Efficient feature selection for sleep staging based on maximal overlap discrete wavelet transform and SVM

Automated Classification of Sleep Stages Using Single-Channel EEG A Machine Learning-Based Method

The main contribution of this paper is to present a novel approach for classifying the sleep stages based on optimal feature selection with ensemble learning stacking model using single-channel EEG signals.To find the suitable features from extracted feature vector, we obtained the ReliefF (ReF), Fisher Score (FS) and Online Stream Feature Selection (OSFS) selection algorithms.The proposed research work was performed on two different subgroups of sleep data of ISRUC-Sleep dataset. The experimental results of the proposed methodology signify that single-channel of EEG signal superior to other machine learning classification models with overall accuracies of 97.93%, 97%, and 95.96% using ISRUC-Sleep subgroup-I (SG-I) data and similarly the proposed model achieved an overall accuracies of 98.16%, 98.78%, and 95.26% using ISRUC-Sleep subgroup-III (SG-III) data with FS, ReF and OSFS respectively.

Comparison of Time-Frequency Analyzes for a Sleep Staging Application with CNN

Sleep staging is the process of acquiring biological signals during sleep and marking them according to the stages of sleep. The procedure is performed by an experienced physician and takes more time. When this process is automated, the processing load will be reduced and the time required to identify disease will also be reduced. In this paper, 8 different transform methods for automatic sleep-staging based on convolutional neural networks (CNNs) were compared to classify sleep stages using single-channel electroencephalogram (EEG) signals. Five different labels were used to stage the sleep. These are Wake (W), Non Rapid Eye Movement (NonREM)-1 (N1), NonREM-2 (N2), NonREM-3 (N3), and REM (R). The classifications were done end-to-end without any hand-crafted features, ie without requiring any feature engineering. Time-Frequency components obtained by Short Time Fourier Transform, Discrete Wavelet Transform, Discrete Cosine Transform, Hilbert-Huang Transform, Discrete Gabor Transform, Fast Walsh-Hadamard Transform, Choi-Williams Distribution, and Wigner-Willie Distribution were classified with a supervised deep convolutional neural network to perform sleep staging. The discrete Cosine Transform-CNN method (DCT-CNN) showed the highest performance among the methods suggested in this paper with an F1 score of 89% and a value of 0.86 kappa. The findings of this study revealed that the transformation techniques utilized for the most accurate representation of input data are far superior to traditional approaches based on manual feature extraction, which acquires time, frequency, or nonlinear characteristics. The results of this article are expected to be useful to researchers in the development of low-cost, and easily portable devices.

Sleep Staging Using Plausibility Score: A Novel Feature Selection Method Based on Metric Learning

As an effective method, feature selection can reduce computational complexity and improve classification performance. A number of criteria exist for feature selection using labeled data, unlabeled data and pairwise constraints, most of which are based on the Euclidean distance. In this paper, we propose a filter method for feature selection with pairwise constraints, aiming to jointly evaluate a feature subset based on metric learning. Two criteria are designed based on the well-known Kullback-Leibler divergence for measuring the difference between must-link constraints and cannot-link constraints that can indicate the feature subset discrimination based on Keep It Simple and Straightforward (KISS) metric learning and Cross-view Quadratic Discriminant Analysis (XQDA) metric learning. To address the challenging feature selection problem, we formulate a sequential search algorithm guided by indicators that are simplified from the proposed criteria. Furthermore, we conducted several experiments on sleep staging based on electroencephalogram (EEG) recordings from the Sleep-EDF Database Expanded. The experimental results demonstrate the effectiveness of the proposed method compared with nine representative feature selection methods. On the data set from healthy volunteers and the data set from volunteers that had mild difficulty falling asleep, the classification average accuracies achieve 97.66% and 93.57% by using the proposed method, respectively.

Motion Imagery-BCI Based on EEG and Eye Movement Data Fusion

Existing studies have demonstrated that eye tracking can be a complementary approach to Electroencephalogram (EEG) based brain-computer interaction (BCI), especially in improving BCI performance in visual perception and cognition. In this paper, we proposed a method to fuse EEG and eye movement data extracted from motor imagery (MI) tasks. The results of the tests showed that on the feature layer, the average MI classification accuracy from the fusion of EEG and eye movement data was higher than that of pure EEG data or pure eye movement data, respectively. Besides, we also found that the average classification accuracy from the fusion on the decision layer was higher than that from the feature layer. Additionally, when EEG data were not available for the shifting of parts of electrodes, we combined EEG data collected from the rest of the electrodes (only 50% of the original) with the eye movement data, and the average MI classification accuracy was only 1.07% lower than that from all available electrodes. This result indicated that eye movement data was feasible to compensate for the loss of the EEG data in the MI scenario. Overall our approach was proved valuable and useful for augmenting MI based BCI applications.

A hybrid self-attention deep learning framework for multivariate sleep stage classification

BACKGROUND

Sleep is a complex and dynamic biological process characterized by different sleep patterns. Comprehensive sleep monitoring and analysis using multivariate polysomnography (PSG) records has achieved significant efforts to prevent sleep-related disorders. To alleviate the time consumption caused by manual visual inspection of PSG, automatic multivariate sleep stage classification has become an important research topic in medical and bioinformatics.

RESULTS

We present a unified hybrid self-attention deep learning framework, namely HybridAtt, to automatically classify sleep stages by capturing channel and temporal correlations from multivariate PSG records. We construct a new multi-view convolutional representation module to learn channel-specific and global view features from the heterogeneous PSG inputs. The hybrid attention mechanism is designed to further fuse the multi-view features by inferring their dependencies without any additional supervision. The learned attentional representation is subsequently fed through a softmax layer to train an end-to-end deep learning model.

CONCLUSIONS

We empirically evaluate our proposed HybridAtt model on a benchmark PSG dataset in two feature domains, referred to as the time and frequency domains. Experimental results show that HybridAtt consistently outperforms ten baseline methods in both feature spaces, demonstrating the effectiveness of HybridAtt in the task of sleep stage classification.

Automated sleep stage classification based on tracheal body sound and actigraphy

The current gold standard for assessment of most sleep disorders is the in-laboratory polysomnography (PSG). This approach produces high costs and inconveniences for the patients. An accessible and simple preliminary screening method to diagnose the most common sleep disorders and to decide whether a PSG is necessary or not is therefore desirable. A minimalistic type-4 monitoring system which utilized tracheal body sound and actigraphy to accurately diagnose the obstructive sleep apnea syndrome was previously developed. To further improve the diagnostic ability of said system, this study aims to examine if it is possible to perform automated sleep staging utilizing body sound to extract cardiorespiratory features and actigraphy to extract movement features.

A linear discriminant classifier based on those features was used for automated sleep staging using the type-4 sleep monitor. For validation 53 subjects underwent a full-night screening at Ulm University Hospital using the developed sleep monitor in addition to polysomnography. To assess sleep stages from PSG, a trained technician manually evaluated EEG, EOG, and EMG recordings.

The classifier reached 86.9% accuracy and a Kappa of 0.69 for sleep/wake classification, 76.3% accuracy and a Kappa of 0.42 for Wake/REM/NREM classification, and 56.5% accuracy and a Kappa of 0.36 for Wake/REM/light sleep/deep sleep classification. For the calculation of sleep efficiency (SE), a coefficient of determination r² of 0.78 is reached. Additionally, subjects were classified into groups of SEs (SE≥40%, SE≥60% and SE≥80%). A Cohen’s Kappa >0.61 was reached for all groups, which is considered as substantial agreement.

The presented method provides satisfactory performance in sleep/wake and wake/REM/NREM sleep staging while maintaining a simple setup and offering high comfort. This minimalistic approach may address the need for a simple yet reliable preliminary sleep screening in an ambulatory setting.

Absence seizure susceptibility correlates with pre-ictal β oscillations

Absence seizures are generalized, cortico-thalamo-cortical (CTC) high power electroencephalographic (EEG) or electrocorticographic (ECoG) events that initiate and terminate suddenly. ECoG recordings of absence seizures in animal models of genetic absence epilepsy show a sudden spike-wave-discharge (SWD) onset that rapidly emerges from normal ECoG activity. However, given that absence seizures occur most often during periods of drowsiness or quiet wakefulness, we wondered whether SWD onset correlates with pre-ictal changes in network activity. To address this, we analyzed ECoG recordings of both spontaneous and induced SWDs in rats with genetic absence epilepsy. We discovered that the duration and intensity of spontaneous SWDs positively correlate with pre-ictal 20-40Hz (β) spectral power and negatively correlate with 4-7Hz (Ø) power. In addition, the output of thalamocortical neurons decreases within the same pre-ictal window of time. In separate experiments we found that the propensity for SWD induction was correlated with pre-ictal β power. These results argue that CTC networks undergo a pre-seizure state transition, possibly due to a functional reorganization of cortical microcircuits, which leads to the generation of absence seizures.

An open-source toolbox for standardized use of PhysioNet Sleep EDF Expanded Database

PhysioNet Sleep EDF database has been the most popular source of data used for developing and testing many automatic sleep staging algorithms. However, the recordings from this database has been used in an inconsistent fashion. For example, arbitrary selection of start and end times from long term recordings, data-hypnogram mismatches, different performance metrics and hypnogram conversion from R&K to AASM. All these differences result in different data sections and performance metrics being used by researchers thereby making any direct comparison between algorithms very difficult. Recently, a superset of this database has been made available on PhysioNet, known as the Sleep EDF Expanded Database which includes 61 recordings. This provides an opportunity to standardize the way in which signals from this database should be used. With this goal in mind, we present in this paper a toolbox for automatically downloading and extracting recordings from the Sleep EDF Expanded database and converting them to a suitable format for use in MATLAB. This toolbox contains functions for selecting appropriate data for sleep analysis (based on our previous recommendations for sleep staging), hypnogram conversion and computation of performance metrics. Its use makes it simpler to start using the new sleep database and also provides a foundation for much-needed standardization in this research field.

Metric learning for automatic sleep stage classification

We introduce in this paper a metric learning approach for automatic sleep stage classification based on single-channel EEG data. We show that learning a global metric from training data instead of using the default Euclidean metric, the k-nearest neighbor classification rule outperforms state-of-the-art methods on Sleep-EDF dataset with various classification settings. The overall accuracy for Awake/Sleep and 4-class classification setting are 98.32% and 94.49% respectively. Furthermore, the superior accuracy is achieved by performing classification on a low-dimensional feature space derived from time and frequency domains and without the need for artifact removal as a preprocessing step.

A two-step automatic sleep stage classification method with dubious range detection

BACKGROUND

The limitations of the current systems of automatic sleep stage classification (ASSC) are essentially related to the similarities between epochs from different sleep stages and the subjects' variability. Several studies have already identified the situations with the highest likelihood of misclassification in sleep scoring. Here, we took advantage of such information to develop an ASSC system based on knowledge of subjects' variability of some indicators that characterize sleep stages and on the American Academy of Sleep Medicine (AASM) rules.

METHODS

An ASSC system consisting of a two-step classifier is proposed. In the first step, epochs are classified using support vector machines (SVMs) spread into different nodes of a decision tree. In the post-processing step, the epochs suspected of misclassification (dubious classification) are tagged, and a new classification is suggested. Identification and correction are based on the AASM rules, and on misclassifications most commonly found/reported in automatic sleep staging. Six electroencephalographic and two electrooculographic channels were used to classify wake, non-rapid eye movement (NREM) sleep--N1, N2 and N3, and rapid eye movement (REM) sleep.

RESULTS

The proposed system was tested in a dataset of 14 clinical polysomnographic records of subjects suspected of apnea disorders. Wake and REM epochs not falling in the dubious range, are classified with accuracy levels compatible with the requirements for clinical applications. The suggested correction assigned to the epochs that are tagged as dubious enhances the global results of all sleep stages.

CONCLUSIONS

This approach provides reliable sleep staging results for non-dubious epochs.

Adaptive automatic sleep stage classification under covariate shift

Current automatic sleep stage classification (ASSC) methods that rely on polysomnographic (PSG) signals suffer from inter-subject differences that make them unreliable in facing with new and different subjects. A novel adaptive sleep scoring method based on unsupervised domain adaptation, aiming to be robust to inter-subject variability, is proposed. We assume that the sleep quality variants follow a covariate shift model, where only the sleep features distribution change in the training and test phases. The maximum overlap discrete wavelet transform (MODWT) is applied to extract relevant features from EEG, EOG and EMG signals. A set of significant features are selected by minimum-redundancy maximum-relevance (mRMR) which is a powerful feature selection method. Finally, an instance-weighting method, namely the importance weighted kernel logistic regression (IWKLR) is applied for the purpose of obtaining adaptation in classification. The classification results using leave one out cross-validation (LOOCV), show that the proposed method performs at the state-of-the art in the field of ASSC.