Extraction of features from sleep EEG for Bayesian assessment of brain development.
Ontology highlight
ABSTRACT:
SUBMITTER: Schetinin V
PROVIDER: S-EPMC5360314 | biostudies-literature | 2017
REPOSITORIES: biostudies-literature
Publications
Extraction of features from sleep EEG for Bayesian assessment of brain development.
Schetinin Vitaly V Jakaite Livija L
PloS one 20170321 3
Brain development can be evaluated by experts analysing age-related patterns in sleep electroencephalograms (EEG). Natural variations in the patterns, noise, and artefacts affect the evaluation accuracy as well as experts' agreement. The knowledge of predictive posterior distribution allows experts to estimate confidence intervals within which decisions are distributed. Bayesian approach to probabilistic inference has provided accurate estimates of intervals of interest. In this paper we propose ...[more]
Similar Datasets
Project description:Sleep has beneficial effects on brain function and learning, which are reflected in plastic changes in the cortex. Early childhood is a time of rapid maturation in fundamental skills-e.g., language, cognitive control, working memory-that are predictive of future functioning. Little is currently known about the interactions between sleep and brain maturation during this developmental period. We propose coherent electroencephalogram (EEG) activity during sleep may provide unique insight into maturational processes of functional brain connectivity. Longitudinal sleep EEG assessments were performed in eight healthy subjects at ages 2, 3 and 5 years. Sleep EEG coherence increased across development in a region- and frequency-specific manner. Moreover, although connectivity primarily decreased intra-hemispherically across a night of sleep, an inter-hemispheric overnight increase occurred in the frequency range of slow waves (0.8-2 Hz), theta (4.8-7.8 Hz) and sleep spindles (10-14 Hz), with connectivity changes of up to 20% across a night of sleep. These findings indicate sleep EEG coherence reflects processes of brain maturation-i.e., programmed unfolding of neuronal networks-and moreover, sleep-related alterations of brain connectivity during the sensitive maturational window of early childhood.
| S-EPMC3925344 | biostudies-literature
Project description:Motor imagery is a common control strategy in EEG-based brain-computer interfaces (BCIs). However, voluntary control of sensorimotor (SMR) rhythms by imagining a movement can be skilful and unintuitive and usually requires a varying amount of user training. To boost the training process, a whole class of BCI systems have been proposed, providing feedback as early as possible while continuously adapting the underlying classifier model. The present work describes a cue-paced, EEG-based BCI system using motor imagery that falls within the category of the previously mentioned ones. Specifically, our adaptive strategy includes a simple scheme based on a common spatial pattern (CSP) method and support vector machine (SVM) classification. The system's efficacy was proved by online testing on 10 healthy participants. In addition, we suggest some features we implemented to improve a system's "flexibility" and "customizability," namely, (i) a flexible training session, (ii) an unbalancing in the training conditions, and (iii) the use of adaptive thresholds when giving feedback.
| S-EPMC5011245 | biostudies-literature
Project description:IntroductionSleep disturbances are common in Alzheimer's disease (AD), with estimates of prevalence as high as 65%. Recent work suggests that specific sleep stages, such as slow-wave sleep (SWS) and rapid eye movement (REM), may directly impact AD pathophysiology. A major limitation to sleep staging is the requirement for clinical polysomnography (PSG), which is often not well tolerated in patients with dementia. We have recently developed a deep learning model to reliably analyze lower quality electroencephalogram (EEG) data obtained from a simple, two-lead EEG headband. Here we assessed whether this methodology would allow for home EEG sleep staging in patients with mild-moderate AD.MethodsA total of 26 mild-moderate AD patients and 24 age-matched, healthy control participants underwent home EEG sleep recordings as well as actigraphy and subjective sleep measures through the Pittsburgh Sleep Quality Index (PSQI). Each participant wore the EEG headband for up to three nights. Sleep was staged using a deep learning model previously developed by our group, and sleep stages were correlated with actigraphy measures as well as PSQI scores.ResultsWe show that home EEG with a headband is feasible and well tolerated in patients with AD. Patients with mild-moderate AD were found to spend less time in SWS compared to healthy control participants. Other sleep stages were not different between the two groups. Actigraphy or the PSQI were not found to predict home EEG sleep stages.DiscussionOur data show that home EEG is well tolerated, and can ascertain reduced SWS in patients with mild-moderate AD. Similar findings have previously been reported, but using clinical PSG not suitable for the home environment. Home EEG will be particularly useful in future clinical trials assessing potential interventions that may target specific sleep stages to alter the pathogenesis of AD.HighlightsHome electroencephalogram (EEG) sleep assessments are important for measuring sleep in patients with dementia because polysomnography is a limited resource not well tolerated in this patient population.Simplified at-home EEG for sleep assessment is feasible in patients with mild-moderate Alzheimer's disease (AD).Patients with mild-moderate AD exhibit less time spent in slow-wave sleep in the home environment, compared to healthy control participants.Compared to healthy control participants, patients with mild-moderate AD spend more time in bed, with decreased sleep efficiency, and more awakenings as measured by actigraphy, but these measures do not correlate with EEG sleep stages.
| S-EPMC9381912 | biostudies-literature
Project description:Background: Previously, we identified sleep-electroencephalography (EEG) spectral power and synchrony features that differed significantly at a population-average level between subjects with and without posttraumatic stress disorder (PTSD). Here, we aimed to examine the extent to which a combination of such features could objectively identify individual subjects with PTSD. Methods: We analyzed EEG data recorded from 78 combat-exposed Veteran men with (n = 31) and without (n = 47) PTSD during two consecutive nights of sleep. To obviate the need for manual assessment of sleep staging and facilitate extraction of features from the EEG data, for each subject, we computed 780 stage-independent, whole-night features from the 10 most commonly used EEG channels. We performed feature selection and trained a logistic regression model using a training set consisting of the first 47 consecutive subjects (18 with PTSD) of the study. Then, we evaluated the model on a testing set consisting of the remaining 31 subjects (13 with PTSD). Results: Feature selection yielded three uncorrelated features that were consistent across the two consecutive nights and discriminative of PTSD. One feature was from the spectral power in the delta band (2-4 Hz) and the other two were from phase synchronies in the alpha (10-12 Hz) and gamma (32-40 Hz) bands. When we combined these features into a logistic regression model to predict the subjects in the testing set, the trained model yielded areas under the receiver operating characteristic curve of at least 0.80. Importantly, the model yielded a testing-set sensitivity of 0.85 and a positive predictive value (PPV) of 0.31. Conclusions: We identified robust stage-independent, whole-night features from EEG signals and combined them into a logistic regression model to discriminate subjects with and without PTSD. On the testing set, the model yielded a high sensitivity and a PPV that was twice the prevalence rate of PTSD in the U.S. Veteran population. We conclude that, using EEG signals collected during sleep, such a model can potentially serve as a means to objectively identify U.S. Veteran men with PTSD.
| S-EPMC7673410 | biostudies-literature
Project description:Brain extraction or skull stripping of magnetic resonance images (MRI) is an essential step in neuroimaging studies, the accuracy of which can severely affect subsequent image processing procedures. Current automatic brain extraction methods demonstrate good results on human brains, but are often far from satisfactory on nonhuman primates, which are a necessary part of neuroscience research. To overcome the challenges of brain extraction in nonhuman primates, we propose a fully-automated brain extraction pipeline combining deep Bayesian convolutional neural network (CNN) and fully connected three-dimensional (3D) conditional random field (CRF). The deep Bayesian CNN, Bayesian SegNet, is used as the core segmentation engine. As a probabilistic network, it is not only able to perform accurate high-resolution pixel-wise brain segmentation, but also capable of measuring the model uncertainty by Monte Carlo sampling with dropout in the testing stage. Then, fully connected 3D CRF is used to refine the probability result from Bayesian SegNet in the whole 3D context of the brain volume. The proposed method was evaluated with a manually brain-extracted dataset comprising T1w images of 100 nonhuman primates. Our method outperforms six popular publicly available brain extraction packages and three well-established deep learning based methods with a mean Dice coefficient of 0.985 and a mean average symmetric surface distance of 0.220?mm. A better performance against all the compared methods was verified by statistical tests (all p-values?<?10-4, two-sided, Bonferroni corrected). The maximum uncertainty of the model on nonhuman primate brain extraction has a mean value of 0.116 across all the 100 subjects. The behavior of the uncertainty was also studied, which shows the uncertainty increases as the training set size decreases, the number of inconsistent labels in the training set increases, or the inconsistency between the training set and the testing set increases.
| S-EPMC6095475 | biostudies-literature
Project description:EEG source localization is an essential tool to reveal the cortical sources underlying brain oscillatory activity. We applied LORETA, a technique of EEG source localization, to identify the principal brain areas involved in the process of falling asleep (sleep onset, SO). We localized the contributing brain areas of activity in the classical frequency bands and tracked their temporal evolution (in 2-min intervals from 2?min prior to SO up to 10?min after SO) during a baseline night and subsequent recovery sleep after total sleep deprivation of 40?h. Delta activity (0.5-5?Hz) gradually increased both in baseline and recovery sleep, starting in frontal areas and finally involving the entire cortex. This increase was steeper in the recovery condition. The evolution of sigma activity (12-16?Hz) resembled an inverted U-shape in both conditions and the activity was most salient in the parietal cortex. In recovery, sigma activity reached its maximum faster than in baseline, but attained lower levels. Theta activity (5-8?Hz) increased with time in large parts of the occipital lobe (baseline and recovery) and in recovery involved additionally frontal areas. Changes in alpha activity (8-12?Hz) at sleep onset involved large areas of the cortex, whereas activity in the beta range (16-24?Hz) was restricted to small cortical areas. The dynamics in recovery could be considered as a "fast-forward version" of the one in baseline. Our results confirm that the process of falling asleep is neither spatially nor temporally a uniform process and that different brain areas might be falling asleep at a different speed potentially reflecting use dependent aspects of sleep regulation.
| S-EPMC6586601 | biostudies-literature
Project description:Aim:This study investigated the neural mechanisms of brain-wave music on sleep quality. Background:Sleep disorders are a common health problem in our society and may result in fatigue, depression, and problems in daytime functioning. Previous studies have shown that brain-wave music generated from electroencephalography (EEG) signals could emotionally affect our nervous system and have positive effects on sleep. However, the neural mechanisms of brain-wave music on the quality of sleep need to be clarified. Methods:A total of 33 young participants were recruited and randomly divided into three groups. The participants listened to rapid eye movement (REM) brain-wave music (Group 1: 13 subjects), slow-wave sleep (SWS) brain-wave music (Group 2: 11 subjects), or white noise (WN) (Control Group: 9 subjects) for 20 min before bedtime for 6 days. EEG and other physiological signals were recorded by polysomnography. Results:We found that the sleep efficiency increased in the SWS group but decreased in REM and WN groups. The sleep efficiency in the SWS group was ameliorated [t(10) = -1.943, p = 0.076]. In the EEG power spectral density analysis, the delta power spectral density in the REM group and in the control group increased, while that in the SWS group decreased [F(2,31) = 7.909, p = 0.005]. In the network analysis, the functional connectivity (FC), assessed with Pearson correlation coefficients, showed that the connectivity strength decreased [t(10) = 1.969, p = 0.073] between the left frontal lobe (F3) and left parietal lobe (C3) in the SWS group. In addition, there was a negative correlation between the FC of the left frontal lobe and the left parietal lobe and sleep latency in the SWS group (r = -0.527, p = 0.064). Conclusion:Slow-wave sleep brain-wave music may have a positive effect on sleep quality, while REM brain-wave music or WN may not have a positive effect. Furthermore, better sleep quality might be caused by a decrease in the power spectral density of the delta band of EEG and an increase in the FC between the left frontal lobe and the left parietal lobe. SWS brain-wave music could be a safe and inexpensive method for clinical use if confirmed by more data.
| S-EPMC7026372 | biostudies-literature
Project description:Epidemiological research reveals that insufficient sleep in children has negative cognitive and emotional consequences; however, the physiological underpinnings of these observations remain understudied. We tested the hypothesis that the topographical distribution of deep sleep slow wave activity during the childhood predicts brain white matter microstructure (myelin) 3.5 y later. Healthy children underwent sleep high-density EEG at baseline (n = 13; ages 2.4-8.0 y) and follow-up (n = 14; ages 5.5-12.2 y). At follow-up, myelin (myelin water fraction) and cortical morphology were also quantified. Our investigation revealed 3 main findings. (1) The Frontal/Occipital (F/O)-ratio at baseline strongly predicted whole brain myelin at follow-up. (2) At follow-up, the F/O-ratio was only minimally (negatively) linked to brain myelin. (3) Cortical morphology was not related to the F/O-ratio, neither at baseline nor at follow-up. Our results support the hypothesis that during child development EEG markers during sleep longitudinally predict brain myelin content. Data extend previous findings reporting a link between EEG markers of sleep need and cortical morphology, by supporting the hypothesis that sleep is a necessary component to underlying processes of brain, and specifically myelin, maturation. In line with the overarching theory that sleep contributes to neurodevelopmental processes, it remains to be investigated whether chronic sleep loss negatively affects white matter myelin microstructure growth during sensitive periods of development.
| S-EPMC6688908 | biostudies-literature
Project description:Understanding how the human brain processes auditory input remains a challenge. Traditionally, a distinction between lower- and higher-level sound features is made, but their definition depends on a specific theoretical framework and might not match the neural representation of sound. Here, we postulate that constructing a data-driven neural model of auditory perception, with a minimum of theoretical assumptions about the relevant sound features, could provide an alternative approach and possibly a better match to the neural responses. We collected electrocorticography recordings from six patients who watched a long-duration feature film. The raw movie soundtrack was used to train an artificial neural network model to predict the associated neural responses. The model achieved high prediction accuracy and generalized well to a second dataset, where new participants watched a different film. The extracted bottom-up features captured acoustic properties that were specific to the type of sound and were associated with various response latency profiles and distinct cortical distributions. Specifically, several features encoded speech-related acoustic properties with some features exhibiting shorter latency profiles (associated with responses in posterior perisylvian cortex) and others exhibiting longer latency profiles (associated with responses in anterior perisylvian cortex). Our results support and extend the current view on speech perception by demonstrating the presence of temporal hierarchies in the perisylvian cortex and involvement of cortical sites outside of this region during audiovisual speech perception.
| S-EPMC7363106 | biostudies-literature
Project description:Late-stage neuropathological hallmarks of Alzheimer's disease (AD) are ?-amyloid (?A) and hyperphosphorylated tau peptides, aggregated into plaques and tangles, respectively. Corresponding phenotypes have been mimicked in existing transgenic mice, however, the translational value of aggressive over-expression has recently been questioned. As controlled gene expression may offer animal models with better predictive validity, we set out to design a transgenic mouse model that circumvents complications arising from pronuclear injection and massive over-expression, by targeted insertion of human mutated amyloid and tau transgenes, under the forebrain- and neurone-specific CaMKII? promoter, termed PLB1(Double). Crossing with an existing presenilin 1 line resulted in PLB1(Triple) mice. PLB1(Triple) mice presented with stable gene expression and age-related pathology of intra-neuronal amyloid and hyperphosphorylated tau in hippocampus and cortex from 6 months onwards. At this early stage, pre-clinical (18)FDG PET/CT imaging revealed cortical hypometabolism with increased metabolic activity in basal forebrain and ventral midbrain. Quantitative EEG analyses yielded heightened delta power during wakefulness and REM sleep, and time in wakefulness was already reliably enhanced at 6 months of age. These anomalies were paralleled by impairments in long-term and short-term hippocampal plasticity and preceded cognitive deficits in recognition memory, spatial learning, and sleep fragmentation all emerging at ?12 months. These data suggest that prodromal AD phenotypes can be successfully modelled in transgenic mice devoid of fibrillary plaque or tangle development. PLB1(Triple) mice progress from a mild (MCI-like) state to a more comprehensive AD-relevant phenotype, which are accessible using translational tools such as wireless EEG and microPET/CT.
| S-EPMC3214038 | biostudies-literature