Sleep stage classification is essential for clinical disease diagnosis and sleep quality assessment. Most of the existing methods for sleep stage classification are based on single-channel or single-modal signal, and extract features using a single-branch, deep convolutional network, which not only hinders the capture of the diversity features related to sleep and increase the computational cost, but also has a certain impact on the accuracy of sleep stage classification. To solve this problem, this paper proposes an end-to-end multi-modal physiological time-frequency feature extraction network (MTFF-Net) for accurate sleep stage classification. First, multi-modal physiological signal containing electroencephalogram (EEG), electrocardiogram (ECG), electrooculogram (EOG) and electromyogram (EMG) are converted into two-dimensional time-frequency images containing time-frequency features by using short time Fourier transform (STFT). Then, the time-frequency feature extraction network combining multi-scale EEG compact convolution network (Ms-EEGNet) and bidirectional gated recurrent units (Bi-GRU) network is used to obtain multi-scale spectral features related to sleep feature waveforms and time series features related to sleep stage transition. According to the American Academy of Sleep Medicine (AASM) EEG sleep stage classification criterion, the model achieved 84.3% accuracy in the five-classification task on the third subgroup of the Institute of Systems and Robotics of the University of Coimbra Sleep Dataset (ISRUC-S3), with 83.1% macro F1 score value and 79.8% Cohen’s Kappa coefficient. The experimental results show that the proposed model achieves higher classification accuracy and promotes the application of deep learning algorithms in assisting clinical decision-making.
The method of using deep learning technology to realize automatic sleep staging needs a lot of data support, and its computational complexity is also high. In this paper, an automatic sleep staging method based on power spectral density (PSD) and random forest is proposed. Firstly, the PSDs of six characteristic waves (K complex wave, δ wave, θ wave, α wave, spindle wave, β wave) in electroencephalogram (EEG) signals were extracted as the classification features, and then five sleep states (W, N1, N2, N3, REM) were automatically classified by random forest classifier. The whole night sleep EEG data of healthy subjects in the Sleep-EDF database were used as experimental data. The effects of using different EEG signals (Fpz-Cz single channel, Pz-Oz single channel, Fpz-Cz + Pz-Oz dual channel), different classifiers (random forest, adaptive boost, gradient boost, Gaussian naïve Bayes, decision tree, K-nearest neighbor), and different training and test set divisions (2-fold cross-validation, 5-fold cross-validation, 10-fold cross-validation, single subject) on the classification effect were compared. The experimental results showed that the effect was the best when the input was Pz-Oz single-channel EEG signal and the random forest classifier was used, no matter how the training set and test set were transformed, the classification accuracy was above 90.79%. The overall classification accuracy, macro average F1 value, and Kappa coefficient could reach 91.94%, 73.2% and 0.845 respectively at the highest, which proved that this method was effective and not susceptible to data volume, and had good stability. Compared with the existing research, our method is more accurate and simpler, and is suitable for automation.
Sleep stage scoring is a hotspot in the field of medicine and neuroscience. Visual inspection of sleep is laborious and the results may be subjective to different clinicians. Automatic sleep stage classification algorithm can be used to reduce the manual workload. However, there are still limitations when it encounters complicated and changeable clinical cases. The purpose of this paper is to develop an automatic sleep staging algorithm based on the characteristics of actual sleep data. In the proposed improved K-means clustering algorithm, points were selected as the initial centers by using a concept of density to avoid the randomness of the original K-means algorithm. Meanwhile, the cluster centers were updated according to the 'Three-Sigma Rule' during the iteration to abate the influence of the outliers. The proposed method was tested and analyzed on the overnight sleep data of the healthy persons and patients with sleep disorders after continuous positive airway pressure (CPAP) treatment. The automatic sleep stage classification results were compared with the visual inspection by qualified clinicians and the averaged accuracy reached 76%. With the analysis of morphological diversity of sleep data, it was proved that the proposed improved K-means algorithm was feasible and valid for clinical practice.
The quality of sleep has a great relationship with health and working efficiency. The result of sleep stage classification is an important indicator to measure the quality of sleep, and it is also an important way to diagnose and treat sleep disorders. In this paper, the method of detrended cross-correlation analysis (DCCA) was used to analyze sleep stage classification, sleep electroencephalograph signals, which were extracted from the MIT-BIH Polysomnographic Database randomly. The results showed that the average DCCA exponent of the awake period is smaller than that of the first stage of non-rapid eye movement (NREM) sleeps. It is well concluded that the method of studying the sleep electroencephalograph with this method is of great significance to improve the quality of sleep, to diagnose and to treat sleep disorders.
Sleep quality is closely related to human health. It is very important to correctly discriminate the sleep stages for evaluating sleep quality, diagnosing and analyzing the sleep-related disorders. Polysomnography (PSG) signals are commonly used to record and analyze sleep stages. Effective feature extraction and representation is one of the most important steps to improve the performance of sleep stage classification. In this work, a collaborative representation (CR) algorithm was adopted to re-represent the original extracted features from electroencephalogram signal, and then the kernel entropy component analysis (KECA) algorithm was further used to reduce the feature dimension of CR-feature. To evaluate the performance of CR-KECA, we compared the original feature, CR feature and readied CR feature (CR-PCA) after principal component analysis (PCA). The experimental results of sleep stage classification indicated that the CR-KECA method achieved the best performance compared with the original feature, CR feature, and CR-PCA feature with the classification accuracy of 68.74±0.46%, sensitivity of 68.76±0.43% and specificity of 92.19±0.11%. Moreover, CR algorithm had low computational complexity, and the feature dimension after KECA was much smaller, which made CR-KECA algorithm suitable for the analysis of large-scale sleep data.
Sleep status is an important indicator to evaluate the health status of human beings. In this paper, we proposed a novel type of unperturbed sleep monitoring system under pillow to identify the pattern change of heart rate variability (HRV) through obtained RR interval signal, and to calculate the corresponding sleep stages combined with hidden Markov model (HMM) under the no-perception condition. In order to solve the existing problems of sleep staging based on HMM, ensemble empirical mode decomposition (EEMD) was proposed to eliminate the error caused by the individual differences in HRV and then to calculate the corresponding sleep stages. Ten normal subjects of different age and gender without sleep disorders were selected from Guangzhou Institute of Respirator Diseases for heart rate monitoring. Comparing sleep stage results based on HMM to that of polysomnography (PSG), the experimental results validate that the proposed noninvasive monitoring system can capture the sleep stages S1–S4 with an accuracy more than 60%, and performs superior to that of the existing sleep staging scheme based on HMM.
The research of sleep staging is not only a basis of diagnosing sleep related diseases but also the precondition of evaluating sleep quality, and has important clinical significance. In recent years, the research of automatic sleep staging based on computer has become a hot spot and got some achievements. The basic knowledge of sleep staging and electroencephalogram (EEG) is introduced in this paper. Then, feature extraction and pattern recognition, two key technologies for automatic sleep staging, are discussed in detail. Wavelet transform and Hilbert-Huang transform, two methods for feature extraction, are compared. Artificial neural network and support vector machine (SVM), two methods for pattern recognition are discussed. In the end, the research status of this field is summarized, and development trends of next phase are pointed out.
Sleep stage classification is a necessary fundamental method for the diagnosis of sleep diseases, which has attracted extensive attention in recent years. Traditional methods for sleep stage classification, such as manual marking methods and machine learning algorithms, have the limitations of low efficiency and defective generalization. Recently, deep neural networks have shown improved results by the capability of learning complex pattern in the sleep data. However, these models ignore the intra-temporal sequential information and the correlation among all channels in each segment of the sleep data. To solve these problems, a hybrid attention temporal sequential network model is proposed in this paper, choosing recurrent neural network to replace traditional convolutional neural network, and extracting temporal features of polysomnography from the perspective of time. Furthermore, intra-temporal attention mechanism and channel attention mechanism are adopted to achieve the fusion of the intra-temporal representation and the fusion of channel-correlated representation. And then, based on recurrent neural network and inter-temporal attention mechanism, this model further realized the fusion of inter-temporal contextual representation. Finally, the end-to-end automatic sleep stage classification is accomplished according to the above hybrid representation. This paper evaluates the proposed model based on two public benchmark sleep datasets downloaded from open-source website, which include a number of polysomnography. Experimental results show that the proposed model could achieve better performance compared with ten state-of-the-art baselines. The overall accuracy of sleep stage classification could reach 0.801, 0.801 and 0.717, respectively. Meanwhile, the macro average F1-scores of the proposed model could reach 0.752, 0.728 and 0.700. All experimental results could demonstrate the effectiveness of the proposed model.
Sleep electroencephalogram (EEG) is an important index in diagnosing sleep disorders and related diseases. Manual sleep staging is time-consuming and often influenced by subjective factors. Existing automatic sleep staging methods have high complexity and a low accuracy rate. A sleep staging method based on support vector machines (SVM) and feature selection using single channel EEG single is proposed in this paper. Thirty-eight features were extracted from the single channel EEG signal. Then based on the feature selection method F-Score's definition, it was extended to multiclass with an added eliminate factor in order to find proper features, which were used as SVM classifier inputs. The eliminate factor was adopted to reduce the negative interaction of features to the result. Research on the F-Score with an added eliminate factor was further accomplished with the data from a standard open source database and the results were compared with none feature selection and standard F-Score feature selection. The results showed that the present method could effectively improve the sleep staging accuracy and reduce the computation time.