Improving Classification of Epileptic and Non-Epileptic EEG Events by Feature Selection

Correctly diagnosing generalized epileptic from non-epileptic episodes, such as psychogenic non epileptic seizures (PNES) and vasovagal or vasodepressor syncope (VVS), despite its importance for the administration of appropriate treatment, life improvement of the patient, and cost reduction for patient and healthcare system, is rarely tackled in the literature. Usually clinicians differentiate between generalized epileptic seizures and PNES based on clinical features and video-EEG. In this work, we investigate the use of machine learning techniques for automatic classification of generalized epileptic and non-epileptic events based only on multi-channel EEG data. For this purpose, we extract the signal patterns in the time domain and in the frequency domain and then combine all features across channels to characterize the spatio-temporal manifestation of seizures. Several classification algorithms are explored and evaluated on EEG epochs from 11 subjects in an inter-subject crossvalidation setting. Due to large number of features feature ranking and selection is performed prior to classification using the ReliefF ranking algorithm within two different voting strategies. The classification models using feature subsets, achieved higher accuracy compared to the models using all features reaching 95% (Bayesian Network), 89% (Random Committee) and 87% (Random Forest) for binary classification (epileptic versus non-epileptic). The results demonstrate the competitiveness of this approach as opposed to previous methods.

Sleep Spindle Detection in EEG Signals Combining HMMs and SVMs

In this paper we present a combined SVM-HMM sleep spindle detection scheme. The proposed scheme takes advantage of the information provided from each of the two prediction models in decision level, in order to provide refined and more accurate spindle detection results. The experimental results showed that the proposed combined scheme achieved an overall detection performance of 90.28%, increasing the best-performing SVM-based model by 2% in terms of absolute performance.

 

Multiresolution Similarity Search in Time Series Data: An Application to EEG Signals

PETRA '13, May 29-31, 2013, Island of Rhodes, Greece. Copyright 2013 ACM 978-1-4503-1300-1/13/05

In this article, we investigate the performance of a seizure detection module for online monitoring of epileptic patients. The module is using as input data streams from electroencephalographic and electrocardiographic recordings. The architecture of the module consists of time and frequency domain feature extraction followed by classification.Four classification algorithms were evaluated on three epileptic subjects. The best performance was achieved by the support vector machine algorithm, with more than 90% for two of the subjects and slightly lower than 90% for the third subject.

 

Online Seizure Detection from EEG and ECG signals for Monitoring of Epileptic Patients

Time series constitute a prevalent data type that arise in several diverse disciplines (e.g., biomedical data, sensor data, images, video data), and therefore analyzing time series is a significant task with a plethora of important applications. In this paper, we study the general problem of similarity search in time series databases and we propose a novel multiresolution indexing (i.e., representation) and retrieval method for time series similarity search. Our approach is motivated by the idea that if we examine a time series at different resolution levels, we could possibly acquire further insights about the data. The proposed algorithm adopts a combined, two-step pruning (filtering) strategy to further reduce data dimensionality by discarding irrelevant time series (i.e., false alarms). At a first level, the time series are represented by line segments and filtered by the triangular inequality property. Then, a Vector Quantization like scheme is applied to encode data and thus to reduce dimensionality. We test and demonstrate the performance of the proposed method, analyzing EEG time series data for retrieval of one of the constituent brain waveforms in EEG recordings, the K-complex, but the method can as well be applied for retrieval of other patterns of interest in time series analysis. The automatic detection and categorization of the EEG patterns will allow the advanced correlation analysis of large amounts of data and will lead to advanced decision making capabilities assisting diagnosis by medical professionals.

 

Seizure detection using EEG and ECG signals for computer-based monitoring, analysis and management of epileptic patients

2014 Elsevier Ltd. All rights reserved.

In this paper a seizure detector using EEG and ECG signals, as a module of a healthcare system, is presented. Specifically, the module is based on short-time analysis with time-domain and frequency-domain features and classification using support vector machines. The seizure detection module was evaluated on three subjects with diagnosed idiopathic generalized epilepsy manifested with absences. The achieved seizure detection accuracy was approximately 90% for all evaluated subjects. Feature ranking investigationand evaluation of the seizure detection module using subsets of features showed that the feature vector composed of approximately the 65%-best ranked parameters provides a good trade-off between computational demands and accuracy. This configurable architecture allows the seizure detection module to operate as part of a healthcare system in offline mode as well as in online mode, where real-time performance is needed.

 

Classification of Epileptic and Non-Epileptic EEG Events

In this paper, the classification of epileptic and nonepileptic events from multi-channel EEG data is investigated using a large number of time and frequency domain features. In contrast to most of the evaluations found in the literature, in this paper the non-epileptic class consists of two types of paroxysmal episodes of loss of consciousness namely the psychogenic non epileptic seizure (PNES) and the vasovagal syncope (VVS). For the classification, several classification algorithms were explored. The classification models were evaluated on EEG epochs from 11 subjects in an inter-subject cross-validation setting and the best among them achieved classification accuracies of 86% (Bayesian Network), 83% (Random Committee) and 74% (Random Forest).

 

Dentate Gyrus Variation along Its Septo-Temporal Axis: Structure and Function in Health and Disease

ISBN: 978-1-63463-371-0, 2015 Nova Science Publishers, Inc.

A bulk of evidence currently suggests that hippocampal formation is a heterogeneous brain structure. Most recent studies recognize a hippocampal pole ( dorsal/septal or posterior in humans) which is primarily related with memory and learning process, and another one ( ventral/temporal or anterior in humans ) which is linked with anxiety, affective or emotional processes. An intermediate region separating the two poles appears to have overlapping characteristics with its neighbors.

 

Absence Seizures

Absence seizures are usually considered to present a short suspension of  consciousness, concomitant with   the 3-Hz spike and wave seen in the EEG. For this reason, mechanisms of generation of spike-and-wave discharges have long been negatively associated with mechanisms of consciousness. We present a review of the various theories that have been developed to explain the generation of spike-and-wave discharges, concluding that hyperexcitable components of the cortico-thalamocortical circuit are likely to explain the EEG discharge. Behavioral analysis of absence seizures points to the possibility that it is not consciousness, as a unitary integrating concept, that is disrupted but rather various components of behavior, such as sensory perception, motor output, attention, and memory, that are suspended. Functional imaging studies of metabolic changes during the discharges point to the involvement of the thalamocortical system but also to the suspension of the default mode network as well as involvement of various cortical regions. Taken altogether, these studies point to the possibility that rather than a diffuse attenuation of consciousness, absence seizures result in the suspension of variable "pieces of consciousness" depending on which cortical, subcortical, and thalamocortical networks are primarily involved. It may be that consciousness itself is more amenable to study if it is conceived in terms of its component pieces rather than a unitary concept.

An intra-K-complex oscillation with independent and labile frequency and topography in NREM sleep

Frontiers in Human Neuroscience, ORIGINAL RESEARCH ARTICLE, Published: 26 April 2013

NREM sleep is characterized by K-complexes (KCs), over the negative phase of which we identified brief activity in the theta range.We recorded high resolution EEG of whole-night sleep from seven healthy volunteers and visually identified 2nd and 3rd stage NREM spontaneous KCs. We identified three major categories: (1) KCs without intra-KC-activity (iKCa), (2) KCs with non-oscillatory iKCa, and (3) KCs with oscillatory iKCa. The latter group of KCs with intra-KC-oscillation (iKCo), was clustered according to the duration of the iKCo. iKCa was observed in most KCs (1150/1522, 75%). iKCos with 2, 3, and 4 waves were observed in 52% (786/1522) of KCs in respective rates of 49% (386/786), 44%, and 7%. Successive waves of iKCos showed on average a shift of their maximal amplitude in the anterio-posterior axis, while the average amplitude of the slow KC showed no spatial shift in time. The iKCo spatial shift was accompanied by transient increases in instantaneous frequency from the theta band toward the alpha band, followed by decreases to upper theta. The study shows that the KC is most often concurrently accompanied by an independent brief iKCo exhibiting topographical relocation of amplitude maxima with every consecutive peak and transient increases in frequency. The iKCo features are potentially reflecting arousing processes taking place during the KC.

Spindle Power Is Not Affected after Spontaneous KComplexes during Human NREM Sleep

Published January 10, 2013

K-complexes and sleep spindles often grouped together characterize the second stage of NREM sleep and interest has been raised on a possible interaction of their underlying mechanisms. The reported inhibition of spindles power for about 15 seconds following evoked K-complexes has implications on their role in arousal. Our objective was to assess this inhibition following spontaneous K-complexes. We used time-frequency analysis of spontaneous K-complexes selected from wholenight EEG recordings of normal subjects. Our results show that spindles are most often observed at the positive phase following the peak of a spontaneous KC (70%). At latencies of 1–3 s following the peak of the K-complex, spindles almost disappear. Compared to long-term effects described for evoked KCs, sleep spindle power is not affected by spontaneous KCs for latencies of 5–15 s. Observation of the recurrence rate of sporadic spindles suggests that the reduction of power at 1–3 s most likely reflects a refractory period of spindles lasting for 1–2 s, rather than an effect of KCs. These results suggest that the mechanisms underlying spontaneous KCs do not affect spindle power as in the case of evoked KCs.

Semi-automatic sleep EEG scoring based on the hypnospectrogram

Published January 10, 2013

Background: Sleep EEG organization is revealed by sleep scoring, a time-consuming process based on strictly defined visual criteria. New method: We explore the possibility of sleep scoring using the whole-night time-frequency analysis, termed hypnospectrogram, with a computer-assisted K-means clustering method. Results: Hypnograms were derived from 10 whole-night sleep EEG recordings using either standard visual scoring under the Rechtshaffen and Kales criteria or semi-automated analysis of the hypnospectrogram derived from a single EEG electrode. We measured substantial agreement between the two approaches with Cohen's kappa considering all 7 stages at 0.61. Comparison with existing methods: A number of existing automated procedures have reached the level of human inter-rater agreement using the standard criteria. However, our approach offers the scorer the opportunity to exploit the information-rich graphic representation of the whole night sleep upon which the automated method works. Conclusion: This work suggests that the hypnospectrogram can be used as an objective graphical representation of sleep architecture upon which sleep scoring can be performed with computer-assisted methods.

Spatiotemporal profiles of focal and generalised spikes in childhood absence epilepsy

Original article with supplemental data, Epileptic Disord 2013; 15 (1): 14-26

TheEEGin childhood absence epilepsy (CAE)maycontain focal and generalised spike-wave discharges (SWDs) with focal, mainly frontal, "lead-in". The term "frontal absence" has been used to imply fast, secondary, 3-Hz generalisation from occult frontal foci with potential impact on clinical EEG interpretation and syndrome classification. The aim of this study was to investigate the relationship between focal and generalised SWDs. We studied five children with CAE and examined a sufficient number of focal ("interictal") and generalised SWDs in order to obtain reliable analysis. All generalised SWDs with focal lead-in were "decomposed" into their "pre-generalisation" focal and "generalised" constituents, which were studied separately. Two types of focal SWD ("interictal" and "pregeneralisation") and generalised SWD were visually clustered into groups, waveform-averaged, and plotted in the 2D-electrode space. Spatiotemporal analysis demonstrated a variety (mean: 4.2 per child; SD: 2.12) of mainly frontalandoccipital locations for pre-generalisation focalSWDswith propagation along the longitudinal axis in either directionandacrosshomologous sites. Interictal focal SWDs demonstrated similar spatiotemporal characteristics. In contrast, the topography and propagation patterns of the first generalised spike of the SWD showed less variability (mean: 2.5 per child; SD: 2.07), mainly involved the fronto-temporal/temporal areas, and correlated poorly (<10%) with that of the pre-generalisation focal SWD. Our findings suggest that the process of generalised epileptogenesis in genetic epilepsies with electrographic "frontal absences" is far more complex than that proposed by the model for occult frontal focus with fast secondary generalisation. (Published with Supplemental data)

Experimental Evaluation of Bluetooth Real Time Capabilities in Communication Intensive Applications

4th International Conference on Wireless Mobile Communication and Healthcare, Athens, Greece, 2014.

During the last years Bluetooth (BT) standard has received increasing acceptance as a prominent communication technology concerning a wide variety of critical applications, posing demanding communication and strict real time constrains. In that respect the main objective of this paper is to offer comprehensive experimental evaluation analyzing BT time constrained performance and critical network parameters. Performance evaluation presented focus on specialized real-time metrics, such as, standard delay deviation, and delay distribution measurements. The study presented, through qualitative and quantitative analysis, reveals significant observations and conclusions. Furthermore, critical trade-offs are presented as far as network conditions are concerned, facilitating optimal network configuration with respect to realistic requirements and parameters posed by critical applications such as medical ones.

Mobile monitoring of epileptic patients using a reconfigurable cyberphysical system that handles multi-parametric data acquisition and analysis.

Bideaux, A.; Anastasopoulou, P.; Cañadas, A.; Fernandez, A.; Hey, S., Proceedings of the 4th International Conference on Wireless Mobile Communication and Healthcare, Athens, Greece, 2014.

A personalized and reconfigurable cyberphysical system to handle multi-parametric data acquisition and analysis for mobile monitoring of epileptic patients.

Bideaux, A.; Anastasopoulou, P.; Cañadas, A.; Fernandez, A.; Hey, S. , 11th Body Sensor Networks Conference, June 2014, Zürich

Spike Detection in EEG by LPP and SVM

7th International Conference on PErvasive Technologies Related to Assistive Environments,27-30 May 2014, Rhodes Greece(Submitted)

Spike Detection in EEG by LPP and SVMThis study presents a computer algorithm to detect epileptiform discharges (spikes) in electroencephalography (EEG) that are manifestations of an epileptogenetic abnormality of the brain. Visual analysis is rater-dependent and time consuming, especially for long-term recordings, such as in sleep studies or in ambulatory EEG. Computerized methods can improve efficiency in reviewing long EEG recordings. The proposed method applies coarse to detailed modeling of the spike waveform and classifies the transients based on Locality Preserving Projections (LPP) and Support Vector Machines (SVM). The method achieves high sensitivity with low false positive rate in a intra-patient crossvalidated setting and thus constitutes a valuable tool for automatic spike assessment.
 

Recent developments of ambulatory assessment methods: an overview of current technologies.  Zeitschrift für Neuropsychologie: "Neuropsychological assessment in the real world: Applications and implications of ambulatory approaches"

Hey, S.; Anastasopoulou, P.; Bideaux, A.; Stork, W. ( submitted )

Evaluation of Time and Frequency Domain Features for Seizure Detection from Combined EEG and ECG signals

IEEE-EMBS International Conferences on Biomedical and Health Informatics (BHI),  1-4  June Valencia, Spain (Submitted)

Evaluation of Time and Frequency Domain Features for Seizure Detection from Combined EEG and ECG signalsIn this paper, a large scale evaluation of time-domain and frequency domain features of electroencephalographic and electrocardiographic signals for seizure detection was performed.
For the classification we relied on the support vector machines algorithm. The seizure detection architecture was evaluated on three subjects and the achieved detection accuracy was more than 90% for two of them and slightly lower than 90% for the third subject.

 

 

 

Elsevier Editorial System(tm) for Journal of Biomedical Informatics

Resource Efficient Data Compression Algorithms for WSN based Demanding Medical ApplicationsChristos P. Antonopoulos, George Chaidos, Nikolaos S. Voros, Elsevier Editorial System(tm) for Journal of Biomedical Informatics, IEEE Transactions on Knowledge and Data Engineering (under revision) 

 

 

 

ClassifΙcation of EEG waveforms by spectral cΙustering

 

ClassifΙcation of EEG waveforms by spectral cΙusteringPattern analysis has often been applied for automated extraction of the consituent brain waves and rhythms in ΕEG. These waveforms, such as the K-complexes, delta waves and spindles, form the sleep microstructure which is examined during diagnosis in sleep studies. Specifically K-complexes have been suggested to protect sleep and also to provide gating functions in idiopathic generalized epilepsies or sleep disorders 1.

 

 

 

Armor Practice Journal (Clustering Workshop on brain and eHealth in Brussels, 5th of November 2013)

Epilepsy, the propensity for recurrent, unprovoked epileptic seizures, is the most common serious neurological disorder, affecting over 50 million people worldwide. Epileptic seizures manifest with a wide variety of motor, cognitive, affective, and autonomic symptoms and signs and associated changes in the electrical activities of the brain (EEG), heart (ECG), muscles (EMG), skin (GSR), as well as changes in other important measurable biological parameters, such as respiration and blood pressure. Their recognition and full understanding is the basis for their optimal management and treatment, but presently is unsatisfactory in many respects. Epileptic seizures occur unpredictably and typically outside hospital and are often misdiagnosed as other episodic disturbances such as syncope, psychogenic and sleep disorders, with which they may co-exist, blurring the clinical presentation; on the other hand, costs of hospital evaluation are substantial, frequently without the desirable results, due to suboptimal monitoring capabilities. More available at http://www.epractice.eu/en/cases/armor , On Twitter .

 

Semi-automatic EEG scoring based on the hypnospectrogram

This work suggests that the hypnospectrogram can be used as an objective graphical representation of sleep architecture upon which sleep scoring can be performed with computer-assisted methods.
More... (Journal of Neuroscience Methods, In Press, Ref.: Ms. No. JNEUMETH-D-13-00346R1)


Absence Seizures

Gotman J. and Kostopoulos GK, chapter 5 in A.E. Cavanna et al. (eds.), Neuroimaging of Consciousness, DOI 10.1007/978-3-642-37580-4_5, © Springer-Verlag Berlin Heidelberg 2013

 

Sleep Spindle Detection in EEG Signals combining HMMs and SVMs

Springer LNCS Proceedings of the 2nd Mining Humanistic Data Workshop (MHDW), Engineering Applications of Neural Networks (EANN), 13-16 Sept. 2013, Greece (to appear).

Sleep Spindle Detection in EEG Signals combining HMMs and SVMsIn this paper we present a combined SVM-HMM sleep spindle detection scheme. The proposed scheme takes advantage of the information provided from each of the two prediction models in decision level, in order to provide refined and more accurate spindle detection results. The experimental results showed that the proposed combined scheme achieved an overall detection performance of 90.28%, increasing the best-performing SVM-based model by 2% in terms of absolute performance.

 

 

 

Design and Implementation of Efficient Reconfigurable Cipher Algorithms for Wireless Sensor Networks

11th IEEE International Conference on Industrial Informatics, Bohum, Germany, 29-31, July, 2013.

Design and Implementation of Efficient Reconfigurable Cipher Algorithms for Wireless Sensor NetworksWireless Sensor Networks (WSN) networks are increasingly utilized in highly demanding scenarios such as medical applications. In such cases hardware designs emerge as a prominent approach to address processing demanding tasks offering significant advantages over respective software based solutions. In such cases reconfiguration capabilities enable the achievement of critical trade-off between adequate performance and resource conservation. In this context the main contribution of this paper is the proposal and performance evaluation of a reconfigurable encryption module with respect to the encryption key size aiming towards WSN utilization. Performance will be presented with respect to encryption process delay, reconfiguration delay and power consumption projecting to energy expenditure.

 

 

One-class classification of temporal EEG patterns for K-complex extraction

35th Int. Conf. of the IEEE Engineering in Medicine and Biology Society (EMBC '13), July 3-7, 2013, Osaka, Japan.

One-class classification of temporal EEG patterns for K-complex extractionThe purpose of this study was to detect one of the constituent brain waveforms in electroencephalography (EEG), the K-complex (KC). The role and significance of the KC include its engagement in information processing, sleep protection, and memory consolidation [1]. The method applies a two-step methodology in which first all the candidate KC waves are extracted based on fundamental morphological features imitating visual criteria. Subsequently each candidate wave is classified as KC or outlier according to its similarity to a set of different patterns (clusters) of annotated KCs. The different clusters are constructed by applying graph partitioning on the training set based on spectral clustering and exhibit temporal similarities in both signal and frequency content. The method was applied in whole-night sleep activity
recorded using multiple EEG electrodes. Cross-validation was performed against visual scoring of singular generalized KCs during all sleep cycles and showed high sensitivity in KC detection.

 

Cortical and sub-cortical excitability during spindles and K-complexes, putative attractors of epileptic activity

30th International Epilepsy Congress, June 23th – 27th, Montreal, 2013.

Cortical and sub-cortical excitability during spindles and K-complexes, putative attractors of epileptic activityThe ARMOR project (www.armor-project.eu) is developing a platform for home monitoring of epilepsy patients. During sleep stage II (SS2) the frequency of specific types of epileptic activity is high. It is believed that large graphoelements of SS2, spindles and K-complexes act as attractors of epileptic activity.

 

 

Multiresolution Similarity Search in Time Series Data: An Application to EEG Signals

PETRA 2013 : 6th International Conference on Pervasive Technologies Related to Assistive Environments, May 29th – 31st, Rhodes Island, Greece, 2013.

Multiresolution Similarity Search in Time Series Data: An Application to EEG SignalsTime series constitute a prevalent data type that arise in several diverse disciplines (e.g., biomedical data, sensor data, images, video data), and therefore analyzing time series is a significant task with a plethora of important applications. In this paper, we study the general problem of similarity search in time series databases and we propose a novel multiresolution indexing (i.e., representation) and retrieval method for time series similarity search. Our approach is motivated by the idea that if we examine a time series at different resolution levels, we could possibly acquire further insights about the data. The proposed algorithm adopts a combined, two-step pruning (filtering) strategy to further reduce data dimensionality by discarding irrelevant time series (i.e., false alarms). At a first level, the time series are represented by line segments and filtered by the triangular inequality property. Then, a Vector Quantization like scheme is applied to encode data and thus to reduce dimensionality.

We test and demonstrate the performance of the proposed method, analyzing EEG time series data for retrieval of one of the  constituent brain waveforms in EEG recordings, the K-complex, but the method can as well be applied for retrieval of other patterns of interest in time series analysis. The automatic detection and categorization of the EEG patterns will allow the advanced correlation analysis of large amounts of data and will lead to advanced decision making capabilities assisting diagnosis by medical professionals.

 

 

Efficient Hardware Based Security Algorithm Implementation for WSN Medical Applications: The ARMOR Perspective

9th International Symposium on Applied Reconfigurable Computing (ARC 2013), March 25th – 27th, Los Angeles, U.S.A, 2013.

Efficient Hardware Based Security Algorithm Implementation for WSN Medical Applications: The ARMOR PerspectiveUtilization of emerging WSN technologies in the field of demanding medical applications comprise one of the most critical and challenging objectives of the ARMOR project. However, contemporary WSN node implementations are notorious for the resource limitations e.g. in
terms of processing and memory capabilities. Therefore, significant effort is devoted in  hardware based implementations of critical components with respect to the project objectives that can alleviate respective re-source limitations drawback.

 

 

 

 

Mobile Multi-parametric Sensor System for Diagnosis of Epilepsy and Brain Related Disorders

Proceedings of 3rd International Conference on Wireless Mobile Communication and Healthcare,November 21st - 23rd,Paris,France,2012.

Mobile Multi-parametric Sensor System for Diagnosis of Epilepsy and Brain Related DisordersEpilepsy is the commonest serious brain disorder, affecting 1-2% of  the general population. Epileptic seizures are usually expressed with a wide range of paroxysmal recurring motor, cognitive, autonomic symptoms and EEG changes. Therefore reliable diagnosis requires state of the art monitoring and communication technologies providing real-time, accurate and continuous brain and body multi-parametric data measurements. The purpose of this paper is to present an adequate mobile system comprising all required sensor types for the everyday life monitoring of patients with epilepsy.

 

 

The Effect of Symmetric Block Ciphers on WSN Performance and Behavior

Proceedings of 5th Fifth IEEE International Workshop on SelectedTopics in Wireless and Mobile computing, October 8th-10th,Barcelona,Spain, 2012.

The Effect of Symmetric Block Ciphers on WSN Performance and BehaviorNowadays Wireless Sensor Networks are increasingly accepted as a reliable solution to highly demanding and critical application scenarios such as military and medical environments, where security support is an absolute prerequisite. However, supporting security implies the execution of cipher algorithms posing significant overheads on WSN nodes, which suffer from scarce resource availability. Moreover, the system wide overheads imposed by requirements related to network performance and behavior are not adequately addressed in current literature. In that direction, this paper intends to evaluate these effects for critical network parameters. The results presented are based on existing results reported in literature measurements concerning the performance overhead imposed by widely utilized encryption algorithms that have been developed for prominent WSN platforms. To evaluate the measurements, the execution performance of three popular cipher algorithms, has been integrated in Omnet++/MiXiM, a well known and widely utilized WSN network simulator. As part of this work, critical insights are provided concerning the system wide effect of deploying security algorithms which are not taken into account when focusing solely in the security algorithm measurements. Furthermore, important trade-offs are provided both qualitatively and quantitatively.

The standard reference text to this publication is: Systems Design for Remote Healthcare, K. Maharatna and S. Bonfiglio (editors), Chapter 6 by A. Krukowski et al, "Patient Health Record (PHR) System", published by Springer New York, November 2014, ISBN: 978-1-4614-8841-5 (Print) 978-1-4614-8842-2 (Online)

 

Spindles and brain activity in the spindle frequency range during human stage 2 sleep

18th AnnualMeeting of the Organization for Human Brain Mapping, June 10th - 14th, Beijing, China, 2012.

Spindles and brain activity in the spindle frequency range during human stage 2 sleepSleep stages are usually defined through quantification of the presence of large graphoelements in the EEG, i.e. NREM stage 2 (NREM2) is defined by the presence of spindles and K-complexes. These large graphoelements appear similar in the EEG as paroxysmal events, and there is more than a suspicion that the same underlying mechanisms may generate the large sleep graphoelements and some forms of epileptic activity [1]. We used whole night sleep MEG recordings to study the quiet (core) periods without large graphoelements of each sleep stage that until then received little attention [2]. The core period of each sleep stage was characterized by changes in activity compared to the awake state in well-defined brain areas either over a wide frequency band (dominated by low frequencies, below the alpha band) or in the gamma band (25 - 90 Hz). The most striking change was the gamma band in the Left medial Dorsal Prefrontal Cortex (L-mDPFC) that was consistently higher during sleep than in the awake state and increased from NREM2 to NREM4 and culminated with highest activity during REM sleep [2]. Recent findings implicate damage to this area in insomnia [3] and in consolidation and other memory related functions [4].

There is a growing interest in NREM stage 2 sleep [5] and specifically sleep spindles. Spindles are often implicated in memory consolidation [6] and more recently as indicators for the integrity of brain function and possibly recovery after stroke and brain injury [7,8]. We report here our recent results for activity during NREM2 in the spindle frequency range (11 - 16 Hz) in normal subjects and stroke patients.

 

Sleep Spindles–As a Biomarker of Brain Function and Plasticity

Spindles appear in the EEG as sinusoidal waves with frequency in the range 11 to 16 Hz. Together with K-complexes they are the hallmarks of NREM sleep and their appearance is taken as evidence of the onset of light sleep. Their specific distribution and exact frequency, changes in early and late sleep during the night. Sleep spindles are also known as "sigma waves" a term initially recommended (1961) but later discouraged by the International Fenderation of Societies for Electroencephalography and Clinical Neurophysiology (IFSECN), and redefined as a "group of rhythmic waves characterized by progressively increasing, then gradually decreasing amplitude".More...