Sleep Stage Classification Research Articles

SLA-MLP: Enhancing Sleep Stage Analysis from EEG Signals Using Multilayer Perceptron Networks

Background/Objectives: Sleep stage analysis is considered to be the key factor for understanding and diagnosing various sleep disorders, as it provides insights into sleep quality and overall health. Methods: Traditional methods of sleep stage classification, such as manual scoring and basic machine learning approaches, often suffer from limitations including subjective biases, limited scalability, and inadequate accuracy. Existing deep learning models have improved the accuracy of sleep stage classification but still face challenges such as overfitting, computational inefficiencies, and difficulties in handling imbalanced datasets. To address these challenges, we propose the Sleep Stage Analysis with Multilayer Perceptron (SLA-MLP) model. Results: SLA-MLP leverages advanced deep learning techniques to enhance the classification of sleep stages from EEG signals. The key steps of this approach include data collection, where diverse and high-quality EEG data are gathered; preprocessing, which involves signal cropping, spectrogram conversion, and normalization to prepare the data for analysis; data balancing, where class weights are adjusted to address any imbalances in the dataset; feature extraction, utilizing Temporal Convolutional Networks (TCNs) to extract meaningful features from the EEG signals; and final classification, applying a Multilayer Perceptron (MLP) to accurately predict sleep stages. Conclusions: SLA-MLP demonstrates superior performance compared to traditional methods by effectively addressing the limitations of existing models. Its robust preprocessing techniques, advanced feature extraction, and adaptive data balancing strategies collectively contribute to obtaining more accurate results, having an accuracy of 97.23% for the S-DSI, 96.23 for the S-DSII and 97.23% for the S-DSIII dataset. This model offers a significant advancement in the field, providing a more precise tool for sleep research and clinical applications.

Exploration of an intrinsically explainable self-attention based model for prototype generation on single-channel EEG sleep stage classification

Prototype-based methods in deep learning offer interpretable explanations for decisions by comparing inputs to typical representatives in the data. This study explores the adaptation of SESM, a self-attention-based prototype method successful in electrocardiogram (ECG) tasks, for electroencephalogram (EEG) signals. The architecture is evaluated on sleep stage classification, exploring its efficacy in predicting stages with single-channel EEG. The model achieves comparable test accuracy compared to EEGNet, a state-of-the-art black-box architecture for EEG classification. The generated prototypical components are exaimed qualitatively and using the area over the perterbation curve (AOPC) indicate some alignment with expected bio-markers for different sleep stages such as alpha spindles and slow waves in non-REM sleep, but the results are severely limited by the model’s ability to only extract and present information in the time-domain. Ablation studies are used to explore the impact of kernel size, number of heads, and diversity threshold on model performance and explainability. This study represents the first application of a self-attention based prototype method to EEG data and provides a step forward in explainable AI for EEG data analysis.

FlexSleepTransformer: a transformer-based sleep staging model with flexible input channel configurations

Clinical sleep diagnosis traditionally relies on polysomnography (PSG) and expert manual classification of sleep stages. Recent advancements in deep learning have shown promise in automating sleep stage classification using a single PSG channel. However, variations in PSG acquisition devices and environments mean that the number of PSG channels can differ across sleep centers. To integrate a sleep staging method into clinical practice effectively, it must accommodate a flexible number of PSG channels. In this paper, we proposed FlexSleepTransformer, a transformer-based model designed to handle varying number of input channels, making it adaptable to diverse sleep staging datasets. We evaluated FlexSleepTransformer using two distinct datasets: the public SleepEDF-78 dataset and the local SleepUHS dataset. Notably, FlexSleepTransformer is the first model capable of simultaneously training on datasets with differing number of PSG channels. Our experiments showed that FlexSleepTransformer trained on both datasets together achieved 98% of the accuracy compared to models trained on each dataset individually. Furthermore, it outperformed models trained exclusively on one dataset when tested on the other dataset. Additionally, FlexSleepTransformer surpassed state-of-the-art CNN and RNN-based models on both datasets. Due to its adaptability with varying channels numbers, FlexSleepTransformer holds significant potential for clinical adoption, especially when trained with data from a wide range of sleep centers.

Enhanced Sleep Stage Classification Using EEG and EOG: A Novel Approach for Feature Selection with Deep Learning and Gaussian Noise Data Augmentation

Enhanced Sleep Stage Classification Using EEG and EOG: A Novel Approach for Feature Selection with Deep Learning and Gaussian Noise Data Augmentation

Automatic Classification of Sleep Stages from EEG Signals Using Riemannian Metrics and Transformer Networks

Automatic Classification of Sleep Stages from EEG Signals Using Riemannian Metrics and Transformer Networks

Unravelling sleep patterns: Supervised contrastive learning with self-attention for sleep stage classification

Sleep data scoring is a crucial and primary step for diagnosing sleep disorders to know the sleep stages from the PSG signals. This study uses supervised contrastive learning with a self-attention mechanism to classify sleep stages. We propose a deep learning framework for automatic sleep stage classification, which involves two training phases: (1) the feature representation learning phase, in which the feature representation network (encoder) learns to extract features from the electroencephalogram (EEG) signals, and (2) the classification network training phase, where a pre-trained encoder (trained during phase I) along with the classifier head is fine-tuned for the classification task. The PSG data shows a non-uniform distribution of sleep stages, with wake (W) (around 30% of total samples) and N2 stages (around 58% and 37% of total samples in Physionet EDF-Sleep 2013 and 2018 datasets, respectively) being more prevalent, leading to an imbalanced dataset. The imbalanced data issue is addressed using a weighted softmax cross-entropy loss function that assigns higher weights to minority sleep stages. Additionally, an oversampling technique (the synthetic minority oversampling technique (SMOTE) (Chawla et al., 2002)[1] ) is applied to generate synthetic samples for minority classes. The proposed model is evaluated on the Physionet EDF-Sleep 2013 and 2018 datasets using Fpz-Cz and Pz-Oz EEG channels. It achieved an overall accuracy of 94.1%, a macro F1 score of 92.64, and a Cohen’s Kappa coefficient of 0.92. Ablation studies demonstrated the importance of triplet loss-based pre-training and oversampling for enhancing performance. The proposed model requires minimal pre-processing, eliminating the need for extensive signal processing expertise, and thus is well-suited for clinicians diagnosing sleep disorders.

Test time adaptation for cross-domain sleep stage classification

Domain shift is a serious problem with existing cross-domain sleep stage classification methods. However, most existing domain adaptation methods require access to a large amount of source domain data containing patient information, which may lead to infringement of patient privacy. At the same time, existing methods require collecting all information in advance when processing target data, which cannot meet the situation where test data arrives in batches or in order in actual scenarios. To this end, we propose a test time adaptation framework based on knowledge distillation for cross-domain sleep stage classification. This teacher-student network framework uses pre-trained model to learn variant features and robust representations. We evaluate our framework in the cross domain scenario composed of three datasets and validate its superiority, real time performance and privacy protection capability.

Optimizing wearable single-channel EEG sleep staging in a heterogeneous sleep-disordered population using transfer learning.

While various wearable EEG devices have been developed, performance evaluation of the devices and their associated automated sleep stage classification models is mostly limited to healthy subjects. A major barrier for applying automated wearable EEG sleep staging in clinical populations is the need for large-scale data for model training. We therefore investigated transfer learning as strategy to overcome limited data availability and optimize automated single-channel EEG sleep staging in people with sleep disorders. We acquired 52 single-channel frontopolar headband EEG recordings from a heterogeneous sleep-disordered population with concurrent PSG. We compared three model training strategies: 'pre-training' (i.e., training on a larger dataset of 901 conventional PSGs), 'training-from-scratch' (i.e., training on wearable headband recordings), and 'fine-tuning' (i.e., training on conventional PSGs, followed by training on headband recordings). Performance was evaluated on all headband recordings using 10-fold cross-validation. Highest performance for 5-stage classification was achieved with fine-tuning (κ = .778), significantly higher than with pre-training (κ = .769) and with training-from-scratch (κ = .733). No significant differences or systematic biases were observed with clinically relevant sleep parameters derived from PSG. All sleep disorder categories showed comparable performance. This study emphasizes the importance of leveraging larger available datasets through deep transfer learning to optimize performance with limited data availability. Findings indicate strong similarity in data characteristics between conventional PSG and headband recordings. Altogether, results suggest the combination of the headband, classification model, and training methodology can be viable for sleep monitoring in the heterogeneous clinical population.

A Novel Sleep Stage Classification Method for Elderly People Using the Sleep EEG Signal and Senet Models

A Novel Sleep Stage Classification Method for Elderly People Using the Sleep EEG Signal and Senet Models

Evaluation of automated pediatric sleep stage classification using U-Sleep: a convolutional neural network.

U-Sleep is a publicly available automated sleep stager, but has not been independently validated using pediatric data. We aimed to a) test the hypothesis that U-Sleep performance is equivalent to trained humans, using a concordance dataset of 50 pediatric polysomnogram excerpts scored by multiple trained scorers, and b) identify clinical and demographic characteristics that impact U-Sleep accuracy, using a clinical dataset of 3114 polysomnograms from a tertiary center. Agreement between U-Sleep and 'gold' 30-second epoch sleep staging was determined across both datasets. Utilizing the concordance dataset, the hypothesis of equivalence between human scorers and U-Sleep was tested using a Wilcoxon two one-sided test (TOST). Multivariable regression and generalized additive modelling were used on the clinical dataset to estimate the effects of age, comorbidities and polysomnographic findings on U-Sleep performance. The median (interquartile range) Cohen's kappa agreement of U-Sleep and individual trained humans relative to "gold" scoring for 5-stage sleep staging in the concordance dataset were similar, kappa=0.79 (0.19) vs 0.78 (0.13) respectively, and satisfied statistical equivalence (TOST p < 0.01). Median (interquartile range) kappa agreement between U-Sleep 2.0 and clinical sleep-staging was kappa=0.69 (0.22). Modelling indicated lower performance for children < 2 years, those with medical comorbidities possibly altering sleep electroencephalography (kappa reduction=0.07-0.15) and those with decreased sleep efficiency or sleep-disordered breathing (kappa reduction=0.1). While U-Sleep algorithms showed statistically equivalent performance to trained scorers, accuracy was lower in children < 2 years and those with sleep-disordered breathing or comorbidities affecting electroencephalography. U-Sleep is suitable for pediatric clinical utilization provided automated staging is followed by expert clinician review.

Automatic sleep stage classification using deep learning: signals, data representation, and neural networks

In clinical practice, sleep stage classification (SSC) is a crucial step for physicians in sleep assessment and sleep disorder diagnosis. However, traditional sleep stage classification relies on manual work by sleep experts, which is time-consuming and labor-intensive. Faced with this obstacle, computer-aided diagnosis (CAD) has the potential to become an intelligent assistant tool for sleep experts, aiding doctors in the assessment and decision-making process. In fact, in recent years, CAD supported by artificial intelligence, especially deep learning (DL) techniques, has been widely applied in SSC. DL offers higher accuracy and lower costs, making a significant impact. In this paper, we will systematically review SSC research based on DL methods (DL-SSC). We explores DL-SSC from several important perspectives, including signal and data representation, data preprocessing, deep learning models, and performance evaluation. Specifically, this paper addresses three main questions: (1) What signals can DL-SSC use? (2) What are the various methods to represent these signals? (3) What are the effective DL models? Through addressing on these questions, this paper will provide a comprehensive overview of DL-SSC.

An enzyme-inspired specificity in deep learning model for sleep stage classification using multi-channel PSG signals input: Separating training approach and its performance on cross-dataset validation for generalizability

Numerous automatic sleep stage classification systems have been developed, but none have become effective assistive tools for sleep technicians due to issues with generalization. Four key factors hinder the generalization of these models are instruments, montage of recording, subject type, and scoring manual factors. This study aimed to develop a deep learning model that addresses generalization problems by integrating enzyme-inspired specificity and employing separating training approaches. Subject type and scoring manual factors were controlled, while the focus was on instruments and montage of recording factors. The proposed model consists of three sets of signal-specific models including EEG-, EOG-, and EMG-specific model. The EEG-specific models further include three sets of channel-specific models. All signal-specific and channel-specific models were established with data manipulation and weighted loss strategies, resulting in three sets of data manipulation models and class-specific models, respectively. These models were CNNs. Additionally, BiLSTM models were applied to EEG- and EOG-specific models to obtain temporal information. Finally, classification task for sleep stage was handled by ‘the-last-dense’ layer. The optimal sampling frequency for each physiological signal was identified and used during the training process. The proposed model was trained on MGH dataset and evaluated using both within dataset and cross-dataset. For MGH dataset, overall accuracy of 81.05 %, MF1 of 79.05 %, Kappa of 0.7408, and per-class F1-scores: W (84.98 %), N1 (58.06 %), N2 (84.82 %), N3 (79.20 %), and REM (88.17 %) can be achieved. Performances on cross-datasets are as follows: SHHS1 200 records reached 79.54 %, 70.56 %, and 0.7078; SHHS2 200 records achieved 76.77 %, 66.30 %, and 0.6632; Sleep-EDF 153 records gained 78.52 %, 72.13 %, and 0.7031; and BCI-MU (local dataset) 94 records achieved 83.57 %, 82.17 %, and 0.7769 for overall accuracy, MF1, and Kappa respectively. Additionally, the proposed model has approximately 9.3 M trainable parameters and takes around 26 s to process one PSG record. The results indicate that the proposed model demonstrates generalizability in sleep stage classification and shows potential as a feasibility tool for real-world applications. Additionally, enzyme-inspired specificity effectively addresses the challenges posed by varying montage of recording, while the identified optimal frequencies mitigate instrument-related issues.

A CNN-Transformer-ConvLSTM-CRF Hybrid Network for Sleep Stage Classification

A CNN-Transformer-ConvLSTM-CRF Hybrid Network for Sleep Stage Classification

MLS-Net: An Automatic Sleep Stage Classifier Utilizing Multimodal Physiological Signals in Mice.

Over the past decades, feature-based statistical machine learning and deep neural networks have been extensively utilized for automatic sleep stage classification (ASSC). Feature-based approaches offer clear insights into sleep characteristics and require low computational power but often fail to capture the spatial-temporal context of the data. In contrast, deep neural networks can process raw sleep signals directly and deliver superior performance. However, their overfitting, inconsistent accuracy, and computational cost were the primary drawbacks that limited their end-user acceptance. To address these challenges, we developed a novel neural network model, MLS-Net, which integrates the strengths of neural networks and feature extraction for automated sleep staging in mice. MLS-Net leverages temporal and spectral features from multimodal signals, such as EEG, EMG, and eye movements (EMs), as inputs and incorporates a bidirectional Long Short-Term Memory (bi-LSTM) to effectively capture the spatial-temporal nonlinear characteristics inherent in sleep signals. Our studies demonstrate that MLS-Net achieves an overall classification accuracy of 90.4% and REM state precision of 91.1%, sensitivity of 84.7%, and an F1-Score of 87.5% in mice, outperforming other neural network and feature-based algorithms in our multimodal dataset.

Automated remote sleep monitoring needs uncertainty quantification.

Wearable electroencephalography devices emerge as a cost-effective and ergonomic alternative to gold-standard polysomnography, paving the way for better health monitoring and sleep disorder screening. Machine learning allows to automate sleep stage classification, but trust and reliability issues have hampered its adoption in clinical applications. Estimating uncertainty is a crucial factor in enhancing reliability by identifying regions of heightened and diminished confidence. In this study, we used an uncertainty-centred machine learning pipeline, U-PASS, to automate sleep staging in a challenging real-world dataset of single-channel electroencephalography and accelerometry collected with a wearable device from an elderly population. We were able to effectively limit the uncertainty of our machine learning model and to reliably inform clinical experts of which predictions were uncertain to improve the machine learning model's reliability. This increased the five-stage sleep-scoring accuracy of a state-of-the-art machine learning model from 63.9% to 71.2% on our dataset. Remarkably, the machine learning approach outperformed the human expert in interpreting these wearable data. Manual review by sleep specialists, without specific training for sleep staging on wearable electroencephalography, proved ineffective. The clinical utility of this automated remote monitoring system was also demonstrated, establishing a strong correlation between the predicted sleep parameters and the reference polysomnography parameters, and reproducing known correlations with the apnea-hypopnea index. In essence, this work presents a promising avenue to revolutionize remote patient care through the power of machine learning by the use of an automated data-processing pipeline enhanced with uncertainty estimation.

A foundational transformer leveraging full night, multichannel sleep study data accurately classifies sleep stages.

To investigate whether a foundational transformer model using 8-hour, multichannel data from polysomnograms can outperform existing artificial intelligence (AI) methods for sleep stage classification. We utilized the Sleep Heart Health Study (SHHS) visits 1 and 2 for training and validation and the Multi-Ethnic Study of Atherosclerosis (MESA) for testing of our model. We trained a self-supervised foundational transformer (called PFTSleep) that encodes 8-hour long sleep studies at 125 Hz with 7 signals including brain, movement, cardiac, oxygen, and respiratory channels. These encodings are used as input for training of an additional model to classify sleep stages, without adjusting the weights of the foundational transformer. We compared our results to existing AI methods that did not utilize 8-hour data or the full set of signals but did report evaluation metrics for the SHHS dataset. We trained and validated a model with 8,444 sleep studies with 7 signals including brain, movement, cardiac, oxygen, and respiratory channels and tested on an additional 2,055 studies. In total, we trained and tested 587,944 hours of sleep study signal data. Area under the precision recall curve (AUPRC) scores were 0.82, 0.40, 0.53, 0.75, and 0.82 and area under the receiving operating characteristics curve (AUROC) scores were 0.99, 0.95, 0.96, 0.98, and 0.99 for wake, N1, N2, N3, and REM, respectively, on the SHHS validation set. For MESA, the AUPRC scores were 0.56, 0.16, 0.40, 0.45, and 0.65 and AUROC scores were 0.94, 0.77, 0.87, 0.91, and 0.96, respectively. Our model was compared to the longest context window state-of-the-art model and showed increases in macro evaluation scores, notably sensitivity (3.7% increase) and multi-class REM (3.39% increase) and wake (0.97% increase) F1 scores. Utilizing full night, multi-channel PSG data encodings derived from a foundational transformer improve sleep stage classification over existing methods.

GRU-powered sleep stage classification with permutation-based EEG channel selection

We present a new approach to classifying the sleep stage that incorporates a computationally inexpensive method based on permutations for channel selection and takes advantage of deep learning power, specifically the gated recurrent unit (GRU) model, along with other deep learning methods. By systematically permuting the electroencephalographic (EEG) channels, different combinations of EEG channels are evaluated to identify the most informative subset for the classification of the 5-class sleep stage. For analysis, we used an EEG dataset that was collected at the International Institute for Integrative Sleep Medicine (WPI-IIIS) at the University of Tsukuba in Japan. The results of these explorations provide many new insights such as the (1) drastic decrease in performance when channels are fewer than 3, (2) 3-random channels selected by permutation provide the same or better prediction than the 3 channels recommended by the American Academy of Sleep Medicine (AASM), (3) N1 class suffers the most in prediction accuracy as the channels drop from 128 to 3 random or 3 AASM, and (4) no single channel provides acceptable levels of accuracy in the prediction of 5 classes. The results obtained show the GRU’s ability to retain essential temporal information from EEG data, which allows capturing the underlying patterns associated with each sleep stage effectively. Using permutation-based channel selection, we enhance or at least maintain as high model efficiency as when using high-density EEG, incorporating only the most informative EEG channels.

A Deep Transfer Learning Approach for Sleep Stage Classification and Sleep Apnea Detection Using Wrist-Worn Consumer Sleep Technologies.

Obstructive sleep apnea (OSA) is a common, underdiagnosed sleep-related breathing disorder with serious health implications Objective - We propose a deep transfer learning approach for sleep stage classification and sleep apnea (SA) detection using wrist-worn consumer sleep technologies (CST). Methods - Our model is based on a deep convolutional neural network (DNN) utilizing accelerometers and photo-plethysmography signals from nocturnal recordings. The DNN was trained and tested on internal datasets that include raw data from clinical and wrist-worn devices; external validation was performed on a hold-out test dataset containing raw data from a wrist-worn CST. Results - Training on clinical data improves performance significantly, and feature enrichment through a sleep stage stream gives only minor improvements. Raw data input outperforms feature-based input in CST datasets. The system generalizes well but performs slightly worse on wearable device data compared to clinical data. However, it excels in detecting events during REM sleep and is associated with arousal and oxygen desaturation. We found; cases that were significantly underestimated were characterized by fewer of such event associations. Conclusion - This study showcases the potential of using CSTs as alternate screening solution for undiagnosed cases of OSA. Significance - This work is significant for its development of a deep transfer learning approach using wrist-worn consumer sleep technologies, offering comprehensive validation for data utilization, and learning techniques, ultimately improving sleep apnea detection across diverse devices.

Deep Neural Networks for Automatic Sleep Stage Classification and Consciousness Assessment in Patients With Disorder of Consciousness

Deep Neural Networks for Automatic Sleep Stage Classification and Consciousness Assessment in Patients With Disorder of Consciousness

Study on the classification of sleep stages in EEG signals based on DoubleLinkSleepCLNet.

The classification of sleep stages based on Electroencephalogram (EEG) changes has significant implications for evaluating sleep quality and sleep status. Most polysomnography (PSG) systems have a limited number of channels and do not achieve optimal classification performance due to a paucity of raw data. To leverage the data characteristics and enhance the classification accuracy, we propose and evaluate a novel dual-link deep neural network model, 'DoubleLinkSleepCLNet'. The DoubleLinkSleepCLNet model performs feature extraction and efficient classification on both the raw EEG and the EEG processed with the Hilbert transform. It leverages the frequency domain and time domain feature modules, resulting in superior performance compared to other models. The DoubleLinkSleepCLNet model, using the 2 Raw/2 Hilbert data modes, achieved the highest classification performance with an accuracy of 88.47%. The average accuracy of the EEG was improved by approximately 4.08% after the application of the Hilbert transform. Additionally, Convolutional Neural Network (CNN) demonstrated superior performance in processing phase information, whereas Long Short-Term Memory (LSTM) excelled in handling time series data. The application of the Hilbert transform to EEG data, followed by processing it with a convolutional neural network, enhances the accuracy of the model. These findings introduce novel concepts for accelerating sleep stage prediction research, suggesting potential applications of these methods to other EEG analyses.