ObjectivesScoring laboratory polysomnography (PSG) data remains a manual task of visually annotating 3 primary categories: sleep stages, sleep disordered breathing, and limb movements. Attempts to automate this process have been hampered by the complexity of PSG signals and physiological heterogeneity between patients. Deep neural networks, which have recently achieved expert-level performance for other complex medical tasks, are ideally suited to PSG scoring, given sufficient training data.MethodsWe used a combination of deep recurrent and convolutional neural networks (RCNN) for supervised learning of clinical labels designating sleep stages, sleep apnea events, and limb movements. The data for testing and training were derived from 10 000 clinical PSGs and 5804 research PSGs.ResultsWhen trained on the clinical dataset, the RCNN reproduces PSG diagnostic scoring for sleep staging, sleep apnea, and limb movements with accuracies of 87.6%, 88.2% and 84.7% on held-out test data, a level of performance comparable to human experts. The RCNN model performs equally well when tested on the independent research PSG database. Only small reductions in accuracy were noted when training on limited channels to mimic at-home monitoring devices: frontal leads only for sleep staging, and thoracic belt signals only for the apnea-hypopnea index.ConclusionsBy creating accurate deep learning models for sleep scoring, our work opens the path toward broader and more timely access to sleep diagnostics. Accurate scoring automation can improve the utility and efficiency of in-lab and at-home approaches to sleep diagnostics, potentially extending the reach of sleep expertise beyond specialty clinics.

The human electroencephalogram (EEG) of sleep undergoes profound changes with age. These changes can be conceptualized as "brain age", which can be compared to an age norm to reflect the deviation from normal aging process. Here, we develop an interpretable machine learning model to predict brain age based on two large sleep EEG datasets: the Massachusetts General Hospital sleep lab dataset (MGH, N = 2,621) covering age 18 to 80; and the Sleep Hearth Health Study (SHHS, N = 3,520) covering age 40 to 80. The model obtains a mean absolute deviation of 8.1 years between brain age and chronological age in the healthy participants in the MGH dataset. As validation, we analyze a subset of SHHS containing longitudinal EEGs 5 years apart, which shows a 5.5 years difference in brain age. Participants with neurological and psychiatric diseases, as well as diabetes and hypertension medications show an older brain age compared to chronological age. The findings raise the prospect of using sleep EEG as a biomarker for healthy brain aging.In total, we identify 2,621 EEGs where 189 of them have neurological or psychiatric diseases. Table 1 provides summary statistics for the dataset.

Training with a large data set enables automated sleep staging that compares favorably with human scorers. Because testing was performed on a large and heterogeneous data set, the performance estimate has low variance and is likely to generalize broadly.

Study Objective:Sleep is reflected not only in the electroencephalogram but also in heart rhythms and breathing patterns. Therefore, we hypothesize that it is possible to accurately stage sleep based on the electrocardiogram (ECG) and respiratory signals.Methods: Using a dataset including 8,682 polysomnographs, we develop deep neural networks to stage sleep from ECG and respiratory signals. Five deep neural networks consisting of convolutional networks and long short-term memory networks are trained to stage sleep using heart and breathing, including the timing of R peaks from ECG, abdominal and chest respiratory effort, and the combinations of these signals.Results: ECG in combination with the abdominal respiratory effort achieve the best performance for staging all five sleep stages with a Cohen's kappa of 0.600 (95% confidence interval 0.599 -0.602); and 0.762 (0.760 -0.763) for discriminating awake vs. rapid eye movement vs. non-rapid eye movement sleep. The performance is better for young participants and for those with a low apnea-hypopnea index, while it is robust for commonly used outpatient medications. Conclusions:Our results validate that ECG and respiratory effort provide substantial information about sleep stages in a large population. It opens new possibilities in sleep research and applications where electroencephalography is not readily available or may be infeasible, such as in critically ill patients. Deep Network ArchitectureWe trained five deep neural networks based on the following input signals and their combinations: 1) ECG; 2) CHEST (chest respiratory effort); 3) ABD (abdominal respiratory effort); 4) ECG+CHEST; and 5) ECG+ABD. Each deep neural network contained a feed-forward convolutional neural network (CNN) which learned features pertaining to each epoch, and a recurrent neural network (RNN), in this case long-short term memory (LSTM), to learn temporal patterns among consecutive epochs.The CNN of the network is similar to that in Hannun et al. 20 . As shown in Figure 1A and Figure 1B, the network for a single type of input signal, i.e. ECG, CHEST or ABD, consists of a convolutional layer, several residual blocks and a final output block. For a network with both ECG and CHEST/ABD as input signals ( Figure 1C), we first fixed the weights of the layers up to the 9 th residual block (gray) for the ECG network and similarly fixed up to the 5 th residual block (gray) for the CHEST/ABD network, concatenated the outputs, and then fed this concatenation into a subnetwork containing five residual blocks and a final output block. The numbers of fixed layers were chosen so that the outputs of layers from different modalities have the same shape (after padding zeros), and were then concatenated.The LSTM of the network has the same structure for different input signals. It is a bi-directional LSTM, where the context cells from the forward and backward directions are concatenated. For the network

A commentary on this article appears in this issue on page 531.

Objective Insomnia is commonly co-morbid with obstructive sleep apnea. Among patients reporting insomnia symptoms, sleep misperception occurs when self-reported sleep duration under-estimates objective measures. Misperception represents a clinical challenge since insomnia management is based entirely on patient self-report. We tested the hypothesis that misperception occurring in sleep apnea patients would improve with subsequent treatment. Methods We compared subjective sleep-wake reports with objective sleep architecture in a cohort of adults with obstructive sleep apnea (n=405) who had two nights of polysomnography (diagnostic and treatment) within a median interval of 92 days. Results Sleep latency was generally over-estimated, while wake after sleep onset and number of awakenings were under-estimated. None of these estimations differed between diagnostic and treatment polysomnography. We observed a large spectrum of total sleep time misperception values during the diagnostic polysomnogram, with one third of the cohort under-estimating their total sleep time by at least 60 minutes. Of those with >60 minute misperception, we observed improved total sleep time perception during treatment polysomnography. Improved perception correlated with improvements in self-reported sleep quality and response confidence. We found no polysomnogram or demographic predictors of total sleep time misperception for the diagnostic polysomnogram, nor did we find objective correlates of improved perception during titration. Conclusion Our results suggest that misperception may improve with treatment of obstructive sleep apnea in patients who also exhibit misperception. Within subject changes in misperception is consistent with misperception being, at least to some extent, a state characteristic, which has implications for management of patients with comorbid insomnia and sleep apnea.

No abstract

Study Objectives: Perception of sleep-wake times may differ from objective measures, although the mechanisms remain elusive. Quantifying the misperception phenotype involves two operational challenges: defining objective sleep latency and treating sleep latency and total sleep time as independent factors. We evaluated a novel approach to address these challenges and test the hypothesis that sleep fragmentation underlies misperception. Methods: We performed a retrospective analysis on patients with or without obstructive sleep apnea during overnight diagnostic polysomnography in our laboratory (n = 391; n = 252). We compared subjective and objective sleep-wake durations to characterize misperception. We introduce a new metric, sleep during subjective latency (SDSL), which captures latency misperception without defining objective sleep latency and allows correction for latency misperception when assessing total sleep time (TST) misperception. Results:The stage content of SDSL is related to latency misperception, but in the opposite manner as our hypothesis: those with > 20 minutes of SDSL had less N1%, more N3%, and lower transition frequency. After adjusting for misperceived sleep during subjective sleep latency, TST misperception was greater in those with longer bouts of REM and N2 stages (OSA patients) as well as N3 (non-OSA patients), which also did not support our hypothesis. Conclusions: Despite the advantages of SDSL as a phenotyping tool to overcome operational issues with quantifying misperception, our results argue against the hypothesis that light or fragmented sleep underlies misperception. Further investigation of sleep physiology utilizing alternative methods than that captured by conventional stages may yield additional mechanistic insights into misperception. I NTRO DUCTI O NAssessment of sleep often includes subjective reports of sleepwake durations, in clinical practice as well as epidemiological studies of sleep duration.1 However, when concurrent objective sleep measurements are available, it is not uncommon to observe that the subjective responses may differ. Underestimation of total sleep time has been previously observed in healthy adults under experimental circumstances, as well as insomnia patients during polysomnography (PSG).2-4 This phenomenon, sometimes termed misperception, may manifest as either overestimating sleep latency (SL) or underestimating total sleep time (TST), or both.4 Some patients with obstructive sleep apnea (OSA) may also exhibit misperception, though perhaps not to the same extent as those with insomnia symptoms. 5-7Several hypotheses have arisen regarding the basis for misperception, although a unified explanation remains elusive. For example, misperception has been linked with anxiety and mood, 3,8 personality traits, and sleep physiology measures such as electroencephalogram patterns of alpha-delta sleep or cyclic alternating pattern and high frequency electroencephalography (EEG) content. [9][10][11][12][13] In addition, we previously investigated possible relat...