Auto-annotating sleep stages based on polysomnographic data.
Ontology highlight
ABSTRACT:
SUBMITTER: Zhang H
PROVIDER: S-EPMC8767308 | biostudies-literature | 2022 Jan
REPOSITORIES: biostudies-literature
Publications
Auto-annotating sleep stages based on polysomnographic data.
Zhang Hanrui H Wang Xueqing X Li Hongyang H Mehendale Soham S Guan Yuanfang Y
Patterns (New York, N.Y.) 20211028 1
Sleep disorders affect the quality of life, and the clinical diagnosis of sleep disorders is a time-consuming and tedious process requiring recording and annotating polysomnographic records. In this work, we developed an auto-annotation algorithm based on polysomnographic records and a deep learning architecture that predicts sleep stages at the millisecond level. The model improves the efficiency of the polysomnographic record annotation process by automatically annotating each record within 3. ...[more]
Similar Datasets
Project description:Obstructive sleep apnea (OSA) is the most common sleep disorder. Sleep bruxism (SB) is a masticatory muscle activity during sleep that commonly co-occurs with OSA. The presented study aimed to assess this relationship and to identify factors affecting this co-occurrence. Adult patients (n = 110) were evaluated for OSA and SB in a sleep laboratory using polysomnography. The episodes of bruxism and respiratory events were scored according to the standards of the American Academy of Sleep Medicine. The prevalence of OSA and SB was found to be 86.37% and 50%, respectively. The bruxism episode index (BEI) was increased in the group with mild and moderate OSA (apnea-hypopnea index (AHI) <30) compared to that in the group with severe OSA (AHI ≥ 30) (5.50 ± 4.58 vs. 1.62 ± 1.28, p < 0.05). A positive correlation between AHI and BEI was observed in the group with AHI < 30. Regression analysis indicated that higher AHI, male gender, and diabetes were independent predictors for the increased BEI in group with AHI < 30. The relationship between OSA and SB depends on the degree of severity of OSA. OSA is correlated with SB in mild and moderate cases of OSA in the group of patients with increased risk of OSA.
| S-EPMC6832407 | biostudies-literature
Project description:Rapid eye movement (REM) sleep behavior disorder is a condition characterized by dream enactment. This condition may accompany neurodegenerative disorders. However, only a few reports from India are available, that too, without any polysomnographic evidence. We are reporting a case of REM sleep behavior disorder with polysomnographic evidence.
| S-EPMC3808072 | biostudies-literature
Project description:In remote areas, visiting a laboratory for sleep testing is inconvenient. We, therefore, developed a Mobile Sleep Lab in a bus powered by fuel cells with two sleep measurement chambers. As the environment in the bus could affect sleep, we examined whether sleep testing in the Mobile Sleep Lab was as feasible as in a conventional sleep laboratory (Human Sleep Lab). We tested 15 healthy adults for four nights using polysomnography (the first two nights at the Human Sleep Lab or Mobile Sleep Lab with a switch to the other facility for the next two nights). Sleep variables of the four measurements were used to assess the discrepancy of different places or different nights. No significant differences were found between the laboratories other than the percentage of total sleep time in stage N3. Next, we analyzed the intraclass correlation coefficient to evaluate the test-retest reliability. The intraclass correlation coefficient between these two measurements: the Human Sleep Lab and Mobile Sleep Lab showed similar reliability for the same sleep variables. The intraclass correlation coefficient revealed that several sleep indexes, such as total sleep time, sleep efficiency, wake after sleep onset, percentage of stage N1, and stage R latency, showed poor reliabilities (<0.5) based on Koo and Li's criteria. In contrast, the percentage of stage N3 showed moderate (0.5-0.75) or good (0.75-0.9) reliabilities. As almost all sleep variables showed no difference and same level of test-retest reliability between the Mobile Sleep Lab and Human Sleep Lab, the Mobile Sleep Lab might be suitable for conducting polysomnography as a conventional sleep laboratory. The reduction in N3 in the Mobile Sleep Lab should be scrutinized in the larger sample, including sleep disorders. Practical application of the Mobile Sleep Lab can transform sleep medicine in remote areas.
| S-EPMC11706495 | biostudies-literature
Project description:BackgroundSleep-wakefulness cycles are an essential diagnostic criterion for disorders of consciousness (DOC), differentiating prolonged DOC from coma. Specific sleep features, like the presence of sleep spindles, are an important marker for the prognosis of recovery from DOC. Based on increasing evidence for a link between sleep and neuronal plasticity, understanding sleep in DOC might facilitate the development of novel methods for rehabilitation. Yet, well-controlled studies of sleep in DOC are lacking. Here, we aimed to quantify, on a reliable evaluation basis, the distribution of behavioral and neurophysiological sleep patterns in DOC over a 24-h period while controlling for environmental factors (by recruiting a group of conscious tetraplegic patients who resided in the same hospital).MethodsWe evaluated the distribution of sleep and wakefulness by means of polysomnography (EEG, EOG, EMG) and video recordings in 32 DOC patients (16 unresponsive wakefulness syndrome [UWS], 16 minimally conscious state [MCS]), and 10 clinical control patients with severe tetraplegia. Three independent raters scored the patients' polysomnographic recordings.ResultsAll but one patient (UWS) showed behavioral and electrophysiological signs of sleep. Control and MCS patients spent significantly more time in sleep during the night than during daytime, a pattern that was not evident in UWS. DOC patients (particularly UWS) exhibited less REM sleep than control patients. Forty-four percent of UWS patients and 12% of MCS patients did not have any REM sleep, while all control patients (100%) showed signs of all sleep stages and sleep spindles. Furthermore, no sleep spindles were found in 62% of UWS patients and 21% of MCS patients. In the remaining DOC patients who had spindles, their number and amplitude were significantly lower than in controls.ConclusionsThe distribution of sleep signs in DOC over 24 h differs significantly from the normal sleep-wakefulness pattern. These abnormalities of sleep in DOC are independent of external factors such as severe immobility and hospital environment.
| S-EPMC7678091 | biostudies-literature
Project description:Study objectivesMany sleep disorders are characterized by increased motor activity during sleep. In contrast, studies on motor activity during physiological sleep are largely lacking. We quantitatively investigated a large range of motor phenomena during polysomnography in physiological sleep.DesignProspective polysomnographic investigation.SettingAcademic referral sleep laboratory.ParticipantsOne hundred healthy sleepers age 19-77 y were strictly selected from a representative population sample by a two-step screening procedure.InterventionsN/A.Measurements and resultsPolysomnography according to American Academy of Sleep Medicine (AASM) standards was performed, and quantitative normative values were established for periodic limb movements in sleep (PLMS), high frequency leg movements (HFLM), fragmentary myoclonus (FM), neck myoclonus (NM), and rapid eye movement (REM)-related electromyographic (EMG) activity. Thirty-six subjects had a PLMS index > 5/h, 18 had a PLMS index > 15/h (90th percentile: 24.8/h). Thirty-three subjects had HFLM (90th percentile: four sequences/night). All subjects had FM (90th percentile 143.7/h sleep). Nine subjects fulfilled AASM criteria for excessive FM. Thirty-five subjects had NM (90th percentile: 8.8/h REM sleep). For REM sleep, different EMG activity measures for the mentalis and flexor digitorum superficialis muscles were calculated: the 90th percentile for phasic mentalis EMG activity for 30-sec epochs according to AASM recommendation was 15.6%, and for tonic mentalis EMG activity 2.6%. Twenty-five subjects exceeded the recently proposed phasic mentalis cutoff of 11%. None of the subjects exceeded the tonic mentalis cutoff of 9.6%.ConclusionQuantification of motor phenomena is a basic prerequisite to develop normative values, and is a first step toward a more precise description of the various motor phenomena present during sleep. Because rates of motor events were unexpectedly high even in physiological sleep, the future use of normative values for both research and clinical routine is essential.
| S-EPMC3972421 | biostudies-literature
Project description:As the number of freight trains on railway networks increases, so does the potential for vibration exposure in dwellings nearby to freight railway lines. Nocturnal trains in particular are of particular importance since night-time exposure may interfere with sleep. The present work investigates the impact of vibration and noise from night-time freight trains on human sleep. In an experimental polysomnographic laboratory study, 24 young healthy volunteers with normal hearing were exposed to simulated freight pass-bys with vibration amplitudes of 0.7 and 1.4 mm/s either 20 or 36 times during the night. Stronger vibrations were associated with higher probabilities of event-related arousals and awakenings (p < 0.001), and sleep stage changes (p < 0.05). Sleep macrostructure was most affected in high vibration nights with 36 events, with increased wakefulness (p < 0.05), reduced continual slow wave sleep (p < 0.05), earlier awakenings (p < 0.05) and an overall increase in sleep stage changes (p < 0.05). Subjects reported sleep disturbance due to vibration (F(4,92) = 25.9, p < 0.001) and noise (F(4,92) = 25.9, p < 0.001), with the number of trains having an effect only for the 0.7 mm/s condition (p < 0.05). The findings show that combined vibration and noise from railway freight affects the natural rhythm of sleep, but extrapolation of significance for health outcomes should be approached with caution.
| S-EPMC4835753 | biostudies-literature
Project description:BackgroundSleep bruxism (SB) is a common behavior that can result in various clinical consequences on human health. Risk factors for SB include among others emotional stress, anxiety, tobacco smoking, and excessive alcohol consumption. Coffee and black tea are among the most commonly consumed beverages worldwide. This study explores the influence of coffee and black tea consumption on bruxism intensity, as observed in polysomnographic examination.MethodsPolysomnographic examination with simultaneous camera recording was conducted in 106 adult subjects. The results were evaluated according to guidelines set out by the American Academy of Sleep Medicine (AASM). The study group was divided according to habitual stimulant usage, as declared by the participants in a self-reported questionnaire. Four groups were identified: coffee drinkers versus non-drinkers and black tea drinkers versus non-drinkers.ResultsThe bruxism episode index (BEI) was increased in coffee-drinkers as opposed to non-drinkers (4.59 ± 3.44 vs. 2.87 ± 1.50, p = 0.011). Sleep fragmentation, measured according to the arousal index, was comparable in coffee drinkers and non-drinkers. Electrolyte and lipid levels were similar in coffee drinkers and non-drinkers. Habitual black tea intake did not affect sleep architecture or bruxism intensity.ConclusionsThe study showed that habitual coffee consumption is a risk factor for the increased intensity of sleep bruxism. Neither coffee nor tea consumption is related to sleep fragmentation in habitual drinkers. Coffee and tea intake does not affect electrolyte and lipid concentrations. Caution should therefore be recommended in drinking coffee in people with sleep bruxism.
| S-EPMC10205497 | biostudies-literature
Project description:Study objectivesTo investigate whether a foundational transformer model using 8-hour, multichannel data from polysomnograms can outperform existing artificial intelligence (AI) methods for sleep stage classification.MethodsWe utilized the Sleep Heart Health Study (SHHS) visits 1 and 2 for training and validation and the Multi-Ethnic Study of Atherosclerosis (MESA) for testing of our model. We trained a self-supervised foundational transformer (called PFTSleep) that encodes 8-hour long sleep studies at 125 Hz with 7 signals including brain, movement, cardiac, oxygen, and respiratory channels. These encodings are used as input for training of an additional model to classify sleep stages, without adjusting the weights of the foundational transformer. We compared our results to existing AI methods that did not utilize 8-hour data or the full set of signals but did report evaluation metrics for the SHHS dataset.ResultsWe trained and validated a model with 8,444 sleep studies with 7 signals including brain, movement, cardiac, oxygen, and respiratory channels and tested on an additional 2,055 studies. In total, we trained and tested 587,944 hours of sleep study signal data. Area under the precision recall curve (AUPRC) scores were 0.82, 0.40, 0.53, 0.75, and 0.82 and area under the receiving operating characteristics curve (AUROC) scores were 0.99, 0.95, 0.96, 0.98, and 0.99 for wake, N1, N2, N3, and REM, respectively, on the SHHS validation set. For MESA, the AUPRC scores were 0.56, 0.16, 0.40, 0.45, and 0.65 and AUROC scores were 0.94, 0.77, 0.87, 0.91, and 0.96, respectively. Our model was compared to the longest context window state-of-the-art model and showed increases in macro evaluation scores, notably sensitivity (3.7% increase) and multi-class REM (3.39% increase) and wake (0.97% increase) F1 scores.ConclusionsUtilizing full night, multi-channel PSG data encodings derived from a foundational transformer improve sleep stage classification over existing methods.
| S-EPMC11326349 | biostudies-literature
Project description:Investigating sleep in mental disorders has the potential to reveal both disorder-specific and transdiagnostic psychophysiological mechanisms. This meta-analysis aimed at determining the polysomnographic (PSG) characteristics of several mental disorders. Relevant studies were searched through standard strategies. Controlled PSG studies evaluating sleep in affective, anxiety, eating, pervasive developmental, borderline and antisocial personality disorders, attention-deficit-hyperactivity disorder (ADHD), and schizophrenia were included. PSG variables of sleep continuity, depth, and architecture, as well as rapid-eye movement (REM) sleep were considered. Calculations were performed with the "Comprehensive Meta-Analysis" and "R" software. Using random effects modeling, for each disorder and each variable, a separate meta-analysis was conducted if at least 3 studies were available for calculation of effect sizes as standardized means (Hedges' g). Sources of variability, that is, sex, age, and mental disorders comorbidity, were evaluated in subgroup analyses. Sleep alterations were evidenced in all disorders, with the exception of ADHD and seasonal affective disorders. Sleep continuity problems were observed in most mental disorders. Sleep depth and REM pressure alterations were associated with affective, anxiety, autism and schizophrenia disorders. Comorbidity was associated with enhanced REM sleep pressure and more inhibition of sleep depth. No sleep parameter was exclusively altered in 1 condition; however, no 2 conditions shared the same PSG profile. Sleep continuity disturbances imply a transdiagnostic imbalance in the arousal system likely representing a basic dimension of mental health. Sleep depth and REM variables might play a key role in psychiatric comorbidity processes. Constellations of sleep alterations may define distinct disorders better than alterations in 1 single variable. (PsycINFO Database Record
| S-EPMC5110386 | biostudies-literature
Project description:BackgroundObstructive sleep apnoea (OSA) is a heterogeneous disorder, and improved understanding of physiologic phenotypes and their clinical implications is needed. We aimed to determine whether routine polysomnographic data can be used to identify OSA phenotypes (clusters) and to assess the associations between the phenotypes and cardiovascular outcomes.MethodsCross-sectional and longitudinal analyses of a multisite, observational US Veteran (n=1247) cohort were performed. Principal components-based clustering was used to identify polysomnographic features in OSA's four pathophysiological domains (sleep architecture disturbance, autonomic dysregulation, breathing disturbance and hypoxia). Using these features, OSA phenotypes were identified by cluster analysis (K-means). Cox survival analysis was used to evaluate longitudinal relationships between clusters and the combined outcome of incident transient ischaemic attack, stroke, acute coronary syndrome or death.ResultsSeven patient clusters were identified based on distinguishing polysomnographic features: 'mild', 'periodic limb movements of sleep (PLMS)', 'NREM and arousal', 'REM and hypoxia', 'hypopnoea and hypoxia', 'arousal and poor sleep' and 'combined severe'. In adjusted analyses, the risk (compared with 'mild') of the combined outcome (HR (95% CI)) was significantly increased for 'PLMS', (2.02 (1.32 to 3.08)), 'hypopnoea and hypoxia' (1.74 (1.02 to 2.99)) and 'combined severe' (1.69 (1.09 to 2.62)). Conventional apnoea-hypopnoea index (AHI) severity categories of moderate (15≤AHI<30) and severe (AHI ≥30), compared with mild/none category (AHI <15), were not associated with increased risk.ConclusionsAmong patients referred for OSA evaluation, routine polysomnographic data can identify physiological phenotypes that capture risk of adverse cardiovascular outcomes otherwise missed by conventional OSA severity classification.
| S-EPMC6693344 | biostudies-literature