Dennis A. Dean

Scaling Up Scientific Discovery in Sleep Medicine: The National Sleep Research Resource.

ABSTRACT Professional sleep societies have identified a need for strategic research in multiple areas that may benefit from access to and aggregation of large, multidimensional datasets. Technological advances provide opportunities to extract and analyze physiological signals and other biomedical information from datasets of unprecedented size, heterogeneity, and complexity. The National Institutes of Health has implemented a Big Data to Knowledge (BD2K) initiative that aims to develop and disseminate state of the art big data access tools and analytical methods. The National Sleep Research Resource (NSRR) is a new National Heart, Lung, and Blood Institute resource designed to provide big data resources to the sleep research community. The NSRR is a web-based data portal that aggregates, harmonizes, and organizes sleep and clinical data from thousands of individuals studied as part of cohort studies or clinical trials and provides the user a suite of tools to facilitate data exploration and data visualization. Each deidentified study record minimally includes the summary results of an overnight sleep study; annotation files with scored events; the raw physiological signals from the sleep record; and available clinical and physiological data. NSRR is designed to be interoperable with other public data resources such as the Biologic Specimen and Data Repository Information Coordinating Center Demographics (BioLINCC) data and analyzed with methods provided by the Research Resource for Complex Physiological Signals (PhysioNet). This article reviews the key objectives, challenges and operational solutions to addressing big data opportunities for sleep research in the context of the national sleep research agenda. It provides information to facilitate further interactions of the user community with NSRR, a community resource.

Taking the lag out of jet lag through model-based schedule design

Travel across multiple time zones results in desynchronization of environmental time cues and the sleep–wake schedule from their normal phase relationships with the endogenous circadian system. Circadian misalignment can result in poor neurobehavioral performance, decreased sleep efficiency, and inappropriately timed physiological signals including gastrointestinal activity and hormone release. Frequent and repeated transmeridian travel is associated with long-term cognitive deficits, and rodents experimentally exposed to repeated schedule shifts have increased death rates. One approach to reduce the short-term circadian, sleep–wake, and performance problems is to use mathematical models of the circadian pacemaker to design countermeasures that rapidly shift the circadian pacemaker to align with the new schedule. In this paper, the use of mathematical models to design sleep–wake and countermeasure schedules for improved performance is demonstrated. We present an approach to designing interventions that combines an algorithm for optimal placement of countermeasures with a novel mode of schedule representation. With these methods, rapid circadian resynchrony and the resulting improvement in neurobehavioral performance can be quickly achieved even after moderate to large shifts in the sleep–wake schedule. The key schedule design inputs are endogenous circadian period length, desired sleep–wake schedule, length of intervention, background light level, and countermeasure strength. The new schedule representation facilitates schedule design, simulation studies, and experiment design and significantly decreases the amount of time to design an appropriate intervention. The method presented in this paper has direct implications for designing jet lag, shift-work, and non-24-hour schedules, including scheduling for extreme environments, such as in space, undersea, or in polar regions.

Periodic Limb Movements During Sleep and Prevalent Hypertension in the Multi-Ethnic Study of Atherosclerosis

Periodic limb movements during sleep (PLMS) are associated with immediate increases in blood pressure. Both PLMS and hypertension have different distributions across racial/ethnic groups. We sought to determine whether PLMS is associated with hypertension among various racial/ethnic groups. A total of 1740 men and women underwent measurement of blood pressure and polysomnography with quantification of PLMS. Hypertension was defined as systolic blood pressure (SBP) ≥140, diastolic BP ≥90, or taking antihypertensive medication. For those taking antihypertensives, an estimated pretreatment SBP value was derived based on observed SBP and medication type/dose. Measures of PLMS, PLMS index, and PLMS arousal index were the main explanatory variables. Hypertension and SBP were modeled with logistic and multivariable regression adjusted for age, sex, body mass index, cardiovascular risk factors, lifestyle/habitual factors, apnea-hypopnea index, and race/ethnicity. In the overall cohort, prevalent hypertension was modestly associated with PLMS index (10 U; odds ratio, 1.05; 95% confidence interval, 1.00–1.10) and PLMS arousal index (1 U; 1.05; 1.01–1.09) after adjusting for confounders. Association in the overall cohort was influenced by large effect sizes in blacks, in whom the odds of prevalent hypertension increased by 21% (1%–45%) for 10 U PLMS index increase and 20% (2%–42%) for 1-U PLMS arousal index increase. In blacks, every 1-U PLMS arousal index increase was associated with SBP 1.01 mm Hg higher (1.01; 0.04–1.98). Associations between PLMS and blood pressure outcomes were also suggested among Chinese-Americans but not in whites or Hispanics. In a multiethnic cohort of community-dwelling men and women, prevalent hypertension and SBP are associated with PLMS frequency in blacks.

A systematic assessment of the association of polysomnographic indices with blood pressure: The multi-ethnic study of atherosclerosis (MESA)

STUDY OBJECTIVE Blood pressure (BP) may be adversely affected by a variety of sleep disturbances, including sleep fragmentation, hypoxemia, respiratory disturbances, and periodic limb movements. We aim to identify which polysomnography indices are most strongly and consistently associated with systolic and diastolic blood pressure (SBP, DBP) levels in a population-based sample. DESIGN Cross-sectional analysis of data from 2,040 participants in the Multi-Ethnic Study of Atherosclerosis (MESA) who underwent polysomnography at MESA Exam 5 in 2011-2013. SETTING Multisite cohort study. PARTICIPANTS Participants were mean age 68 y (54% females; 28% African American, 24% Hispanic, 11% Chinese). MEASUREMENTS Thirty-two candidate polysomnography predictors were identified representing the domains of breathing disturbance frequency, hypoxemia, sleep architecture, and periodic limb movements. Cluster analysis was used for variable reduction. Statistical models, adjusted for potential confounders, were derived using stepwise regression. Final models were selected using cross-validation techniques. RESULTS The apnea-hypopnea index (AHI) defined using a 4% desaturation hypopnea criterion (AHI4P) was most consistently associated with SBP level. The AHI and periodic limb movement index (associated with arousals; PLMIA) were significantly associated with DBP. Estimated adjusted differences in SBP and DBP levels between an individual with no sleep apnea (AHI4P = 0) and one with moderately severe sleep apnea (AHI4P = 30) were 2.2 mm Hg and 1.1 mm Hg, respectively. Each 10-unit increase in the PLMIA was associated with an increase in DBP of 1.2 mm Hg. CONCLUSION Our results support the use of a currently recommended apnea-hypopnea index definition as a marker of blood pressure risk and indicate that measurement of limb movements with arousals is also independently associated with diastolic blood pressure.

Developing Mathematical Models of Neurobehavioral Performance for the "Real World"

Work-related operations requiring extended wake durations, night, or rotating shifts negatively affect worker neurobehavioral performance and health. These types of work schedules are required in many industries, including the military, transportation, and health care. These industries are increasingly using or considering the use of mathematical models of neurobehavioral performance as a means to predict the neurobehavioral deficits due to these operational demands, to develop interventions that decrease these deficits, and to provide additional information to augment existing decision-making processes. Recent advances in mathematical modeling have allowed its application to real-world problems. Developing application-specific expertise is necessary to successfully apply mathematical models, in part because development of new algorithms and methods linking the models to the applications may be required. During a symposium, “Modeling Human Neurobehavioral Performance II: Towards Operational Readiness,” at the 2006 SIAM-SMB Conference on the Life Sciences, examples of the process of applying mathematical models, including model construction, model validation, or developing model-based interventions, were presented. The specific applications considered included refining a mathematical model of sleep/wake patterns of airline flight crew, validating a mathematical model using railroad operations data, and adapting a mathematical model to develop appropriate countermeasure recommendations based on known constraints. As mathematical models and their associated analytical methods continue to transition into operational settings, such additional development will be required. However, major progress has been made in using mathematical model outputs to inform those individuals making schedule decisions for their workers.

Comparison of Amplitude Recovery Dynamics of Two Limit Cycle Oscillator Models of the Human Circadian Pacemaker

At an organism level, the mammalian circadian pacemaker is a two‐dimensional system. For these two dimensions, phase (relative timing) and amplitude of the circadian pacemaker are commonly used. Both the phase and the amplitude (A) of the human circadian pacemaker can be observed within multiple physiological measures—including plasma cortisol, plasma melatonin, and core body temperature (CBT)—all of which are also used as markers of the circadian system. Although most previous work has concentrated on changes in phase of the circadian system, critically timed light exposure can significantly reduce the amplitude of the pacemaker. The rate at which the amplitude recovers to its equilibrium level after reduction can have physiological significance. Two mathematical models that describe the phase and amplitude dynamics of the pacemaker have been reported. These models are essentially equivalent in predictions of phase and in predictions of amplitude recovery for small changes from an equilibrium value (A=1), but are markedly different in the prediction of recovery rates when A<0.6. To determine which dynamic model best describes the amplitude recovery observed in experimental data; both models were fit to CBT data using a maximum likelihood procedure and compared using Akaikes Information Criterion (AIC). For all subjects, the model with the lower recovery rate provided a better fit to data in terms of AIC, supporting evidence that the amplitude recovery of the endogenous pacemaker is slow at low amplitudes. Experiments derived from model predictions are proposed to test the influence of low amplitude recovery on the physiological and neurobehavioral functions.

Development and Validation of Computational Models for Mammalian Circadian Oscillators

Circadian rhythms are endogenous rhythms with a cycle length of approximately 24 h. Rhythmic production of specific proteins within pacemaker structures is the basis for these physiological and behavioral rhythms. Prior work on mathematical modeling of molecular circadian oscillators has focused on the fruit fly, Drosophila melanogaster. Recently, great advances have been made in our understanding of the molecular basis of circadian rhythms in mammals. Mathematical models of the mammalian circadian oscillator are needed to piece together diverse data, predict experimental results, and help us understand the clock as a whole. Our objectives are to develop mathematical models of the mammalian circadian oscillator, generate and test predictions from these models, gather information on the parameters needed for model development, integrate the molecular model with an existing model of the influence of light and rhythmicity on human performance, and make models available in BioSpice so that they can be easily used by the general community. Two new mammalian models have been developed, and experimental data are summarized. These studies have the potential to lead to new strategies for resetting the circadian clock. Manipulations of the circadian clock can be used to optimize performance by promoting alertness and physiological synchronization.

Biological Time Series Analysis Using a Context Free Language: Applicability to Pulsatile Hormone Data

We present a novel approach for analyzing biological time-series data using a context-free language (CFL) representation that allows the extraction and quantification of important features from the time-series. This representation results in Hierarchically AdaPtive (HAP) analysis, a suite of multiple complementary techniques that enable rapid analysis of data and does not require the user to set parameters. HAP analysis generates hierarchically organized parameter distributions that allow multi-scale components of the time-series to be quantified and includes a data analysis pipeline that applies recursive analyses to generate hierarchically organized results that extend traditional outcome measures such as pharmacokinetics and inter-pulse interval. Pulsicons, a novel text-based time-series representation also derived from the CFL approach, are introduced as an objective qualitative comparison nomenclature. We apply HAP to the analysis of 24 hours of frequently sampled pulsatile cortisol hormone data, which has known analysis challenges, from 14 healthy women. HAP analysis generated results in seconds and produced dozens of figures for each participant. The results quantify the observed qualitative features of cortisol data as a series of pulse clusters, each consisting of one or more embedded pulses, and identify two ultradian phenotypes in this dataset. HAP analysis is designed to be robust to individual differences and to missing data and may be applied to other pulsatile hormones. Future work can extend HAP analysis to other time-series data types, including oscillatory and other periodic physiological signals.

Associations between quantitative sleep EEG and subsequent cognitive decline in older women

The pathophysiological processes of Alzheimers dementia predate its clinical manifestation. Sleep disturbances can accelerate the aging process and are common features of dementia. This study examined whether quantitative sleep electroencephalogram changes predate the clinical development of mild cognitive impairment and/or incident dementia. We collected data from a nested case‐control sample of women (mean age 83 years) from the Sleep and Cognition Study, an ancillary study to the longitudinal Study of Osteoporotic Fractures, who were characterized as cognitively normal at the time of a baseline polysomnography study (Study of Osteoporotic Fractures visit 8) based on a Mini‐Mental Status Exam (MMSE) score >24. Cases (n = 85) were women who developed new mild cognitive impairment or dementia by objective cognitive testing 5 years after polysomnography. Controls were women with no mild cognitive impairment/dementia (n = 85) at baseline or at follow‐up. Differences in electroencephalogram absolute and relative power density were observed between the two groups. Specifically, higher electroencephalogram power values were found in the dementia/mild cognitive impairment group, for the alpha (p = .01) and theta bands (p = .04) in non‐rapid eye movement sleep, as well as alpha (p = .04) and sigma (p = .04) bands in rapid eye movement sleep. In contrast, there were no group differences in traditional polysomnography measures of sleep architecture and sleep stage distribution, as well as sleep apnea and periodic limb movement indices. Our results provide evidence for quantitative electroencephalogram changes, which precede the clinical onset of cognitive decline and the diagnosis of dementia in elderly women, and support the application of quantitative sleep electroencephalogram analysis as a promising biomarker for imminent cognitive decline.

RREV: Reconfigurable Rendering Engine for visualization of clinically annotated polysomnograms

In sleep medicine, clinical studies often use their own data dictionaries for capturing clinical sleep events using proprietary signal analysis software [1][2]. Visualization of polysomnograms and their associated events from multiple distinct studies, such as for the National Sleep Research Resource (NSRR)[3], is an unresolved issue. Currently, there is no known visualization software for the European Data Format (EDF) that can be dynamically configured to support rendering of sleep events for multiple vendor formats. To address this challenge, domain ontology has been developed as a part of NSRR to model all sleep medicine terms and concepts to provide a common schema for addressing the structural and semantic heterogeneity of multiple vendor formats [4]. A Reconfigurable Rendering Engine using Abstract Factory pattern [5] and domain ontology provides a standard interface for accessing ontology-enabled clinical events for the visualization of electrophysiological signals. About 11,078 polysomnograms (8,444 SHHS, 860 CHAT, 591 HeartBEAT, 730 CFS, 453 SOF) [12] in EDF have been processed resulting in 1.1TB of web-accessible and reusable PSGs with NSRR standardized event annotations.