Estimation of the cortical functional connectivity with the multimodal integration of high-resolution EEG and fMRI data by directed transfer function
Introduction
Nowadays, there are several types of brain imaging device that are able to provide images of the functional activity of the cerebral cortex based on hemodynamic, metabolic, or electromagnetic measurements. However, static images of brain regions activated during particular tasks do not convey a sufficient amount of information with respect to the central issue of how these regions communicated with each other. The concept of brain connectivity now plays a central role in neuroscience as a way to understand one possible ‘code ’ of the functioning brain, as well as the organized behavior of cortical regions beyond the simple mapping of their activity (David et al., 2004, Horwitz, 2003, Lee et al., 2003). Different approaches for the estimate of cortical connectivity have already been exploited in the literature based on hemodynamic or metabolic measurements (Buchel and Friston, 1997), electroencephalography (EEG) scalp potentials (Brovelli et al., 2002, Gevins et al., 1989, Urbano et al., 1998), and magnetoencephalographic (MEG) fields (Taniguchi et al., 2000).
Structural equation models have been used to investigate cortical connectivity in the human brain by means of functional magnetic resonance imaging (fMRI, Buchel and Friston, 1997, McIntosh and Gonzalez-Lima, 1994, Schlosser et al., 2003). However, the temporal dynamics of the hemodynamic process (on the scale of seconds) makes it problematic to follow the transient activity of the neural populations that develops in the order of tens of milliseconds. Indeed, EEG and MEG are known as useful techniques for the study of brain dynamics due to their high temporal resolution (milliseconds; Nunez, 1981, Nunez, 1995). However, in the EEG case, the different electrical conductivities of the brain, skull, and scalp markedly blur the EEG potential distributions and render the localization of the underlying cortical generators problematic. In last decade, techniques known as high-resolution EEG had allowed to estimate precisely the cortical activity from noninvasive EEG measurements (Babiloni et al., 1997, Gevins, 1989, Gevins et al., 1991, Gevins et al., 1999, He and Lian, 2002, He et al., 2002, Nunez, 1995). This body of EEG techniques includes the use of a large number of scalp electrodes, realistic models of the head derived from structural magnetic resonance images (MRIs), and advanced processing methodologies related to the solution of the linear inverse problem. These methodologies allow the estimation of cortical current density from sensor measurements (Babiloni et al., 2000, Grave de Peralta Menendez and Gonzalez Andino, 1999, Pascual-Marqui, 1995). Nowadays, all the connectivity estimations performed on cerebral electromagnetic signals have been computed between the signals gathered from the electric or magnetic sensors (Gevins et al., 1989, Urbano et al., 1998). On the other hand, the relation between the observed spatial patterns in the sensor space and those in the source space is complicated by the spreading of the potential from the cortex to the sensors. Recently, a methodology being able to compute the coherence between cortical areas has been introduced, and applications to MEG data gathered from normal and Parkinson's disease patients were provided (Gross et al., 2001, Gross et al., 2003). This methodology was applied to cortical signals estimated from MEG measurements, thus improving the spatial details available with respect to the computation of coherence from signals derived directly from the sensors. However, in this case, the direction of the information flow between the cortical areas was not available due to the nondirectional nature of coherence computation.
Here, we present a novel computational approach for the estimation of cortical connectivity based on directed transfer function (DTF), a technique used to estimate the direction of information flow between signals gathered from EEG sensors (Kaminski and Blinowska, 1991, Kaminski et al., 2001). We applied the DTF approach on the cortical signals estimated from high-resolution EEG recordings by using realistic head models and a cortical reconstruction algorithm on an average of 5000 dipoles uniformly disposed along the cortical surface. The estimation of the cortical activity was obtained by application of the linear inverse procedure (Grave de Peralta Menendez and Gonzalez Andino, 1999, Pascual-Marqui, 1995). In this estimation procedure, several a priori information were used in order to increase the quality of the cortical current density estimates. The first a priori information used was the anatomical constraints by placing the current dipoles orthogonally to the reconstructed cortical surface. An additional constraint was to force the dipoles to explain the recorded data with a minimum or a low amount of energy (minimum-norm solutions; Dale and Sereno, 1993, Hämäläinen and Ilmoniemi, 1984).
The solution space can be further reduced by using information derived from hemodynamic measures (i.e., fMRI–BOLD phenomenon) recorded during the same task. The rationale of this multimodal approach is that neural activity, modulating neuronal firing and generating synchronized and coherent EEG potentials, increases glucose and oxygen demands (Dale et al., 2000, Liu et al., 1998, Magistretti et al., 1999). This results in an increase in the local hemodynamic response that can be measured by fMRI (Grinvald et al., 1986, Puce et al., 1997). Hence, fMRI responses and cortical sources of EEG data can be spatially related (Logothetis et al., 2001). Furthermore, numerical simulations have shown that the use of fMRI priors increases the quality of the cortical current estimations (Babiloni et al., 2003, Liu, 2000, Liu et al., 1998). In the present study, this integrated approach is proposed for the estimation of cortical connectivity from combined electromagnetic and hemodynamic measurements in humans, and tested by analyzing visually paced finger movements executed by four healthy subjects.
Section snippets
Subject and experimental design
Four normal right-handed subjects (one male, three females; mean age 23 ± 0.2 years) participated in the study after informed consent was obtained, according to the Institutional Review Board at the University of Illinois at Chicago. Subjects were seated comfortably in an armchair with both arms relaxed and resting on pillows and were requested to perform fast repetitive right finger movements that were cued by visual stimuli. Ten to fifteen blocks of 2-Hz thumb oppositions for right hands were
Results
A selection of the gathered ERPs related to the visually paced right finger tapping task performed by Subject #1 is shown on the upper panel of Fig. 1. These waveforms are relative to the signals gathered from the standard electrode leads of the augmented 10–20 international system that are represented on the realistic geometry scalp reconstruction of the subject (center of the figure). The waveforms shown resulted from the average of the artifact-free trials aligned on the EMG onset.
Methodological considerations
We have presented a body of techniques to unveil changes in functional connectivity between cortical ROIs, depicted on realistic geometry models of the head volume conductor, based on high-resolution EEG data. The connectivity estimations have been performed on ROIs depicted along Brodman areas (BAs) identified on individual cortical model. This strategy uses a priori information according to the role of the BAs in the brain functions. The presented technique could also be applied by drawing
Acknowledgments
This work was partially supported by grant NSF BES-0218736 and by a grant from the IRIB Program.
References (57)
- et al.
High resolution EEG: a new model-dependent spatial deblurring method using a realistically shaped MR-constructed subject's head model
Electroencephalogr. Clin. Neurophysiol.
(1997) - et al.
Multimodal integration of high-resolution EEG and functional magnetic resonance imaging data: a simulation study
NeuroImage
(2003) - et al.
Spatiotemporal mapping of brain activity by integration of multiple imaging modalities
Curr. Opin. Neurobiol.
(2001) - et al.
Dynamic statistical parametric mapping: combining fMRI and MEG for high-resolution imaging of cortical activity
Neuron.
(2000) - et al.
Evaluation of different measures of functional connectivity using a neural mass model
NeuroImage
(2004) - et al.
Event-related covariances during a bimanual visuomotor task II. Preparation and feedback
Electroencephalogr. Clin. Neurophysiol.
(1989) - et al.
Properties of MEG tomographic maps obtained with spatial filtering
NeuroImage
(2003) - et al.
Boundary element method based cortical potential imaging of somatosensory evoked potentials using subjects' magnetic resonance images
NeuroImage
(2002) The elusive concept of brain connectivity
NeuroImage
(2003)- et al.
The functional brain connectivity workshop: report and commentary
NeuroImage
(2003)
Information flow from the sensorimotor cortex to muscle in humans
Clin. Neurophysiol.
Computerized processing of EEG-EOG-EMG artifacts for multi-centric studies in EEG oscillations and event-related potentials
Int. J. Psychophysiol.
Event-related EEG/MEG synchronization and desynchronization: basic principles
Clin. Neurophysiol.
The organization of the cortical motor system: new concepts
Electroencephalogr. Clin. Neurophysiol.
Functional segregation of movement-related rhythmic activity in the human brain
NeuroImage
Altered effective connectivity during working memory performance in schizophrenia: a study with fMRI and structural equation modeling
NeuroImage
Movement-related desynchronization of the cerebral cortex studied with spatially filtered magnetoencephalography
NeuroImage
Testing for nonlinearity in time series: the method of surrogate data
Physica D
Dynamic functional coupling of high resolution EEG potentials related to unilateral internally triggered one-digit movements
Electroencephalogr. Clin. Neurophysiol.
Significance testing of many variables. Problems and solutions
Neuropsychobiology
Assessing time-varying cortical connectivity by high resolution EEG and directed transfer function: simulations and application to finger tapping data
High-resolution electroencephalogram: source estimates of Laplacian-transformed somatosensory-evoked potentials using a realistic subject head model constructed from magnetic resonance images
Med. Biol. Eng. Comput.
Medium-range oscillatory network and the 20-hz sensorimotor induced potential
NeuroImage
Modulation of connectivity in visual pathways by attention: cortical interactions evaluated with structural equation modelling and fMRI
Cereb. Cortex
Integrative visuomotor behavior is associated with interregionally coherent oscillations in the human brain
J. Neurophysiol.
Improved localization of cortical activity by combining EEG and MEG with MRI cortical surface reconstruction: a linear approach
J. Cogn. Neurosci.
Functional coupling and regional activation of human cortical motor areas during simple, internally paced and externally paced finger movements
Brain
Dynamic functional topography of cognitive task
Brain Topogr.
Cited by (354)
Dementia classification using a graph neural network on imaging of effective brain connectivity
2024, Computers in Biology and MedicineEffect of low concentrations of indoor toluene on cognitive performance: EEG-based evidence
2022, Building and Environment