Sleep Stage Classification Using Non-Invasive Bed Sensing and Deep Learning

Sleep is vital for our body’s physical restoration, but sleep disorders can cause various problems. Determining sleep stages is essential for diagnosing and curing such disorders. Polysomnography (PSG) signals are recordings of brain activity, eye movements, muscle activity and other physiological signals that are collected during a sleep study. Insomnia, Sleep Apnea, and Restless Legs Syndrome are some of the sleep problems that can be identified using these signals. However, analysing PSG signals manually can be time-consuming and prone to errors. Deep Learning Models such as Convolutional Neural Networks (CNN), can be used to automate the analysis of PSG signals. CNN is followed by Long-Short Term Memory (LSTM) and CNN are used as a stack ensemble method to recognize patterns in the signals that correspond to different sleep stages and events. By training these models on large datasets of PSG signals, they can detect the disorders. The dataset is collected from PhysioNet Sleep-EDF dataset that consists of PSG signals. The accuracies obtained using different training and testing data using CNN and CNN-LSTM are 95.15% and 83.9% respectively, and using metadata classifier the overall accuracy is increased by 1%. The future enhancement of the paper can be done by considering Heart rate, EEG Pz-oz signals and EEG Pz-oz along with EEG Fpz-cz.

Sleep stage classification is important to accurately predict and diagnose patients with sleep disorders. Though various deep learning approaches have been implemented to classify sleep classes, these consist limitations that impact the accuracy and processing time of the classification model. The aim of this research is to enhance the accuracy and minimize the training time of the deep learning classification model. The proposed system consists of One Dimensional Convolutional Neural Network (CNN) with enhanced loss function to improve the accuracy of scoring of five different sleep classes. Preprocessing, Feature Extraction and Classification are the main components of the proposed system. Initially, EEG signals are fed to an adaptive filter for preprocessing, in order to remove any noise in signal. Thereafter, feature is extracted through multiple convolutional and pooling layers, and finally the classification is done by fully connected layer using softmax activation with enhanced loss function. The proposed solution is tested on data samples from multiple datasets with five classes of Sleep classification. Based on the obtained results, the proposed solution has found to achieve an accuracy of 96.26% which is almost 4.2% higher than the state-of-the-art solution which is 92.76%. Furthermore, the processing time has been reduced by 11 milliseconds against the state-of-the-art solution. The proposed system focused on classifying sleep stages in five classes using EEG signals with deep learning approach. It enhances the loss function in order to minimize errors in the prediction of sleep classes and improves the accuracy of the model. Furthermore, the training speed of the model has also been reduced by applying batch normalization techniques inside the model. In the future, larger datasets of different sleep disorder patients with varying features can be used for training and implementing the proposed solution. The datasets can also be pre-processed using additional techniques to refine the data before feeding to the neural network model.

Cardiovascular Disease and Sleep Disordered Breathing: Are Children Vulnerable?

The clinical polysomnographic (PSG) reports of 2,650 consecutive adults studied during 41 months were reviewed retrospectively to identify all patients treated with fluoxetine or tricyclic antidepressants. The PSG reports of four other adult groups were also reviewed: periodic limb movement (PLM) disorder (n = 28); sleep terror/sleepwalking (ST/SW) (n = 54); rapid eye movement (REM) sleep behavior disorder (RBD) (n = 70); patients with clinically unremarkable sleep during two consecutive PSG studies (n = 30). Standard PSG recording and scoring methods were employed. A total of 1.5% (n = 41) and 2.0% (n = 52) of patients were receiving fluoxetine or tricyclics (amitriptyline or nortriptyline, n = 31; imipramine or desipramine, n = 16; protriptyline or trimipramine, n = 5). A selective association between fluoxetine and extensive, prominent eye movements in nonrapid eye movement (NREM) sleep was detected, utilizing Fisher's exact one-tailed statistic (p less than 0.00001 for each comparison). The detection rates were fluoxetine, 48.8% (20/41); tricyclics, 5.8% (3/52); RBD, 4.3% (3/70); objectively normal sleepers, 3.3% (1/30); PLM, ST/SW, 0% (0/82). These groups had similar mean ages (31.5-45.4 years) and gender distributions (50.0-60.7% male), apart from RBD. The effect of fluoxetine, a potent and specific serotonin reuptake inhibitor, on NREM eye movements is postulated to derive from potentiation of serotonergic neurons that inhibit brainstem "omnipause neurons", which, in turn, inhibit saccadic eye movements, thus resulting in disinhibited release of saccades. In addition, a 31-year-old man with obsessive-compulsive disorder developed RBD soon after starting fluoxetine therapy, which persisted at PSG study 19 months after fluoxetine discontinuation.

Primary headaches are closely related to sleep. Modifications in the patterns of arousal during sleep have been reported in migraine, especially in the nights preceding a headache attack. We aimed at evaluating the pattern of arousal from sleep in a group of patients affected by sleep-related migraine. We enrolled 10 consecutive patients, three males and seven females, aged between 20 and 62 years, who presented frequent attacks of migraine without aura (more than five per month), closely related to sleep (more than one-half of the attacks occurred during sleep, causing an awakening). A control group was studied, matched for age and sex. Patients and controls underwent a full-night polysomnographic study, following adaptation; arousal pattern was studied by the scoring of the high-frequency EEG arousal and by the cyclic alternating pattern (CAP). Migraineurs showed a lower CAP rate in non-rapid eye movement (NREM) sleep and, in particular, a lower number of A1 phases (low-frequency, high-amplitude EEG bursts) compared with the controls. Migraineurs also showed a lower index of high-frequency EEG arousals during rapid eye movement (REM) sleep. The reduction in the CAP rate indicates a lower level of arousal fluctuation in NREM sleep. The reduced arousal index in REM suggests a dysfunction in neural structures involved in both the control of REM sleep and the pathophysiology of migraine, such as the hypothalamus and the brainstem.

Introduction Most of the current automatic polysomnography sleep staging methods use multi-signal data and require a sequence of preceding and following epochs to score the stage of a specific epoch, which may not be desirable for analysis in real-time and/or in free-living conditions. We developed a deep learning-based sleep staging algorithm, namely SleepInceptionNet, that is designed to score each epoch using only single-channel electroencephalogram (EEG) data within that specific epoch. Methods Polysomnography data of 883 participants (937,975 thirty-second epochs) in the Multi-Ethnic Study of Atherosclerosis (MESA) obtained from the National Sleep Research Resource (NSRR) were randomly divided into a separate training/validation set of 194 participants and a test set of 689 participants. Each 30-second raw central EEG channel signal was transformed to time-frequency domain images using continuous wavelet transform method. Sleep stage in each epoch was obtained using SleepInceptionNet, in which the InceptionV3 convolutional neural network structure was trained and tuned on the time-frequency images of the training set to classify each epoch into one of the five stages of Wake, N1, N2, N3, or rapid eye movement (REM) sleep. Results Compared to the ground truth manually scored polysomnography sleep stages, SleepInceptionNet achieved an overall kappa agreement of 0.690 and overall weighted accuracy of 0.897. The model showed accuracy (mean±SD across the test set participants) of 0.940±0.067 in detecting Wake, 0.883±0.047 in detecting N1, 0.845±0.055 in detecting N2, 0.939±0.038 in detecting N3, and 0.930±0.038 in detecting REM epochs, in reference to manually scored polysomnography. Conclusion SleepInceptionNet showed a high agreement with manually scored polysomnography in epoch-by-epoch classification of sleep stages. This study demonstrates the viability of real-time, accurate sleep staging using a single-channel EEG, which could have a variety of applications such as delivery of on-demand interventions during specific sleep stages in free-living conditions. Support (If Any) MESA Sleep Ancillary study was funded by NIH-NHLBI Association of Sleep Disorders with Cardiovascular Health Across Ethnic Groups (RO1 HL098433). MESA is supported by NHLBI (HHSN268201500003I, N01-HC-95159, N01-HC-95160, N01-HC-95161, N01-HC-95162, N01-HC-95163, N01-HC-95164, N01- HC-95165, N01-HC-95166, N01-HC-95167, N01-HC-95168, and N01-HC-95169) and NCATS (UL1-TR-000040, UL1-TR-001079, and UL1-TR-001420). NSRR was supported by NHLBI (R24 HL114473, 75N92019R002).

PurposeObstructive sleep apnea (OSA) is a prevalent sleep-related breathing disorder. Research conducted on patients with OSA using electroencephalography (EEG) has revealed a noticeable shift in the overnight polysomnography (PSG) power spectrum. To better quantify the effects of OSA on brain function and to identify the most reliable predictors of pathological cortical activation, this study quantified the PSG power and its association with the degree of hypoxia in OSA patients.Patients and MethodsThis retrospective study recruited 93 patients with OSA. OSA patients were divided into three groups based on their apnea-hypopnea index (AHI) scores. The clinical characteristics and sleep macrostructure of these patients were examined, followed by an analysis of PSG signals. Power spectral density (PSD) in five frequency bands was analyzed during nonrapid eye movement (NREM) sleep, rapid eye movement (REM) sleep, and wakefulness. Finally, correlation analysis was conducted to assess the relationships among PSD, PSG parameters, and serum levels of S100β and uric acid.ResultsObstructive sleep apnea occurred during both the NREM and REM sleep phases. Except for a decrease in the duration of N2 sleep and an increase in the microarousal index, there were no significant differences in sleep architecture based on disease severity. Compared to the mild OSA group, the theta and alpha band PSD in the frontal and occipital regions during NREM sleep and wakefulness were significantly decreased in the moderate and severe OSA groups. Correlation analysis revealed that theta PSD in N1 and N3 stages were negatively correlated the AHI, oxygen desaturation index, SaO2<90% and microarousal index.ConclusionThese findings imply that patients with more severe OSA exhibited considerable NREM hypoxia and abnormal brain activity in the frontal and occipital regions. Therefore, sleep EEG oscillation may be a useful neurophysiological indicator for assessing brain function and disease severity in patients with OSA.

Electroencephalogram (EEG) signal is often used to assess sleep quality and treat sleep disorders. Many existing methods usually obtain high accuracy through a large number of feature preprocessing and feature extraction of EEG signals, which need a lot of prior knowledge as the basis. In this paper, a novel sleep stage classification framework, named FRL&S2SL, is proposed. The framework combines fast representation learning (FRL) and semantic-to-signal learning (S2SL) and uses single-channel EEG without any preprocessing of EEG signals. In the proposed framework, we utilize convolutional neural networks (CNN) to extract time-invariant features and bidirectional long short-term memory (BiLSTM) models to extract temporal features. Furthermore, auxiliary classifier generative adversarial network (ACGAN) is used to embed semantic features into signal features and to extract knowledge domain features of EEG signals for the first time. According to the American Academy of Sleep Medicine (AASM), sleep is divided into five stages: awake, rapid eye movement (REM) and three non-rapid eye movement (N1/N2/N3). We evaluated our framework using single-channel EEG (Fpz-Oz) from Sleep-EDF dataset, which is subject to the standards specified by AASM. The results show that our framework has achieved state-of-the-art in many evaluation metrics.

<h2>Abstract</h2> Sleep is vital for our physical and mental health, and sound sleep can help us to focus on daily activities. Therefore, a sleep study that includes sleep patterns and disorders is crucial to enhancing our knowledge about individual health status. The findings on sleep stages and sleep disorders relied on Polysomnography and self-report measures, and then the study went through clinical assessments by expert physicians. Artificial Intelligence has made the evaluation process of sleep stages and sleep disorders classification more efficient. Many studies have focused on analyzing various datasets using advanced techniques and algorithms to improve computational ease and accuracy. This review aims to comprehensively analyze recent literature on different approaches and outcomes in sleep studies, specifically on ‘sleep stages classification' and ‘sleep disorder detection' using Artificial Intelligence. Initially, 185 articles were selected from top journals, and eventually, 81 of them were reviewed in detail, covering the period from 2016 to 2023. Brain waves are the most commonly used body signals for studying sleep patterns and disorders. Almost 36 % of the research exclusively used brain activity signals, and 80 % combined them with other body parameters in sleep staging. The Neural Network algorithms are the most popular, having 47 % of the total usage. At the same time, Long Short-Term Memory, Ensemble Learning, Support Vector Machine, and Random Forest accounted for 15 %, 12 %, 7 %, and 6 % of usage, respectively. For evaluating AI model performance, accuracy or precision is used in 86.42 % of cases, followed by an F1 score of 46.91 %, Kappa of 39.51 %, Specificity of 30.86 %, Sensitivity of 29.63 %, along with other metrics. This article would help physicians and researchers get the gist of Artificial Intelligence's contribution to sleep studies and the feasibility of their intended work.

Poor quality and poor duration of sleep have been associated with cognitive decline, diseases, and disorders. Therefore, sleep studies are imperative to recapitulate phenotypes associated with poor sleep quality and uncover mechanisms contributing to psychopathology. Classification of sleep stages, vigilance state bout durations, and number of transitions amongst vigilance states serves as a proxy for evaluating sleep quality in preclinical studies. Currently, the gold standard for sleep staging is expert human inspection of polysomnography (PSG) obtained from preclinical rodent models and this approach is immensely time consuming. To accelerate the analysis, we developed a deep-learning-based software tool for automated sleep stage classification in rats. This study aimed to develop an automated method for classifying three sleep stages in rats (REM/paradoxical sleep, NREM/slow-wave sleep, and wakefulness) using a deep learning approach based on single-channel EEG data. Single-channel EEG data were acquired from 16 rats, each undergoing two 24 h recording sessions. The data were labeled by human experts in 10 s epochs corresponding to three stages: REM/paradoxical sleep, NREM/slow-wave sleep, and wakefulness. A deep neural network (DNN) model was designed and trained to classify these stages using the raw temporal data from the EEG. The DNN achieved strong performance in predicting the three sleep stages, with an average F1 score of 87.6% over a cross-validated test set. The algorithm was able to predict key parameters of sleep architecture, including total bout duration, average bout duration, and number of bouts, with significant accuracy. Our deep learning model effectively automates the classification of sleep stages using single-channel EEG data in rats, reducing the need for labor-intensive manual annotation. This tool enables high-throughput sleep studies and may accelerate research into sleep-related pathologies. Furthermore, we provide over 700 h of expert-scored sleep data, available for public use in future research studies.

Rapid eye movement sleep behaviour disorder is characterized by dream-enacting behaviour and impaired motor inhibition during rapid eye movement sleep. Rapid eye movement sleep behaviour disorder is commonly associated with neurodegenerative disorders, but also reported in narcolepsy with cataplexy. Most narcolepsy with cataplexy patients lack the sleep-wake, and rapid eye movement sleep, motor-regulating hypocretin neurons in the lateral hypothalamus. In contrast, rapid eye movement sleep behaviour disorder and hypocretin deficiency are rare in narcolepsy without cataplexy. We hypothesized that rapid eye movement sleep behaviour disorder coexists with cataplexy in narcolepsy due to hypocretin deficiency. In our study, rapid eye movement sleep behaviour disorder was diagnosed by the International Classification of Sleep Disorders (2nd edition) criteria in 63 narcolepsy patients with or without cataplexy. Main outcome measures were: rapid eye movement sleep behaviour disorder symptoms; short and long muscle activations per hour rapid eye movement and non-rapid eye movement sleep; and periodic and non-periodic limb movements per hour rapid eye movement and non-rapid eye movement sleep. Outcome variables were analysed in relation to cataplexy and hypocretin deficiency with uni- and multivariate logistic/linear regression models, controlling for possible rapid eye movement sleep behaviour disorder biasing factors (age, gender, disease duration, previous anti-cataplexy medication). Only hypocretin deficiency independently predicted rapid eye movement sleep behaviour disorder symptoms (relative risk = 3.69, P = 0.03), long muscle activations per hour rapid eye movement sleep (ln-coefficient = 0.81, P < 0.01), and short muscle activations per hour rapid eye movement sleep (ln-coefficient = 1.01, P < 0.01). Likewise, periodic limb movements per hour rapid eye movement and non-rapid eye movement sleep were only associated with hypocretin deficiency (P < 0.01). A significant association between hypocretin deficiency and cataplexy was confirmed (P < 0.01). In a sub-analysis, hypocretin deficiency suggested the association of periodic limb movements and rapid eye movement sleep behaviour disorder outcomes (symptoms, non-periodic short and long muscle activity) in rapid eye movement sleep. Our results support the hypothesis that hypocretin deficiency is independently associated with rapid eye movement sleep behaviour disorder in narcolepsy. Thus, hypocretin deficiency is linked to the two major disturbances of rapid eye movement sleep motor regulation in narcolepsy: rapid eye movement sleep behaviour disorder and cataplexy. Hypocretin deficiency is also significantly associated with periodic limb movements in rapid eye movement and non-rapid eye movement sleep, and provides a possible pathophysiological link between rapid eye movement sleep behaviour disorder and periodic limb movements in narcolepsy. The study supports the hypothesis that an impaired hypocretin system causes a general instability of motor regulation during wakefulness, rapid eye movement and non-rapid eye movement sleep in human narcolepsy.

Ischemic stroke selectively inhibits REM sleep of rats

Automatic sleep stage monitoring is an essential tool for the diagnosis and treatment of sleep-related disorders effectively. Although extensive studies focusing on single source data or information, such as single channel electroencephalogram (EEG), electrooculogram (EOG), and electromyogram (EMG), with machine learning (ML) and deep learning (DL) have been presented in this field of automatic sleep staging, there are very limited literature's that have investigated the impact of multisource information fusion on ML or DL based schemes for the sleep stage classification. In addition, exploiting recurrent neural network (RNN) to learn temporal information from the sequences of multi-source inputs can bring significant improvement in classifying the sleep stage. Therefore, we aim to develop an multi-source data fusion, more specifically two source's signal fusion, enabled end-to-end convolutional recurrent neural network (CRNN) in order to perform automatic sleep stage classification in this article. In the proposed scheme, no hand crafted features is used to train the model and hence the proposed model can be called an end-to-end approach. The proposed CRNN model can extract time-invariant features from raw data of two input sources and can fuse them to learn temporally correlated features. Sleep-EDF expanded benchmark dataset is used in this study and we have employed Fpz-Cz EEG channel signal and EOG signal as input with 3 variants of RNN layer in model architecture separately in order to evaluate and investigate the performance. The proposed CRNN model with standard RNN layers can classify the Sleep-EDF dataset with maximum average accuracy of 90.30% and maximum Cohen's kappa coefficient of 86.86 compared other two RNN variant's layer (e.g., gated recurrent unit and long short term memory) based models which is very promising in the studies on classification of sleep stages as well as can be regarded as an alternative to conventional sleep staging.

Sleep monitoring typically requires the uncomfortable and expensive polysomnography (PSG) test to determine the sleep stages. Body movement and cardiopulmonary signals provide an alternative way to perform sleep staging. In recent years, long-short term memory (LSTM) networks and convolutional neural networks (CNN) have dominated automatic sleep staging due to their better learning ability than machine learning classifiers. However, LSTM may lose information when dealing with long sequences, while CNN is not good at sequence modeling. As an improvement, we develop a hierarchical attention-based deep learning method for sleep staging using body movement, electrocardiogram (ECG), and abdominal breathing signals. We apply the multi-head self-attention to model the global context of feature sequences and coupled it with CNN to achieve a hierarchical self-attention weight assignment. We evaluate the performance of the method using two public datasets. Our method outperforms other baselines in the three sleep stages, achieving an accuracy of 84.3%, an F1 score of 0.8038, and a Cohen's Kappa coefficient of 0.7036. The result demonstrates the effectiveness of the hierarchical self-attention mechanism when processing feature sequences in the sleep stage classification problem. This paper provides new possibilities for long-term sleep monitoring using movement and cardiopulmonary signals obtained from non-invasive devices. The code can be found at: https://github.com/scutrd/attention-sleep-staging.

Sleep quality in patients studied with laboratory-based polysomnography may differ from sleep quality in patients studied at home but remains clinically relevant and important to describe. We assessed objective sleep quality and explored factors associated with poor sleep in patients undergoing laboratory-based polysomnography. We reviewed diagnostic polysomnography studies from a 10-year period at a single sleep center. Total sleep time (TST) and sleep efficiency (SE) were assessed as markers of sleep quality. Poor sleep was defined as TST ≤ 4 hours or SE ≤ 50%. Multivariable analysis was performed to determine associations between objective sleep quality as an outcome and multiple candidate predictors including age, sex, race, body mass index, comorbidities, severity of obstructive sleep apnea, and central nervous system medications. Among 4957 patients (age 53 ± 15 years), average TST and median SE were 5.8 hours and 79%, respectively. There were 556 (11%) and 406 (8%) patients who had poor sleep based on TST and SE, respectively. In multivariable analysis, those who were older (per 10 years: 1.48 [1.34, 1.63]), male (1.38 [1.14,1.68]), and had severe obstructive sleep apnea (1.76 [1.28, 2.43]) were more likely to have short sleep. Antidepressant use was associated with lower odds of short sleep (0.77 [0.59,1.00]). Older age (per 10 years: 1.48 [1.34, 1.62]), male sex (1.34 [1.07,1.68]), and severe obstructive sleep apnea (2.16 [1.47, 3.21]) were associated with higher odds of poor SE. We describe TST and SE from a single sleep center cohort. Multiple demographic characteristics were associated with poor objective sleep in patients during laboratory-based polysomnography. Harrison EI, Roth RH, Lobo JM, et al. Sleep time and efficiency in patients undergoing laboratory-based polysomnography. J Clin Sleep Med. 2021;17(8):1591-1598.