Sleep spindle detection through amplitude–frequency normal modelling

The Effect of Targeted Memory Reactivation on Dogs' Visuospatial Memory

The role of sleep in memory consolidation is a widely discussed but still debated area of research. In light of the fact that memory consolidation during sleep is an evolutionary adaptive function, investigating the same phenomenon in nonhuman model species is highly relevant for its understanding. One such species, which has acquired human-analog sociocognitive skills through convergent evolution, is the domestic dog. Family dogs have surfaced as an outstanding animal model in sleep research, and their learning skills (in a social context) are subject to sleep-dependent memory consolidation. These results, however, are correlational, and the next challenge is to establish causality. In the present study, we aimed to adapt a TMR (targeted memory reactivation) paradigm in dogs and investigate its effect on sleep parameters. Dogs (N = 16) learned new commands associated with different locations and afterward took part in a sleep polysomnography recording when they were re-exposed to one of the previously learned commands. The results did not indicate a cueing benefit on choice performance. However, there was evidence for a decrease in choice latency after sleep, while the density (occurrence/minute) of fast sleep spindles was also notably higher during TMR recordings than adaptation recordings from the same animals and even compared with a larger reference sample from a previous work. Our study provides empirical evidence that TMR is feasible with family dogs, even during a daytime nap. Furthermore, the present study highlights several methodological and conceptual challenges for future research.

Sleep-spindles as a marker of attention and intelligence in dogs

The sleep spindle-generating thalamo-cortical circuitry supports attention capacity in awake humans and animals, but using sleep spindles to predict differences in attention has not been tried in either. Of the more commonly examined cognitive correlates of spindle occurrence and amplitude, post-sleep recall, and general intelligence, only post-sleep recall had been studied in dogs, rats and mice. Here, we examined a sample of companion dogs (N = 58) for whom polysomnographic recordings and several cognitive tests were performed on two occasions each, with a three-month break in-between. Five of the tests were used to extract a factor analogous to human g (general mental ability). A sixth test in the battery measured sustained attention. Both attention and g-factor scores were linked to higher slow spindle occurrence and absolute sigma power detected in polysomnographic recordings over the central electrode. These effects persisted across measurement occasions. Higher intrinsic spindle frequency was, in turn, linked to lower g-factor scores but displayed no relationship with attention scores. The overlap in localization and direction for the effects of slow spindle density (spindles/minute) and sigma power supports that they tap into the same underlying cognition-relevant aspects of spindling. Given earlier large sample and meta-analysis validations of sigma power as a reliable predictor of cognitive performance in humans, we thus conclude that the currently handled method for quantifying spindle density in dogs indeed measures cognition-relevant spindle activity by virtue of its agreement with the sigma power alternative.

The "Twin Peaks" method of automated Spike-Wave detection: A two-step, two-criteria Matlab application

BACKGROUND

There are many automated spike-wave discharge detectors, but the known weaknesses of otherwise good methods and the varying working conditions of different research groups (mainly the access to hardware and software) invite further exploration into alternative approaches.

NEW METHOD

The algorithm combines two criteria, one in the time-domain and one in the frequency-domain, exploiting morphological asymmetry and the presence of harmonics, respectively. The time-domain criterion is additionally adjusted by normal modelling between the first and second iterations.

RESULTS

We report specificity, sensitivity and accuracy values for 20 recordings from 17 mature, male WAG/Rij rats. In addition, performance was preliminary tested with different hormones, pharmacological injections and species (mice) in a smaller sample. Accuracy and specificity were consistently above 91 %. The number of automatically detected spike-wave discharges was strongly correlated with the numbers derived from visual inspection. Sensitivity varied more strongly than specificity, but high values were observed in both rats and mice.

COMPARISON WITH EXISTING METHODS

The algorithm avoids low-voltage movement artifacts, displays a lower false positive rate than many predecessors and appears to work across species, i.e. while designed initially with data from the WAG/Rij rat, the algorithm can pick up seizure activity in the mouse of considerably lower inter-spike frequency. Weaknesses of the proposed method include a lower sensitivity than several predecessors.

CONCLUSION

The algorithm excels in being a selective and flexible (based on e.g. its performance across rats and mice) spike-wave discharge detector. Future work could attempt to increase the sensitivity of this approach.

Spectral and temporal characterization of sleep spindles-methodological implications

Objective. Nested into slow oscillations (SOs) and modulated by their up-states, spindles are electrophysiological hallmarks of N2 sleep stage that present a complex hierarchical architecture. However, most studies have only described spindles in basic statistical terms, which were limited to the spindle itself without analyzing the characteristics of the pre-spindle moments in which the SOs are originated. The aim of this study was twofold: (a) to apply spectral and temporal measures to the pre-spindle and spindle periods, as well as analyze the correlation between them, and (b) to evaluate the potential of these spectral and temporal measures in future automatic detection algorithms.Approach. An automatic spindle detection algorithm was applied to the overnight electroencephalographic recordings of 26 subjects. Ten complementary features (five spectral and five temporal parameters) were computed in the pre-spindle and spindle periods after their segmentation. These features were computed independently in each period and in a time-resolved way (sliding window). After the statistical comparison of both periods, a correlation analysis was used to assess their interrelationships. Finally, a receiver operating-characteristic (ROC) analysis along with a bootstrap procedure was conducted to further evaluate the degree of separability between the pre-spindle and spindle periods.Main results. The results show important time-varying changes in spectral and temporal parameters. The features calculated in pre-spindle and spindle periods are strongly and significantly correlated, demonstrating the association between the pre-spindle characteristics and the subsequent spindle. The ROC analysis exposes that the typical feature used in automatic spindle detectors, i.e. the power in the sigma band, is outperformed by other features, such as the spectral entropy in this frequency range.Significance. The novel features applied here demonstrate their utility as predictors of spindles that could be incorporated into novel algorithms of automatic spindle detectors, in which the analysis of the pre-spindle period becomes relevant for improving their performance. From the clinical point of view, these features may serve as novel precision therapeutic targets to enhance spindle production with the aim of improving memory, cognition, and sleep quality in healthy and clinical populations. The results evidence the need for characterizing spindles in terms beyond power and the spindle period itself to more dynamic measures and the pre-spindle period. Physiologically, these findings suggest that spindles are more than simple oscillations, but nonstable oscillatory bursts embedded in the complex pre-spindle dynamics.

Shared and unique features of mammalian sleep spindles - insights from new and old animal models

Sleep spindles are phasic events observed in mammalian non-rapid eye movement sleep. They are relevant today in the study of memory consolidation, sleep quality, mental health and ageing. We argue that our advanced understanding of their mechanisms has not exhausted the utility and need for animal model work. This is both because some topics, like cognitive ageing, have not yet been addressed sufficiently in comparative efforts and because the evolutionary history of this oscillation is still poorly understood. Comparisons across species often are either limited to referencing the classical cat and rodent models, or are over-inclusive, uncritically including reports of sleep spindles in rarely studied animals. In this review, we discuss the emergence of new (dog and sheep) models for sleep spindles and compare the strengths and shortcomings of new and old models based on the three validation criteria for animal models - face, predictive, and construct validity. We conclude that an emphasis on cognitive ageing might dictate the future of comparative sleep spindle studies, a development that is already becoming visible in studies on dogs. Moreover, reconstructing the evolutionary history of sleep spindles will require more stringent criteria for their identification, across more species. In particular, a stronger emphasis on construct and predictive validity can help verify if spindle-like events in other species are actual sleep spindles. Work in accordance with such stricter validation suggests that sleep spindles display more universally shared features, like defining frequency, than previously thought.

Averaging sleep spindle occurrence in dogs predicts learning performance better than single measures

Although a positive link between sleep spindle occurrence and measures of post-sleep recall (learning success) is often reported for humans and replicated across species, the test–retest reliability of the effect is sometimes questioned. The largest to date study could not confirm the association, however methods for automatic spindle detection diverge in their estimates and vary between studies. Here we report that in dogs using the same detection method across different learning tasks is associated with observing a positive association between sleep spindle density (spindles/minute) and learning success. Our results suggest that reducing measurement error by averaging across measurements of density and learning can increase the visibility of this effect, implying that trait density (estimated through averaged occurrence) is a more reliable predictor of cognitive performance than estimates based on single measures.

Possible association between spindle frequency and reversal-learning in aged family dogs

In both humans and dogs sleep spindle occurrence between acquisition and recall of a specific memory correlate with learning performance. However, it is not known whether sleep spindle characteristics are also linked to performance beyond the span of a day, except in regard to general mental ability in humans. Such a relationship is likely, as both memory and spindle expression decline with age in both species (in dogs specifically the density and amplitude of slow spindles). We investigated if spindle amplitude, density (spindles/minute) and/or frequency (waves/second) correlate with performance on a short-term memory and a reversal-learning task in old dogs (> 7 years), when measurements of behavior and EEG were on average a month apart. Higher frequencies of fast (≥ 13 Hz) spindles on the frontal and central midline electrodes, and of slow spindles (≤ 13 Hz) on the central midline electrode were linked to worse performance on a reversal-learning task. The present findings suggest a role for spindle frequency as a biomarker of cognitive aging across species: Changes in spindle frequency are associated with dementia risk and onset in humans and declining learning performance in the dog.

Sleep Spindle Detection Using RUSBoost and Synchrosqueezed Wavelet Transform

Sleep spindles are important electroencephalographic (EEG) waveforms in sleep medicine; however, it is burdensome even for experts to detect spindles, so automatic spindle detection methodologies have been investigated. Conventional methods utilize waveforms template matching or machine learning for detecting spindles. In the former approach, it is necessary to tune thresholds for individual adaptation, while the latter approach has the problem of imbalanced data because the amount of sleep spindles is small compared with the entire EEG data. The present work proposes a sleep spindle detection method that combines wavelet synchrosqueezed transform (SST) and random under-sampling boosting (RUSBoost). SST is a time-frequency analysis method suitable for extracting features of spindle waveforms. RUSBoost is a framework for coping with the imbalanced data problem. The proposed SST-RUS can deal with the imbalanced data in spindle detection and does not require threshold tuning because RUSBoost uses majority voting of weak classifiers for discrimination. The performance of SST-RUS was validated using an open-access database called the Montreal archives of sleep studies cohort 1 (MASS-C1), which showed an F-measure of 0.70 with a sensitivity of 76.9% and a positive predictive value of 61.2%. The proposed method can reduce the burden of PSG scoring.

Sleep Spindles: Mechanisms and Functions

Sleep spindles are burstlike signals in the electroencephalogram (EEG) of the sleeping mammalian brain and electrical surface correlates of neuronal oscillations in thalamus. As one of the most inheritable sleep EEG signatures, sleep spindles probably reflect the strength and malleability of thalamocortical circuits that underlie individual cognitive profiles. We review the characteristics, organization, regulation, and origins of sleep spindles and their implication in non-rapid-eye-movement sleep (NREMS) and its functions, focusing on human and rodent. Spatially, sleep spindle-related neuronal activity appears on scales ranging from small thalamic circuits to functional cortical areas, and generates a cortical state favoring intracortical plasticity while limiting cortical output. Temporally, sleep spindles are discrete events, part of a continuous power band, and elements grouped on an infraslow time scale over which NREMS alternates between continuity and fragility. We synthesize diverse and seemingly unlinked functions of sleep spindles for sleep architecture, sensory processing, synaptic plasticity, memory formation, and cognitive abilities into a unifying sleep spindle concept, according to which sleep spindles 1) generate neural conditions of large-scale functional connectivity and plasticity that outlast their appearance as discrete EEG events, 2) appear preferentially in thalamic circuits engaged in learning and attention-based experience during wakefulness, and 3) enable a selective reactivation and routing of wake-instated neuronal traces between brain areas such as hippocampus and cortex. Their fine spatiotemporal organization reflects NREMS as a physiological state coordinated over brain and body and may indicate, if not anticipate and ultimately differentiate, pathologies in sleep and neurodevelopmental, -degenerative, and -psychiatric conditions.

Age-related differences and sexual dimorphism in canine sleep spindles

Non-REM bursts of activity in the sigma range (9-16 Hz) typical of sleep spindles predict learning in dogs, similar to humans and rats. Little is known, however, about the age-related changes in amplitude, density (spindles/minute) and frequency (waves/second) of canine spindles. We investigated a large sample (N = 155) of intact and neutered pet dogs of both sexes, varying in breed and age, searching for spindles in segments of non-REM sleep. We recorded EEG from both a frontal midline electrode (Fz) and a central midline electrode (Cz) in 55.5% of the dogs, in the remaining animals only the Fz electrode was active (bipolar derivation). A similar topography was observed for fast (≥13 Hz) spindle occurrence as in humans (fast spindle number, density on Cz > Fz). For fast spindles, density was higher in females, and increased with age. These effects were more pronounced among intact animals and on Fz. Slow spindle density declined and fast spindle frequency increased with age on Cz, while on Fz age-related amplitude decline was observed. The frequency of fast spindles on Fz and slow spindles on Cz was linked to both sex and neutering, suggesting modulation by sexual hormones. Intact females displayed higher frequencies than males and neutered females. Our findings support the argument that sigma bursts in the canine non-REM sleep are analogous to human sleep spindles, and suggest that slow and fast spindles display different trajectories related to age, of which an increase in frontal fast spindles is unique to dogs.

A deep learning approach for real-time detection of sleep spindles

OBJECTIVE

Sleep spindles have been implicated in memory consolidation and synaptic plasticity during NREM sleep. Detection accuracy and latency in automatic spindle detection are critical for real-time applications.

APPROACH

Here we propose a novel deep learning strategy (SpindleNet) to detect sleep spindles based on a single EEG channel. While the majority of spindle detection methods are used for off-line applications, our method is well suited for online applications.

MAIN RESULTS

Compared with other spindle detection methods, SpindleNet achieves superior detection accuracy and speed, as demonstrated in two publicly available expert-validated EEG sleep spindle datasets. Our real-time detection of spindle onset achieves detection latencies of 150-350 ms (~two-three spindle cycles) and retains excellent performance under low EEG sampling frequencies and low signal-to-noise ratios. SpindleNet has good generalization across different sleep datasets from various subject groups of different ages and species.

SIGNIFICANCE

SpindleNet is ultra-fast and scalable to multichannel EEG recordings, with an accuracy level comparable to human experts, making it appealing for long-term sleep monitoring and closed-loop neuroscience experiments.

Spindler: a framework for parametric analysis and detection of spindles in EEG with application to sleep spindles

OBJECTIVE

EEG spindles, narrow-band oscillatory signal bursts, are widely-studied biomarkers of subject state and neurological function. Most existing methods for spindle detection select algorithm parameters by optimizing agreement with expert labels. We propose a new framework for selecting algorithm parameters based on stability of spindle properties and elucidate the dependence of these properties on parameter selection for several algorithms.

APPROACH

To demonstrate this approach we developed a new algorithm (Spindler) that decomposes the signal using matching pursuit with Gabor atoms and computes the spindles for each point in a fine grid of parameter values. After computing characteristic surfaces as a function of parameters, Spindler selects algorithm parameters based on the stability of characteristic surface geometry.

MAIN RESULTS

Spindler performs well relative to several common supervised and unsupervised EEG sleep spindle detection methods. Spindler is available as an open-source MATLAB toolbox (https://github.com/VisLab/EEG-Spindles). In addition to Spindler, the toolbox provides implementations of several other spindle detection algorithms as well as standardized methods for matching ground truth to predictions and a framework for understanding algorithm parameter surfaces.

SIGNIFICANCE

This work demonstrates that parameter selection based on physical constraints rather than labelled data can provide effective, fully-automated, unsupervised spindle detection. This work also exposes the dangers of applying cross-validation without considering the dependence of spindle properties on parameters. Parameters selected to optimize one performance metric or matching method are not optimized for others. Furthermore, elucidation of the stability of predicted indicators with respect to algorithm parameter selection is critical to practical application of these algorithms.

Sleep spindle detection using deep learning: A validation study based on crowdsourcing

Sleep spindles are significant transient oscillations observed on the electroencephalogram (EEG) in stage 2 of non-rapid eye movement sleep. Deep belief network (DBN) gaining great successes in images and speech is still a novel method to develop sleep spindle detection system. In this paper, crowdsourcing replacing gold standard was applied to generate three different labeled samples and constructed three classes of datasets with a combination of these samples. An F1-score measure was estimated to compare the performance of DBN to other three classifiers on classifying these samples, with the DBN obtaining an result of 92.78%. Then a comparison of two feature extraction methods based on power spectrum density was made on same dataset using DBN. In addition, the DBN trained in dataset was applied to detect sleep spindle from raw EEG recordings and performed a comparable capacity to expert group consensus.

EEG Transients in the Sigma Range During non-REM Sleep Predict Learning in Dogs

Sleep spindles are phasic bursts of thalamo-cortical activity, visible in the cortex as transient oscillations in the sigma range (usually defined in humans as 12-14 or 9-16 Hz). They have been associated with sleep-dependent memory consolidation and sleep stability in humans and rodents. Occurrence, frequency, amplitude and duration of sleep spindles co-vary with age, sex and psychiatric conditions. Spindle analogue activity in dogs has been qualitatively described, but never quantified and related to function. In the present study we used an adjusted version of a detection method previously validated in children to test whether detections in the dogs show equivalent functional correlates as described in the human literature. We found that the density of EEG transients in the 9-16 Hz range during non-REM sleep relates to memory and is characterized by sexual dimorphism similarly as in humans. The number of transients/minute was larger in the learning condition and for female dogs, and correlated with the increase of performance during recall. It can be concluded that in dogs, automatic detections in the 9-16 Hz range, in particular the slow variant (<13 Hz), are functional analogues of human spindles.

Multichannel sleep spindle detection using sparse low-rank optimization

BACKGROUND

Automated single-channel spindle detectors, for human sleep EEG, are blind to the presence of spindles in other recorded channels unlike visual annotation by a human expert.

NEW METHOD

We propose a multichannel spindle detection method that aims to detect global and local spindle activity in human sleep EEG. Using a non-linear signal model, which assumes the input EEG to be the sum of a transient and an oscillatory component, we propose a multichannel transient separation algorithm. Consecutive overlapping blocks of the multichannel oscillatory component are assumed to be low-rank whereas the transient component is assumed to be piecewise constant with a zero baseline. The estimated oscillatory component is used in conjunction with a bandpass filter and the Teager operator for detecting sleep spindles.

RESULTS AND COMPARISON WITH OTHER METHODS

The proposed method is applied to two publicly available databases and compared with 7 existing single-channel automated detectors. F₁ scores for the proposed spindle detection method averaged 0.66 (0.02) and 0.62 (0.06) for the two databases, respectively. For an overnight 6 channel EEG signal, the proposed algorithm takes about 4min to detect sleep spindles simultaneously across all channels with a single setting of corresponding algorithmic parameters.

CONCLUSIONS

The proposed method attempts to mimic and utilize, for better spindle detection, a particular human expert behavior where the decision to mark a spindle event may be subconsciously influenced by the presence of a spindle in EEG channels other than the central channel visible on a digital screen.

Fast and Stable Signal Deconvolution via Compressible State-Space Models

OBJECTIVE

Common biological measurements are in the form of noisy convolutions of signals of interest with possibly unknown and transient blurring kernels. Examples include EEG and calcium imaging data. Thus, signal deconvolution of these measurements is crucial in understanding the underlying biological processes. The objective of this paper is to develop fast and stable solutions for signal deconvolution from noisy, blurred, and undersampled data, where the signals are in the form of discrete events distributed in time and space.

METHODS

We introduce compressible state-space models as a framework to model and estimate such discrete events. These state-space models admit abrupt changes in the states and have a convergent transition matrix, and are coupled with compressive linear measurements. We consider a dynamic compressive sensing optimization problem and develop a fast solution, using two nested expectation maximization algorithms, to jointly estimate the states as well as their transition matrices. Under suitable sparsity assumptions on the dynamics, we prove optimal stability guarantees for the recovery of the states and present a method for the identification of the underlying discrete events with precise confidence bounds.

RESULTS

We present simulation studies as well as application to calcium deconvolution and sleep spindle detection, which verify our theoretical results and show significant improvement over existing techniques.

CONCLUSION

Our results show that by explicitly modeling the dynamics of the underlying signals, it is possible to construct signal deconvolution solutions that are scalable, statistically robust, and achieve high temporal resolution.

SIGNIFICANCE

Our proposed methodology provides a framework for modeling and deconvolution of noisy, blurred, and undersampled measurements in a fast and stable fashion, with potential application to a wide range of biological data.

Sleep Spindles as an Electrographic Element: Description and Automatic Detection Methods

Sleep spindle is a peculiar oscillatory brain pattern which has been associated with a number of sleep (isolation from exteroceptive stimuli, memory consolidation) and individual characteristics (intellectual quotient). Oddly enough, the definition of a spindle is both incomplete and restrictive. In consequence, there is no consensus about how to detect spindles. Visual scoring is cumbersome and user dependent. To analyze spindle activity in a more robust way, automatic sleep spindle detection methods are essential. Various algorithms were developed, depending on individual research interest, which hampers direct comparisons and meta-analyses. In this review, sleep spindle is first defined physically and topographically. From this general description, we tentatively extract the main characteristics to be detected and analyzed. A nonexhaustive list of automatic spindle detection methods is provided along with a description of their main processing principles. Finally, we propose a technique to assess the detection methods in a robust and comparable way.

Automated sleep spindle detection using IIR filters and a Gaussian Mixture Model

Sleep spindle detection using modern signal processing techniques such as the Short-Time Fourier Transform and Wavelet Analysis are common research methods. These methods are computationally intensive, especially when analysing data from overnight sleep recordings. The authors of this paper propose an alternative using pre-designed IIR filters and a multivariate Gaussian Mixture Model. Features extracted with IIR filters are clustered using a Gaussian Mixture Model without the use of any subject independent thresholds. The Algorithm was tested on a database consisting of overnight sleep PSG of 5 subjects and an online public spindles database consisting of six 30 minute sleep excerpts. An overall sensitivity of 57% and a specificity of 98.24% was achieved in the overnight database group and a sensitivity of 65.19% at a 16.9% False Positive proportion for the 6 sleep excerpts.

Using a quadratic parameter sinusoid model to characterize the structure of EEG sleep spindles

Sleep spindles are essentially non-stationary signals that display time and frequency-varying characteristics within their envelope, which makes it difficult to accurately identify its instantaneous frequency and amplitude. To allow a better parameterization of the structure of spindle, we propose modeling spindles using a Quadratic Parameter Sinusoid (QPS). The QPS is well suited to model spindle activity as it utilizes a quadratic representation to capture the inherent duration and frequency variations within spindles. The effectiveness of our proposed model and estimation technique was quantitatively evaluated in parameter determination experiments using simulated spindle-like signals and real spindles in the presence of background EEG. We used the QPS parameters to predict the energy and frequency of spindles with a mean accuracy of 92.34 and 97.73% respectively. We also show that the QPS parameters provide a quantification of the amplitude and frequency variations occurring within sleep spindles that can be observed visually and related to their characteristic "waxing and waning" shape. We analyze the variations in the parameters values to present how they can be used to understand the inter- and intra-participant variations in spindle structure. Finally, we present a comparison of the QPS parameters of spindles and non-spindles, which shows a substantial difference in parameter values between the two classes.

Stage-independent, single lead EEG sleep spindle detection using the continuous wavelet transform and local weighted smoothing

Sleep spindles are critical in characterizing sleep and have been associated with cognitive function and pathophysiological assessment. Typically, their detection relies on the subjective and time-consuming visual examination of electroencephalogram (EEG) signal(s) by experts, and has led to large inter-rater variability as a result of poor definition of sleep spindle characteristics. Hitherto, many algorithmic spindle detectors inherently make signal stationarity assumptions (e.g., Fourier transform-based approaches) which are inappropriate for EEG signals, and frequently rely on additional information which may not be readily available in many practical settings (e.g., more than one EEG channels, or prior hypnogram assessment). This study proposes a novel signal processing methodology relying solely on a single EEG channel, and provides objective, accurate means toward probabilistically assessing the presence of sleep spindles in EEG signals. We use the intuitively appealing continuous wavelet transform (CWT) with a Morlet basis function, identifying regions of interest where the power of the CWT coefficients corresponding to the frequencies of spindles (11-16 Hz) is large. The potential for assessing the signal segment as a spindle is refined using local weighted smoothing techniques. We evaluate our findings on two databases: the MASS database comprising 19 healthy controls and the DREAMS sleep spindle database comprising eight participants diagnosed with various sleep pathologies. We demonstrate that we can replicate the experts' sleep spindles assessment accurately in both databases (MASS database: sensitivity: 84%, specificity: 90%, false discovery rate 83%, DREAMS database: sensitivity: 76%, specificity: 92%, false discovery rate: 67%), outperforming six competing automatic sleep spindle detection algorithms in terms of correctly replicating the experts' assessment of detected spindles.

A comparison of two sleep spindle detection methods based on all night averages: individually adjusted vs. fixed frequencies

Sleep spindles are frequently studied for their relationship with state and trait cognitive variables, and they are thought to play an important role in sleep-related memory consolidation. Due to their frequent occurrence in NREM sleep, the detection of sleep spindles is only feasible using automatic algorithms, of which a large number is available. We compared subject averages of the spindle parameters computed by a fixed frequency (FixF) (11–13 Hz for slow spindles, 13–15 Hz for fast spindles) automatic detection algorithm and the individual adjustment method (IAM), which uses individual frequency bands for sleep spindle detection. Fast spindle duration and amplitude are strongly correlated in the two algorithms, but there is little overlap in fast spindle density and slow spindle parameters in general. The agreement between fixed and manually determined sleep spindle frequencies is limited, especially in case of slow spindles. This is the most likely reason for the poor agreement between the two detection methods in case of slow spindle parameters. Our results suggest that while various algorithms may reliably detect fast spindles, a more sophisticated algorithm primed to individual spindle frequencies is necessary for the detection of slow spindles as well as individual variations in the number of spindles in general.

Learning machines and sleeping brains: Automatic sleep stage classification using decision-tree multi-class support vector machines

BACKGROUND

Sleep staging is a critical step in a range of electrophysiological signal processing pipelines used in clinical routine as well as in sleep research. Although the results currently achievable with automatic sleep staging methods are promising, there is need for improvement, especially given the time-consuming and tedious nature of visual sleep scoring.

NEW METHOD

Here we propose a sleep staging framework that consists of a multi-class support vector machine (SVM) classification based on a decision tree approach. The performance of the method was evaluated using polysomnographic data from 15 subjects (electroencephalogram (EEG), electrooculogram (EOG) and electromyogram (EMG) recordings). The decision tree, or dendrogram, was obtained using a hierarchical clustering technique and a wide range of time and frequency-domain features were extracted. Feature selection was carried out using forward sequential selection and classification was evaluated using k-fold cross-validation.

RESULTS

The dendrogram-based SVM (DSVM) achieved mean specificity, sensitivity and overall accuracy of 0.92, 0.74 and 0.88 respectively, compared to expert visual scoring. Restricting DSVM classification to data where both experts' scoring was consistent (76.73% of the data) led to a mean specificity, sensitivity and overall accuracy of 0.94, 0.82 and 0.92 respectively.

COMPARISON WITH EXISTING METHODS

The DSVM framework outperforms classification with more standard multi-class "one-against-all" SVM and linear-discriminant analysis.

CONCLUSION

The promising results of the proposed methodology suggest that it may be a valuable alternative to existing automatic methods and that it could accelerate visual scoring by providing a robust starting hypnogram that can be further fine-tuned by expert inspection.

Amplitude normalization applied to an artificial neural network-based automatic sleep spindle detection system

Sleep spindles are significant rhythmic transients present in the sleep electroencephalogram (EEG) of non-rapid eye movement (NREM) sleep. Automatic sleep spindle detection techniques are sought for the automation of sleep staging and the detailed study of sleep spindle patterns, of possible physiological significance. A deficiency of many of the available automatic detection techniques is their reliance on the amplitude level of the recorded EEG voltage values. In the present work, an automatic sleep spindle detection system that has been previously proposed, using a Multi-Layer Perceptron (MLP) Artificial Neural Network (ANN), was evaluated using a voltage amplitude normalization procedure, with the aim of making the performance of the ANN independent of the absolute voltage level of the individual subjects' recordings. The application of the normalization procedure led to a reduction in the false positive rate (FPR) as well as in the sensitivity. When the ANN was trained on a combination of data from healthy subjects, the reduction of FPR was from 42.6% to 19%, while the sensitivity of the ANN was kept at acceptable levels, i.e., 73.4% for the normalized procedure vs 84.6% for the non-normalized procedure.

Evaluating the use of line length for automatic sleep spindle detection

Sleep spindles are transient waveforms observed on the electroencephalogram (EEG) during the N2 stage of sleep. In this paper we evaluate the use of line length, an efficient and low-complexity time domain feature, for automatic detection of sleep spindles. We use this feature with a simple algorithm to detect spindles achieving sensitivity of 83.6% and specificity of 87.9%. We also present a comparison of these results with other spindle detection methods evaluated on the same dataset. Further, we implemented the algorithm on a MSP430 microcontroller achieving a power consumption of 56.7 μW. The overall detection performance, combined with the low power consumption show that line length could be a useful feature for detecting sleep spindles in wearable and resource-constrained systems.