Automated EEG sleep staging in the term-age baby using a generative modelling approach

Resting state electroencephalographic brain activity in neonates can predict age and is indicative of neurodevelopmental outcome

OBJECTIVE

Electroencephalography (EEG) can be used to estimate neonates' biological brain age. Discrepancies between postmenstrual age and brain age, termed the brain age gap, can potentially quantify maturational deviation. Existing brain age EEG models are not well suited to clinical cot-side use for estimating neonates' brain age gap due to their dependency on relatively large data and pre-processing requirements.

METHODS

We trained a deep learning model on resting state EEG data from preterm neonates with normal neurodevelopmental Bayley Scale of Infant and Toddler Development (BSID) outcomes, using substantially reduced data requirements. We subsequently tested this model in two independent datasets from two clinical sites.

RESULTS

In both test datasets, using only 20 min of resting-state EEG activity from a single channel, the model generated accurate age predictions: mean absolute error = 1.03 weeks (p-value = 0.0001) and 0.98 weeks (p-value = 0.0001). In one test dataset, where 9-month follow-up BSID outcomes were available, the average neonatal brain age gap in the severe abnormal outcome group was significantly larger than that of the normal outcome group: difference in mean brain age gap = 0.50 weeks (p-value = 0.04).

CONCLUSIONS

These findings demonstrate that the deep learning model generalises to independent datasets from two clinical sites, and that the model's brain age gap magnitudes differ between neonates with normal and severe abnormal follow-up neurodevelopmental outcomes.

SIGNIFICANCE

The magnitude of neonates' brain age gap, estimated using only 20 min of resting state EEG data from a single channel, can encode information of clinical neurodevelopmental value.

A Sequential End-to-End Neonatal Sleep Staging Model with Squeeze and Excitation Blocks and Sequential Multi-Scale Convolution Neural Networks

Automatic sleep staging offers a quick and objective assessment for quantitatively interpreting sleep stages in neonates. However, most of the existing studies either do not encompass any temporal information, or simply apply neural networks to exploit temporal information at the expense of high computational overhead and modeling ambiguity. This limits the application of these methods to multiple scenarios. In this paper, a sequential end-to-end sleep staging model, SeqEESleepNet, which is competent for parallelly processing sequential epochs and has a fast training rate to adapt to different scenarios, is proposed. SeqEESleepNet consists of a sequence epoch generation (SEG) module, a sequential multi-scale convolution neural network (SMSCNN) and squeeze and excitation (SE) blocks. The SEG module expands independent epochs into sequential signals, enabling the model to learn the temporal information between sleep stages. SMSCNN is a multi-scale convolution neural network that can extract both multi-scale features and temporal information from the signal. Subsequently, the followed SE block can reassign the weights of features through mapping and pooling. Experimental results exhibit that in a clinical dataset, the proposed method outperforms the state-of-the-art approaches, achieving an overall accuracy, F1-score, and Kappa coefficient of 71.8%, 71.8%, and 0.684 on a three-class classification task with a single channel EEG signal. Based on our overall results, we believe the proposed method could pave the way for convenient multi-scenario neonatal sleep staging methods.

Sensory event-related potential morphology predicts age in premature infants

OBJECTIVE

We investigated whether sensory-evoked cortical potentials could be used to estimate the age of an infant. Such a model could be used to identify infants who deviate from normal neurodevelopment.

METHODS

Infants aged between 28- and 40-weeks post-menstrual age (PMA) (166 recording sessions in 96 infants) received trains of visual and tactile stimuli. Neurodynamic response functions for each stimulus were derived using principal component analysis and a machine learning model trained and validated to predict infant age.

RESULTS

PMA could be predicted accurately from the magnitude of the evoked responses (training set mean absolute error and 95% confidence intervals: 1.41 [1.14; 1.74] weeks,p = 0.0001; test set mean absolute error: 1.55 [1.21; 1.95] weeks,p = 0.0002). Moreover, we show that their predicted age (their brain age) is correlated with a measure known to relate to maturity of the nervous system and is linked to long-term neurodevelopment.

CONCLUSIONS

Sensory-evoked potentials are predictive of age in premature infants and brain age deviations are related to biologically and clinically meaningful individual differences in nervous system maturation.

SIGNIFICANCE

This model could be used to detect abnormal development of infants' response to sensory stimuli in their environment and may be predictive of neurodevelopmental outcome.

Build a bridge between ECG and EEG signals for atrial fibrillation diagnosis using AI methods

Atrial fibrillation (AF) is a very common type of cardiac arrhythmia. The main characteristic of AF is an abnormally rapid and disordered atrial rhythm causing an atrial dysfunction, which can be visualized on an electrocardiograph (ECG) and distinguished by irregular fluctuations. Despite continuous and considerable efforts to analyze the pathophysiology of AF, it is challenging to determine the underlying pathogenesis of the disease in individual patients. This study aims to build a bridge between ECG and electroencephalogram (EEG) signals to probe the strong influence between human brain activity and AF by AI methods. We first found that the one-second data fragment shows the most excellent performance in our time window configuration. Secondly, in our proposed measurement, most cortical potentials were partly associated with AF. Thirdly, we found that only a few channels of data were sufficient for analysis. Finally, our experiment shows δ wave has the best performance compared to other wave bands. By AI methods, the paper contributes to concluding that δ wave band of EEG is the most associated brain wave type with AF. By EEG signals from typical regions, the central region, parietal and Occipital might be the most associated encephalic regions with AF. The clinical trial registration number for our study is ChiCTR2300068625.

Automatic neonatal sleep stage classification: A comparative study

Sleep is an essential feature of living beings. For neonates, it is vital for their mental and physical development. Sleep stage cycling is an important parameter to assess neonatal brain and physical development. Therefore, it is crucial to administer newborn's sleep in the neonatal intensive care unit (NICU). Currently, Polysomnography (PSG) is used as a gold standard method for classifying neonatal sleep patterns, but it is expensive and requires a lot of human involvement. Over the last two decades, multiple researchers are working on automatic sleep stage classification algorithms using electroencephalography (EEG), electrocardiography (ECG), and video. In this study, we present a comprehensive review of existing algorithms for neonatal sleep, their limitations and future recommendations. Additionally, a brief comparison of the extracted features, classification algorithms and evaluation parameters is reported in the proposed study.

A convolutional neural network-based decision support system for neonatal quiet sleep detection

Sleep plays an important role in neonatal brain and physical development, making its detection and characterization important for assessing early-stage development. In this study, we propose an automatic and computationally efficient algorithm to detect neonatal quiet sleep (QS) using a convolutional neural network (CNN). Our study used 38-hours of electroencephalography (EEG) recordings, collected from 19 neonates at Fudan Children's Hospital in Shanghai, China (Approval No. (2020) 22). To train and test the CNN, we extracted 12 prominent time and frequency domain features from 9 bipolar EEG channels. The CNN architecture comprised two convolutional layers with pooling and rectified linear unit (ReLU) activation. Additionally, a smoothing filter was applied to hold the sleep stage for 3 minutes. Through performance testing, our proposed method achieved impressive results, with 94.07% accuracy, 89.70% sensitivity, 94.40% specificity, 79.82% F1-score and a 0.74 kappa coefficient when compared to human expert annotations. A notable advantage of our approach is its computational efficiency, with the entire training and testing process requiring only 7.97 seconds. The proposed algorithm has been validated using leave one subject out (LOSO) validation, which demonstrates its consistent performance across a diverse range of neonates. Our findings highlight the potential of our algorithm for real-time neonatal sleep stage classification, offering a fast and cost-effective solution. This research opens avenues for further investigations in early-stage development monitoring and the assessment of neonatal health.

A two-branch trade-off neural network for balanced scoring sleep stages on multiple cohorts

Introduction

Automatic sleep staging is a classification process with severe class imbalance and suffers from instability of scoring stage N1. Decreased accuracy in classifying stage N1 significantly impacts the staging of individuals with sleep disorders. We aim to achieve automatic sleep staging with expert-level performance in both N1 stage and overall scoring.

Methods

A neural network model combines an attention-based convolutional neural network and a classifier with two branches is developed. A transitive training strategy is employed to balance universal feature learning and contextual referencing. Parameter optimization and benchmark comparisons are conducted using a large-scale dataset, followed by evaluation on seven datasets in five cohorts.

Results

The proposed model achieves an accuracy of 88.16%, Cohen's kappa of 0.836, and MF1 score of 0.818 on the SHHS1 test set, also with comparable performance to human scorers in scoring stage N1. Incorporating multiple cohort data improves its performance. Notably, the model maintains high performance when applied to unseen datasets and patients with neurological or psychiatric disorders.

Discussion

The proposed algorithm demonstrates strong performance and generalizablility, and its direct transferability is noteworthy among similar studies on automated sleep staging. It is publicly available, which is conducive to expanding access to sleep-related analysis, especially those associated with neurological or psychiatric disorders.

A multi-task and multi-channel convolutional neural network for semi-supervised neonatal artefact detection

Objective. Automated artefact detection in the neonatal electroencephalogram (EEG) is crucial for reliable automated EEG analysis, but limited availability of expert artefact annotations challenges the development of deep learning models for artefact detection. This paper proposes a semi-supervised deep learning approach for artefact detection in neonatal EEG that requires few labelled data by training a multi-task convolutional neural network (CNN).Approach. An unsupervised and a supervised objective were jointly optimised by combining an autoencoder and an artefact classifier in one multi-output model that processes multi-channel EEG inputs. The proposed semi-supervised multi-task training strategy was compared to a classical supervised strategy and other existing state-of-the-art models. The models were trained and tested separately on two different datasets, which contained partially annotated multi-channel neonatal EEG. Models were evaluated using the F1-statistic and the relevance of the method was investigated in the context of a functional brain age (FBA) prediction model.Main results. The proposed multi-task and multi-channel CNN methods outperformed state-of-the-art methods, reaching F1 scores of 86.2% and 95.7% on two separate datasets. The proposed semi-supervised multi-task training strategy was shown to be superior to a classical supervised training strategy when the amount of labels in the dataset was artificially reduced. Finally, we found that the error of a brain age prediction model correlated with the amount of automatically detected artefacts in the EEG segment.Significance. Our results show that the proposed semi-supervised multi-task training strategy can train CNNs successfully even when the amount of labels in the dataset is limited. Therefore, this method is a promising semi-supervised technique for developing deep learning models with scarcely labelled data. Moreover, a correlation between the error of FBA estimates and the amount of detected artefacts in the corresponding EEG segments indicates the relevance of artefact detection for robust automated EEG analysis.

Sleep state organisation of moderate to late preterm infants in the neonatal unit

BACKGROUND

Sleep supports neurodevelopment and sleep architecture reflects brain maturation. This prospective observational study describes the nocturnal sleep architecture of healthy moderate to late preterm (MLP) infants in the neonatal unit at 36 weeks post menstrual age (PMA).

METHODS

MLP infants, in the neonatal unit of a tertiary hospital in Ireland from 2017 to 2018, had overnight continuous electroencephalography (cEEG) with video for a minimum 12 h at 36 weeks PMA. The total sleep time (TST) including periods of active sleep (AS), quiet sleep (QS), indeterminate sleep (IS), wakefulness and feeding were identified, annotated and quantified.

RESULTS

A total of 98 infants had cEEG with video monitoring suitable for analysis. The median (IQR) of TST in the 12 h period was 7.09 h (IQR 6.61-7.76 h), 4.58 h (3.69-5.09 h) in AS, 2.02 h (1.76-2.36 h) in QS and 0.65 h (0.48-0.89 h) in IS. The total duration of AS was significantly lower in infants born at lower GA (p = 0.007) whilst the duration of individual QS periods was significantly higher (p = 0.001).

CONCLUSION

Overnight cEEG with video at 36 weeks PMA showed that sleep state architecture is dependent on birth GA. Infants with a lower birth GA have less AS and more QS that may have implications for later neurodevelopment.

IMPACT

EEG provides objective information about the sleep organisation of the moderate to late preterm (MLP) infant. Quantitative changes in sleep states occur with each week of advancing gestational age (GA). Active sleep (AS) is the dominant sleep state that was significantly lower in infants born at lower GA. MLP infants who were exclusively fed orally had a shorter total sleep time and less AS compared to infants who were fed via nasogastric tube.

Examining time-frequency mechanisms of full-fledged deep sleep development in newborns of different gestational age in the first days of their postnatal development

Early age-related changes in EEG time-frequency characteristics during the restful sleep of newborns of different gestational ages result in the development of conventional EEG signs of deep sleep already during the first postnatal week of their life. Allocating newborns to different groups based on their gestational age and duration of postnatal period allowed demonstrating substantial intergroup differences in brain activity during sleep and wakefulness, along with significant variability in the time-frequency characteristics of brain activity. The process of conventional deep sleep development in infants born prior to the week 35 of gestation is associated with an increase in the power of alpha activity in the sensorimotor cortex of the brain.

Sleep State Trend (SST), a bedside measure of neonatal sleep state fluctuations based on single EEG channels

OBJECTIVE

To develop and validate an automated method for bedside monitoring of sleep state fluctuations in neonatal intensive care units.

METHODS

A deep learning-based algorithm was designed and trained using 53 EEG recordings from a long-term (a)EEG monitoring in 30 near-term neonates. The results were validated using an independent dataset from 30 polysomnography recordings. In addition, we constructed Sleep State Trend (SST), a bedside-ready means for visualizing classifier outputs.

RESULTS

The accuracy of quiet sleep detection in the training data was 90%, and the accuracy was comparable (85-86 %) in all bipolar derivations available from the 4-electrode recordings. The algorithm generalized well to a polysomnography dataset, showing 81% overall accuracy despite different signal derivations. SST allowed an intuitive, clear visualization of the classifier output.

CONCLUSIONS

Fluctuations in sleep states can be detected at high fidelity from a single EEG channel, and the results can be visualized as a transparent and intuitive trend in the bedside monitors.

SIGNIFICANCE

The Sleep State Trend (SST) may provide caregivers and clinical studies a real-time view of sleep state fluctuations and its cyclicity.

A-PASS: An automated pipeline to analyze simultaneously acquired EEG-fMRI data for studying brain activities during sleep

OBJECTIVE

Concurrent electroencephalography and functional magnetic resonance imaging (EEG-fMRI) signals can be used to uncover the nature of brain activities during sleep. However, analyzing simultaneously acquired EEG-fMRI data is extremely time consuming and experience dependent. Thus, we developed a pipeline, which we named A-PASS, to automatically analyze simultaneously acquired EEG-fMRI data for studying brain activities during sleep.

APPROACH

A deep learning model was trained on a sleep EEG-fMRI dataset from 45 subjects and used to perform sleep stage scoring. Various fMRI indices can be calculated with A-PASS to depict the neurophysiological characteristics across different sleep stages. We tested the performance of A-PASS on an independent sleep EEG-fMRI dataset from 28 subjects. Statistical maps regarding the main effect of sleep stages and differences between each pair of stages of fMRI indices were generated and compared using both A-PASS and manual processing methods.

MAIN RESULTS

The deep learning model implemented in A-PASS achieved both an accuracy and F1-score higher than 70% for sleep stage classification on EEG data acquired during fMRI scanning. The statistical maps generated from A-PASS largely resembled those produced from manually scored stages plus a combination of multiple software programs.

SIGNIFICANCE

A-PASS allowed efficient EEG-fMRI data processing without manual operation and could serve as a reliable and powerful tool for simultaneous EEG-fMRI studies on sleep.

Automated detection and removal of flat line segments and large amplitude fluctuations in neonatal electroencephalography

Background

Artefact removal in neonatal electroencephalography (EEG) by visual inspection generally depends on the expertise of the operator, is time consuming and is not a consistent pre-processing step to the pipeline for the automated EEG analysis. Therefore, there is the need for the automated detection and removal of artefacts in neonatal EEG, especially of distinct and predominant artefacts such as flat line segments (mainly caused by instrumental error where contact between electrodes and head box is lost) and large amplitude fluctuations (related to neonatal movements).

Method

A threshold-based algorithm for the automated detection and removal of flat line segments and large amplitude fluctuations in neonatal EEG of infants at term-equivalent age is developed. The algorithm applies thresholds to the absolute second difference, absolute amplitude, absolute first difference and the ratio between the frequency content above 50 Hz and the frequency content across all frequencies.

Results

The algorithm reaches a median accuracy of 0.91, a median hit rate of 0.91 and a median false discovery rate of 0.37. Also, a significant improvement (≈10%) in the performance of a four-stage sleep classifier is observed after artefact removal with the proposed algorithm as compared to before its application.

Significance

An automated artefact removal method contributes to the pipeline of automated EEG analysis. The proposed algorithm has shown to have good performance and to be effective in neonatal EEG applications.

Supported Diagnosis of Attention Deficit and Hyperactivity Disorder from EEG Based on Interpretable Kernels for Hidden Markov Models

As a neurodevelopmental pathology, Attention Deficit Hyperactivity Disorder (ADHD) mainly arises during childhood. Persistent patterns of generalized inattention, impulsivity, or hyperactivity characterize ADHD that may persist into adulthood. The conventional diagnosis relies on clinical observational processes yielding high rates of overdiagnosis due to varying interpretations among specialists or missing information. Although several studies have designed objective behavioral features to overcome such an issue, they lack significance. Despite electroencephalography (EEG) analyses extracting alternative biomarkers using signal processing techniques, the nonlinearity and nonstationarity of EEG signals restrain performance and generalization of hand-crafted features. This work proposes a methodology to support ADHD diagnosis by characterizing EEG signals from hidden Markov models (HMM), classifying subjects based on similarity measures for probability functions, and spatially interpreting the results using graphic embeddings of stochastic dynamic models. The methodology learns a single HMM for EEG signal from each patient, so favoring the inter-subject variability. Then, the Probability Product Kernel, specifically developed for assessing the similarity between HMMs, fed a support vector machine that classifies subjects according to their stochastic dynamics. Lastly, the kernel variant of Principal Component Analysis provided a means to visualize the EEG transitions in a two-dimensional space, evidencing dynamic differences between ADHD and Healthy Control children. From the electrophysiological perspective, we recorded EEG under the Stop Signal Task modified with reward levels, which considers cognitive features of interest as insufficient motivational circuits recruitment. The methodology compares the supported diagnosis in two EEG channel setups (whole channel set and channels of interest in frontocentral area) and four frequency bands (Theta, Alpha, Beta rhythms, and a wideband). Results evidence an accuracy rate of 97.0% in the Beta band and in the channels where previous works found error-related negativity events. Such accuracy rate strongly supports the dual pathway hypothesis and motivational deficit concerning the pathophysiology of ADHD. It also demonstrates the utility of joining inhibitory and motivational paradigms with dynamic EEG analysis into a noninvasive and affordable diagnostic tool for ADHD patients.

A Deep Shared Multi-Scale Inception Network Enables Accurate Neonatal Quiet Sleep Detection with Limited EEG Channels

In this paper, we introduce a new variation of the Convolutional Neural Network Inception block, called Sinc, for sleep stage classification in premature newborn babies using electroencephalogram (EEG). In practice, there are many medical centres where only a limited number of EEG channels are recorded. Existing automated algorithms mainly use multi-channel EEGs which perform poorly when fewer numbers of channels are available. The proposed Sinc utilizes multi-scale analysis to place emphasis on the temporal EEG information to be less dependent on the number of EEG channels. In Sinc, we increase the receptive fields through Inception while by additionally sharing the filters that have similar receptive fields, overfitting is controlled and the number of trainable parameters dramatically reduced. To train and test this model, 96 longitudinal EEG recordings from 26 premature infants are used. The Sinc-based model significantly outperforms state-of-the-art neonatal quiet sleep detection algorithms, with mean Kappa 0.77 0.01 (with 8-channel EEG) and 0.75 0.01 (with a single bipolar channel EEG). This is the first study using Inception-based networks for EEG analysis that utilizes filter sharing to improve efficiency and trainability. The suggested network can successfully detect quiet sleep stages with even a single EEG channel making it more practical especially in the hospital setting where cerebral function monitoring is predominantly used.

Investigation of low dimensional feature spaces for automatic sleep staging

BACKGROUND

Automatic sleep stage classification depends crucially on the selection of a good set of descriptive features. However, the selection of a feature set with an appropriate low computational cost without compromising classification performance is still a challenge. This study attempts to represent sleep EEG patterns using a minimum number of features, without significant performance loss.

METHODS

Three feature selection algorithms were applied to a high dimensional feature space comprising 84 features. These methods were based on a bootstrapping approach guided by Gini ranking and mutual information between the features. The algorithms were tested on three scalp electroencephalography (EEG) and one ear-EEG datasets. The relationship between the information carried by different features was investigated using mutual information and illustrated by a graphical clustering tool.

RESULTS

The minimum number of features that can represent the whole feature set without performance loss was found to range between 5 and 11 for different datasets. In ear-EEG, 7 features based on Continuous Wavelet Transform (CWT) resulted in similar performance as the whole set whereas in two scalp EEG datasets, the difference between minimal CWT set and the whole set was statistically significant (0.008 and 0.017 difference in average kappa). Features were divided into groups according to the type of information they carry. The group containing relative power features was identified as the most informative feature group in sleep stage classification, whereas the group containing non-linear features was found to be the least informative.

CONCLUSIONS

This study showed that EEG sleep staging can be performed based on a low dimensional feature space without significant decrease in sleep staging performance. This is especially important in the case of wearable devices like ear-EEG where low computational complexity is needed. The division of the feature space into groups of features, and the analysis of the distribution of feature groups for different feature set sizes, is helpful in the selection of an appropriate feature set.

Machine and Deep Learning in Molecular and Genetic Aspects of Sleep Research

Epidemiological sleep research strives to identify the interactions and causal mechanisms by which sleep affects human health, and to design intervention strategies for improving sleep throughout the lifespan. These goals can be advanced by further focusing on the environmental and genetic etiology of sleep disorders, and by development of risk stratification algorithms, to identify people who are at risk or are affected by, sleep disorders. These studies rely on comprehensive sleep-related data which often contains complex multi-dimensional physiological and molecular measurements across multiple timepoints. Thus, sleep research is well-suited for the application of computational approaches that can handle high-dimensional data. Here, we survey recent advances in machine and deep learning together with the availability of large human cohort studies with sleep data that can jointly drive the next breakthroughs in the sleep-research field. We describe sleep-related data types and datasets, and present some of the tasks in the field that can be targets for algorithmic approaches, as well as the challenges and opportunities in pursuing them.

A perinatal stress calculator for the neonatal intensive care unit: an unobtrusive approach

OBJECTIVE

Early experience of pain and stress in the neonatal intensive care unit is known to have an effect on the neurodevelopment of the infant. However, an automated method to quantify the procedural pain or perinatal stress in premature patients does not exist.

APPROACH

In the current study, EEG and ECG data were collected for more than 3 hours from 136 patients in order to quantify stress exposure. Specifically, features extracted from the EEG and heart-rate variability in both quiet and non-quiet sleep segments were used to develop a subspace linear-discriminant analysis stress classifier.

MAIN RESULTS

The main novelty of the study lies in the absence of intrusive methods or pain elicitation protocols to develop the stress classifier. Three main findings can be reported. First, we developed different stress classifiers for the different age groups and stress intensities, obtaining an area under the curve in the range [0.78-0.93] for non-quiet sleep and [0.77-0.96] for quiet sleep. Second, a dysmature EEG was found in patients under stress. Third, an enhanced cortical connectivity and increased brain-heart communication was correlated with a higher stress load, while the autonomic activity did not seem to be associated to stress exposure.

SIGNIFICANCE

The results shed a light on the pain and stress processing in preterm neonates, suggesting that software tools to investigate dysmature EEG might be helpful to assess stress load in premature patients. These results could be the foundation to assess the impact of stress on infants' development and to tune preventive care.

Identifying tracé alternant activity in neonatal EEG using an inter-burst detection approach

Electroencephalography (EEG) is an important clinical tool for reviewing sleep-wake cycling in neonates in intensive care. Tracé alternant (TA)-a characteristic pattern of EEG activity during quiet sleep in term neonates-is defined by alternating periods of short-duration, high-voltage activity (bursts) separated by lower-voltage activity (inter-bursts). This study presents a novel approach for detecting TA activity by first detecting the inter-bursts and then processing the temporal map of the bursts and inter-bursts. EEG recordings from 72 healthy term neonates were used to develop and evaluate performance of 1) an inter-burst detection method which is then used for 2) detection of TA activity. First, multiple amplitude and spectral features were combined using a support vector machine (SVM) to classify bursts from inter-bursts within TA activity, resulting in a median area under the operating characteristic curve (AUC) of 0.95 (95% confidence interval, CI: 0.93 to 0.98). Second, post-processing of the continuous SVM output, the confidence score, was used to produce a TA envelope. This envelope was used to detect TA activity within the continuous EEG with a median AUC of 0.84 (95% CI: 0.80 to 0.88). These results validate how an inter-burst detection approach combined with post processing can be used to classify TA activity. Detecting the presence or absence of TA will help quantify disruption of the clinically important sleep-wake cycle.

Proposing a convolutional neural network for stress assessment by means of derived heart rate from functional near infrared spectroscopy

BACKGROUND

Stress is known as one of the major factors threatening human health. A large number of studies have been performed in order to either assess or relieve stress by analyzing the brain and heart-related signals.

METHOD

In this study, a method based on the Convolutional Neural Network (CNN) approach is proposed to assess stress induced by the Montreal Imaging Stress Task. The proposed model is trained on the heart rate signal derived from functional Near-Infrared Spectroscopy (fNIRS), which is referred to as HRF. In this regard, fNIRS signals of 20 healthy volunteers were recorded using a configuration of 23 channels located on the prefrontal cortex. The proposed deep learning system consists of two main parts where in the first part, the one-dimensional convolutional neural network is employed to build informative activation maps, and then in the second part, a stack of deep fully connected layers is used to predict the stress existence probability. Thereafter, the employed CNN method is compared with the Dense Neural Network, Support Vector Machine, and Random Forest regarding various classification metrics.

RESULTS

Results clearly showed the superiority of CNN over all other methods. Additionally, the trained HRF model significantly outperforms the model trained on the filtered fNIRS signals, where the HRF model could achieve 98.69 ± 0.45% accuracy, which is 10.09% greater than the accuracy obtained by the fNIRS model.

CONCLUSIONS

Employment of the proposed deep learning system trained on the HRF measurements leads to higher stress classification accuracy than the accuracy reported in the existing studies where the same experimental procedure has been done. Besides, the proposed method suggests better stability with lower variation in prediction. Furthermore, its low computational cost opens up the possibility to be applied in real-time monitoring of stress assessment.

Applying a data-driven approach to quantify EEG maturational deviations in preterms with normal and abnormal neurodevelopmental outcomes

Premature babies are subjected to environmental stresses that can affect brain maturation and cause abnormal neurodevelopmental outcome later in life. Better understanding this link is crucial to developing a clinical tool for early outcome estimation. We defined maturational trajectories between the Electroencephalography (EEG)-derived ‘brain-age’ and postmenstrual age (the age since the last menstrual cycle of the mother) from longitudinal recordings during the baby’s stay in the Neonatal Intensive Care Unit. Data consisted of 224 recordings (65 patients) separated for normal and abnormal outcome at 9–24 months follow-up. Trajectory deviations were compared between outcome groups using the root mean squared error (RMSE) and maximum trajectory deviation (δ^max). 113 features were extracted (per sleep state) to train a data-driven model that estimates brain-age, with the most prominent features identified as potential maturational and outcome-sensitive biomarkers. RMSE and δ^max showed significant differences between outcome groups (cluster-based permutation test, p < 0.05). RMSE had a median (IQR) of 0.75 (0.60–1.35) weeks for normal outcome and 1.35 (1.15–1.55) for abnormal outcome, while δ^max had a median of 0.90 (0.70–1.70) and 1.90 (1.20–2.90) weeks, respectively. Abnormal outcome trajectories were associated with clinically defined dysmature and disorganised EEG patterns, cementing the link between early maturational trajectories and neurodevelopmental outcome.

Sleep State Modulates Resting-State Functional Connectivity in Neonates

The spontaneous cerebral activity that gives rise to resting-state networks (RSNs) has been extensively studied in infants in recent years. However, the influence of sleep state on the presence of observable RSNs has yet to be formally investigated in the infant population, despite evidence that sleep modulates resting-state functional connectivity in adults. This effect could be extremely important, as most infant neuroimaging studies rely on the neonate to remain asleep throughout data acquisition. In this study, we combine functional near-infrared spectroscopy with electroencephalography to simultaneously monitor sleep state and investigate RSNs in a cohort of healthy term born neonates. During active sleep (AS) and quiet sleep (QS) our newborn neonates show functional connectivity patterns spatially consistent with previously reported RSN structures. Our three independent functional connectivity analyses revealed stronger interhemispheric connectivity during AS than during QS. In turn, within hemisphere short-range functional connectivity seems to be enhanced during QS. These findings underline the importance of sleep state monitoring in the investigation of RSNs.

Neonatal sleep stage identification using long short-term memory learning system

Neonatal sleep analysis at the neonatal intensive care units (NICU) is critical for the diagnosis of any brain growth risks during the early stages of life. In this paper, an investigation is carried out on the use of a long short-term memory (LSTM) learning system in automatic sleep stage scoring in neonates. The developed algorithm automatically classifies sleep stages based on inputs from a single channel EEG recording. Up to this date, only a single study have developed an approach for automatic sleep stage scoring in neonatal sleep signals using deep neural network (DNN). A total of 5095 sleep stages signals acquired from EEG recordings of the University of Pittsburgh are used in this study. The sleep stages are annotated by a medical doctor from the Pediatric Neurology Department of Case Western Reserve University for three neonatal sleep stages including the awake (W), active sleep (AS), and quiet sleep (QS) stages on every 60-s epoch. The signals are pre-processed through normalization and filtering. The resulted signals are divided following 4-, 6-, and 10-fold cross-validation schemes. The training and classification process is done using a bi-directional LSTM network classifier built with pre-defined training parameters. At the end, the developed algorithm is evaluated along with a complete summary table that reports the results of this study and other state-of-the-art studies. The current study achieved high levels of Cohen's kappa (κ), accuracy, and F1 score with 91.37%, 96.81%, and 94.43%, respectively. Based on the confusion matrix, the overall true positives percentage reached 95.21%. The developed algorithm gave promising results in automatic sleep stage scoring in neonatal sleep signals. Future work include LSTM architecture and training parameters improvements to enhance the overall accuracy of the classifier.

The future of sleep health: a data-driven revolution in sleep science and medicine

In recent years, there has been a significant expansion in the development and use of multi-modal sensors and technologies to monitor physical activity, sleep and circadian rhythms. These developments make accurate sleep monitoring at scale a possibility for the first time. Vast amounts of multi-sensor data are being generated with potential applications ranging from large-scale epidemiological research linking sleep patterns to disease, to wellness applications, including the sleep coaching of individuals with chronic conditions. However, in order to realise the full potential of these technologies for individuals, medicine and research, several significant challenges must be overcome. There are important outstanding questions regarding performance evaluation, as well as data storage, curation, processing, integration, modelling and interpretation. Here, we leverage expertise across neuroscience, clinical medicine, bioengineering, electrical engineering, epidemiology, computer science, mHealth and human-computer interaction to discuss the digitisation of sleep from a inter-disciplinary perspective. We introduce the state-of-the-art in sleep-monitoring technologies, and discuss the opportunities and challenges from data acquisition to the eventual application of insights in clinical and consumer settings. Further, we explore the strengths and limitations of current and emerging sensing methods with a particular focus on novel data-driven technologies, such as Artificial Intelligence.

A convolutional neural network outperforming state-of-the-art sleep staging algorithms for both preterm and term infants

OBJECTIVE

To classify sleep states using electroencephalogram (EEG) that reliably works over a wide range of preterm ages, as well as term age.

APPROACH

A convolutional neural network is developed to perform 2- and 4-class sleep classification in neonates. The network takes as input an 8-channel 30 s EEG segment and outputs the sleep state probabilities. Apart from simple downsampling of the input and smoothing of the output, the suggested network is an end-to-end algorithm that avoids the need for hand-crafted feature selection or complex pre/post processing steps. To train and test this method, 113 EEG recordings from 42 infants are used.

MAIN RESULTS

For quiet sleep detection (the 2-class problem), mean kappa between the network estimate and the ground truth annotated by EEG human experts is 0.76. The sensitivity and specificity are 90% and 88%, respectively. For 4-class classification, mean kappa is 0.64. The averaged sensitivity and specificity (1 versus all) respectively equal 72% and 91%. The results outperform current state-of-the-art methods for which kappa ranges from 0.66 to 0.70 in preterm and from 0.51 to 0.61 in term infants, based on training and testing using the same database.

SIGNIFICANCE

The proposed method has the highest reported accuracy for EEG sleep state classification for both preterm and term age neonates.

A review of automated sleep stage scoring based on physiological signals for the new millennia

BACKGROUND AND OBJECTIVE

Sleep is an important part of our life. That importance is highlighted by the multitude of health problems which result from sleep disorders. Detecting these sleep disorders requires an accurate interpretation of physiological signals. Prerequisite for this interpretation is an understanding of the way in which sleep stage changes manifest themselves in the signal waveform. With that understanding it is possible to build automated sleep stage scoring systems. Apart from their practical relevance for automating sleep disorder diagnosis, these systems provide a good indication of the amount of sleep stage related information communicated by a specific physiological signal.

METHODS

This article provides a comprehensive review of automated sleep stage scoring systems, which were created since the year 2000. The systems were developed for Electrocardiogram (ECG), Electroencephalogram (EEG), Electrooculogram (EOG), and a combination of signals.

RESULTS

Our review shows that all of these signals contain information for sleep stage scoring.

CONCLUSIONS

The result is important, because it allows us to shift our research focus away from information extraction methods to systemic improvements, such as patient comfort, redundancy, safety and cost.

Temporal evolution of quantitative EEG within 3 days of birth in early preterm infants

For the premature newborn, little is known about changes in brain activity during transition to extra-uterine life. We aim to quantify these changes in relation to the longer-term maturation of the developing brain. We analysed EEG for up to 72 hours after birth from 28 infants born <32 weeks of gestation. These infants had favourable neurodevelopment at 2 years of age and were without significant neurological compromise at time of EEG monitoring. Quantitative EEG was generated using features representing EEG power, discontinuity, spectral distribution, and inter-hemispheric connectivity. We found rapid changes in cortical activity over the 3 days distinct from slower changes associated with gestational age: for many features, evolution over 1 day after birth is equivalent to approximately 1 to 2.5 weeks of maturation. Considerable changes in the EEG immediately after birth implies that postnatal adaption significantly influences cerebral activity for early preterm infants. Postnatal age, in addition to gestational age, should be considered when analysing preterm EEG within the first few days after birth.

A Bayesian parametric model for quantifying brain maturation from sleep-EEG in the vulnerable newborn baby

Newborn babies, particularly preterms, can exhibit early deviations in sleep maturation as seen by Electroencephalogram (EEG) recordings. This may be indicative of cognitive problems by school-age. The current 'clinically-driven' approach uses separate algorithms to first extract sleep states and then predict EEG 'brain-age'. Maturational deviations are identified when the brain-age no longer matches the Postmenstrual Age (PMA, the age since the last menstrual cycle of the mother). However, the PMA range where existing sleep staging algorithms perform optimally, is limited, which subsequently limits the PMA range for brain-age prediction. We introduce a Bayesian Parametric Model (BPM) as a single end-to-end solution to directly estimate brain-age, modelling for sleep state maturation without requiring a separately optimized sleep staging algorithm. Comparison of this model with a traditional multi-stage approach, yields a similar Krippendorff's $\alpha = 0.92$ (a performance measure ranging from 0 (chance agreement) to 1 (perfect agreement)) with the BPM performing better at younger ages <30 weeks PMA. The BPM's potential to detect maturational deviations is also explored on a few preterm babies who were abnormal at 9 months follow-up.