz sampling rate. The system's performance and resistance to motion is demonstrated by long term quantitative phantom tests, Valsalva maneuver tests, and multiparameter monitoring during parabolic flight and high altitude hiking. To the best of our knowledge, this is the first report of multiparameter aNIRS monitoring and its application in parabolic flight.

Denoised P300 and machine learning-based concealed information test method.

In this paper, a novel P300-based concealed information test (CIT) method was proposed to improve the efficiency of differentiating deception and truth-telling. Thirty subjects including the guilty and innocent performed the paradigm based on three types of stimuli. In order to reduce the influence from the occasional variability of cognitive states on the CIT, several single-trials from Pz in probe stimuli within each subject were first averaged. Then the three groups of features were extracted from these averaged single-trials. Finally, two classes of feature samples were used to train a support vector machine (SVM) classifier. Meanwhile, the optimal number of averaged Pz waveforms and some other parameter values in the classifiers were determined by the cross validation procedures. Results show that if choosing accuracy of 90% as a detecting standard of P3 component to classify a subject's status (guilty or innocent), our method can achieve individual diagnostic rate of 100%. The individual diagnostic rate of our method was higher than the results of the other related reports. The presented method improves efficiency of CIT, and is more practical, lower fatigue and less countermeasure behavior in comparison with previous report methods, which could extend the laboratory study to the practical application.

Nonlinear feature extraction of sleeping EEG signals.

This study calculated the spectrum entropy (SE), approximate entropy (ApEn), and Lem-Ziv complexity (LZC) of sleeping EEG signals of eight healthy adults. The statistical results show that all the three nonlinear features can clearly reflect sleeping stage. Among them, the SE is easy to calculate and traces varying sleeping periods fairly and consistently, while the ApEn performs even better but is relatively complicated. The LZC is also simple but loses information details in its preprocessing of original measurement data, which consequently down grades its consistency. Based on a tradeoff of efficiency and efficacy, we consider the SE would be a good feature for real-time tracing sleep stages. Some conclusions are reported based on this study.