Radar-based sleep stage classification in children undergoing polysomnography: a pilot-study

Unobtrusive Bed Monitor State of the Art

On average, people spend more than a quarter of their day in bed. If physiological information could be collected automatically while we sleep, it would be effective not only for health management but also for disease prevention. Unobtrusive bed monitoring devices have been developed over the past 30 years or so to detect physiological information without awareness, and this method attracted attention again in the 2020s, with the proliferation of deep learning, AI, and IoT. This section describes the current state of the art.

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

STUDY OBJECTIVES

U-Sleep is a publicly available automated sleep stager, but has not been independently validated using pediatric data. We aimed to (1) 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 (2) identify clinical and demographic characteristics that impact U-Sleep accuracy, using a clinical dataset of 3,114 polysomnograms from a tertiary center.

METHODS

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 2 1-sided test. Multivariable regression and generalized additive modeling were used on the clinical dataset to estimate the effects of age, comorbidities, and polysomnographic findings on U-Sleep performance.

RESULTS

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 (2 1-sided test P < .01). Median (interquartile range) kappa agreement between U-Sleep 2.0 and clinical sleep-staging was kappa = 0.69 (0.22). Modeling 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).

CONCLUSIONS

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.

CITATION

Kevat A, Steinkey R, Suresh S, et al. Evaluation of automated pediatric sleep stage classification using U-Sleep: a convolutional neural network. J Clin Sleep Med. 2025;21(2):277-285.

LANMAO sleep recorder versus polysomnography in neonatal EEG recording and sleep analysis

BACKGROUND

The field of neonatal sleep analysis is burgeoning with devices that purport to offer alternatives to polysomnography (PSG) for monitoring sleep patterns. However, the majority of these devices are limited in their capacity, typically only distinguishing between sleep and wakefulness. This study aims to assess the efficacy of a novel wearable electroencephalographic (EEG) device, the LANMAO Sleep Recorder, in capturing EEG data and analyzing sleep stages, and to compare its performance against the established PSG standard.

METHODS

The study involved concurrent sleep monitoring of 34 neonates using both PSG and the LANMAO device. Initially, the study verified the consistency of raw EEG signals captured by the LANMAO device, employing relative spectral power analysis and Pearson correlation coefficients (PCC) for validation. Subsequently, the LANMAO device's integrated automated sleep staging algorithm was evaluated by comparing its output with expert-generated sleep stage classifications.

RESULTS

Analysis revealed that the PCC between the relative spectral powers of various frequency bands during different sleep stages ranged from 0.28 to 0.48. Specifically, the correlation for delta waves was recorded at 0.28. The automated sleep staging algorithm of the LANMAO device demonstrated an overall accuracy of 79.60 %, Cohen kappa of 0.65, and F1 Score of 76.93 %. Individual accuracy for Wake at 87.20 %, NREM at 85.70 %, and REM Sleep at 81.30 %.

CONCLUSION

While the LANMAO Sleep Recorder's automated sleep staging algorithm necessitates further refinement, the device shows promise in accurately recording neonatal EEG during sleep. Its potential for minimal invasiveness makes it an appealing option for monitoring sleep conditions in newborns, suggesting a novel approach in the field of neonatal sleep analysis.

Noncontact respiration monitoring techniques in young children: A scoping review

Pediatric sleep-related breathing disorders, or sleep-disordered breathing (SDB), cover a range of conditions, including obstructive sleep apnea, central sleep apnea, sleep-related hypoventilation disorders, and sleep-related hypoxemia disorder. Pediatric SDB is often underdiagnosed, potentially due to difficulties associated with performing the gold standard polysomnography in children. This scoping review aims to: (1) provide an overview of the studies reporting on safe, noncontact monitoring of respiration in young children, (2) describe the accuracy of these techniques, and (3) highlight their respective advantages and limitations. PubMed and EMBASE were searched for studies researching techniques in children <12 years old. Both quantitative data and the quality of the studies were analyzed. The evaluation of study quality was conducted using the QUADAS-2 tool. A total of 19 studies were included. Techniques could be grouped into bed-based methods, microwave radar, video, infrared (IR) cameras, and garment-embedded sensors. Most studies either measured respiratory rate (RR) or detected apneas; n = 2 aimed to do both. At present, bed-based approaches are at the forefront of research in noncontact RR monitoring in children, boasting the most sophisticated algorithms in this field. Yet, despite extensive studies, there remains no consensus on a definitive method that outperforms the rest. The accuracies reported by these studies tend to cluster within a similar range, indicating that no single technique has emerged as markedly superior. Notably, all identified methods demonstrate capability in detecting body movements and RR, with reported safety for use in children across the board. Further research into contactless alternatives should focus on cost-effectiveness, ease-of-use, and widespread availability.

An electroencephalography-based sleep index and supervised machine learning as a suitable tool for automated sleep classification in children

STUDY OBJECTIVES

Although sleep is frequently disrupted in the pediatric intensive care unit, it is currently not possible to perform real-time sleep monitoring at the bedside. In this study, spectral band powers of electroencephalography data are used to derive a simple index for sleep classification.

METHODS

Retrospective study at Erasmus MC Sophia Children's Hospital, using hospital-based polysomnography recordings obtained in non-critically ill children between 2017 and 2021. Six age categories were defined: 6-12 months, 1-3 years, 3-5 years, 5-9 years, 9-13 years, and 13-18 years. Candidate index measures were derived by calculating spectral band powers in different frequent frequency bands of smoothed electroencephalography. With the best performing index, sleep classification models were developed for two, three, and four states via decision tree and five-fold nested cross-validation. Model performance was assessed across age categories and electroencephalography channels.

RESULTS

In total 90 patients with polysomnography were included, with a mean (standard deviation) recording length of 10.3 (1.1) hours. The best performance was obtained with the gamma to delta spectral power ratio of the F4-A1 and F3-A1 channels with smoothing. Balanced accuracy was 0.88, 0.74, and 0.57 for two-, three-, and four-state classification. Across age categories, balanced accuracy ranged between 0.83 and 0.92 and 0.72 and 0.77 for two- and three-state classification, respectively.

CONCLUSIONS

We propose an interpretable and generalizable sleep index derived from single-channel electroencephalography for automated sleep monitoring at the bedside in non-critically ill children ages 6 months to 18 years, with good performance for two- and three-state classification.

CITATION

van Twist E, Hiemstra FW, Cramer ABG, et al. An electroencephalography-based sleep index and supervised machine learning as a suitable tool for automated sleep classification in children. J Clin Sleep Med. 2024;20(3):389-397.

Developing a deep learning model for sleep stage prediction in obstructive sleep apnea cohort using 60 GHz frequency-modulated continuous-wave radar

Given the significant impact of sleep on overall health, radar technology offers a promising, non-invasive, and cost-effective avenue for the early detection of sleep disorders, even prior to relying on polysomnography (PSG)-based classification. In this study, we employed an attention-based bidirectional long short-term memory (Attention Bi-LSTM) model to accurately predict sleep stages using 60 GHz frequency-modulated continuous-wave (FMCW) radar. Our dataset comprised 78 participants from an ongoing obstructive sleep apnea (OSA) cohort, recruited between July 2021 and November 2022, who underwent overnight polysomnography alongside radar sensor monitoring. The dataset encompasses comprehensive polysomnography recordings, spanning both sleep and wakefulness states. The predictions achieved a Cohen's kappa coefficient of 0.746 and an overall accuracy of 85.2% in classifying wakefulness, rapid-eye-movement (REM) sleep, and non-REM (NREM) sleep (N1 + N2 + N3). The results demonstrated that the models incorporating both Radar 1 and Radar 2 data consistently outperformed those using only Radar 1 data, indicating the potential benefits of utilising multiple radars for sleep stage classification. Although the performance of the models tended to decline with increasing OSA severity, the addition of Radar 2 data notably improved the classification accuracy. These findings demonstrate the potential of radar technology as a valuable screening tool for sleep stage classification.

Optimizing sleep staging on multimodal time series: Leveraging borderline synthetic minority oversampling technique and supervised convolutional contrastive learning

Sleep is an important research area in nutritional medicine that plays a crucial role in human physical and mental health restoration. It can influence diet, metabolism, and hormone regulation, which can affect overall health and well-being. As an essential tool in the sleep study, the sleep stage classification provides a parsing of sleep architecture and a comprehensive understanding of sleep patterns to identify sleep disorders and facilitate the formulation of targeted sleep interventions. However, the class imbalance issue is typically salient in sleep datasets, which severely affects classification performances. To address this issue and to extract optimal multimodal features of EEG, EOG, and EMG that can improve the accuracy of sleep stage classification, a Borderline Synthetic Minority Oversampling Technique (B-SMOTE)-Based Supervised Convolutional Contrastive Learning (BST-SCCL) is proposed, which can avoid the risk of data mismatch between various sleep knowledge domains (varying health conditions and annotation rules) and strengthening learning characteristics of the N1 stage from the pair-wise segments comparison strategy. The lightweight residual network architecture with a novel truncated cross-entropy loss function is designed to accommodate multimodal time series and boost the training speed and performance stability. The proposed model has been validated on four well-known public sleep datasets (Sleep-EDF-20, Sleep-EDF-78, ISRUC-1, and ISRUC-3) and its superior performance (overall accuracy of 91.31-92.34%, MF1 of 88.21-90.08%, and Cohen's Kappa coefficient k of 0.87-0.89) has further demonstrated its effectiveness. It shows the great potential of contrastive learning for cross-domain knowledge interaction in precision medicine.

Ultra-Wideband Radar for Simultaneous and Unobtrusive Monitoring of Respiratory and Heart Rates in Early Childhood: A Deep Transfer Learning Approach

Unobtrusive monitoring of children's heart rate (HR) and respiratory rate (RR) can be valuable for promoting the early detection of potential health issues, improving communication with healthcare providers and reducing unnecessary hospital visits. A promising solution for wireless vital sign monitoring is radar technology. This paper presents a novel approach for the simultaneous estimation of children's RR and HR utilizing ultra-wideband (UWB) radar using a deep transfer learning algorithm in a cohort of 55 children. The HR and RR are calculated by processing radar signals via spectrogram from time epochs of 10 s (25 sample length of hamming window with 90% overlap) and then transforming the resultant representation into 2-dimensional images. These images were fed into a pre-trained Visual Geometry Group-16 (VGG-16) model (trained on ImageNet dataset), with weights of five added layers fine-tuned using the proposed data. The prediction on the test data achieved a mean absolute error (MAE) of 7.3 beats per minute (BPM < 6.5% of average HR) and 2.63 breaths per minute (BPM < 7% of average RR). We also achieved a significant Pearson's correlation of 77% and 81% between true and extracted for HR and RR, respectively. HR and RR samples are extracted every 10 s.

Sleep Stage Classification in Children Using Self-Attention and Gaussian Noise Data Augmentation

The analysis of sleep stages for children plays an important role in early diagnosis and treatment. This paper introduces our sleep stage classification method addressing the following two challenges: the first is the data imbalance problem, i.e., the highly skewed class distribution with underrepresented minority classes. For this, a Gaussian Noise Data Augmentation (GNDA) algorithm was applied to polysomnography recordings to seek the balance of data sizes for different sleep stages. The second challenge is the difficulty in identifying a minority class of sleep stages, given their short sleep duration and similarities to other stages in terms of EEG characteristics. To overcome this, we developed a DeConvolution- and Self-Attention-based Model (DCSAM) which can inverse the feature map of a hidden layer to the input space to extract local features and extract the correlations between all possible pairs of features to distinguish sleep stages. The results on our dataset show that DCSAM based on GNDA obtains an accuracy of 90.26% and a macro F1-score of 86.51% which are higher than those of our previous method. We also tested DCSAM on a well-known public dataset-Sleep-EDFX-to prove whether it is applicable to sleep data from adults. It achieves a comparable performance to state-of-the-art methods, especially accuracies of 91.77%, 92.54%, 94.73%, and 95.30% for six-stage, five-stage, four-stage, and three-stage classification, respectively. These results imply that our DCSAM based on GNDA has a great potential to offer performance improvements in various medical domains by considering the data imbalance problems and correlations among features in time series data.

The past, present, and future of sleep quality assessment and monitoring

Sleep quality is considered to be an individual's self-satisfaction with all aspects of the sleep experience. Good sleep not only improves a person's physical, mental and daily functional health, but also improves the quality-of-life level to some extent. In contrast, chronic sleep deprivation can increase the risk of diseases such as cardiovascular diseases, metabolic dysfunction and cognitive and emotional dysfunction, and can even lead to increased mortality. The scientific evaluation and monitoring of sleep quality is an important prerequisite for safeguarding and promoting the physiological health of the body. Therefore, we have compiled and reviewed the existing methods and emerging technologies commonly used for subjective and objective evaluation and monitoring of sleep quality, and found that subjective sleep evaluation is suitable for clinical screening and large-scale studies, while objective evaluation results are more intuitive and scientific, and in the comprehensive evaluation of sleep, if we want to get more scientific monitoring results, we should combine subjective and objective monitoring and dynamic monitoring.

Deep learning-based algorithm accurately classifies sleep stages in preadolescent children with sleep-disordered breathing symptoms and age-matched controls

Introduction

Visual sleep scoring has several shortcomings, including inter-scorer inconsistency, which may adversely affect diagnostic decision-making. Although automatic sleep staging in adults has been extensively studied, it is uncertain whether such sophisticated algorithms generalize well to different pediatric age groups due to distinctive EEG characteristics. The preadolescent age group (10-13-year-olds) is relatively understudied, and thus, we aimed to develop an automatic deep learning-based sleep stage classifier specifically targeting this cohort.

Methods

A dataset (n = 115) containing polysomnographic recordings of Icelandic preadolescent children with sleep-disordered breathing (SDB) symptoms, and age and sex-matched controls was utilized. We developed a combined convolutional and long short-term memory neural network architecture relying on electroencephalography (F4-M1), electrooculography (E1-M2), and chin electromyography signals. Performance relative to human scoring was further evaluated by analyzing intra- and inter-rater agreements in a subset (n = 10) of data with repeat scoring from two manual scorers.

Results

The deep learning-based model achieved an overall cross-validated accuracy of 84.1% (Cohen's kappa κ = 0.78). There was no meaningful performance difference between SDB-symptomatic (n = 53) and control subgroups (n = 52) [83.9% (κ = 0.78) vs. 84.2% (κ = 0.78)]. The inter-rater reliability between manual scorers was 84.6% (κ = 0.78), and the automatic method reached similar agreements with scorers, 83.4% (κ = 0.76) and 82.7% (κ = 0.75).

Conclusion

The developed algorithm achieved high classification accuracy and substantial agreements with two manual scorers; the performance metrics compared favorably with typical inter-rater reliability between manual scorers and performance reported in previous studies. These suggest that our algorithm may facilitate less labor-intensive and reliable automatic sleep scoring in preadolescent children.

Development of a non-contact sleep monitoring system for children

Daily monitoring is important, even for healthy children, because sleep plays a critical role in their development and growth. Polysomnography is necessary for sleep monitoring. However, measuring sleep requires specialized equipment and knowledge and is difficult to do at home. In recent years, smartwatches and other devices have been developed to easily measure sleep. However, they cannot measure children's sleep, and contact devices may disturb their sleep.

A non-contact method of measuring sleep is the use of video during sleep. This is most suitable for the daily monitoring of children’s sleep, as it is simple and inexpensive. However, the algorithms have been developed only based on adult sleep, whereas children’s sleep is known to differ considerably from that of adults.

For this reason, we conducted a non-contact estimation of sleep stages for children using video. The participants were children between the ages of 0–6 years old. We estimated the four stages of sleep using the body movement information calculated from the videos recorded. Six parameters were calculated from body movement information. As children’s sleep is known to change significantly as they grow, estimation was divided into two groups (0–2 and 3–6 years).

The results show average estimation accuracies of 46.7 ± 6.6 and 49.0 ± 4.8% and kappa coefficients of 0.24 ± 0.11 and 0.28 ± 0.06 in the age groups of 0–2 and 3–6 years, respectively. This performance is comparable to or better than that reported in previous adult studies.

Development of a classifier to screen for severe sleep disorders in children

This study aimed to develop an automatic classifier for the identification of severe sleep disorders that require immediate intervention in children. Our study assessed 7,008 children (age: 0-83 months) in Japan, whose parents and nursery teachers recorded their 14-day sleep patterns. Sleep quality was assessed by pediatricians and scored as 1 (no severe sleep disorder) or 0 (severe sleep disorder). Discriminant analysis was performed for each age group using sleep quality (0 or 1) as the dependent variable and variables in the 14-day sleep log as independent variables. A stepwise method was used to select the independent variables to build the best model. The accuracy of the discriminant analysis for the age groups ranged from 71.3 to 97.3%. In summary, we developed an automatic classifier with sufficient application value to screen for severe sleep disorders in children. In the future, this classifier can be used to rapidly determine the presence or absence of severe sleep disorders in children based on their 14-day sleep logs, thus allowing immediate intervention.

Non-contact Sleep/Wake Monitoring Using Impulse-Radio Ultrawideband Radar in Neonates

Background: The gold standard for sleep monitoring, polysomnography (PSG), is too obtrusive and limited for practical use with tiny infants or in neonatal intensive care unit (NICU) settings. The ability of impulse-radio ultrawideband (IR-UWB) radar, a non-contact sensing technology, to assess vital signs and fine movement asymmetry in neonates was recently demonstrated. The purpose of this study was to investigate the possibility of quantitatively distinguishing and measuring sleep/wake states in neonates using IR-UWB radar and to compare its accuracy with behavioral observation-based sleep/wake analyses using video recordings. Methods: One preterm and three term neonates in the NICU were enrolled, and voluntary movements and vital signs were measured by radar at ages ranging from 2 to 27 days. Data from a video camcorder, amplitude-integrated electroencephalography (aEEG), and actigraphy were simultaneously recorded for reference. Radar signals were processed using a sleep/wake decision algorithm integrated with breathing signals and movement features. Results: The average recording time for the analysis was 13.0 (7.0-20.5) h across neonates. Compared with video analyses, the sleep/wake decision algorithm for neonates correctly classified 72.2% of sleep epochs and 80.6% of wake epochs and achieved a final Cohen's kappa coefficient of 0.49 (0.41-0.59) and an overall accuracy of 75.2%. Conclusions: IR-UWB radar can provide considerable accuracy regarding sleep/wake decisions in neonates, and although current performance is not yet sufficient, this study demonstrated the feasibility of its possible use in the NICU for the first time. This unobtrusive, non-contact radar technology is a promising method for monitoring sleep/wake states with vital signs in neonates.