A digital approach to asthma self-management in adults: Protocol for a pragmatic randomized controlled trial.

Asthma self-management can improve symptom control, but adherence to established self-management behaviors is often poor. With adult asthma uncontrolled in over 60% of U.S. cases, there is a need for scalable, cost-effective tools to improve asthma outcomes. Here we describe a protocol for the Asthma Digital Study, a 24-month, decentralized, pragmatic, open-label, randomized controlled trial investigating the impact of a digital asthma self-management (DASM) program on asthma outcomes in adults. The program leverages consumer-grade devices with a smartphone app to provide "smart nudges," symptom logging, trigger tracking, and other features. Participants are recruited (target N = 900) from throughout the U.S., and randomized to a DASM or control arm (1:1). Co-primary outcomes at one year are a) asthma-associated costs for acute care and b) change from baseline in Asthma Control Test™ scores. Findings may inform decisions around adoption of digital tools for asthma self-management. Trial registration:clinicaltrials.gov identifier: NCT04609644. Registered: Oct 30, 2020.

Pilot Study of a Brief Sleep Intervention for Suicidal Ideation in Bipolar Disorder.

OBJECTIVE: We investigated the feasibility and acceptability of a brief sleep intervention and sleep monitoring device in bipolar disorder (BD), as well as the intervention's effectiveness in reducing suicidal ideation through improved sleep quality and duration. METHODS: Participants (N=13) with BD received 4 sessions of cognitive-behavioral therapy for sleep disturbance and completed assessments of the intervention and sleep device acceptability, mood, suicidal ideation, and sleep at pretreatment and posttreatment. RESULTS: Feasibility and acceptability of the intervention were high at both pretreatment and posttreatment and did not significantly change by the intervention's conclusion, although participants reported being significantly more likely to recommend the intervention to others at posttreatment. The sleep device was easy to understand, wear, and did not interfere with participants' sleep. Suicidal ideation and depressive symptoms significantly decreased from pretreatment to posttreatment (P<0.05). There were small, but not significant, improvements in the percentage of time spent in stable sleep and total average nightly sleep. CONCLUSION: The potential of this intervention for reducing suicidal ideation and improving sleep is promising, but future research is warranted.

Multicomponent Analysis of Sleep Using Electrocortical, Respiratory, Autonomic and Hemodynamic Signals Reveals Distinct Features of Stable and Unstable NREM and REM Sleep.

A new concept of non-rapid eye movement (NREM) and rapid eye movement (REM) sleep is proposed, that of multi-component integrative states that define stable and unstable sleep, respectively, NREMS, NREMUS REMS, and REMUS. Three complementary data sets are used: obstructive sleep apnea (20), healthy subjects (11), and high loop gain sleep apnea (50). We use polysomnography (PSG) with beat-to-beat blood pressure monitoring, and electrocardiogram (ECG)-derived cardiopulmonary coupling (CPC) analysis to demonstrate a bimodal, rather than graded, characteristic of NREM sleep. Stable NREM (NREMS) is characterized by high probability of occurrence of the <1 Hz slow oscillation, high delta power, stable breathing, blood pressure dipping, strong sinus arrhythmia and vagal dominance, and high frequency CPC. Conversely, unstable NREM (NREMUS) has the opposite features: a fragmented and discontinuous <1 Hz slow oscillation, non-dipping of blood pressure, unstable respiration, cyclic variation in heart rate, and low frequency CPC. The dimension of NREM stability raises the possibility of a comprehensive integrated multicomponent network model of NREM sleep which captures sleep onset (e.g., ventrolateral preoptic area-based sleep switch) processes, synaptic homeostatic delta power kinetics, and the interaction of global and local sleep processes as reflected in the spatiotemporal evolution of cortical "UP" and "DOWN" states, while incorporating the complex dynamics of autonomic-respiratory-hemodynamic systems during sleep. Bimodality of REM sleep is harder to discern in health. However, individuals with combined obstructive and central sleep apnea allows ready recognition of REMS and REMUS (stable and unstable REM sleep, respectively), especially when there is a discordance of respiratory patterns in relation to conventional stage of sleep.

Cerebral Small Vessel Diseases and Sleep Related Strokes.

BACKGROUND AND PURPOSE: Sleep related Stroke (SRS) is common and has been associated with cerebral small vessel diseases (SVD) in ischemic strokes (ISs). We tested the hypothesis that SRS is associated with SVD in both ischemic and hemorrhagic stroke. METHODS: Prospectively collected data from patients consecutively enrolled after intracerebral hemorrhage (ICH) related to SVD or after IS were analyzed. Symptom onset was recorded as SRS versus awake. Each ICH was grouped according to lobar and deep locations. The IS cohort was etiologically characterized based on the Causative Classification of Stroke system. Frequencies of SRS within and between ICH and IS cohorts as well as its associations (etiology, risk factors) were analyzed. RESULTS: We analyzed 1812 IS (mean age 67.9 years ± 15.9 years, 46.4% female) and 1038 ICH patients (mean age 72.5 years ± 13.0 years, 45.4% female). SRS was significantly more common among SVD-related ICH patients (n = 276, 26.6%) when compared to all IS (n = 363, 20.0%, P < .001) and in both, small artery occlusion (SAO) related IS and lobar ICH within the respective IS and ICH cohorts (16.3% SRS versus 9.1% awake for SAO within all IS, P < .001; and 57.1% SRS versus 47.7% awake for lobar bleeds within all ICH, P = .008). These associations remained significant after controlling for age, sex and risk factors. CONCLUSIONS: SRS was associated with SVD. The SAO etiology and cerebral amyloid angiopathy related lobar ICH suggest that the presence of SVD can interact with sleep or arousal related hemodynamic changes to cause ischemic and hemorrhagic stroke.

Brain age from the electroencephalogram of sleep.

The human electroencephalogram (EEG) of sleep undergoes profound changes with age. These changes can be conceptualized as "brain age (BA)," which can be compared to chronological age to reflect the degree of deviation from normal aging. Here, we develop an interpretable machine learning model to predict BA based on 2 large sleep EEG data sets: the Massachusetts General Hospital (MGH) sleep lab data set (N = 2532; ages 18-80); and the Sleep Heart Health Study (SHHS, N = 1974; ages 40-80). The model obtains a mean absolute deviation of 7.6 years between BA and chronological age (CA) in healthy participants in the MGH data set. As validation, a subset of SHHS containing longitudinal EEGs 5.2 years apart shows an average of 5.4 years increase in BA. Participants with significant neurological or psychiatric disease exhibit a mean excess BA, or "brain age index" (BAI = BA-CA) of 4 years relative to healthy controls. Participants with hypertension and diabetes have a mean excess BA of 3.5 years. The findings raise the prospect of using the sleep EEG as a potential biomarker for healthy brain aging.

A novel adhesive biosensor system for detecting respiration, cardiac, and limb movement signals during sleep: validation with polysomnography.

BACKGROUND: Although in-lab polysomnography (PSG) remains the gold standard for objective sleep monitoring, the use of at-home sensor systems has gained popularity in recent years. Two categories of monitoring, autonomic and limb movement physiology, are increasingly recognized as critical for sleep disorder phenotyping, yet at-home options remain limited outside of research protocols. The purpose of this study was to validate the BiostampRC® sensor system for respiration, electrocardiography (ECG), and leg electromyography (EMG) against gold standard PSG recordings. METHODS: We report analysis of cardiac and respiratory data from 15 patients and anterior tibialis (AT) data from 19 patients undergoing clinical PSG for any indication who simultaneously wore BiostampRC® sensors on the chest and the bilateral AT muscles. BiostampRC® is a flexible, adhesive, wireless sensor capable of capturing accelerometry, ECG, and EMG. We compared BiostampRC® data and feature extractions with those obtained from PSG. RESULTS: The heart rate extracted from BiostampRC® ECG showed strong agreement with the PSG (cohort root mean square error of 5 beats per minute). We found the thoracic BiostampRC® respiratory waveform, derived from its accelerometer, accurately calculated the respiratory rate (mean average error of 0.26 and root mean square error of 1.84 breaths per minute). The AT EMG signal supported periodic limb movement detection, with area under the curve of the receiver operating characteristic curve equaling 0.88. Upon completion, 88% of subjects indicated willingness to wear BiostampRC® at home on an exit survey. CONCLUSION: The results demonstrate that BiostampRC® is a tolerable and accurate method for capturing respiration, ECG, and AT EMG time series signals during overnight sleep when compared with simultaneous PSG recordings. The signal quality sufficiently supports analytics of clinical relevance. Larger longitudinal in-home studies are required to support the role of BiostampRC® in clinical management of sleep disorders involving the autonomic nervous system and limb movements.

Expert-level sleep scoring with deep neural networks.

Objectives: Scoring 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. Methods: We 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. Results: When 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. Conclusions: By 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.

Effects of post-exposure naps on exposure therapy for social anxiety.

Exposure therapy for social anxiety disorder (SAD) utilizes fear extinction, a memory process enhanced by sleep. We investigated whether naps following exposure sessions might improve symptoms and biomarkers in response to social stress in adults undergoing 5-week exposure-based group SAD therapy. Thirty-two participants aged 18-39 (18 females) with SAD were randomized. Before and after treatment, participants completed the Liebowitz Social Anxiety Scale (LSAS) and underwent a Trier Social Stress Test with psychophysiological monitoring (mpTSST) that included skin conductance (SCL), electromyographic (EMG) and electrocardiographic recording, and an auditory startle procedure while anticipating the social stressor. At sessions 3 and 4, exposure was followed by either a 120-min polysomnographically monitored sleep opportunity (Nap, N = 17) or wakefulness (Wake, N = 15). Primary hypotheses about SAD symptom change (LSAS) and EMG blink-startle response failed to differ with naps, despite significant symptom improvement (LSAS) with therapy. Some secondary biomarkers, however, provided preliminary support for enhanced extinction learning with naps, with trend-level Time (pre-, post-treatment) × Arm interactions and significant reduction from pre- to post treatment in the Nap arm alone for mpTSST SCL and salivary cortisol rise. Because of the small sample size and limited sleep duration, additional well-powered studies with more robust sleep interventions are indicated.

Antiepileptic drug treatment after an unprovoked first seizure: A decision analysis.

OBJECTIVE: To compare the expected quality-adjusted life-years (QALYs) in adult patients undergoing immediate vs deferred antiepileptic drug (AED) treatment after a first unprovoked seizure. METHODS: We constructed a simulated clinical trial (Markov decision model) to compare immediate vs deferred AED treatment after a first unprovoked seizure in adults. Three base cases were considered, representing patients with varying degrees of seizure recurrence risk and effect of seizures on quality of life (QOL). Cohort simulation was performed to determine which treatment strategy would maximize the patient's expected QALYs. Sensitivity analyses were guided by clinical data to define decision thresholds across plausible measurement ranges, including seizure recurrence rate, effect of seizure recurrence on QOL, and efficacy of AEDs. RESULTS: For patients with a moderate risk of recurrent seizures (52.0% over 10 years after first seizure), immediate AED treatment maximized QALYs compared to deferred treatment. Sensitivity analyses showed that for the preferred choice to change to deferred AED treatment, key clinical measures needed to reach implausible values were 10-year seizure recurrence rate ≤38.0%, QOL reduction with recurrent seizures ≤0.06, and efficacy of AEDs on lowering seizure recurrence rate ≤16.3%. CONCLUSION: Our model determined that immediate AED treatment is preferable to deferred treatment in adult first-seizure patients over a wide and clinically relevant range of variables. Furthermore, our analysis suggests that the 10-year seizure recurrence rate that justifies AED treatment (38.0%) is substantially lower than the 60% threshold used in the current definition of epilepsy.

Repetitive transcranial magnetic stimulation of the right parietal cortex for comorbid generalized anxiety disorder and insomnia: A randomized, double-blind, sham-controlled pilot study.

BACKGROUND: Repetitive transcranial magnetic stimulation (rTMS) has been considered to be a promising technique for the treatment of neuropsychiatric disorders. However, little is known about the effectiveness of rTMS in the treatment of generalized anxiety disorder (GAD). Moreover, treatment data on comorbid GAD and insomnia remain lacking. The aim of this study was to examine the therapeutic effects of 1 Hz rTMS applied over the right parietal lobe on both anxiety and insomnia symptoms in patients with comorbid GAD and insomnia. METHODS: 36 patients were randomized to either sham or active rTMS group (n = 18 each group). The rTMS was administered over the right posterior parietal cortex (P4 electrode site) at a frequency of 1 Hz and an intensity of 90% of the resting motor threshold. RESULTS: Ten days of 1 Hz rTMS to the right parietal lobe significantly improved both anxiety and insomnia symptoms in the active group. Although the anxiety severity was not significantly correlated with insomnia severity at baseline, the improvement in the Hamilton Rating Scale for Anxiety (HRSA) scores were positively correlated with improvement in the Pittsburgh Sleep Quality Index (PSQI) scores. CONCLUSIONS: The present study is the first randomized sham-controlled study to assess the effectiveness of low frequency rTMS on the right parietal lobe in patients with comorbid GAD and insomnia. Our results suggested that 1 Hz low frequency rTMS administered over the parietal cortex is effective for both anxiety and insomnia symptoms in patients with comorbid GAD and insomnia.

Heart rate phenotypes and clinical correlates in a large cohort of adults without sleep apnea.

BACKGROUND: Normal sleep is associated with typical physiological changes in both the central and autonomic nervous systems. In particular, nocturnal blood pressure dipping has emerged as a strong marker of normal sleep physiology, whereas the absence of dipping or reverse dipping has been associated with cardiovascular risk. However, nocturnal blood pressure is not measured commonly in clinical practice. Heart rate (HR) dipping in sleep may be a similar important marker and is measured routinely in at-home and in-laboratory sleep testing. METHODS: We performed a retrospective cross-sectional analysis of diagnostic polysomnography in a clinically heterogeneous cohort of n=1047 adults without sleep apnea. RESULTS: We found that almost half of the cohort showed an increased HR in stable nonrapid eye movement sleep (NREM) compared to wake, while only 13.5% showed a reduced NREM HR of at least 10% relative to wake. The strongest correlates of HR dipping were younger age and male sex, whereas the periodic limb movement index (PLMI), sleep quality, and Epworth Sleepiness Scale (ESS) scores were not correlated with HR dipping. PLMI was however significantly correlated with metrics of impaired HR variability (HRV): increased low-frequency power and reduced high-frequency power. HRV metrics were unrelated to sleep quality or the ESS value. Following the work of Vgontzas et al, we also analyzed the sub-cohort with insomnia symptoms and short objective sleep duration. Interestingly, the sleep-wake stage-specific HR values depended upon insomnia symptoms more than sleep duration. CONCLUSION: While our work demonstrates heterogeneity in cardiac metrics (HR and HRV), the population analysis suggests that pathological signatures of HR (nondipping and elevation) are common even in this cohort selected for the absence of sleep apnea. Future prospective work in clinical populations will further inform risk stratification and set the stage for testing interventions.

Sleep deficiency and motor vehicle crash risk in the general population: a prospective cohort study.

BACKGROUND: Insufficient sleep duration and obstructive sleep apnea, two common causes of sleep deficiency in adults, can result in excessive sleepiness, a well-recognized cause of motor vehicle crashes, although their contribution to crash risk in the general population remains uncertain. The objective of this study was to evaluate the relation of sleep apnea, sleep duration, and excessive sleepiness to crash risk in a community-dwelling population. METHODS: This was a prospective observational cohort study nested within the Sleep Heart Health Study, a community-based study of the health consequences of sleep apnea. The participants were 1745 men and 1456 women aged 40-89 years. Sleep apnea was measured by home polysomnography and questionnaires were used to assess usual sleep duration and daytime sleepiness. A follow-up questionnaire 2 years after baseline ascertained driving habits and motor vehicle crash history. Logistic regression analysis was used to examine the relation of sleep apnea and sleep duration at baseline to the occurrence of motor vehicle crashes during the year preceding the follow-up visit, adjusting for relevant covariates. The population-attributable fraction of motor vehicle crashes was estimated from the sample proportion of motor vehicle crashes and the adjusted odds ratios for motor vehicle crash within each exposure category. RESULTS: Among 3201 evaluable participants, 222 (6.9%) reported at least one motor vehicle crash during the prior year. A higher apnea-hypopnea index (p < 0.01), fewer hours of sleep (p = 0.04), and self-reported excessive sleepiness (p < 0.01) were each significantly associated with crash risk. Severe sleep apnea was associated with a 123% increased crash risk, compared to no sleep apnea. Sleeping 6 hours per night was associated with a 33% increased crash risk, compared to sleeping 7 or 8 hours per night. These associations were present even in those who did not report excessive sleepiness. The population-attributable fraction of motor vehicle crashes was 10% due to sleep apnea and 9% due to sleep duration less than 7 hours. CONCLUSIONS: Sleep deficiency due to either sleep apnea or insufficient sleep duration is strongly associated with motor vehicle crashes in the general population, independent of self-reported excessive sleepiness.

Estimation of adaptive ventilation success and failure using polysomnogram and outpatient therapy biomarkers.

Study Objectives: Adaptive servo-ventilation (ASV) devices provide anticyclic pressure support for the treatment of central and/or complex sleep apnea, including heart failure patients. Variability in responses in the clinic and negative clinical trials motivated assessment of standard and novel signal biomarkers for ASV efficacy. Methods: Multiple clinical databases were queried to assess potential signal biomarkers of ASV effectiveness, including the following: (1) attended laboratory adaptive ventilation titrations: 108, of which 66 had mainstream ETCO2 measurements; (2) AirView data in 98 participants, (3) complete data, from diagnostic polysomnogram (PSG) through review and prospective analysis of on-therapy data using SleepyHead freeware in 44 participants; and (4) hemodynamic data in the form of beat-to-beat blood pressure during ASV titration, using a Finometer in five participants. Results: Signal biomarkers of reduced ASV efficacy were noted as follows: (1) an arousal index which markedly exceeded the respiratory event index during positive pressure titration; (2) persistent pressure cycling during long-term ASV therapy, visible in online review systems or reviewing data using freeware; (3) the ASV-associated pressure cycling induced arousals, sleep fragmentation, and blood pressure surges; and (4) elevated ratios of 95th percentile to median tidal volume, minute ventilation, and respiratory rate were associated with pressure cycling. High intraclass coefficients (>0.8) for machine apnea-hypopnea index and other extractable metrics were consistent with stability of patterns over multiple nights of use. Global clinical outcomes correlated negatively with pressure cycling. Conclusions: Potential polysomnographic- and device-related signal biomarkers of ASV efficacy are described and may allow improved estimation of therapeutic effectiveness of adaptive ventilation.

ADARRI: a novel method to detect spurious R-peaks in the electrocardiogram for heart rate variability analysis in the intensive care unit.

We developed a simple and fully automated method for detecting artifacts in the R-R interval (RRI) time series of the ECG that is tailored to the intensive care unit (ICU) setting. From ECG recordings of 50 adult ICU-subjects we selected 60 epochs with valid R-peak detections and 60 epochs containing artifacts leading to missed or false positive R-peak detections. Next, we calculated the absolute value of the difference between two adjacent RRIs (adRRI), and obtained the empirical probability distributions of adRRI values for valid R-peaks and artifacts. From these, we calculated an optimal threshold for separating adRRI values arising from artifact versus non-artefactual data. We compared the performance of our method with the methods of Berntson and Clifford on the same data. We identified 257,458 R-peak detections, of which 235,644 (91.5%) were true detections and 21,814 (8.5%) arose from artifacts. Our method showed superior performance for detecting artifacts with sensitivity 100%, specificity 99%, precision 99%, positive likelihood ratio of 100 and negative likelihood ratio <0.001 compared to Berntson's and Clifford's method with a sensitivity, specificity, precision and positive and negative likelihood ratio of 99%, 78%, 82%, 4.5, 0.013 for Berntson's method and 55%, 98%, 96%, 27.5, 0.460 for Clifford's method, respectively. A novel algorithm using a patient-independent threshold derived from the distribution of adRRI values in ICU ECG data identifies artifacts accurately, and outperforms two other methods in common use. Furthermore, the threshold was calculated based on real data from critically ill patients and the algorithm is easy to implement.

Sleep devices: wearables and nearables, informational and interventional, consumer and clinical.

The field of sleep is in many ways ideally positioned to take full advantage of advancements in technology and analytics that is fueling the mobile health movement. Combining hardware and software advances with increasingly available big datasets that contain scored data obtained under gold standard sleep laboratory conditions completes the trifecta of this perfect storm. This review highlights recent developments in consumer and clinical devices for sleep, emphasizing the need for validation at multiple levels, with the ultimate goal of using personalized data and advanced algorithms to provide actionable information that will improve sleep health.

Cardiopulmonary coupling spectrogram as an ambulatory clinical biomarker of sleep stability and quality in health, sleep apnea, and insomnia.

STUDY OBJECTIVES: Ambulatory tracking of sleep and sleep pathology is rapidly increasing with the introduction of wearable devices. The objective of this study was to evaluate a wearable device which used novel computational analysis of the electrocardiogram (ECG), collected over multiple nights, as a method to track the dynamics of sleep quality in health and disease. METHODS: This study used the ECG as a primary signal, a wearable device, the M1, and an analysis of cardiopulmonary coupling to estimate sleep quality. The M1 measures trunk movements, the ECG, body position, and snoring vibrations. Data from three groups of patients were analyzed: healthy participants and people with sleep apnea and insomnia, obtained from multiple nights of recording. Analysis focused on summary measures and night-to-night variability, specifically the intraclass coefficient. RESULTS: Data were collected from 10 healthy participants, 18 people with positive pressure-treated sleep apnea, and 20 people with insomnia, 128, 65, and 121 nights, respectively. In any participant, all nights were consecutive. High-frequency coupling (HFC), the signal biomarker of stable breathing and stable sleep, showed high intraclass coefficients (ICCs) in healthy participants and people with sleep apnea (0.83, 0.89), but only 0.66 in people with insomnia. The only statistically significant difference between weekday and weekend in healthy subjects was HFC duration: 242.8 ± 53.8 vs. 275.8 ± 57.1 minutes (89 vs. 39 total nights), F(1,126) = 9.86, p = .002. CONCLUSIONS: The M1 and similar wearable devices provide new opportunities to measure sleep in dynamic ways not possible before. These measurements can yield new biological insights and aid clinical management.

Dexmedetomidine promotes biomimetic non-rapid eye movement stage 3 sleep in humans: A pilot study.

OBJECTIVES: Sleep, which comprises of rapid eye movement (REM) and non-REM stages 1-3 (N1-N3), is a natural occurring state of decreased arousal that is crucial for normal cardiovascular, immune and cognitive function. The principal sedative drugs produce electroencephalogram beta oscillations, which have been associated with neurocognitive dysfunction. Pharmacological induction of altered arousal states that neurophysiologically approximate natural sleep, termed biomimetic sleep, may eliminate drug-induced neurocognitive dysfunction. METHODS: We performed a prospective, single-site, three-arm, randomized-controlled, crossover polysomnography pilot study (n = 10) comparing natural, intravenous dexmedetomidine- (1-µg/kg over 10 min [n = 7] or 0.5-µg/kg over 10 min [n = 3]), and zolpidem-induced sleep in healthy volunteers. Sleep quality and psychomotor performance were assessed with polysomnography and the psychomotor vigilance test, respectively. Sleep quality questionnaires were also administered. RESULTS: We found that dexmedetomidine promoted N3 sleep in a dose dependent manner, and did not impair performance on the psychomotor vigilance test. In contrast, zolpidem extended release was associated with decreased theta (∼5-8 Hz; N2 and N3) and increased beta oscillations (∼13-25 Hz; N2 and REM). Zolpidem extended release was also associated with increased lapses on the psychomotor vigilance test. No serious adverse events occurred. CONCLUSIONS: Pharmacological induction of biomimetic N3 sleep with psychomotor sparing benefits is feasible. SIGNIFICANCE: These results suggest that α2a adrenergic agonists may be developed as a new class of sleep enhancing medications with neurocognitive sparing benefits.

Large-Scale Automated Sleep Staging.

Study Objectives: Automated sleep staging has been previously limited by a combination of clinical and physiological heterogeneity. Both factors are in principle addressable with large data sets that enable robust calibration. However, the impact of sample size remains uncertain. The objectives are to investigate the extent to which machine learning methods can approximate the performance of human scorers when supplied with sufficient training cases and to investigate how staging performance depends on the number of training patients, contextual information, model complexity, and imbalance between sleep stage proportions. Methods: A total of 102 features were extracted from six electroencephalography (EEG) channels in routine polysomnography. Two thousand nights were partitioned into equal (n = 1000) training and testing sets for validation. We used epoch-by-epoch Cohen's kappa statistics to measure the agreement between classifier output and human scorer according to American Academy of Sleep Medicine scoring criteria. Results: Epoch-by-epoch Cohen's kappa improved with increasing training EEG recordings until saturation occurred (n = ~300). The kappa value was further improved by accounting for contextual (temporal) information, increasing model complexity, and adjusting the model training procedure to account for the imbalance of stage proportions. The final kappa on the testing set was 0.68. Testing on more EEG recordings leads to kappa estimates with lower variance. Conclusion: 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.