Functional brain network analysis using minimum spanning trees in Multiple Sclerosis: an MEG source-space study.

Cognitive dysfunction in Multiple Sclerosis (MS) is closely related to altered functional brain network topology. Conventional network analyses to compare groups are hampered by differences in network size, density and suffer from normalization problems. We therefore computed the Minimum Spanning Tree (MST), a sub-graph of the original network, to counter these problems. We hypothesize that functional network changes analysed with MSTs are important for understanding cognitive changes in MS and that changes in MST topology also represent changes in the critical backbone of the original brain networks. Here, resting-state magnetoencephalography (MEG) recordings from 21 early MS patients and 17 age-, gender-, and education-matched controls were projected onto atlas-based regions-of-interest (ROIs) using beamforming. The phase lag index was applied to compute functional connectivity between regions, from which a graph and subsequently the MST was constructed. Results showed lower global integration in the alpha2 (10-13Hz) and beta (13-30Hz) bands in MS patients, whereas higher global integration was found in the theta band. Changes were most pronounced in the alpha2 band where a loss of hierarchical structure was observed, which was associated with poorer cognitive performance. Finally, the MST in MS patients as well as in healthy controls may represent the critical backbone of the original network. Together, these findings indicate that MST network analyses are able to detect network changes in MS patients, which may correspond to changes in the core of functional brain networks. Moreover, these changes, such as a loss of hierarchical structure, are related to cognitive performance in MS.

Structural degree predicts functional network connectivity: a multimodal resting-state fMRI and MEG study.

Communication between neuronal populations in the human brain is characterized by complex functional interactions across time and space. Recent studies have demonstrated that these functional interactions depend on the underlying structural connections at an aggregate level. Multiple imaging modalities can be used to investigate the relation between the structural connections between brain regions and their functional interactions at multiple timescales. We investigated if consistent modality-independent functional interactions take place between brain regions, and whether these can be accounted for by underlying structural properties. We used functional MRI (fMRI) and magnetoencephalography (MEG) recordings from a population of healthy adults together with a previously described structural network. A high overlap in resting-state functional networks was found in fMRI and especially alpha band MEG recordings. This overlap was characterized by a strongly interconnected functional core network in temporo-posterior brain regions. Anatomically realistically coupled neural mass models revealed that this strongly interconnected functional network emerges near the threshold for global synchronization. Most importantly, this functional core network could be explained by a trade-off between the product of the degrees of structurally-connected regions and the Euclidean distance between them. For both fMRI and MEG, the product of the degrees of connected regions was the most important predictor for functional network connectivity. Therefore, irrespective of the modality, these results indicate that a functional core network in the human brain is especially shaped by communication between high degree nodes of the structural network.

Optimizing parameter choice for FSL-Brain Extraction Tool (BET) on 3D T1 images in multiple sclerosis.

BACKGROUND: Brain atrophy studies often use FSL-BET (Brain Extraction Tool) as the first step of image processing. Default BET does not always give satisfactory results on 3DT1 MR images, which negatively impacts atrophy measurements. Finding the right alternative BET settings can be a difficult and time-consuming task, which can introduce unwanted variability. AIM: To systematically analyze the performance of BET in images of MS patients by varying its parameters and options combinations, and quantitatively comparing its results to a manual gold standard. METHODS: Images from 159 MS patients were selected from different MAGNIMS consortium centers, and 16 different 3DT1 acquisition protocols at 1.5 T or 3T. Before running BET, one of three pre-processing pipelines was applied: (1) no pre-processing, (2) removal of neck slices, or (3) additional N3 inhomogeneity correction. Then BET was applied, systematically varying the fractional intensity threshold (the "f" parameter) and with either one of the main BET options ("B" - bias field correction and neck cleanup, "R" - robust brain center estimation, or "S" - eye and optic nerve cleanup) or none. For comparison, intracranial cavity masks were manually created for all image volumes. FSL-FAST (FMRIB's Automated Segmentation Tool) tissue-type segmentation was run on all BET output images and on the image volumes masked with the manual intracranial cavity masks (thus creating the gold-standard tissue masks). The resulting brain tissue masks were quantitatively compared to the gold standard using Dice overlap coefficient (DOC). Normalized brain volumes (NBV) were calculated with SIENAX. NBV values obtained using for SIENAX other BET settings than default were compared to gold standard NBV with the paired t-test. RESULTS: The parameter/preprocessing/options combinations resulted in 20,988 BET runs. The median DOC for default BET (f=0.5, g=0) was 0.913 (range 0.321-0.977) across all 159 native scans. For all acquisition protocols, brain extraction was substantially improved for lower values of "f" than the default value. Using native images, optimum BET performance was observed for f=0.2 with option "B", giving median DOC=0.979 (range 0.867-0.994). Using neck removal before BET, optimum BET performance was observed for f=0.1 with option "B", giving median DOC 0.983 (range 0.844-0.996). Using the above BET-options for SIENAX instead of default, the NBV values obtained from images after neck removal with f=0.1 and option "B" did not differ statistically from NBV values obtained with gold-standard. CONCLUSION: Although default BET performs reasonably well on most 3DT1 images of MS patients, the performance can be improved substantially. The removal of the neck slices, either externally or within BET, has a marked positive effect on the brain extraction quality. BET option "B" with f=0.1 after removal of the neck slices seems to work best for all acquisition protocols.

Dynamics underlying spontaneous human alpha oscillations: a data-driven approach.

Although the cognitive and clinical correlates of spontaneous human alpha oscillations as recorded with electroencephalography (EEG) or magnetoencephalography (MEG) are well documented, the dynamics underlying these oscillations is still a matter of debate. This study proposes a data-driven method to reveal the dynamics of these oscillations. It demonstrates that spontaneous human alpha oscillations as recorded with MEG can be viewed as noise-perturbed damped harmonic oscillations. This provides evidence for the hypothesis that these oscillations reflect filtered noise and hence do not possess limit-cycle dynamics. To illustrate the use of the model, we apply it to two data-sets in which a decrease in alpha power can be observed across conditions. The associated differences in the estimated model parameters show that observed decreases in alpha power are associated with different kinds of changes in the dynamics. Thus, the model parameters are useful dynamical biomarkers for spontaneous human alpha oscillations.

Cortico-spinal synchronization reflects changes in performance when learning a complex bimanual task.

Motor performance is accompanied by neural activity in various cortical and sub-cortical areas. This intricate network has to be delicately orchestrated. We analyzed the role of beta synchronization in motor learning using magneto-encephalography combined with electromyography. Cortico-spinal synchronization in the beta band was found to be of particular importance in establishing bimanual movement patterns in the context of a 3:2 polyrhythmic (isometric) force production task. Its dynamics correlated highly with the learning of this complex bimanual motor skill. We submit that the cortical dynamics entrains the spinal motor system by which cortico-spinal beta synchrony serves higher-level motor control functions as primary means of information transfer along the neural axis.

Graph theoretical analysis of magnetoencephalographic functional connectivity in Alzheimer's disease.

In this study we examined changes in the large-scale structure of resting-state brain networks in patients with Alzheimer's disease compared with non-demented controls, using concepts from graph theory. Magneto-encephalograms (MEG) were recorded in 18 Alzheimer's disease patients and 18 non-demented control subjects in a no-task, eyes-closed condition. For the main frequency bands, synchronization between all pairs of MEG channels was assessed using a phase lag index (PLI, a synchronization measure insensitive to volume conduction). PLI-weighted connectivity networks were calculated, and characterized by a mean clustering coefficient and path length. Alzheimer's disease patients showed a decrease of mean PLI in the lower alpha and beta band. In the lower alpha band, the clustering coefficient and path length were both decreased in Alzheimer's disease patients. Network changes in the lower alpha band were better explained by a 'Targeted Attack' model than by a 'Random Failure' model. Thus, Alzheimer's disease patients display a loss of resting-state functional connectivity in lower alpha and beta bands even when a measure insensitive to volume conduction effects is used. Moreover, the large-scale structure of lower alpha band functional networks in Alzheimer's disease is more random. The modelling results suggest that highly connected neural network 'hubs' may be especially at risk in Alzheimer's disease.

Functional and effective whole brain connectivity using magnetoencephalography to identify monozygotic twin pairs.

Resting-state functional connectivity patterns are highly stable over time within subjects. This suggests that such 'functional fingerprints' may have strong genetic component. We investigated whether the functional (FC) or effective (EC) connectivity patterns of one monozygotic twin could be used to identify the co-twin among a larger sample and determined the overlap in functional fingerprints within monozygotic (MZ) twin pairs using resting state magnetoencephalography (MEG). We included 32 cognitively normal MZ twin pairs from the Netherlands Twin Register who participate in the EMIF-AD preclinAD study (average age 68 years). Combining EC information across multiple frequency bands we obtained an identification rate over 75%. Since MZ twin pairs are genetically identical these results suggest a high genetic contribution to MEG-based EC patterns, leading to large similarities in brain connectivity patterns between two individuals even after 60 years of life or more.

Resting state oscillatory brain dynamics in Parkinson's disease: an MEG study.

OBJECTIVE: The pathophysiological mechanisms of cognitive dysfunction and dementia in Parkinson's disease (PD) are still poorly understood. Altered resting state oscillatory brain activity may reflect underlying neuropathological changes. The present study using magneto encephalography (MEG) was set up to study differences in the pattern of resting state oscillatory brain activity in groups of demented and non-demented PD patients and healthy, elderly controls. METHODS: The pattern of MEG background oscillatory activity was studied in 13 demented PD patients, 13 non-demented PD patients and 13 healthy controls. Whole head MEG recordings were obtained in the morning in an eyes closed and an eyes open, resting state condition. Relative spectral power was calculated using Fast Fourier Transformation in delta, theta, alpha, beta and gamma frequency bands. RESULTS: In the non-demented PD patients, relative theta power was diffusely increased and beta power concomitantly decreased relative to controls. gamma Power was decreased in central and parietal channels. In the demented PD patients, a diffuse increase in relative delta and to lesser extent theta power and a decrease in relative alpha, beta and to lesser extent gamma power were found in comparison to the non-demented PD group. In addition, reactivity to eye opening was much reduced in the demented PD group. CONCLUSIONS: Parkinson's disease is characterized by a slowing of resting state brain activity involving theta, beta and gamma frequency bands. Dementia in PD is associated with a further slowing of resting state brain activity, additionally involving delta and alpha bands, as well as a reduction in reactivity to eye-opening. SIGNIFICANCE: The differential patterns of slowing of resting state brain activity in demented and non-demented PD patients suggests that, in conjunction with a progression of the pathological changes already present in non-demented patients, additional mechanisms are involved in the development of dementia in PD.

Color opponency in cone-driven horizontal cells in carp retina. Aspecific pathways between cones and horizontal cells.

The spectral and dynamic properties of cone-driven horizontal cells in carp retina were evaluated with silent substitution stimuli and/or saturating background illumination. The aim of this study was to describe the wiring underlying the spectral sensitivity of these cells. We will present electrophysiological data that indicate that all cone-driven horizontal cell types receive input from all spectral cone types, and we will present evidence that all cone-driven horizontal cell types feedback to all spectral cone types. These two findings are the basis for a model for the spectral and dynamic behavior of all cone-driven horizontal cells in carp retina. The model can account for the spectral as well as the dynamic behavior of the horizontal cells. It will be shown that the strength of the feedforward and feedback pathways between a horizontal cell and a particular spectral cone type are roughly proportional. This model is in sharp contrast to the Stell model, where the spectral behavior of the three horizontal cell types is explained by a cascade of feedforward and feedback pathways between cones and horizontal cells. The Stell model accounts for the spectral but not for the dynamic behavior of the horizontal cells.

Lateral feedback from monophasic horizontal cells to cones in carp retina. II. A quantitative model.

About half of the monophasic horizontal cells in carp retina receive input from both red- and green-sensitive cones. Since the horizontal cells feed back to cones, the color and feedback pathways result in wavelength- and intensity-dependent changes of the dynamics and of the receptive field amplitude profile of the horizontal cell responses. In this paper we present a quantitative model that describes adequately the color and spatial coding and the dynamics of the responses from monophasic horizontal cells in carp. Lateral feedback plays a distinct role in this model.

Lateral feedback from monophasic horizontal cells to cones in carp retina. I. Experiments.

The spatial and color coding of the monophasic horizontal cells were studied in light- and dark-adapted retinae. Slit displacement experiments revealed differences in integration area for the different cone inputs of the monophasic horizontal cells. The integration area measured with a 670-nm stimulus was larger than that measured with a 570-nm stimulus. Experiments in which the diameter of the test spot was varied, however, revealed at high stimulus intensities a larger summation area for 520-nm stimuli than for 670-nm stimuli. The reverse was found for low stimulus intensities. To investigate whether these differences were due to interaction between the various cone inputs to the monophasic horizontal cell, adaptation experiments were performed. It was found that the various cone inputs were not independent. Finally, some mechanisms for the spatial and color coding will be discussed.

MEG-compatible force sensor.

By use of an insulating material we constructed a strain gauge based sensor to measure isometric forces in parallel with magneto-encephalographic recordings (i.e. without interference). The sensor can be used in different geometries to measure force production in different dimensions. Furthermore, it can easily be adapted or modified for specific experimental applications. Finally, on-line processing of the recorded forces, e.g., for the purpose of feedback, can be realized using standard MEG equipment.

Opportunities and methodological challenges in EEG and MEG resting state functional brain network research.

Electroencephalogram (EEG) and magnetoencephalogram (MEG) recordings during resting state are increasingly used to study functional connectivity and network topology. Moreover, the number of different analysis approaches is expanding along with the rising interest in this research area. The comparison between studies can therefore be challenging and discussion is needed to underscore methodological opportunities and pitfalls in functional connectivity and network studies. In this overview we discuss methodological considerations throughout the analysis pipeline of recording and analyzing resting state EEG and MEG data, with a focus on functional connectivity and network analysis. We summarize current common practices with their advantages and disadvantages; provide practical tips, and suggestions for future research. Finally, we discuss how methodological choices in resting state research can affect the construction of functional networks. When taking advantage of current best practices and avoid the most obvious pitfalls, functional connectivity and network studies can be improved and enable a more accurate interpretation and comparison between studies.

Ethambutol changes the color coding of carp retinal ganglion cells reversibly.

The influence of ethambutol on retinal function was studied by recording ganglion cell responses in isolated carp retinas superfused with a Ringer solution containing different concentrations of ethambutol (0 mg/liter, 10 mg/liter, 20 mg/liter, 30 mg/liter). The results indicate that ethambutol reversibly affects color opponency, without changing the sensitivity of the underlying receptor processes. The amacrine and bipolar cells are the most likely candidates to be affected by ethambutol.

Motion-onset visual-evoked potentials as a function of retinal eccentricity in man.

Visual-evoked potentials were elicited by the motion-onset of a black-and-white square-wave grating of 2.4 cycles/deg that drifted from right to left at a velocity of 3 deg/s. The center of the 2 x 2 deg stimulus field was binocularly viewed either foveally or at eccentricities of 6, 12, or 20 deg in the lower visual field along the vertical meridian. Peak-to-peak amplitudes P1-N2 and N2-P2 were found to decrease non-linearly as a function of eccentricity. The VEP-amplitudes were standardized by setting each foveal value to 100%, and a relative measure was derived for peripheral values given by the ratio of the peripheral to the foveal values. The decrease of the relative VEP-values with eccentricity was significantly smaller than that of the relative cortical magnification factor of striate cortex in man, whereas it agreed fairly well with that of the relative point-image size of the area MT in Macaque monkey. In this respect, the motion-onset VEP is distinct from the pattern-reversal VEP, the amplitude of which decreases much more rapidly with retinal eccentricity; hence, it may involve different generating structures of the brain.

Spatial patterns of visual cortical fast EEG during conditioned reflex in a rhesus monkey.

A preliminary assay was made of the existence of time-space coherence patterns of fast EEG activity in the visual cortex of a Rhesus monkey. The primary intent of the present study was to evaluate the similarities and differences in relation to the olfactory bulb, where such coherences have been described and have been demonstrated to be associated with behaviour. Segments 1.5 s in duration were recorded simultaneously without averaging from 16 to 35 subdural electrodes fixed over the left occipital lobe in an array 3.6 cm X 2.8 cm. Each segment was taken during the delivery of a visual conditioned stimulus (CS) and the performance of a conditioned response (CR) by a well-trained Rhesus monkey. The EEGs appeared chaotic with irregular bursts lasting 75-200 ms, resembling those in the olfactory EEG but with lower peak frequencies. Fourier spectra showed broad distributions of power resembling '1/f noise' with multiple peaks in the range of 20-40 Hz. Time intervals were selected where coherent activity seemed to be present at a number of electrodes. A dominant component waveform that was common to all channels was extracted by principal components analysis (PCA) of each segment. The distribution of the power of this component across the electrodes (the factor loadings) was used to describe the spatial pattern of the coherent cortical activity. Statistical analyses suggested that different patterns could be associated to the CS and the CR, as has been found in the olfactory system. These patterns remained stable over a 6 week recording interval. The patterns can be better discriminated, when the factor loadings of each channel are normalized to zero mean and unit variance, to discard a basic pattern of power distribution, which may reflect anatomical and electrode positioning factors that are related to behavioral information processing by the cortex. The wide spatial distribution of the common patterns found suggests that EEG patterns that manifest differing states of the visual cortex may also be accessible with scalp electrodes.

Visual stimulation reduces EEG activity in man.

Data are presented which show that background electric activity of the human brain is reduced by visual stimulation. Occipital EEG amplitude decreases 5-15% for all frequencies analyzed (0.2-40 Hz) upon pattern stimulation. The reduction is stimulus-specific, i.e. is the strongest for stimuli that activate a large number of visual cortical neurons.

Magnetoencephalographic analysis of cortical activity in Alzheimer's disease: a pilot study.

OBJECTIVES: In the present study, MEG was used to analyze spectral power and reference-free coherence in patients with probable Alzheimer's disease (AD). METHODS: Sixty-one channel MEG was recorded in 5 AD patients and 5 age-matched controls at rest with eyes open and eyes closed, as well as during the performance of two different mental tasks. Artefact-free epochs were selected for the analysis of power and coherence values in each of 5 4-Hz wide frequency bands ranging from 2 to 22 Hz. RESULTS: In AD patients, the absolute low frequency magnetic power was significantly and rather diffusely increased relative to controls with a fronto-central maximum. High frequency power values were significantly decreased over the occipital and temporal areas. Reactivity to eye-opening and mental tasks was reduced in the patient group. Relative to controls, a general decrease of MEG coherence values, including all frequencies analyzed, was found in AD patients. CONCLUSIONS: These observations confirm the pattern of changes in spectral power and reactivity known from EEG studies and suggest that coherence decreases in AD patients are widespread and include frequencies outside the alpha band.

The localization of spontaneous brain activity: first results in patients with cerebral tumors.

OBJECTIVE: From EEG studies, it is known that structural brain lesions are accompanied by abnormal rhythmic electric activity. With the better spatial resolution of MEG, MEG dipole analysis can extend the knowledge based on EEG power spectra. This study presents the first results of a completely automatic analysis method applied to spontaneous MEG. METHODS: Spontaneous MEG data of 5 patients with cerebral brain tumors and 4 controls were collected using a whole-head MEG system. Signals were bandpass-filtered with cut-off frequencies according to standard EEG bands. A moving dipole model was fitted to samples with at least twice the average sample power. Dipoles explaining 90% or more of the magnetic variance were projected onto a matched MR scan. RESULTS: In controls, dipole distributions are symmetrical with respect to the mid-sagittal plane whereas distributions in patients often are asymmetrical to it. Dipoles describing gamma activity were located contralateral, and dipoles describing delta and theta activity were located ipsilateral to lesions. CONCLUSIONS: The automatic method gives plausible 3-dimensional information about generator foci of abnormal slow waves and other rhythms with respect to lesion foci and thereby adds physiological knowledge to that derived from EEG power spectra.

A neural complexity measure applied to MEG data in Alzheimer's disease.

OBJECTIVE: A measure of neural complexity (C(N)) (Proc. Natl Acad. Sci. USA 91 (1994) 5033) was applied to magnetoencephalography (MEG) data to test the hypothesis that C(N) decreases when information processing in the brain is impaired, as is the case in patients with Alzheimer's Disease (AD). METHODS: One hundred and fifty-one channel MEGs were recorded in 20 AD patients and 20 healthy age-matched controls in a resting condition with eyes open (EO) and eyes closed (EC). Artifact-free epochs of 117 channels were selected for analysis. C(N) and D(2) were computed in different frequency bands, and correlated with the MMSE. RESULTS: The Group x Frequency band interaction was significant for both C(N) and D(2). C(N) was higher in AD, as compared with controls, in the 2-4 and 4-8Hz bands, and D(2) was higher in AD patients in the 14-20 and 20-30Hz bands. The C(N) was higher in the EC condition compared to the EO condition, whereas the D(2) was higher in the EO condition. CONCLUSIONS: The hypothesis of Tononi et al. (Proc. Natl Acad. Sci. USA 91 (1994) 5033) that the neural complexity decreases in AD patients has to be rejected. However, both neural complexity and the correlation dimension did show differences between controls and AD patients which depended on frequency band.