MetaSleepLearner: A Pilot Study on Fast Adaptation of Bio-Signals-Based Sleep Stage Classifier to New Individual Subject Using Meta-Learning

Motion Cognitive Decoding of Cross-Subject Motor Imagery Guided on Different Visual Stimulus Materials

BACKGROUND

Motor imagery (MI) plays an important role in brain-computer interfaces, especially in evoking event-related desynchronization and synchronization (ERD/S) rhythms in electroencephalogram (EEG) signals. However, the procedure for performing a MI task for a single subject is subjective, making it difficult to determine the actual situation of an individual's MI task and resulting in significant individual EEG response variations during motion cognitive decoding.

METHODS

To explore this issue, we designed three visual stimuli (arrow, human, and robot), each of which was used to present three MI tasks (left arm, right arm, and feet), and evaluated differences in brain response in terms of ERD/S rhythms. To compare subject-specific variations of different visual stimuli, a novel cross-subject MI-EEG classification method was proposed for the three visual stimuli. The proposed method employed a covariance matrix centroid alignment for preprocessing of EEG samples, followed by a model agnostic meta-learning method for cross-subject MI-EEG classification.

RESULTS AND CONCLUSION

The experimental results showed that robot stimulus materials were better than arrow or human stimulus materials, with an optimal cross-subject motion cognitive decoding accuracy of 79.04%. Moreover, the proposed method produced robust classification of cross-subject MI-EEG signal decoding, showing superior results to conventional methods on collected EEG signals.

A Feature Fusion Model Based on Temporal Convolutional Network for Automatic Sleep Staging Using Single-Channel EEG

Sleep staging is a crucial task in sleep monitoring and diagnosis, but clinical sleep staging is both time-consuming and subjective. In this study, we proposed a novel deep learning algorithm named feature fusion temporal convolutional network (FFTCN) for automatic sleep staging using single-channel EEG data. This algorithm employed a one-dimensional convolutional neural network (1D-CNN) to extract temporal features from raw EEG, and a two-dimensional CNN (2D-CNN) to extract time-frequency features from spectrograms generated through continuous wavelet transform (CWT) at the epoch level. These features were subsequently fused and further fed into a temporal convolutional network (TCN) to classify sleep stages at the sequence level. Moreover, a two-step training strategy was used to enhance the model's performance on an imbalanced dataset. Our proposed method exhibits superior performance in the 5-class classification task for healthy subjects, as evaluated on the SHHS-1, Sleep-EDF-153, and ISRUC-S1 datasets. This work provided a straightforward and promising method for improving the accuracy of automatic sleep staging using only single-channel EEG, and the proposed method exhibited great potential for future applications in professional sleep monitoring, which could effectively alleviate the workload of sleep technicians.

Motor imagery decoding using source optimized transfer learning based on multi-loss fusion CNN

Transfer learning is increasingly used to decode multi-class motor imagery tasks. Previous transfer learning ignored the optimizability of the source model, weakened the adaptability to the target domain and limited the performance. This paper first proposes the multi-loss fusion convolutional neural network (MF-CNN) to make an optimizable source model. Then we propose a novel source optimized transfer learning (SOTL), which optimizes the source model to make it more in line with the target domain's features to improve the target model's performance. We transfer the model trained from 16 healthy subjects to 16 stroke patients. The average classification accuracy achieves 51.2 ± 0.17% in the four types of unilateral upper limb motor imagery tasks, which is significantly higher than the classification accuracy of deep learning (p < 0.001) and transfer learning (p < 0.05). In this paper, an MI model from the data of healthy subjects can be used for the classification of stroke patients and can demonstrate good classification results, which provides experiential support for the study of transfer learning and the modeling of stroke rehabilitation training.

Personalization of Automatic Sleep Scoring: How Best to Adapt Models to Personal Domains in Wearable EEG

Wearable EEG enables us to capture large amounts of high-quality sleep data for diagnostic purposes. To make full use of this capacity we need high-performance automatic sleep scoring models. To this end, it has been noted that domain mismatch between recording equipment can be considerable, e.g. PSG to wearable EEG, but a previously observed benefit from personalizing models to individual subjects further indicates a personal domain in sleep EEG. In this work, we have investigated the extent of such a personal domain in wearable EEG, and review supervised and unsupervised approaches to personalization as found in the literature. We investigated the personalization effect of the unsupervised Adversarial Domain Adaptation and implemented an unsupervised method based on statistics alignment. No beneficial personalization effect was observed using these unsupervised methods. We find that supervised personalization leads to a substantial performance improvement on the target subject ranging from 15% Cohen's Kappa for subjects with poor performance ( ) to roughly 2% on subjects with high performance ( ). This improvement was present for models trained on both small and large data sets, indicating that even high-performance models benefit from supervised personalization. We found that this personalization can be beneficially regularized using Kullback-Leibler regularization, leading to lower variance with negligible cost to improvement. Based on the experiments, we recommend model personalization using Kullback-Leibler regularization.

PSEENet: A Pseudo-Siamese Neural Network Incorporating Electroencephalography and Electrooculography Characteristics for Heterogeneous Sleep Staging

Sleep staging plays a critical role in evaluating the quality of sleep. Currently, most studies are either suffering from dramatic performance drops when coping with varying input modalities or unable to handle heterogeneous signals. To handle heterogeneous signals and guarantee favorable sleep staging performance when a single modality is available, a pseudo-siamese neural network (PSN) to incorporate electroencephalography (EEG), electrooculography (EOG) characteristics is proposed (PSEENet). PSEENet consists of two parts, spatial mapping modules (SMMs) and a weight-shared classifier. SMMs are used to extract high-dimensional features. Meanwhile, joint linkages among multi-modalities are provided by quantifying the similarity of features. Finally, with the cooperation of heterogeneous characteristics, associations within various sleep stages can be established by the classifier. The evaluation of the model is validated on two public datasets, namely, Montreal Archive of Sleep Studies (MASS) and SleepEDFX, and one clinical dataset from Huashan Hospital of Fudan University (HSFU). Experimental results show that the model can handle heterogeneous signals, provide superior results under multimodal signals and show good performance with single modality. PSEENet obtains accuracy of 79.1%, 82.1% with EEG, EEG and EOG on Sleep-EDFX, and significantly improves the accuracy with EOG from 73.7% to 76% by introducing similarity information.

Toward Interpretable Sleep Stage Classification Using Cross-Modal Transformers

Accurate sleep stage classification is significant for sleep health assessment. In recent years, several machine-learning based sleep staging algorithms have been developed, and in particular, deep-learning based algorithms have achieved performance on par with human annotation. Despite improved performance, a limitation of most deep-learning based algorithms is their black-box behavior, which have limited their use in clinical settings. Here, we propose a cross-modal transformer, which is a transformer-based method for sleep stage classification. The proposed cross-modal transformer consists of a cross-modal transformer encoder architecture along with a multi-scale one-dimensional convolutional neural network for automatic representation learning. The performance of our method is on-par with the state-of-the-art methods and eliminates the black-box behavior of deep-learning models by utilizing the interpretability aspect of the attention modules. Furthermore, our method provides considerable reductions in the number of parameters and training time compared to the state-of-the-art methods. Our code is available at https://github.com/Jathurshan0330/Cross-Modal-Transformer. A demo of our work can be found at https://bit.ly/Cross_modal_transformer_demo.

MVF-SleepNet: Multi-View Fusion Network for Sleep Stage Classification

Sleep stage classification is of great importance in human health monitoring and disease diagnosing. Clinically, visual-inspected classifying sleep into different stages is quite time consuming and highly relies on the expertise of sleep specialists. Many automated models for sleep stage classification have been proposed in previous studies but their performances still exist a gap to the real clinical application. In this work, we propose a novel multi-view fusion network named MVF-SleepNet based on multi-modal physiological signals of electroencephalography (EEG), electrocardiography (ECG), electrooculography (EOG), and electromyography (EMG). To capture the relationship representation among multi-modal physiological signals, we construct two views of Time-frequency images (TF images) and Graph-learned graphs (GL graphs). To learn the spectral-temporal representation from sequentially timed TF images, the combination of VGG-16 and GRU networks is utilized. To learn the spatial-temporal representation from sequentially timed GL graphs, the combination of Chebyshev graph convolution and temporal convolution networks is employed. Fusing the spectral-temporal representation and spatial-temporal representation can further boost the performance of sleep stage classification. A large number of experiment results on the publicly available datasets of ISRUC-S1 and ISRUC-S3 show that the MVF-SleepNet achieves overall accuracy of 0.821, F1 score of 0.802 and Kappa of 0.768 on ISRUC-S1 dataset, and accuracy of 0.841, F1 score of 0.828 and Kappa of 0.795 on ISRUC-S3 dataset. The MVF-SleepNet achieves competitive results on both datasets of ISRUC-S1 and ISRUC-S3 for sleep stage classification compared to the state-of-the-art baselines. The source code of MVF-SleepNet is available on Github (https://github.com/YJPai65/MVF-SleepNet).

Transfer Learning for Automatic Sleep Staging Using a Pre-Gelled Electrode Grid

Novel sensor solutions for sleep monitoring at home could alleviate bottlenecks in sleep medical care as well as enable selective or continuous observation over long periods of time and contribute to new insights in sleep medicine and beyond. Since especially in the latter case the sensor data differ strongly in signal, number and extent of sensors from the classical polysomnography (PSG) sensor technology, an automatic evaluation is essential for the application. However, the training of an automatic algorithm is complicated by the fact that the development phase of the new sensor technology, extensive comparative measurements with standardized reference systems, is often not possible and therefore only small datasets are available. In order to circumvent high system-specific training data requirements, we employ pre-training on large datasets with finetuning on small datasets of new sensor technology to enable automatic sleep phase detection for small test series. By pre-training on publicly available PSG datasets and finetuning on 12 nights recorded with new sensor technology based on a pre-gelled electrode grid to capture electroencephalography (EEG), electrooculography (EOG) and electromyography (EMG), an F1 score across all sleep phases of 0.81 is achieved (wake 0.84, N1 0.62, N2 0.81, N3 0.87, REM 0.88), using only EEG and EOG. The analysis additionally considers the spatial distribution of the channels and an approach to approximate classical electrode positions based on specific linear combinations of the new sensor grid channels.

Dual-Teacher Feature Distillation: A Transfer Learning Method for Insomniac PSG Staging

Insomnia is the most common sleep disorder linked with adverse long-term medical and psychiatric outcomes. Automatic sleep staging plays a crucial role in aiding doctors to diagnose insomnia disorder. Only a few studies have been conducted to develop automatic sleep staging methods for insomniacs, and most of them have utilized transfer learning methods, which involve pre-training models on healthy individuals and then fine-tuning them on insomniacs. Unfortunately, significant differences in feature distribution between the two subject groups impede the transfer performance, highlighting the need to effectively integrate the features of healthy subjects and insomniacs. In this paper, we propose a dual-teacher cross-domain knowledge transfer method based on the feature-based knowledge distillation to improve the performance of sleep staging for insomniacs. Specifically, the insomnia teacher directly learns from insomniacs and feeds the corresponding domain-specific features into the student network, while the health domain teacher guide the student network to learn domain-generic features. During the training process, we adopt the OFD (Overhaul of Feature Distillation) method to build the health domain teacher. We conducted the experiments to validate the proposed method, using the Sleep-EDF database as the source domain and the CAP-Database as the target domain. The results demonstrate that our method surpasses advanced techniques, achieving an average sleep staging accuracy of 80.56% on the CAP-Database. Furthermore, our method exhibits promising performance on the private dataset.

Automatic classification of sleep stages using EEG signals and convolutional neural networks

Sleep stages classification is one of the new topics in studying human life quality because it plays a crucial role in getting a healthy lifestyle. Abnormal changes or absence of normal sleep may lead to different diseases such as heart-related diseases, diabetes, and obesity. In general, sleep staging analysis can be performed using electroencephalography (EEG) signals. This study proposes a convolutional neural network (CNN) based methodology for sleep stage classification using EEG signals taken by six channels and transformed into time-frequency analysis images. The proposed methodology consists of three major steps: (i) segment the EEG signal into epochs with 30 seconds in length, (ii) convert epochs into 2D representation using time-frequency analysis, and (iii) feed the 2D time-frequency analysis to the 2D CNN. The results showed that the proposed methodology is robust and achieved a very high accuracy of 99.39% for channel C4-A1. All other channels have accuracy values above 98.5%, which indicates that any channel can be used for sleep stage classification with high accuracy. The proposed methodology outperformed the methods in the literature in terms of overall accuracy or single channel accuracy. It is expected to provide a great benefit for physicians, especially neurologists; by providing them with a new powerful tool to support the clinical diagnosis of sleep-related diseases.

Few-shot EEG sleep staging based on transductive prototype optimization network

Electroencephalography (EEG) is a commonly used technology for monitoring brain activities and diagnosing sleep disorders. Clinically, doctors need to manually stage sleep based on EEG signals, which is a time-consuming and laborious task. In this study, we propose a few-shot EEG sleep staging termed transductive prototype optimization network (TPON) method, which aims to improve the performance of EEG sleep staging. Compared with traditional deep learning methods, TPON uses a meta-learning algorithm, which generalizes the classifier to new classes that are not visible in the training set, and only have a few examples for each new class. We learn the prototypes of existing objects through meta-training, and capture the sleep features of new objects through the "learn to learn" method of meta-learning. The prototype distribution of the class is optimized and captured by using support set and unlabeled high confidence samples to increase the authenticity of the prototype. Compared with traditional prototype networks, TPON can effectively solve too few samples in few-shot learning and improve the matching degree of prototypes in prototype network. The experimental results on the public SleepEDF-2013 dataset show that the proposed algorithm outperform than most advanced algorithms in the overall performance. In addition, we experimentally demonstrate the feasibility of cross-channel recognition, which indicates that there are many similar sleep EEG features between different channels. In future research, we can further explore the common features among different channels and investigate the combination of universal features in sleep EEG. Overall, our method achieves high accuracy in sleep stage classification, demonstrating the effectiveness of this approach and its potential applications in other medical fields.

Decentralized Data-Privacy Preserving Deep-Learning Approaches for Enhancing Inter-Database Generalization in Automatic Sleep Staging

Automatic sleep staging has been an active field of development. Despite multiple efforts, the area remains a focus of research interest. Indeed, while promising results have reported in past literature, uptake of automatic sleep scoring in the clinical setting remains low. One of the current issues regards the difficulty to generalization performance results beyond the local testing scenario, i.e. across data from different clinics. Issues derived from data-privacy restrictions, that generally apply in the medical domain, pose additional difficulties in the successful development of these methods. We propose the use of several decentralized deep-learning approaches, namely ensemble models and federated learning, for robust inter-database performance generalization and data-privacy preservation in automatic sleep staging scenario. Specifically, we explore four ensemble combination strategies (max-voting, output averaging, size-proportional weighting, and Nelder-Mead) and present a new federated learning algorithm, so-called sub-sampled federated stochastic gradient descent (ssFedSGD). To evaluate generalization capabilities of such approaches, experimental procedures are carried out using a leaving-one-database-out direct-transfer scenario on six independent and heterogeneous public sleep staging databases. The resulting performance is compared with respect to two baseline approaches involving single-database and centralized multiple-database derived models. Our results show that proposed decentralized learning methods outperform baseline local approaches, and provide similar generalization results to centralized database-combined approaches. We conclude that these methods are more preferable choices, as they come with additional advantages concerning improved scalability, flexible design, and data-privacy preservation.

An effective hybrid feature selection using entropy weight method for automatic sleep staging

Objective. Sleep staging is the basis for sleep quality assessment and diagnosis of sleep-related disorders. In response to the inadequacy of traditional manual judgement of sleep stages, using machine learning techniques for automatic sleep staging has become a hot topic. To improve the performance of sleep staging, numerous studies have extracted a large number of sleep-related characteristics. However, there are redundant and irrelevant features in the high-dimensional features that reduce the classification accuracy. To address this issue, an effective hybrid feature selection method based on the entropy weight method is proposed in this paper for automatic sleep staging.Approach. Firstly, we preprocess the four modal polysomnography (PSG) signals, including electroencephalogram (EEG), electrooculogram (EOG), electrocardiogram (ECG) and electromyogram (EMG). Secondly, the time domain, frequency domain and nonlinear features are extracted from the preprocessed signals, with a total of 185 features. Then, in order to acquire characteristics of the multi-modal signals that are highly correlated with the sleep stages, the proposed hybrid feature selection method is applied to choose effective features. This method is divided into two stages. In stage I, the entropy weight method is employed to combine two filter methods to build a subset of features. This stage evaluates features based on information theory and distance metrics, which can quickly obtain a subset of features and retain the relevant features. In stage II, Sequential Forward Selection is used to evaluate the subset of features and eliminate redundant features. Further more, to achieve better performance of classification, an ensemble model based on support vector machine, K-nearest neighbor, random forest and multilayer perceptron is finally constructed for classifying sleep stages.Main results. The experiment using the Cyclic Alternating Pattern (CAP) sleep database is performed to assess the performance of the method proposed in this paper. The proposed hybrid feature selection method chooses only 30 features highly correlated to sleep stages. The accuracy, F1 score and Kappa coefficient of 6 class sleep staging reach 88.86%, 83.15% and 0.8531%, respectively.Significance. Experimental results show the effectiveness of the proposed method compared to the existing state-of-the-art studies. It greatly reduces the number of features required while achieving outstanding auto-sleep staging results.

Challenges of Applying Automated Polysomnography Scoring at Scale

Automatic polysomnography analysis can be leveraged to shorten scoring times, reduce associated costs, and ultimately improve the overall diagnosis of sleep disorders. Multiple and diverse strategies have been attempted for implementation of this technology at scale in the routine workflow of sleep centers. The field, however, is complex and presents unsolved challenges in a number of areas. Recent developments in computer science and artificial intelligence are nevertheless closing the gap. Technological advances are also opening new pathways for expanding our current understanding of the domain and its analysis.

Diagnosis of Obstructive Sleep Apnea Using Feature Selection, Classification Methods, and Data Grouping Based Age, Sex, and Race

Obstructive sleep apnea (OSA) is a prevalent sleep disorder that affects approximately 3-7% of males and 2-5% of females. In the United States alone, 50-70 million adults suffer from various sleep disorders. OSA is characterized by recurrent episodes of breathing cessation during sleep, thereby leading to adverse effects such as daytime sleepiness, cognitive impairment, and reduced concentration. It also contributes to an increased risk of cardiovascular conditions and adversely impacts patient overall quality of life. As a result, numerous researchers have focused on developing automated detection models to identify OSA and address these limitations effectively and accurately. This study explored the potential benefits of utilizing machine learning methods based on demographic information for diagnosing the OSA syndrome. We gathered a comprehensive dataset from the Torr Sleep Center in Corpus Christi, Texas, USA. The dataset comprises 31 features, including demographic characteristics such as race, age, sex, BMI, Epworth score, M. Friedman tongue position, snoring, and more. We devised a novel process encompassing pre-processing, data grouping, feature selection, and machine learning classification methods to achieve the research objectives. The classification methods employed in this study encompass decision tree (DT), naive Bayes (NB), k-nearest neighbor (kNN), support vector machine (SVM), linear discriminant analysis (LDA), logistic regression (LR), and subspace discriminant (Ensemble) classifiers. Through rigorous experimentation, the results indicated the superior performance of the optimized kNN and SVM classifiers for accurately classifying sleep apnea. Moreover, significant enhancements in model accuracy were observed when utilizing the selected demographic variables and employing data grouping techniques. For instance, the accuracy percentage demonstrated an approximate improvement of 4.5%, 5%, and 10% with the feature selection approach when applied to the grouped data of Caucasians, females, and individuals aged 50 or below, respectively. Furthermore, a comparison with prior studies confirmed that effective data grouping and proper feature selection yielded superior performance in OSA detection when combined with an appropriate classification method. Overall, the findings of this research highlight the importance of leveraging demographic information, employing proper feature selection techniques, and utilizing optimized classification models for accurate and efficient OSA diagnosis.

Single-Channel Ecg-Based Sleep Stage Classification With End-To-End Trainable Deep Neural Networks

The demand for automatic sleep stage classification using easily obtainable signals like electrocardiograms (ECGs) is rising due to the growing number of sleep disorder cases. Our study examined the potential of using single-channel ECG signals for user-friendly automatic sleep stage classification. Unlike previous studies that relied on manual features such as heart rate and variability, we propose using fully neural network-based features. The proposed model uses a ContextNet-based feature encoder applied to the ECG spectrogram, and a Transformer model to capture the temporal properties of sleep cycles over the course of the night.

Design of a Sleep Modulation System with FPGA-Accelerated Deep Learning for Closed-loop Stage-Specific In-Phase Auditory Stimulation

Closed-loop sleep modulation is an emerging research paradigm to treat sleep disorders and enhance sleep benefits. However, two major barriers hinder the widespread application of this research paradigm. First, subjects often need to be wire-connected to rack-mount instrumentation for data acquisition, which negatively affects sleep quality. Second, conventional real-time sleep stage classification algorithms give limited performance. In this work, we conquer these two limitations by developing a sleep modulation system that supports closed-loop operations on the device. Sleep stage classification is performed using a lightweight deep learning (DL) model accelerated by a low-power field-programmable gate array (FPGA) device. The DL model uses a single channel electroencephalogram (EEG) as input. Two convolutional neural networks (CNNs) are used to capture general and detailed features, and a bidirectional long-short-term memory (LSTM) network is used to capture time-variant sequence features. An 8-bit quantization is used to reduce the computational cost without compromising performance. The DL model has been validated using a public sleep database containing 81 subjects, achieving a state-of-the-art classification accuracy of 85.8% and a F1-score of 79%. The developed model has also shown the potential to be generalized to different channels and input data lengths. Closed-loop in-phase auditory stimulation has been demonstrated on the test bench.

A novel semi-supervised meta learning method for subject-transfer brain-computer interface

The brain-computer interface (BCI) provides a direct communication pathway between the human brain and external devices. However, the models trained for existing subjects perform poorly on new subjects, which is termed the subject calibration problem. In this paper, we propose a semi-supervised meta learning (SSML) method for subject-transfer calibration. The proposed SSML learns a model-agnostic meta learner with existing subjects and then fine-tunes the meta learner in a semi-supervised learning manner, i.e. using a few labelled samples and many unlabelled samples of the target subject for calibration. It is significant for BCI applications in which labelled data are scarce or expensive while unlabelled data are readily available. Three different BCI paradigms are tested: event-related potential detection, emotion recognition and sleep staging. The SSML achieved classification accuracies of 0.95, 0.89 and 0.83 in the benchmark datasets of three paradigms. The runtime complexity of SSML grows linearly as the number of samples of target subject increases so that is possible to apply it in real-time systems. This study is the first attempt to apply semi-supervised model-agnostic meta learning methodology for subject calibration. The experimental results demonstrated the effectiveness and potential of the SSML method for subject-transfer BCI applications.

A Sparse Multiclass Motor Imagery EEG Classification Using 1D-ConvResNet

Multiclass motor imagery classification is essential for brain–computer interface systems such as prosthetic arms. The compressive sensing of EEG helps classify brain signals in real-time, which is necessary for a BCI system. However, compressive sensing is limited, despite its flexibility and data efficiency, because of its sparsity and high computational cost in reconstructing signals. Although the constraint of sparsity in compressive sensing has been addressed through neural networks, its signal reconstruction remains slow, and the computational cost increases to classify the signals further. Therefore, we propose a 1D-Convolutional Residual Network that classifies EEG features in the compressed (sparse) domain without reconstructing the signal. First, we extract only wavelet features (energy and entropy) from raw EEG epochs to construct a dictionary. Next, we classify the given test EEG data based on the sparse representation of the dictionary. The proposed method is computationally inexpensive, fast, and has high classification accuracy as it uses a single feature to classify without preprocessing. The proposed method is trained, validated, and tested using multiclass motor imagery data of 109 subjects from the PhysioNet database. The results demonstrate that the proposed method outperforms state-of-the-art classifiers with 96.6% accuracy.

A novel automated robust dual-channel EEG-based sleep scoring system using optimal half-band pair linear-phase biorthogonal wavelet filter bank

Nowadays, the hectic work life of people has led to sleep deprivation. This may further result in sleep-related disorders and adverse physiological conditions. Therefore, sleep study has become an active research area. Sleep scoring is crucial for detecting sleep-related disorders like sleep apnea, insomnia, narcolepsy, periodic leg movement (PLM), and restless leg syndrome (RLS). Sleep is conventionally monitored in a sleep laboratory using polysomnography (PSG) which is the recording of various physiological signals. The traditional sleep stage scoring (SSG) done by professional sleep scorers is a tedious, strenuous, and time-consuming process as it is manual. Hence, developing a machine-learning model for automatic SSG is essential. In this study, we propose an automated SSG approach based on the biorthogonal wavelet filter bank's (BWFB) novel least squares (LS) design. We have utilized a huge Wisconsin sleep cohort (WSC) database in this study. The proposed study is a pioneering work on automatic sleep stage classification using the WSC database, which includes good sleepers and patients suffering from various sleep-related disorders, including apnea, insomnia, hypertension, diabetes, and asthma. To investigate the generalization of the proposed system, we evaluated the proposed model with the following publicly available databases: cyclic alternating pattern (CAP), sleep EDF, ISRUC, MIT-BIH, and the sleep apnea database from St. Vincent's University. This study uses only two unipolar EEG channels, namely O1-M2 and C3-M2, for the scoring. The Hjorth parameters (HP) are extracted from the wavelet subbands (SBS) that are obtained from the optimal BWFB. To classify sleep stages, the HP features are fed to several supervised machine learning classifiers. 12 different datasets have been created to develop a robust model. A total of 12 classification tasks (CT) have been conducted employing various classification algorithms. Our developed model achieved the best accuracy of 83.2% and Cohen's Kappa of 0.7345 to reliably distinguish five sleep stages, using an ensemble bagged tree classifier with 10-fold cross-validation using WSC data. We also observed that our system is either better or competitive with existing state-of-art systems when we tested with the above-mentioned five databases other than WSC. This method yielded promising results using only two EEG channels using a huge WSC database. Our approach is simple and hence, the developed model can be installed in home-based clinical systems and wearable devices for sleep scoring.

A Temporal-Spectral Fused and Attention-Based Deep Model for Automatic Sleep Staging

Sleep staging is a vital process for evaluating sleep quality and diagnosing sleep-related diseases. Most of the existing automatic sleep staging methods focus on time-domain information and often ignore the transformation relationship between sleep stages. To deal with the above problems, we propose a Temporal-Spectral fused and Attention-based deep neural Network model (TSA-Net) for automatic sleep staging, using a single-channel electroencephalogram (EEG) signal. The TSA-Net is composed of a two-stream feature extractor, feature context learning, and conditional random field (CRF). Specifically, the two-stream feature extractor module can automatically extract and fuse EEG features from time and frequency domains, considering that both temporal and spectral features can provide abundant distinguishing information for sleep staging. Subsequently, the feature context learning module learns the dependencies between features using the multi-head self-attention mechanism and outputs a preliminary sleep stage. Finally, the CRF module further applies transition rules to improve classification performance. We evaluate our model on two public datasets, Sleep-EDF-20 and Sleep-EDF-78. In terms of accuracy, the TSA-Net achieves 86.64% and 82.21% on the Fpz-Cz channel, respectively. The experimental results illustrate that our TSA-Net can optimize the performance of sleep staging and achieve better staging performance than state-of-the-art methods.

RAPIDEST: A Framework for Obstructive Sleep Apnea Detection

The traditional polysomnography (PSG) examination for Obstructive Sleep Apnea (OSA) diagnosis needs to measure several signals, such as EEG, ECG, EMG, EOG and the oxygen level in blood, of a patient who may have to wear many sensors during sleep. After the PSG examination, the Apnea-Hypopnea Index (AHI) is calculated based on the measured data to evaluate the severity of apnea and hypopnea for the patient. This process is obviously complicated and inconvenient. In this paper, we propose an AI-based framework, called RAre Pattern Identification and DEtection for Sleep-stage Transitions (RAPIDEST), to detect OSA based on the sequence of sleep stages from which a novel rarity score is defined to capture the unusualness of the sequence of sleep stages. More importantly, under this framework, we only need EEG signals, thus significantly simplifying the signal collection process and reducing the complexity of the severity determination of apnea and hypopnea. We have conducted extensive experiments to verify the relationship between the rarity score and AHI and demonstrate the effectiveness of our proposed approach.

A Novel Approach for Sleep Arousal Disorder Detection Based on the Interaction of Physiological Signals and Metaheuristic Learning

The vast majority of sleep disturbances are caused by various types of sleep arousal. To diagnose sleep disorders and prevent health problems such as cardiovascular disease and cognitive impairment, sleep arousals must be accurately detected. Consequently, sleep specialists must spend considerable time and effort analyzing polysomnography (PSG) recordings to determine the level of arousal during sleep. The development of an automated sleep arousal detection system based on PSG would considerably benefit clinicians. We quantify the EEG-ECG by using Lyapunov exponents, fractals, and wavelet transforms to identify sleep stages and arousal disorders. In this paper, an efficient hybrid-learning method is introduced for the first time to detect and assess arousal incidents. Modified drone squadron optimization (mDSO) algorithm is used to optimize the support vector machine (SVM) with radial basis function (RBF) kernel. EEG-ECG signals are preprocessed samples from the SHHS sleep dataset and the PhysioBank challenge 2018. In comparison to other traditional methods for identifying sleep disorders, our physiological signals correlation innovation is much better than similar approaches. Based on the proposed model, the average error rate was less than 2%-7%, respectively, for two-class and four-class issues. Additionally, the proper classification of the five sleep stages is determined to be accurate 92.3% of the time. In clinical trials of sleep disorders, the hybrid-learning model technique based on EEG-ECG signal correlation features is effective in detecting arousals.

Personalized Blood Glucose Prediction for Type 1 Diabetes Using Evidential Deep Learning and Meta-Learning

The availability of large amounts of data from continuous glucose monitoring (CGM), together with the latest advances in deep learning techniques, have opened the door to a new paradigm of algorithm design for personalized blood glucose (BG) prediction in type 1 diabetes (T1D) with superior performance. However, there are several challenges that prevent the widespread implementation of deep learning algorithms in actual clinical settings, including unclear prediction confidence and limited training data for new T1D subjects. To this end, we propose a novel deep learning framework, Fast-adaptive and Confident Neural Network (FCNN), to meet these clinical challenges. In particular, an attention-based recurrent neural network is used to learn representations from CGM input and forward a weighted sum of hidden states to an evidential output layer, aiming to compute personalized BG predictions with theoretically supported model confidence. The model-agnostic meta-learning is employed to enable fast adaptation for a new T1D subject with limited training data. The proposed framework has been validated on three clinical datasets. In particular, for a dataset including 12 subjects with T1D, FCNN achieved a root mean square error of 18.64±2.60 mg/dL and 31.07±3.62 mg/dL for 30 and 60-minute prediction horizons, respectively, which outperformed all the considered baseline methods with significant improvements. These results indicate that FCNN is a viable and effective approach for predicting BG levels in T1D. The well-trained models can be implemented in smartphone apps to improve glycemic control by enabling proactive actions through real-time glucose alerts.

Sleep Staging Framework with Physiologically Harmonized Sub-Networks

Sleep screening is an important tool for both healthcare and neuroscientific research. Automatic sleep scoring is an alternative to the time-consuming gold-standard manual scoring procedure. Recently there have seen promising results on automatic stage scoring by extracting spatio-temporal features via deep neural networks from electroencephalogram (EEG). However, such methods fail to consistently yield good performance due to a missing piece in data representation: the medical criterion of the sleep scoring task on top of EEG features. We argue that capturing stage-specific features that satisfy the criterion of sleep medicine is non-trivial for automatic sleep scoring. This paper considers two criteria: Transient stage marker and Overall profile of EEG features, then we propose a physiologically meaningful framework for sleep stage scoring via mixed deep neural networks. The framework consists of two sub-networks: feature extraction networks, constructed in consideration of the physiological characteristics of sleep, and an attention-based scoring decision network. Moreover, we quantize the framework for potential use under an IoT setting. For proof-of-concept, the performance of the proposed framework is demonstrated by introducing multiple sleep datasets with the largest comprising 42,560 h recorded from 5,793 subjects. From the experiment results, the proposed method achieves a competitive stage scoring performance, especially for Wake, N2, and N3, with higher F1 scores of 0.92, 0.86, and 0.88, respectively. Moreover, the feasibility analysis of framework quantization provides a potential for future implementation in the edge computing field and clinical settings.

Comparison of deep transfer learning algorithms and transferability measures for wearable sleep staging

Background

Obtaining medical data using wearable sensors is a potential replacement for in-hospital monitoring, but the lack of data for such sensors poses a challenge for development. One solution is using in-hospital recordings to boost performance via transfer learning. While there are many possible transfer learning algorithms, few have been tested in the domain of EEG-based sleep staging. Furthermore, there are few ways for determining which transfer learning method will work best besides exhaustive testing. Measures of transferability do exist, but are typically used for selection of pre-trained models rather than algorithms and few have been tested on medical signals. We tested several supervised transfer learning algorithms on a sleep staging task using a single channel of EEG (AF7-Fpz) captured from an in-home commercial system.

Results

Two neural networks—one bespoke and another state-of-art open-source architecture—were pre-trained on one of six source datasets comprising 11,561 subjects undergoing clinical polysomnograms (PSGs), then re-trained on a target dataset of 75 full-night recordings from 24 subjects. Several transferability measures were then tested to determine which is most effective for assessing performance on unseen target data. Performance on the target dataset was improved using transfer learning, with re-training the head layers being the most effective in the majority of cases (up to 63.9% of cases). Transferability measures generally provided significant correlations with accuracy (up to \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$r = -0.53$$\end{document}r=-0.53).

Conclusion

Re-training the head layers provided the largest performance boost. Transferability measures are useful indicators of transfer learning effectiveness.

Applying Meta-Learning and Iso Principle for Development of EEG-Based Emotion Induction System

Music is often used for emotion induction. ince the emotions felt when listening to it vary from person to person, customized music is required. Our previous work designed a music generation system that created personalized music based on participants' emotions predicted from EEG data. Although our system effectively induced emotions, unfortunately, it suffered from two problems. The first is that a long EEG recording is required to train emotion prediction models. In this paper, we trained models with a small amount of EEG data. We proposed emotion prediction with meta-learning and compared its performance with two other training methods. The second problem is that the generated music failed to consider the participants' emotions before they listened to music. We solved this challenge by constructing a system that adapted an iso principle that gradually changed the music from close to the participants' emotions to the target emotion. Our results showed that emotion prediction with meta-learning had the lowest RMSE among three methods (p < 0.016). Both a music generation system based on the iso principle and our conventional music generation system more effectively induced emotion than music generation that was not based on the emotions of the participants (p < 0.016).

CSAC-Net: Fast Adaptive sEMG Recognition through Attention Convolution Network and Model-Agnostic Meta-Learning

Gesture recognition through surface electromyography (sEMG) provides a new method for the control algorithm of bionic limbs, which is a promising technology in the field of human–computer interaction. However, subject specificity of sEMG along with the offset of the electrode makes it challenging to develop a model that can quickly adapt to new subjects. In view of this, we introduce a new deep neural network called CSAC-Net. Firstly, we extract the time-frequency feature from the raw signal, which contains rich information. Secondly, we design a convolutional neural network supplemented by an attention mechanism for further feature extraction. Additionally, we propose to utilize model-agnostic meta-learning to adapt to new subjects and this learning strategy achieves better results than the state-of-the-art methods. By the basic experiment on CapgMyo and three ablation studies, we demonstrate the advancement of CSAC-Net.

Automatic sleep staging of EEG signals: recent development, challenges, and future directions

Modern deep learning holds a great potential to transform clinical practice on human sleep. Teaching a machine to carry out routine tasks would be a tremendous reduction in workload for clinicians. Sleep staging, a fundamental step in sleep practice, is a suitable task for this and will be the focus in this article. Recently, automatic sleep staging systems have been trained to mimic manual scoring, leading to similar performance to human sleep experts, at least on scoring of healthy subjects. Despite tremendous progress, we have not seen automatic sleep scoring adopted widely in clinical environments. This review aims to give a shared view of the authors on the most recent state-of-the-art development in automatic sleep staging, the challenges that still need to be addressed, and the future directions for automatic sleep scoring to achieve clinical value.

Transfer learning for non-image data in clinical research: A scoping review

BACKGROUND

Transfer learning is a form of machine learning where a pre-trained model trained on a specific task is reused as a starting point and tailored to another task in a different dataset. While transfer learning has garnered considerable attention in medical image analysis, its use for clinical non-image data is not well studied. Therefore, the objective of this scoping review was to explore the use of transfer learning for non-image data in the clinical literature.

METHODS AND FINDINGS

We systematically searched medical databases (PubMed, EMBASE, CINAHL) for peer-reviewed clinical studies that used transfer learning on human non-image data. We included 83 studies in the review. More than half of the studies (63%) were published within 12 months of the search. Transfer learning was most often applied to time series data (61%), followed by tabular data (18%), audio (12%) and text (8%). Thirty-three (40%) studies applied an image-based model to non-image data after transforming data into images (e.g. spectrograms). Twenty-nine (35%) studies did not have any authors with a health-related affiliation. Many studies used publicly available datasets (66%) and models (49%), but fewer shared their code (27%).

CONCLUSIONS

In this scoping review, we have described current trends in the use of transfer learning for non-image data in the clinical literature. We found that the use of transfer learning has grown rapidly within the last few years. We have identified studies and demonstrated the potential of transfer learning in clinical research in a wide range of medical specialties. More interdisciplinary collaborations and the wider adaption of reproducible research principles are needed to increase the impact of transfer learning in clinical research.

SleepTransformer: Automatic Sleep Staging with Interpretability and Uncertainty Quantification

BACKGROUND

Black-box skepticism is one of the main hindrances impeding deep-learning-based automatic sleep scoring from being used in clinical environments.

METHODS

Towards interpretability, this work proposes a sequence-to-sequence sleep staging model, namely SleepTransformer. It is based on the transformer backbone and offers interpretability of the models decisions at both the epoch and sequence level. We further propose a simple yet efficient method to quantify uncertainty in the models decisions. The method, which is based on entropy, can serve as a metric for deferring low-confidence epochs to a human expert for further inspection.

RESULTS

Making sense of the transformers self-attention scores for interpretability, at the epoch level, the attention scores are encoded as a heat map to highlight sleep-relevant features captured from the input EEG signal. At the sequence level, the attention scores are visualized as the influence of different neighboring epochs in an input sequence (i.e. the context) to recognition of a target epoch, mimicking the way manual scoring is done by human experts.

CONCLUSION

Additionally, we demonstrate that SleepTransformer performs on par with existing methods on two databases of different sizes.

SIGNIFICANCE

Equipped with interpretability and the ability of uncertainty quantification, SleepTransformer holds promise for being integrated into clinical settings.

Unsupervised Domain Adaptation by Statistics Alignment for Deep Sleep Staging Networks

Deep sleep staging networks have reached top performance on large-scale datasets. However, these models perform poorer when training and testing on small sleep cohorts due to data inefficiency. Transferring well-trained models from large-scale datasets (source domain) to small sleep cohorts (target domain) is a promising solution but still remains challenging due to the domain-shift issue. In this work, an unsupervised domain adaptation approach, domain statistics alignment (DSA), is developed to bridge the gap between the data distribution of source and target domains. DSA adapts the source models on the target domain by modulating the domain-specific statistics of deep features stored in the Batch Normalization (BN) layers. Furthermore, we have extended DSA by introducing cross-domain statistics in each BN layer to perform DSA adaptively (AdaDSA). The proposed methods merely need the well-trained source model without access to the source data, which may be proprietary and inaccessible. DSA and AdaDSA are universally applicable to various deep sleep staging networks that have BN layers. We have validated the proposed methods by extensive experiments on two state-of-the-art deep sleep staging networks, DeepSleepNet+ and U-time. The performance was evaluated by conducting various transfer tasks on six sleep databases, including two large-scale databases, MASS and SHHS, as the source domain, four small sleep databases as the target domain. Thereinto, clinical sleep records acquired in Huashan Hospital, Shanghai, were used. The results show that both DSA and AdaDSA could significantly improve the performance of source models on target domains, providing novel insights into the domain generalization problem in sleep staging tasks.

Practical Lessons on 12-Lead ECG Classification: Meta-Analysis of Methods From PhysioNet/Computing in Cardiology Challenge 2020

Cardiovascular diseases (CVDs) are one of the most fatal disease groups worldwide. Electrocardiogram (ECG) is a widely used tool for automatically detecting cardiac abnormalities, thereby helping to control and manage CVDs. To encourage more multidisciplinary researches, PhysioNet/Computing in Cardiology Challenge 2020 (Challenge 2020) provided a public platform involving multi-center databases and automatic evaluations for ECG classification tasks. As a result, 41 teams successfully submitted their solutions and were qualified for rankings. Although Challenge 2020 was a success, there has been no in-depth methodological meta-analysis of these solutions, making it difficult for researchers to benefit from the solutions and results. In this study, we aim to systematically review the 41 solutions in terms of data processing, feature engineering, model architecture, and training strategy. For each perspective, we visualize and statistically analyze the effectiveness of the common techniques, and discuss the methodological advantages and disadvantages. Finally, we summarize five practical lessons based on the aforementioned analysis: (1) Data augmentation should be employed and adapted to specific scenarios; (2) Combining different features can improve performance; (3) A hybrid design of different types of deep neural networks (DNNs) is better than using a single type; (4) The use of end-to-end architectures should depend on the task being solved; (5) Multiple models are better than one. We expect that our meta-analysis will help accelerate the research related to ECG classification based on machine-learning models.

Evaluating Convolutional Neural Networks as a Method of EEG-EMG Fusion

Wearable robotic exoskeletons have emerged as an exciting new treatment tool for disorders affecting mobility; however, the human–machine interface, used by the patient for device control, requires further improvement before robotic assistance and rehabilitation can be widely adopted. One method, made possible through advancements in machine learning technology, is the use of bioelectrical signals, such as electroencephalography (EEG) and electromyography (EMG), to classify the user's actions and intentions. While classification using these signals has been demonstrated for many relevant control tasks, such as motion intention detection and gesture recognition, challenges in decoding the bioelectrical signals have caused researchers to seek methods for improving the accuracy of these models. One such method is the use of EEG–EMG fusion, creating a classification model that decodes information from both EEG and EMG signals simultaneously to increase the amount of available information. So far, EEG–EMG fusion has been implemented using traditional machine learning methods that rely on manual feature extraction; however, new machine learning methods have emerged that can automatically extract relevant information from a dataset, which may prove beneficial during EEG–EMG fusion. In this study, Convolutional Neural Network (CNN) models were developed using combined EEG–EMG inputs to determine if they have potential as a method of EEG–EMG fusion that automatically extracts relevant information from both signals simultaneously. EEG and EMG signals were recorded during elbow flexion–extension and used to develop CNN models based on time–frequency (spectrogram) and time (filtered signal) domain image inputs. The results show a mean accuracy of 80.51 ± 8.07% for a three-class output (33.33% chance level), with an F-score of 80.74%, using time–frequency domain-based models. This work demonstrates the viability of CNNs as a new method of EEG–EMG fusion and evaluates different signal representations to determine the best implementation of a combined EEG–EMG CNN. It leverages modern machine learning methods to advance EEG–EMG fusion, which will ultimately lead to improvements in the usability of wearable robotic exoskeletons.

Multi-View Spatial-Temporal Graph Convolutional Networks With Domain Generalization for Sleep Stage Classification

Sleep stage classification is essential for sleep assessment and disease diagnosis. Although previous attempts to classify sleep stages have achieved high classification performance, several challenges remain open: 1) How to effectively utilize time-varying spatial and temporal features from multi-channel brain signals remains challenging. Prior works have not been able to fully utilize the spatial topological information among brain regions. 2) Due to the many differences found in individual biological signals, how to overcome the differences of subjects and improve the generalization of deep neural networks is important. 3) Most deep learning methods ignore the interpretability of the model to the brain. To address the above challenges, we propose a multi-view spatial-temporal graph convolutional networks (MSTGCN) with domain generalization for sleep stage classification. Specifically, we construct two brain view graphs for MSTGCN based on the functional connectivity and physical distance proximity of the brain regions. The MSTGCN consists of graph convolutions for extracting spatial features and temporal convolutions for capturing the transition rules among sleep stages. In addition, attention mechanism is employed for capturing the most relevant spatial-temporal information for sleep stage classification. Finally, domain generalization and MSTGCN are integrated into a unified framework to extract subject-invariant sleep features. Experiments on two public datasets demonstrate that the proposed model outperforms the state-of-the-art baselines.

Transferring Structured Knowledge in Unsupervised Domain Adaptation of a Sleep Staging Network

Automatic sleep staging based on deep learning (DL) has been attracting attention for analyzing sleep quality and determining treatment effects. It is challenging to acquire long-term sleep data from numerous subjects and manually labeling them even though most DL-based models are trained using large-scale sleep data to provide state-of-the-art performance. One way to overcome this data shortage is to create a pre-trained network with an existing large-scale dataset (source domain) that is applicable to small cohorts of datasets (target domain); however, discrepancies in data distribution between the domains prevent successful refinement of this approach. In this paper, we propose an unsupervised domain adaptation method for sleep staging networks to reduce discrepancies by realigning the domains in the same space and producing domain-invariant features. Specifically, in addition to a classical domain discriminator, we introduce local dis-criminators-subject and stage-to maintain the intrinsic structure of sleep data to decrease local misalignments while using adversarial learning to play a minimax game between the feature extractor and discriminators. Moreover, we present several optimization schemes during training because the conventional adversarial learning is not effective to our training scheme. We evaluate the performance of the proposed method by examining the staging performances of a baseline network compared with direct transfer (DT) learning in various conditions. The experimental results demonstrate that the proposed domain adaptation significantly improves the performance though it needs no labeled sleep data in target domain.

Estimation of Apnea-Hypopnea Index using Deep Learning on 3D Craniofacial Scans

Obstructive sleep apnea (OSA) is characterized by decreased breathing events that occur through the night, with severity reported as the apnea-hypopnea index (AHI), which is associated with certain craniofacial features. In this study, we used data from 1366 patients collected as part of Stanford Technology Analytics and Genomics in Sleep (STAGES) across 11 US and Canadian sleep clinics and analyzed 3D craniofacial scans with the goal of predicting AHI, as measured using gold standard nocturnal polysomnography (PSG). First, the algorithm detects pre-specified landmarks on mesh objects and aligns scans in 3D space. Subsequently, 2D images and depth maps are generated by rendering and rotating scans by 45-degree increments. Resulting images were stacked as channels and used as input to multi-view convolutional neural networks, which were trained and validated in a supervised manner to predict AHI values derived from PSGs. The proposed model achieved a mean absolute error of 11.38 events/hour, a Pearson correlation coefficient of 0.4, and accuracy for predicting OSA of 67% using 10-fold cross-validation. The model improved further by adding patient demographics and variables from questionnaires. We also show that the model performed at the level of three sleep medicine specialists, who used clinical experience to predict AHI based on 3D scan displays. Finally, we created topographic displays of the most important facial features used by the model to predict AHI, showing importance of the neck and chin area. The proposed algorithm has potential to serve as an inexpensive and efficient screening tool for individuals with suspected OSA.

SleepPoseNet: Multi-View Learning for Sleep Postural Transition Recognition Using UWB

Recognizing movements during sleep is crucial for the monitoring of patients with sleep disorders, and the utilization of ultra-wideband (UWB) radar for the classification of human sleep postures has not been explored widely. This study investigates the performance of an off-the-shelf single antenna UWB in a novel application of sleep postural transition (SPT) recognition. The proposed Multi-View Learning, entitled SleepPoseNet or SPN, with time series data augmentation aims to classify four standard SPTs. SPN exhibits an ability to capture both time and frequency features, including the movement and direction of sleeping positions. The data recorded from 38 volunteers displayed that SPN with a mean accuracy of 73.7 ±0.8 % significantly outperformed the mean accuracy of 59.9 ±0.7 % obtained from deep convolution neural network (DCNN) in recent state-of-the-art work on human activity recognition using UWB. Apart from UWB system, SPN with the data augmentation can ultimately be adopted to learn and classify time series data in various applications.

Automatic Detection of Obstructive Sleep Apnea Events Using a Deep CNN-LSTM Model

Obstructive sleep apnea (OSA) is a common sleep-related respiratory disorder. Around the world, more and more people are suffering from OSA. Because of the limitation of monitor equipment, many people with OSA remain undetected. Therefore, we propose a sleep-monitoring model based on single-channel electrocardiogram using a convolutional neural network (CNN), which can be used in portable OSA monitor devices. To learn different scale features, the first convolution layer comprises three types of filters. The long short-term memory (LSTM) is used to learn the long-term dependencies such as the OSA transition rules. The softmax function is connected to the final fully connected layer to obtain the final decision. To detect a complete OSA event, the raw ECG signals are segmented by a 10 s overlapping sliding window. The proposed model is trained with the segmented raw signals and is subsequently tested to evaluate its event detection performance. According to experiment analysis, the proposed model exhibits Cohen's kappa coefficient of 0.92, a sensitivity of 96.1%, a specificity of 96.2%, and an accuracy of 96.1% with respect to the Apnea-ECG dataset. The proposed model is significantly higher than the results from the baseline method. The results prove that our approach could be a useful tool for detecting OSA on the basis of a single-lead ECG.