A hierarchical sequential neural network with feature fusion for sleep staging based on EOG and RR signals

Abstract

Objective. Currently, the automatic sleep staging methods mainly face two problems: the first problem is that although the algorithms which use electroencephalogram (EEG) signals perform well, acquiring EEG signals is complicated and uncomfortable; the second problem is that if the methods utilize physiological signals collected by user-friendly devices, such as cardiorespiratory signals, whose accuracies are hard to be accepted by clinicians, although the employed signals are easy and comfortable to acquire. Approach. To overcome the two issues, an automatic sleep staging method is proposed by developing a hierarchical sequential neural network to process only the electrooculogram (EOG) and R-R interval (RR) signals. The two signals are convenient and comfortable to acquire. The proposed network mainly contains two parts: comprehensive feature learning and sequence learning. The first part extracts hand-crafted features, and network trained features are simultaneously learned by a two-scale network. Then the two kinds of features are fused. The second part utilized a two-flow recurrent neural network (RNN) to learn temporal information between sleep epochs. Main results. The proposed method was evaluated on 86 subjects from two public databases, the Montreal archive of sleep studies (MASS) and sleep apnea (SA). The proposed method can discriminate five sleep stages with the F1-score of 0.781 and 0.740 for MASS and SA, respectively. And discriminate four stages with the F1-score of 0.858 and 0.802 for MASS and SA, respectively. Significance. The proposed method can achieve comparable performance as using EEG signals for sleep staging and have better performance compared to five related state-of-the-art methods. Model analysis displayed that the network can learn effective features and sequence information from EOG and RR signals. In summary, the proposed method is promising to enable new sleep monitoring in a more convenient way while having a good performance on sleep staging.

Cross-Comparison of EMG-to-Force Methods for Multi-DoF Finger Force Prediction Using One-DoF Training

“…The network trained features can preserve the morphological information since the convolution kernels (filters) are 2-D tensors with a certain length (filter size), which try to match the spatial characteristics in the high-density EMG signals. Besides, the LSTM can grasp time series information in the EMG signals [30], [31]. However, one limitation of the proposed neural networks is that the training procedure is time-consuming (approximately 30 minutes for each subject).…”

Automatic Sleep Stage Classification Based on Two-channel EOG and One-channel EMG

“…The EOG makes the main contribution for sleep scoring derived from the combination of EOG and EMG. The placements of EOG electrodes are close to the placements of Fp1 and Fp2 EEG electrodes [19]. Hence, the EOG recordings are highly in uenced by a portion of EEG activities [19].…”

“…The placements of EOG electrodes are close to the placements of Fp1 and Fp2 EEG electrodes [19]. Hence, the EOG recordings are highly in uenced by a portion of EEG activities [19]. During NREM sleep, EOG has a similar waveform with the EEG signals recorded at the frontal poles [19].…”

“…Eye movements tend to slow down with the depth of sleep. EOG signals record the movement of the eyes is a fundamental indicator to distinguish between REM and NREM stages [19]. As shown in During N3 sleep in Fig.…”

“…The brain provides the most useful information about sleep regulation. EEG is the most important signal that directly re ects the state of the brain in sleep [19]. However, due to the structure of the scalp and the effects from the hair, novel materials or electrodes are still hard to obtain that can be used for EEG acquisition with comfort and high signal-to-noise ratio [19].…”