Stationarity stopping criterion for matching pursuit��framework and encephalographic illustration

We present a new stopping criterion for the matching pursuit (MP) algorithm, based on evaluating stationarity of the residua of the consecutive MP iterations. The new stopping criterion is based on a model of a nonstationary signal, which assumes that the part of the signal that is of interest is nonstationary and contaminated by a weakly stationary noise. Mean- and variance-stationarity of the residua obtained from each step of MP is evaluated by means of dedicated statistical tests-the Kwiatkowski-Phillips-Schmidt-Shin (KPSS) test and the White test, respectively. We illustrate the proposed concept by an example in which we analyse magnetoencephalographic (MEG) data.     

Test of statistical stability of the electroencephalogram

The stability of the EEG during eyes closed and eyes open were tested by the statistical method. The EEG records of 25 s were segmented into ten 2.5 s intervals for each state. The mean and variance and power spectra (2.5 to 17.5 Hz at resolution of 2.5 Hz) were calculated for each segment of 2.5 s interval. Thus, the sequences of mean values, variances and power at each frequency component were obtained for 25 s epoch. The stationarity of these sequences were tested by the run test and the trend test. The stationarity of mean, variance and power spectra were not rejected for 25 s records by any of two tests. In the records of 50 s, about 10–20% of records tested were rejected. The nonstationarity of the EEG appeared for 50 s records. This means that the EEG during period the eyes were closed, or opened can be regarded as stationary over time period as long as at least 25 s.     

EEG-MEG Integration Enhances the Characterization of Functional and Effective Connectivity in the Resting State Network

At the sensor level many aspects, such as spectral power, functional and effective connectivity as well as relative-power-ratio ratio (RPR) and spatial resolution have been comprehensively investigated through both electroencephalography (EEG) and magnetoencephalography (MEG). Despite this, differences between both modalities have not yet been systematically studied by direct comparison. It remains an open question as to whether the integration of EEG and MEG data would improve the information obtained from the above mentioned parameters. Here, EEG (64-channel system) and MEG (275 sensor system) were recorded simultaneously in conditions with eyes open (EO) and eyes closed (EC) in 29 healthy adults. Spectral power, functional and effective connectivity, RPR, and spatial resolution were analyzed at five different frequency bands (delta, theta, alpha, beta and gamma). Networks of functional and effective connectivity were described using a spatial filter approach called the dynamic imaging of coherent sources (DICS) followed by the renormalized partial directed coherence (RPDC). Absolute mean power at the sensor level was significantly higher in EEG than in MEG data in both EO and EC conditions. At the source level, there was a trend towards a better performance of the combined EEG+MEG analysis compared with separate EEG or MEG analyses for the source mean power, functional correlation, effective connectivity for both EO and EC. The network of coherent sources and the spatial resolution were similar for both the EEG and MEG data if they were analyzed separately. Results indicate that the combined approach has several advantages over the separate analyses of both EEG and MEG. Moreover, by a direct comparison of EEG and MEG, EEG was characterized by significantly higher values in all measured parameters in both sensor and source level. All the above conclusions are specific to the resting state task and the specific analysis used in this study to have general conclusion multi-center studies would be helpful.     

Combined EEG/MEG Can Outperform Single Modality EEG or MEG Source Reconstruction in Presurgical Epilepsy Diagnosis

We investigated two important means for improving source reconstruction in presurgical epilepsy diagnosis. The first investigation is about the optimal choice of the number of epileptic spikes in averaging to (1) sufficiently reduce the noise bias for an accurate determination of the center of gravity of the epileptic activity and (2) still get an estimation of the extent of the irritative zone. The second study focuses on the differences in single modality EEG (80-electrodes) or MEG (275-gradiometers) and especially on the benefits of combined EEG/MEG (EMEG) source analysis. Both investigations were validated with simultaneous stereo-EEG (sEEG) (167-contacts) and low-density EEG (ldEEG) (21-electrodes). To account for the different sensitivity profiles of EEG and MEG, we constructed a six-compartment finite element head model with anisotropic white matter conductivity, and calibrated the skull conductivity via somatosensory evoked responses. Our results show that, unlike single modality EEG or MEG, combined EMEG uses the complementary information of both modalities and thereby allows accurate source reconstructions also at early instants in time (epileptic spike onset), i.e., time points with low SNR, which are not yet subject to propagation and thus supposed to be closer to the origin of the epileptic activity. EMEG is furthermore able to reveal the propagation pathway at later time points in agreement with sEEG, while EEG or MEG alone reconstructed only parts of it. Subaveraging provides important and accurate information about both the center of gravity and the extent of the epileptogenic tissue that neither single nor grand-averaged spike localizations can supply.     

Frequency Change Detection in Human Auditory Cortex

We offer a model of how human cortex detects changes in the auditory environment. Auditory change detection has recently been the object of intense investigation via the mismatch negativity (MMN). MMN is a preattentive response to sudden changes in stimulation, measured noninvasively in the electroencephalogram (EEG) and the magnetoencephalogram (MEG). It is elicited in the oddball paradigm, where infrequent deviant tones intersperse a series of repetitive standard tones. However, little apart from the participation of tonotopically organized auditory cortex is known about the neural mechanisms underlying change detection and the MMN. In the present study, we investigate how poststimulus inhibition might account for MMN and compare the effects of adaptation with those of lateral inhibition in a model describing tonotopically organized cortex. To test the predictions of our model, we performed MEG and EEG measurements on human subjects and used both small- (<1/3 octave) and large- (>5 octaves) frequency differences between the standard and deviant tones. The experimental results bear out the prediction that MMN is due to both adaptation and lateral inhibition. Finally, we suggest that MMN might serve as a probe of what stimulus features are mapped by human auditory cortex.     

Revealing Time-Unlocked Brain Activity from MEG Measurements by Common Waveform Estimation

Brain activities related to cognitive functions, such as attention, occur with unknown and variable delays after stimulus onsets. Recently, we proposed a method (Common Waveform Estimation, CWE) that could extract such brain activities from magnetoencephalography (MEG) or electroencephalography (EEG) measurements. CWE estimates spatiotemporal MEG/EEG patterns occurring with unknown and variable delays, referred to here as unlocked waveforms, without hypotheses about their shapes. The purpose of this study is to demonstrate the usefulness of CWE for cognitive neuroscience. For this purpose, we show procedures to estimate unlocked waveforms using CWE and to examine their role. We applied CWE to the MEG epochs during Go trials of a visual Go/NoGo task. This revealed unlocked waveforms with interesting properties, specifically large alpha oscillations around the temporal areas. To examine the role of the unlocked waveform, we attempted to estimate the strength of the brain activity of the unlocked waveform in various conditions. We made a spatial filter to extract the component reflecting the brain activity of the unlocked waveform, applied this spatial filter to MEG data under different conditions (a passive viewing, a simple reaction time, and Go/NoGo tasks), and calculated the powers of the extracted components. Comparing the powers across these conditions suggests that the unlocked waveforms may reflect the inhibition of the task-irrelevant activities in the temporal regions while the subject attends to the visual stimulus. Our results demonstrate that CWE is a potential tool for revealing new findings of cognitive brain functions without any hypothesis in advance.     

Analysis of induced components in electroencephalograms using a multiple correlation method

Background  

Evoked and induced activities are two typical components in the EEG and MEG time series after a stimulation. While evoked activity is phase-locked to the stimulus, induced activity is not. Present analysis methods are able to detect non-phase-locked parts of the signal, however, they do not improve the signal-to-noise ratio (SNR) of these signal components.     

Non-linear analysis of the electroencephalogram in Creutzfeldt-Jakob disease

Creutzfeldt-Jakob disease is a rare, neurological, dementing disorder characterised by periodic sharp waves in the electroencephalogram (EEG). Non-linear analysis of these EEG changes may provide insight into the abnormal dynamics of cortical neural networks in this disorder. Babloyantz et al. have suggested that the periodic sharp waves reflect low-dimensional chaotic dynamics in the brain. In the present study this hypothesis was re-examined using newly developed techniques for non-linear time series analysis. We analysed the EEG of a patient with autopsy-proven Creutzfeldt-Jakob disease using the method of non-linear forecasting as introduced by Sugihara and May, and we tested for non-linearity with amplitude-adjusted, phase-randomised surrogate data. Two epochs with generalised periodic sharp waves showed clear evidence for non-linearity. These epochs could be predicted better and further ahead in time than most of the irregular background activity. Testing against cycle-randomised surrogate data and close inspection of the periodograms showed that the non-linearity of the periodic sharp waves may be better explained by quasi-periodicity than by low-dimensional chaos. The EEG further displayed at least one example of a sudden, large qualitative change in the dynamics, highly suggestive of a bifurcation. The presence of quasi-periodicity and bifurcations strongly argues for the use of a non-linear model to describe the EEG in Creutzfeldt-Jakob disease. Received: 28 October 1996 / Accepted in revised form: 8 July 1997     

Accounting for Linear Transformations of EEG and MEG Data in Source Analysis

Analyses of electro- and magnetoencephalography (EEG, MEG) data often involve a linear modification of signals at the sensor level. Examples include re-referencing of the EEG, computation of synthetic gradiometer in MEG, or the removal of artifactual independent components to clean EEG and MEG data. A question of practical relevance is, if such modifications must be accounted for by adapting the physical forward model (leadfield) before subsequent source analysis. Here, we show that two scenarios need to be differentiated. In the first scenario, which corresponds to re-referencing the EEG and synthetic gradiometer computation in MEG, the leadfield must be adapted before source analysis. In the second scenario, which corresponds to removing artifactual components to ‘clean’ the data, the leadfield must not be changed. We demonstrate and discuss the consequences of wrongly modifying the leadfield in the latter case for an example. Future EEG and MEG studies employing source analyses should carefully consider whether and, if so, how the leadfield must be modified as explicated here.     

Comparative power spectral analysis of simultaneous elecroencephalographic and magnetoencephalographic recordings in humans suggests non-resistive extracellular media

The resistive or non-resistive nature of the extracellular space in the brain is still debated, and is an important issue for correctly modeling extracellular potentials. Here, we first show theoretically that if the medium is resistive, the frequency scaling should be the same for electroencephalogram (EEG) and magnetoencephalogram (MEG) signals at low frequencies (<10 Hz). To test this prediction, we analyzed the spectrum of simultaneous EEG and MEG measurements in four human subjects. The frequency scaling of EEG displays coherent variations across the brain, in general between 1/f and 1/f ², and tends to be smaller in parietal/temporal regions. In a given region, although the variability of the frequency scaling exponent was higher for MEG compared to EEG, both signals consistently scale with a different exponent. In some cases, the scaling was similar, but only when the signal-to-noise ratio of the MEG was low. Several methods of noise correction for environmental and instrumental noise were tested, and they all increased the difference between EEG and MEG scaling. In conclusion, there is a significant difference in frequency scaling between EEG and MEG, which can be explained if the extracellular medium (including other layers such as dura matter and skull) is globally non-resistive.     

Singing of <Emphasis Type="Italic">Neoconocephalus robustus</Emphasis> as an example of deterministic chaos in insects

We use nonlinear time series analysis methods to analyse the dynamics of the sound-producing apparatus of the katydid Neoconocephalus robustus. We capture the dynamics by analysing a recording of the singing activity. First, we reconstruct the phase space from the sound recording and test it against determinism and stationarity. After confirming determinism and stationarity, we show that the maximal Lyapunov exponent of the series is positive, which is a strong indicator for the chaotic behaviour of the system. We discuss that methods of nonlinear time series analysis can yield instructive insights and foster the understanding of acoustic communication among insects.     

A review of EEG and MEG for brainnetome research

The majority of brain activities are performed by functionally integrating separate regions of the brain. Therefore, the synchronous operation of the brain’s multiple regions or neuronal assemblies can be represented as a network with nodes that are interconnected by links. Because of the complexity of brain interactions and their varying effects at different levels of complexity, one of the corresponding authors of this paper recently proposed the brainnetome as a new –ome to explore and integrate the brain network at different scales. Because electroencephalography (EEG) and magnetoencephalography (MEG) are noninvasive and have outstanding temporal resolution and because they are the primary clinical techniques used to capture the dynamics of neuronal connections, they lend themselves to the analysis of the neural networks comprising the brainnetome. Because of EEG/MEG’s applicability to brainnetome analyses, the aim of this review is to identify the procedures that can be used to form a network using EEG/MEG data in sensor or source space and to promote EEG/MEG network analysis for either neuroscience or clinical applications. To accomplish this aim, we show the relationship of the brainnetome to brain networks at the macroscale and provide a systematic review of network construction using EEG and MEG. Some potential applications of the EEG/MEG brainnetome are to use newly developed methods to associate the properties of a brainnetome with indices of cognition or disease conditions. Associations based on EEG/MEG brainnetome analysis may improve the comprehension of the functioning of the brain in neuroscience research or the recognition of abnormal patterns in neurological disease.     

Bispectral analysis does not differentiate between anaesthesia EEG and a linear random process]

Bispectral analysis of the electroencephalogram (EEG) has been used to monitor depth of anaesthesia. In the majority of publications this has involved the use of the so called BIS-Index TM (Aspect Medical Systems, Inc.). The exact relationship of this index to such bispectral parameters as the bispectrum and bicoherence has not yet been reported. If the EEG is considered as a linear random process, bicoherence is trivial, i.e. it is independent of the EEG frequency. The aim of this study was to determine the proportion of EEG epochs with non-trivial bicoherence during isoflurane/N20 anaesthesia. We reanalyzed 25.5 hours of digitalised EEG signal from 9 patients undergoing gynaecological surgery. The test developed by Hinich for Gaussian distribution and linearity was then applied. The test was validated using various synthetic surrogate data: Gaussian random data, the z-component of the Lorenz attractor, the phase randomized EEG and the phase randomized z-component of the Lorenz attractor. The percentage of epochs (8.192 s, 1024 data points) with non-trivial bicoherence was: Lorenz data 95.4%, phase randomized Lorenz data 9.4%, synthetic Gaussian data 14.8%, original EEG 9.1%, phase randomized EEG 5.1%. The original EEG data were not found to contain a higher percentage of epochs with non-trivial bicoherence than the phase randomized data and the synthetic Gaussian data. We conclude that bispectral analysis does not substantially add to the information obtained with other methods of quantitative EEG analysis.     

Replicability of MEG and EEG measures of the auditory N1/N1m-response

We investigated the replicability of the source location, amplitude and latency measures of the auditory evoked N1 (EEG) and N1m (MEG) responses. Each of the 5 subjects was measured 6 times in two recording sessions. Responses to monaural stimuli were recorded from 122 MEG and 64 EEG channels simultaneously. The EEG data were modeled with a symmetrically-located dipole pair. For the MEG data, one dipole in each hemisphere was located independently using a subset of channels. Standard deviation (SD) was used as a measure for replicability. The average SD of the x, y and z coordinates of the contralateral N1m dipole was about 2 mm, whereas the corresponding figures for the ipsilateral N1m and the contra- and ipsilateral N1 were about twice as large. The SDs of the dipole amplitudes and latencies were almost equal with MEG and EEG. The amplitude and latency measures of the MEG field gradient waveforms were almost as replicable as those of the dipole models. The results suggest that both MEG and EEG can be used for investigating the simultaneous activity of the left and right auditory cortices independently, MEG being superior in certain experimental setups.     

Population Level Inference for Multivariate MEG Analysis

Multivariate analysis is a very general and powerful technique for analysing Magnetoencephalography (MEG) data. An outstanding problem however is how to make inferences that are consistent over a group of subjects as to whether there are condition-specific differences in data features, and what are those features that maximise these differences. Here we propose a solution based on Canonical Variates Analysis (CVA) model scoring at the subject level and random effects Bayesian model selection at the group level. We apply this approach to beamformer reconstructed MEG data in source space. CVA estimates those multivariate patterns of activation that correlate most highly with the experimental design; the order of a CVA model is then determined by the number of significant canonical vectors. Random effects Bayesian model comparison then provides machinery for inferring the optimal order over the group of subjects. Absence of a multivariate dependence is indicated by the null model being the most likely. This approach can also be applied to CVA models with a fixed number of canonical vectors but supplied with different feature sets. We illustrate the method by identifying feature sets based on variable-dimension MEG power spectra in the primary visual cortex and fusiform gyrus that are maximally discriminative of data epochs before versus after visual stimulation.     

Electro/magnetoencephalography beamforming with spatial restrictions based on sparsity

We present a source localization method for electroencephalographic (EEG) and magnetoencephalographic (MEG) data which is based on an estimate of the sparsity obtained through the eigencanceler (EIG), which is a spatial filter whose weights are constrained to lie in the noise subspace. The EIG provides rejection of directional interferences while minimizing noise contributions and maintaining specified beam pattern constraints. In our case, the EIG is used to estimate the sparsity of the signal as a function of the position, then we use this information to spatially restrict the neural sources to locations out of the sparsity maxima. As proof of the concept, we incorporate this restriction in the “classical” linearly constrained minimum variance (LCMV) source localization approach in order to enhance its performance. We present numerical examples to evaluate the proposed method using realistically simulated EEG/MEG data for different signal-to-noise (SNR) conditions and various levels of correlation between sources, as well as real EEG/MEG measurements of median nerve stimulation. Our results show that the proposed method has the potential of reducing the bias on the search of neural sources in the classical approach, as well as making it more effective in localizing correlated sources.     

Modelling and detecting deep brain activity with MEG and EEG

We introduce an anatomical and electrophysiological model of deep brain structures dedicated to magnetoencephalography (MEG) and electroencephalography (EEG) source imaging. So far, most imaging inverse models considered that MEG/EEG surface signals were predominantly produced by cortical, hence superficial, neural currents. Here we question whether crucial deep brain structures such as the basal ganglia and the hippocampus may also contribute to distant, scalp MEG and EEG measurements. We first design a realistic anatomical and electrophysiological model of these structures and subsequently run Monte-Carlo experiments to evaluate the respective sensitivity of the MEG and EEG to signals from deeper origins. Results indicate that MEG/EEG may indeed localize these deeper generators, which is confirmed here from experimental MEG data reporting on the modulation of alpha (10–12 Hz) brain waves.     

Spontaneous electroencephalographic and magneto-encephalographic activity as a marker of Alzheimer's disease and mild cognitive impairment

The association between the electroencephalogram (EEG) and the study of cognitive impairment was observed from the beginning of using this technique. The introduction of the magnetoencephalogram (MEG) has enabled new lines of research to be developed with a potential for significant clinical applicability. Both techniques have a series of advantages, such as the direct detection of neuronal activity. The EEG-MEG spectral variations in Alzheimer's disease (AD) and mild cognitive impairment (MCI) are also linked to essential physiological processes, such as neuronal disconnection or the deficiency in certain neurotransmitters. These spectral variations are basically characterised by a slowing down of the trace when spontaneous activity is registered, with an increase in the power of low frequency bands (delta and theta) and a decrease in the high frequency bands (alpha, beta, gamma). The spectral analysis gives sensitivity-specificity results of around 80%. By using MEG, it has been possible to specifically locate the sources of this low frequency activity, which has enabled the sensitivity of the test to be increased to 93.3%, particularly when combined with the nuclear resonance data. However, the most promising results come from longitudinal studies which attempt to predict those MCI patients with a higher risk of developing AD. In this case, some EEG studies have shown a sensitivity of 85% when detecting these patients. Even more important, some longitudinal MEG studies have been able to determine that high parietal delta activity increases the relative risk of developing AD by 350%.     

Hemispheric asymmetries in cortical electrical activity during sleep. I

The hypothesis of a predominance of the right hemisphere in stage REM as compared to NREM has been tested through a spectral analysis of the EEG recorded from left (T3) and right (T4) temporal sites in 5 young healthy right-handed male subjects. Variations in the asymmetry coefficient R - L/R + L in different sleep stages have been analyzed by one way ANOVAs and Sheffé's tests. The hypothesis of a progressive increase in left hemisphere activity throughout different REM cycles as one approaches final awakenings have been investigated by comparing variations in the asymmetry coefficient for epochs of REM and stage 2 NREM sampled in different phases of the REM cycle. EEG results do not support either the hypothesized stage dependent or cycle dependent variation in EEG activity during sleep. We question whether variations in EEG amplitude and synchronization can be used as indices of hemispheric asymmetries during sleep.     

Coupled oscillators for modeling and analysis of EEG/MEG oscillations.

This study presents three EEG/MEG applications in which the modeling of oscillatory signal components offers complementary analysis and an improved explanation of the underlying generator structures. Coupled oscillator networks were used for modeling. Parameters of the corresponding ordinary coupled differential equation (ODE) system are identified using EEG/MEG data and the resulting solution yields the modeled signals. This model-related analysis strategy provides information about the coupling quantity and quality between signal components (example 1, neonatal EEG during quiet sleep), allows identification of the possible contribution of hidden generator structures (example 2, 600-Hz MEG oscillations in somatosensory evoked magnetic fields), and can explain complex signal characteristics such as amplitude-frequency coupling and frequency entrainment (example 3, EEG burst patterns in sedated patients).