Source analysis of EEG oscillations using high-resolution EEG and MEG

doi:10.1016/S0079-6123(06)59003-X

Progress in Brain Research

Volume 159, 2006, Pages 29-42

https://doi.org/10.1016/S0079-6123(06)59003-X Get rights and content

Abstract

We investigated spatial properties of the source distributions that generate scalp electroencephalographic (EEG) oscillations. The inherent complexity of the spatio-temporal dynamics of EEG oscillations indicates that conceptual models that view source activity as consisting only of a few “equivalent dipoles” are inadequate. We present an approach that uses volume conduction models to characterize the distinct spatial filtering of cortical source activity by average reference EEG, high-resolution EEG, and magnetoencephalography (MEG). By comparing these three measures, we can make inferences about the sources of EEG oscillations without having to make prior assumptions about the sources. We apply this approach to spontaneous EEG oscillations observed with eyes closed at rest. Both EEG and MEG recordings show robust alpha rhythms over posterior regions of the cortex; however, the dominant frequency of these rhythms varies between EEG and MEG recordings. Frontal alpha and theta rhythms are generated almost exclusively by superficial radial dipole layers that generate robust EEG signals but very little MEG signals; these sources are presumably mainly in the gyral crowns of frontal cortex. MEG and high-resolution EEG estimates of alpha rhythms provide evidence of local tangential and radial sources in the posterior cortex, lying mainly on sulcal and gyral surfaces. Despite the detailed information about local radial and tangential sources potentially afforded by high-resolution EEG and MEG, it is also evident that the alpha and theta rhythms receive contributions from non-local source activity, for instance large dipole layers distributed over lobeal or (potentially) even larger spatial scales.

Section snippets

Spatial properties of EEG oscillations

Electroencephalography (EEG) is a very large-scale measure of brain source activity, apparently recording synaptic activity synchronized over macroscopic (centimeter), regional, and even whole brain spatial scales (Nunez, 2000; Nunez and Srinivasan, 2006). Synchrony among neural populations in compact regions of the brain produces localized dipole current sources. Synchrony among neural populations distributed across the cortex can give rise to regional or global networks consisting of many

Source models for EEG oscillations

The dipole approximation to cortical current sources provides the basis for any realistic EEG source model. It is based on the idea that at large distances, any complex current distribution in a small region of the cortex can be approximated by a “dipole” or more accurately, a dipole moment per unit volume. A “large distance” in this case is at least three or four times the distance between the effective poles of the dipole. Superficial gyral surfaces are located at roughly 1.5 cm from scalp

Spatial filtering of scalp potentials

The relationship between scalp potentials $Φ (r, t)$ and the intermediate scale meso-source function P(r, t) or dipole moment per unit volume (μA/mm²) can be written in terms of a Green's function G_E(r,r′) that describes the head volume coductor: $Φ (r, t) = \int_{B} G_{E} (r, r^{'}) P (r^{'}, t) d r^{'}$ The Green's function expresses the relationship between a unit source at location r′ and the measurement point on the scalp surface r; it depends only on the properties of volume conduction in the head. The potential anywhere in

Source analysis by high-resolution EEG methods

High-resolution EEG methods are based on a conceptual framework that differs substantially from that of source localization. The current source distribution in the cranial volume cannot be estimated uniquely using only data recorded on the scalp surface. Assumptions must be applied by all source localization methods to arrive at source location estimates. By contrast, high-resolution EEG methods do not require any assumptions about the sources; instead they enhance the sensitivity of each

MEG

Magnetoencephalography is a relatively new technology that provides useful insights into sources when used as an adjunct to EEG measurements. One common but entirely erroneous idea is that MEG spatial resolution is superior to EEG spatial resolution, and that MEG should simply replace EEG to extract more accurate information about sources. A much more accurate view is that MEG and EEG are selectively sensitive to different sources; this can be either an advantage or disadvantage to MEG,

Simultaneous EEG and MEG recording

We describe here a preliminary study of spontaneous EEG sources; conventional average reference EEG, high-resolution EEG estimates (obtained with the New Orleans spline-Laplacian algorithm, Nunez and Srinivasan, 2006), and simultaneous MEG recordings are compared so as to obtain inferences about the underlying sources. The approach described here can be extended to event-related dynamics, evoked potentials, or any other scalp recorded EEG signals.

Spontaneous EEG exhibited by two of six subjects

Spectra of average reference EEG, high-resolution EEG, and MEG

Figure 4 shows RMS amplitude spectra at all electrodes or MEG sensors for one subject (SL) at rest with eyes closed. The upper plot shows the average reference EEG amplitude spectra; the middle plot shows the Laplacian spectra; and the lower plot shows the MEG amplitude spectra. Since these measures have units, we focused on frequencies with clear peaks in the spectrum of each channel. This subject shows a robust alpha rhythm with all three experimental measures.

In the average reference EEG

Spatial filtering implies temporal filtering

MEG, EEG, and Laplacian EEG each reflect cortical source dynamics that has been spatially filtered; the properties of the spatial filter are given by the appropriate Green's function. The simulations presented here and in Nunez and Srinivasan (2006) emphasize the critical point that each of these measures is preferentially sensitive to different kinds of source distributions. Scalp potentials emphasize distributed sources encompassing large regions of the cortex on lobeal or even larger scales.

A methodological framework for interpreting the sources of EEG oscillations

We have emphasized that interpreting the EEG or MEG in terms of the underlying source characteristics requires information about the selective sensitivity of different measures; such information is obtained by modeling volume conduction in the head. With sufficiently dense electrode arrays (64–128) we are able to make detailed estimates of superficial, localized sources using high-resolution EEG, in this case the surface Laplacian. However, the sensitivity of the surface Laplacian is

Acknowledgments

This research was supported by a grant from the NIMH MH68004.

References (24)

C. Andrew et al.
Event-related coherence as a tool for studying dynamic interaction of brain regions
Electroencephalogr. Clin. Neurophysiol
(1996)
C. Andrew et al.
On the existence of different alpha band rhythms in the hand area of man
Neurosci. Lett
(1997)
W. Klimesch et al.
‘Paradoxical’ alpha synchronization in a memory task
Cogn. Brain Res.
(1999)
F.A. Lopes da Silva
Functional localization using EEG and/or MEG data: volume conductor and source models
Magn. Reson. Imaging,
(2004)
P.L. Nunez
The brain's magnetic field: some effects of multiple sources on localization methods
Electroencephalogr. Clin. Neurophysiol.
(1986)
P.L. Nunez et al.
A theoretical and experimental study of high resolution EEG based on surface Laplacians and cortical imaging
Electroencephalogr. Clin. Neurophysiol.
(1994)
P.L. Nunez et al.
EEG coherence II: experimental measures of multiple coherence measures
Electroencephalogr. Clin. Neurophysiol.
(1999)
Y. Okada et al.
Comparison of MEG and EEG on the basis of somatic evoked responses elicited by stimulation of the snout in the juvenile swine
Clin. Neurophysiol.
(1999)
G. Pfurtscheller et al.
Event-related EEG/MEG synchronization and desynchronization: basic principles
Electroencephalogr. Clin. Neurophysiol.
(1999)
B. Schack et al.
Phase-coupling of theta-gamma EEG rhythms during short-term memory processing
Int. J. Psychophysiol.
(2002)

R. Srinivasan

Spatial structure of the human alpha rhythm: global correlation in adults and local correlation in children

Clin. Neurophysiol.

(1999)

H.D. Jasper et al.

Electrocorticograms in man. Effects of voluntary movement upon the electrical activity of the precentral gyrus

Arch. Fur Psychiatrie und Zeitschrift Neurologie,

(1949)

Cited by (90)

Take your time: Slow brain rhythms predict fluid intelligence
2023, Intelligence
Evidence is mixed whether fluid intelligence (Gf) is associated with increased or decreased alpha and beta band activity (7–30 Hz). Moreover, the Gf relationship with the delta and theta band activity (1–7 Hz) is unknown. We recorded electroencephalographic (EEG) data in 160 healthy adults solving Raven's Advanced Progressive Matrices with a randomized item order to control for item difficulty unaffected by sequential effects. The participants studied each matrix for 30 s before the response bank onset, so we could track the time course of neural activity during problem solving. We measured Gf using six tests. Gf positively correlated with the delta band power, while there was no correlation with the theta band power. For almost all of the participants, we identified the specific slow rhythm frequency, which varied in power as a function of item difficulty. We observed that the lower this frequency, the higher Gf, but only in men. Finally, the alpha and low-beta activity correlated negatively with Gf after we had filtered out the activity during idle intervals (the latter reflecting waiting for the response bank). Overall, the brain activity in the delta, alpha, and beta bands explained 22.6% of Gf variance.
Measurements of the lateral cerebellar hemispheres using near-infrared spectroscopy through comparison between autism spectrum disorder and typical development
2023, Neuroscience Letters
The cerebellum plays a vital role in cognition, communication with the cerebral cortex, and fine motor coordination. Near-infrared spectroscopy (NIRS) is a portable, less restrictive, and noninvasive functional brain imaging method that can capture brain activity during movements by measuring the relative oxyhemoglobin (oxy-Hb) concentrations in the blood. However, the feasibility of using NIRS to measure cerebellar activity requires discussion. We compared NIRS responses between areas assumed to be the cerebellum and the occipital lobe during a fine motor task (tying a bow knot) and a visual task. Our results showed that the oxy-Hb concentration increased more in the occipital lobe than in the cerebellum during the visual task (p =.034). In contrast, during the fine motor task, the oxy-Hb concentration decreased in the occipital lobe but increased significantly in the cerebellum, indicating a notable difference (p =.015). These findings suggest that we successfully captured cerebellar activity associated with processing, particularly fine motor coordination. Moreover, the observed responses did not differ between individuals with autism spectrum disorder and those with typical development. Our study demonstrates the meaningful utility of NIRS as a method for measuring cerebellar activity during movements.
Validating MEG source imaging of resting state oscillatory patterns with an intracranial EEG atlas
2023, NeuroImage
Magnetoencephalography (MEG) is a widely used non-invasive tool to estimate brain activity with high temporal resolution. However, due to the ill-posed nature of the MEG source imaging (MSI) problem, the ability of MSI to identify accurately underlying brain sources along the cortical surface is still uncertain and requires validation.
We validated the ability of MSI to estimate the background resting state activity of 45 healthy participants by comparing it to the intracranial EEG (iEEG) atlas (https://mni-open-ieegatlas.research.mcgill.ca/). First, we applied wavelet-based Maximum Entropy on the Mean (wMEM) as an MSI technique. Next, we converted MEG source maps into intracranial space by applying a forward model to the MEG-reconstructed source maps, and estimated virtual iEEG (ViEEG) potentials on each iEEG channel location; we finally quantitatively compared those with actual iEEG signals from the atlas for 38 regions of interest in the canonical frequency bands.
The MEG spectra were more accurately estimated in the lateral regions compared to the medial regions. The regions with higher amplitude in the ViEEG than in the iEEG were more accurately recovered. In the deep regions, MEG-estimated amplitudes were largely underestimated and the spectra were poorly recovered. Overall, our wMEM results were similar to those obtained with minimum norm or beamformer source localization. Moreover, the MEG largely overestimated oscillatory peaks in the alpha band, especially in the anterior and deep regions. This is possibly due to higher phase synchronization of alpha oscillations over extended regions, exceeding the spatial sensitivity of iEEG but detected by MEG. Importantly, we found that MEG-estimated spectra were more comparable to spectra from the iEEG atlas after the aperiodic components were removed.
This study identifies brain regions and frequencies for which MEG source analysis is likely to be reliable, a promising step towards resolving the uncertainty in recovering intracerebral activity from non-invasive MEG studies.
Parietal alpha underlies slower cognitive responses during interference processing in adolescents
2023, Behavioural Brain Research
Adolescence is a critical period when cognitive control is rapidly maturing across several core dimensions. Here, we evaluated how healthy adolescents (13–17 years of age, n = 44) versus young adults (18–25 years of age, n = 49) differ across a series of cognitive assessments with simultaneous electroencephalography (EEG) recordings. Cognitive tasks included selective attention, inhibitory control, working memory, as well as both non-emotional and emotional interference processing. We found that adolescents displayed significantly slower responses than young adults specifically on the interference processing tasks. Evaluation of EEG event-related spectral perturbations (ERSPs) on the interference tasks showed that adolescents consistently had greater event-related desynchronization in alpha/beta frequencies in parietal regions. Midline frontal theta activity was also greater in the flanker interference task in adolescents, suggesting greater cognitive effort. Parietal alpha activity predicted age-related speed differences during non-emotional flanker interference processing, and frontoparietal connectivity, specifically midfrontal theta – parietal alpha functional connectivity predicted speed effects during emotional interference. Overall, our neuro-cognitive results illustrate developing cognitive control in adolescents particularly for interference processing, predicted by differential alpha band activity and connectivity in parietal brain regions.
A review of critical challenges in MI-BCI: From conventional to deep learning methods
2023, Journal of Neuroscience Methods
Brain-computer interfaces (BCIs) have achieved significant success in controlling external devices through the Electroencephalogram (EEG) signal processing. BCI-based Motor Imagery (MI) system bridges brain and external devices as communication tools to control, for example, wheelchair for people with disabilities, robotic control, and exoskeleton control. This success largely depends on the machine learning (ML) approaches like deep learning (DL) models. DL algorithms provide effective and powerful models to analyze compact and complex EEG data optimally for MI-BCI applications. DL models with CNN network have revolutionized computer vision through end-to-end learning from raw data. Meanwhile, RNN networks have been able to decode EEG signals by processing sequences of time series data. However, many challenges in the MI-BCI field have affected the performance of DL models. A major challenge is the individual differences in the EEG signal of different subjects. Therefore, the model must be retrained from the scratch for each new subject, which leads to computational costs. Analyzing the EEG signals is challenging due to its low signal to noise ratio and non-stationary nature. Additionally, limited size of existence datasets can lead to overfitting which can be prevented by using transfer learning (TF) approaches. The main contributions of this study are discovering major challenges in the MI-BCI field by reviewing the state of art machine learning models and then suggesting solutions to address these challenges by focusing on feature selection, feature extraction and classification methods.
Tensor decomposition of human narrowband oscillatory brain activity in frequency, space and time
2022, Biological Psychology
Citation Excerpt :
We denote these two sets as SNRlow and SNRhigh, respectively. After setting the activation at all dipoles, the forward model was applied and EEG time series at 64 electrodes arranged according to the 10-10 system (Seeck et al., 2017) were generated. This was done separately for every SNR setting.
Many brain processes in health and disease are associated with modulation of narrowband brain oscillations (NBOs) in the scalp-recorded EEG, which exhibit specific frequency spectra and scalp topography. Isolating and tracking NBOs over time using algorithms is useful in domains such as brain-computer interfaces or when measuring the EEG effects of experimental manipulations. Previously, we successfully applied modified tensor methods for identifying and tracking NBO activity over time or conditions. We introduced frequency and spatial constraints that greatly improved their physiological plausibility. In this paper we rigorously demonstrate the power and precision of tensor methods to separate, isolate and track NBOs using sources simulated with an anatomical forward model. This allows us to control the attributes of NBOs and validate tensor solutions. We find that tensor methods can accurately identify, separate and track NBOs over time, using realistic sources either alone or in combination, and compare favorably to well-known spatio-spectral decomposition methods for NBO estimation.

View all citing articles on Scopus

View full text

Source analysis of EEG oscillations using high-resolution EEG and MEG

Abstract

Section snippets

Spatial properties of EEG oscillations

Source models for EEG oscillations

Spatial filtering of scalp potentials

Source analysis by high-resolution EEG methods

MEG

Simultaneous EEG and MEG recording

Spectra of average reference EEG, high-resolution EEG, and MEG

Spatial filtering implies temporal filtering

A methodological framework for interpreting the sources of EEG oscillations

Acknowledgments

Electroencephalogr. Clin. Neurophysiol

Neurosci. Lett

Cogn. Brain Res.

Magn. Reson. Imaging,

Electroencephalogr. Clin. Neurophysiol.

Electroencephalogr. Clin. Neurophysiol.

Electroencephalogr. Clin. Neurophysiol.

Clin. Neurophysiol.

Electroencephalogr. Clin. Neurophysiol.

Int. J. Psychophysiol.

Clin. Neurophysiol.

Electrocorticograms in man. Effects of voluntary movement upon the electrical activity of the precentral gyrus

Arch. Fur Psychiatrie und Zeitschrift Neurologie,