EEG and fMRI Coregistration to Investigate the Cortical Oscillatory Activities During Finger Movement

Electroencephalography combined with functional magnetic resonance imaging (EEG-fMRI) may be used to identify blood oxygenation level dependent (BOLD) signal changes associated with physiological and pathological EEG event. In this study we used EEG-fMRI to determine the possible correlation between topographical movement-related EEG changes in brain oscillatory activity recorded from EEG electrodes over the scalp and fMRI-BOLD cortical responses in motor areas during finger movement. Thirty-two channels of EEG were recorded in 9 subjects during eyes-open condition inside a 1.5 T magnetic resonance (MR) scanner using a MR-compatible EEG recording system. Off-line MRI artifact subtraction software was applied to obtain continuous EEG data during␣fMRI acquisition. For EEG data analysis we used the event-related-synchronization/desynchronization (ERS/ERD) approach to investigate where movement-related decreases in alpha and beta power are located. For image statistical analysis we used a general linear model (GLM) approach. There was a significant correlation between the positive-negative ratio of BOLD signal peaks and ERD values in the electrodes over the region of activation. We conclude that combined EEG-fMRI may be used to investigate movement-related oscillations of the human brain inside an MRI scanner and the movement-related changes in the EMG or EEG signals are useful to identify the brain activation sources responsible for BOLD-signal changes.     

On the relationship between slow cortical potentials and BOLD signal changes in humans.

This review summarizes experimental studies that investigated the relationship between DC-recorded slow event-related potentials (slow waves) of the electroencephalogram (EEG) and the hemodynamic BOLD response, as measured with functional magnetic resonance imaging (fMRI). Slow waves have been found to accompany a large number of cognitive processes in a systematic and topographically specific way, and have thus been successfully employed in psychophysiological experiments to dissociate cognitive functions by means of their slow wave topography. Recently, however, several independent studies, using different experimental paradigms, suggest the existence of another feature of slow waves, i.e., a close relationship with the fMRI BOLD response. Some of these studies found couplings between slow waves and BOLD signals in various brain regions, using simultaneous EEG-fMRI recordings. Others found similar task-related activation patterns of slow waves (i.e., scalp topographies) and BOLD responses (i.e., activated voxel profiles), as well as corresponding parametric increases of signal strength with increasing task difficulty. The close relationship between slow waves and BOLD responses reported here concerns a low frequency range of the EEG signal (<1 Hz) that has so far received less attention in studies on the correspondence between EEG and fMRI than the higher frequencies, and therefore adds to the various findings obtained at higher EEG frequencies. Implications for the use of slow waves for neuroscientific research are discussed.     

Cortical and subcortical correlates of electroencephalographic alpha rhythm modulation

Neural correlates of electroencephalographic (EEG) alpha rhythm are poorly understood. Here, we related EEG alpha rhythm in awake humans to blood-oxygen-level-dependent (BOLD) signal change determined by functional magnetic resonance imaging (fMRI). Topographical EEG was recorded simultaneously with fMRI during an open versus closed eyes and an auditory stimulation versus silence condition. EEG was separated into spatial components of maximal temporal independence using independent component analysis. Alpha component amplitudes and stimulus conditions served as general linear model regressors of the fMRI signal time course. In both paradigms, EEG alpha component amplitudes were associated with BOLD signal decreases in occipital areas, but not in thalamus, when a standard BOLD response curve (maximum effect at approximately 6 s) was assumed. The part of the alpha regressor independent of the protocol condition, however, revealed significant positive thalamic and mesencephalic correlations with a mean time delay of approximately 2.5 s between EEG and BOLD signals. The inverse relationship between EEG alpha amplitude and BOLD signals in primary and secondary visual areas suggests that widespread thalamocortical synchronization is associated with decreased brain metabolism. While the temporal relationship of this association is consistent with metabolic changes occurring simultaneously with changes in the alpha rhythm, sites in the medial thalamus and in the anterior midbrain were found to correlate with short time lag. Assuming a canonical hemodynamic response function, this finding is indicative of activity preceding the actual EEG change by some seconds.     

Resting state networks in human cervical spinal cord observed with fMRI

It has been reported that spontaneous fluctuations of blood oxygen level dependent (BOLD) signals can be detected in human brain and constitute resting state networks. It has not been reported whether resting state networks also exist in human spinal cord. In the present study, we investigate spontaneous BOLD signal changes in human cervical spinal cord during resting state. fMRI data were analyzed with independent component analysis and SPM software package. Acceptable reproducibility of spatial maps of BOLD signal changes was found across sessions, with the highest correlation values ranging from 0.18 to 0.44. The dominant frequency of signal changes from independent components with the highest correlation values was approximately the frequency range of the respiratory circle. Activities of spinal motor neurons innervating the scalenes were considered as a major factor in the production of BOLD signal fluctuations were observed in this study. Our findings suggest that BOLD fMRI can be applied to study the features of low-frequency rhythmic activities and corresponding mechanisms in the spinal cord during resting state.     

Two-Dimensional Temporal Clustering Analysis for Patients with Epilepsy: Detecting Epilepsy-Related Information in EEG-fMRI Concordant,Discordant and Spike-Less Patients

EEG acquired simultaneously with fMRI (EEG-fMRI) is a multimodal method that has shown promise in mapping the seizure onset zone in patients with focal epilepsy. However, there are many instances when this method is unsuccessful or not applicable, and other data driven fMRI methods may be utilized. One such method is the two-dimensional temporal clustering analysis (2dTCA). In this study we compared the classic EEG-fMRI and 2dTCA performance in mapping regions related to the seizure onset region in 18 focal epilepsy patients (12 presenting interictal epileptiform discharges (IEDs), during EEG-fMRI acquisition) with Engel I or II surgical outcome. Activation maps of both 2dTCA timing outputs (positive and negative histograms) and EEG detected IEDs were computed and compared to the region of epilepsy surgical resection. Patients were evaluated in three categories based on frequency of EEG detected spiking during the MRI. EEG-fMRI maps were concordant to the epilepsy region in 5/12 subjects, four with frequent IEDs on EEG. The 2dTCA was successful in mapping 13/18 patients including 3/6 with no IEDs detected (10/12 with IEDs detected). The epilepsy-related activities were successfully mapped by both methods in only 4/12 patients. This work suggests that the epilepsy-related information detected by each method may be different: while EEG-fMRI is more accurate in patients with high rather than lower numbers of EEG detected IEDs; 2dTCA can be useful in evaluating patients even when no concurrent EEG spikes are detected or EEG-fMRI is not effective. Therefore, our results support that 2dTCA might be an alternative for mapping epilepsy-related BOLD activity in negative EEG-fMRI (6/7 patients) and spike-less patients.     

Quantitative functional imaging of the brain: towards mapping neuronal activity by BOLD fMRI.

Quantitative magnetic resonance imaging (MRI) and spectroscopy (MRS) measurements of energy metabolism (i.e. cerebral metabolic rate of oxygen consumption, CMR(O2)), blood circulation (i.e. cerebral blood flow, CBF, and volume, CBV), and functional MRI (fMRI) signal over a wide range of neuronal activity and pharmacological treatments are used to interpret the neurophysiologic basis of blood oxygenation level dependent (BOLD) image-contrast at 7 T in glutamatergic neurons of rat cerebral cortex. Multi-modal MRI and MRS measurements of CMR(O2), CBF, CBV and BOLD signal (both gradient-echo and spin-echo) are used to interpret the neuroenergetic basis of BOLD image-contrast. Since each parameter that can influence the BOLD image-contrast is measured quantitatively and separately, multi-modal measurements of changes in CMR(O2), CBF, CBV, BOLD fMRI signal allow calibration and validation of the BOLD image-contrast. Good agreement between changes in CMR(O2) calculated from BOLD theory and measured by (13)C MRS, reveals that BOLD fMRI signal-changes at 7 T are closely linked with alterations in neuronal glucose oxidation, both for activation and deactivation paradigms. To determine the neurochemical basis of BOLD, pharmacological treatment with lamotrigine, which is a neuronal voltage-dependent Na(+) channel blocker and neurotransmitter glutamate release inhibitor, is used in a rat forepaw stimulation model. Attenuation of the functional changes in CBF and BOLD with lamotrigine reveals that the fMRI signal is associated with release of glutamate from neurons, which is consistent with a link between neurotransmitter cycling and energy metabolism. Comparisons of CMR(O2) and CBF over a wide dynamic range of neuronal activity provide insight into the regulation of energy metabolism and oxygen delivery in the cerebral cortex. The current results reveal the energetic and physiologic components of the BOLD fMRI signal and indicate the required steps towards mapping neuronal activity quantitatively by fMRI at steady-state. Consequences of these results from rat brain for similar calibrated BOLD fMRI studies in the human brain are discussed.     

fMRI brain mapping during motion capture and FES induced motor tasks: signal to noise ratio assessment

Functional Electrical Stimulation (FES) is a well known clinical rehabilitation procedure, however the neural mechanisms that underlie this treatment at Central Nervous System (CNS) level are still not completely understood. Functional magnetic resonance imaging (fMRI) is a suitable tool to investigate effects of rehabilitative treatments on brain plasticity. Moreover, monitoring the effective executed movement is needed to correctly interpret activation maps, most of all in neurological patients where required motor tasks could be only partially accomplished. The proposed experimental set-up includes a 1.5 T fMRI scanner, a motion capture system to acquire kinematic data, and an electro-stimulation device. The introduction of metallic devices and of stimulation current in the MRI room could affect fMRI acquisitions so as to prevent a reliable activation maps analysis. What we are interested in is that the Blood Oxygenation Level Dependent (BOLD) signal, marker of neural activity, could be detected within a given experimental condition and set-up. In this paper we assess temporal Signal to Noise Ratio (SNR) as image quality index. BOLD signal change is about 1–2% as revealed by a 1.5 T scanner. This work demonstrates that, with this innovative set-up, in the main cortical sensorimotor regions 1% BOLD signal change can be detected at least in the 93% of the sub-volumes, and almost 100% of the sub-volumes are suitable for 2% signal change detection. The integrated experimental set-up will therefore allows to detect FES induced movements fMRI maps simultaneously with kinematic acquisitions so as to investigate FES-based rehabilitation treatments contribution at CNS level.     

Frontal theta EEG activity correlates negatively with the default mode network in resting state.

We used simultaneously recorded EEG and fMRI to investigate in which areas the BOLD signal correlates with frontal theta power changes, while subjects were quietly lying resting in the scanner with their eyes open. To obtain a reliable estimate of frontal theta power we applied ICA on band-pass filtered (2-9 Hz) EEG data. For each subject we selected the component that best matched the mid-frontal scalp topography associated with the frontal theta rhythm. We applied a time-frequency analysis on this component and used the time course of the frequency bin with the highest overall power to form a regressor that modeled spontaneous fluctuations in frontal theta power. No significant positive BOLD correlations with this regressor were observed. Extensive negative correlations were observed in the areas that together form the default mode network. We conclude that frontal theta activity can be seen as an EEG index of default mode network activity.     

Investigation of human brain hemodynamics by simultaneous near-infrared spectroscopy and functional magnetic resonance imaging

The aim of this study was to compare functional cerebral hemodynamic signals obtained simultaneously by near infrared spectroscopy (NIRS) and by functional magnetic resonance imaging (fMRI). The contribution of superficial layers (skin and skull) to the NIRS signal was also assessed. Both methods were used to generate functional maps of the motor cortex area during a periodic sequence of stimulation by finger motion and rest. In all subjects we found a good collocation of the brain activity centers revealed by both methods. We also found a high temporal correlation between the BOLD signal (fMRI) and the deoxy-hemoglobin concentration (NIRS) in the subjects who exhibited low fluctuations in superficial head tissues.     

Optimisation of a post-processing method to remove the pulse artifact from EEG data recorded during fMRI: an application to P300 recordings during e-fMRI

In functional cerebral studies, it has been established that co-registered electroencephalography (EEG) measurements and functional magnetic resonance imaging (fMRI) were complementary. However, EEG data recorded inside an MRI scanner are heavily distorted, mainly by the most prominent artifact, the cardiac pulse artifact (PA). We describe an original algorithm which yields a high-quality PA filter and demonstrates how this tool can be used to improve the quality of P300 ERP measurements during event-related fMRI (e-fMRI) experiments. EEG data were acquired in interleaved mode during e-fMRI while six healthy volunteers performed a visual odd-ball task, involving Distractors, Target and Novel stimuli, to elicit P300 components. The PA was corrected with the original algorithm. The temporal variations in the PA were evidenced using a principal component analysis (PCA), on each EEG channel. The procedure yielded several PA templates, which were regressed from the EEG data. The PA removal procedure was optimised, and then implemented to improve the measured P300 components. Regressing the most adequate PA template resulted in a high-quality reduction in spectral power at frequencies associated with the cardiac PA. More reliable P300 component measurements were obtained, evidencing higher amplitudes for Novels (9.76-11.20 microV) than for to Targets (6.3-9.09 microV) in centro-parietal and prefrontal areas. The improvement of the processing of EEG data acquired simultaneously with fMRI data provides a new tool and casts perspectives to study the functional organisation of the brain.     

Online Reduction of Artifacts in EEG of Simultaneous EEG-fMRI Using Reference Layer Adaptive Filtering (RLAF)

Simultaneous electroencephalography (EEG) and functional magnetic resonance imaging (fMRI) allow us to study the active human brain from two perspectives concurrently. Signal processing based artifact reduction techniques are mandatory for this, however, to obtain reasonable EEG quality in simultaneous EEG-fMRI. Current artifact reduction techniques like average artifact subtraction (AAS), typically become less effective when artifact reduction has to be performed on-the-fly. We thus present and evaluate a new technique to improve EEG quality online. This technique adds up with online AAS and combines a prototype EEG-cap for reference recordings of artifacts, with online adaptive filtering and is named reference layer adaptive filtering (RLAF). We found online AAS?+?RLAF to be highly effective in improving EEG quality. Online AAS?+?RLAF outperformed online AAS and did so in particular online in terms of the chosen performance metrics, these being specifically alpha rhythm amplitude ratio between closed and opened eyes (3–45% improvement), signal-to-noise-ratio of visual evoked potentials (VEP) (25–63% improvement), and VEPs variability (16–44% improvement). Further, we found that EEG quality after online AAS?+?RLAF is occasionally even comparable with the offline variant of AAS at a 3T MRI scanner. In conclusion RLAF is a very effective add-on tool to enable high quality EEG in simultaneous EEG-fMRI experiments, even when online artifact reduction is necessary.     

Isolating the sources of widespread physiological fluctuations in functional near-infrared spectroscopy signals

Physiological fluctuations at low frequency (<0.1 Hz) are prominent in functional near-infrared spectroscopy (fNIRS) measurements in both resting state and functional task studies. In this study, we used the high spatial resolution and full brain coverage of functional magnetic resonance imaging (fMRI) to understand the origins and commonalities of these fluctuations. Specifically, we applied a newly developed method, regressor interpolation at progressive time delays, to analyze concurrently recorded fNIRS and fMRI data acquired both in a resting state study and in a finger tapping study. The method calculates the voxelwise correlations between blood oxygen level dependent (BOLD) fMRI and fNIRS signals with different time shifts and localizes the areas in the brain that highly correlate with the fNIRS signal recorded at the surface of the head. The results show the wide spatial distribution of this physiological fluctuation in BOLD data, both in task and resting states. The brain areas that are highly correlated with global physiological fluctuations observed by fNIRS have a pattern that resembles the venous system of the brain, indicating the blood fluctuation from veins on the brain surface might strongly contribute to the overall fNIRS signal.     

全面性棘慢复合波相关神经网络的EEG-fMRI研究

全面性癫痫的脑电图(EEG)呈全面性棘慢复合波放电(GSWDs),近来研究表明GSWDs并非"全面地"累及整个大脑,而是与特异的神经网络相关。同步脑电图联合功能磁共振(EEG-fMRI)技术可无创、高时空分辨率地了解痫性电活动时各脑区的代谢功能变化。EEG-fMRI研究已发现丘脑-基底节-皮质网络在GSWDs的产生与维持中起重要作用。与GSWDs相关的功能磁共振(fMRI)信号变化规律:前额叶、额顶叶、后扣带回、楔前叶出现相似的负激活(deactivation)信号,丘脑出现较为一致的激活(activation)信号,基底节出现负激活信号。进一步研究发现,fMRI信号随GSWDs出现的时程动态变化,皮质神经元活动可能先于丘脑出现,但不同EEG-fMRI研究的皮质激活部位存在个体化表现,散在地分布于额叶、顶叶等皮质区域。进一步发展更加复杂的分析方法揭示fMRI信号的时程变化规律,研究GSWDs产生的起始、传导和维持涉及的神经网络,揭示全面性癫痫的行为表现、药物治疗反应与神经网络的关系是未来EEG-fMRI研究的方向。     

青少年肌阵挛癫痫发作间期EEG-fMRI的研究

采用同步脑电与功能磁共振(Simultaneous electroencephalography-correlated functional magnetic resonance imaging,EEG-fMRI)技术,研究青少年肌阵挛癫痫患者发作间期痫样放电时脑部血氧水平依赖(Blood oxygen level-dependent,BOLD)信号变化。结果发现:双侧大脑半球的激活及失活信号变化普遍对称且各自独立存在,信号由枕顶至额区逐渐减少。阳性激活区有:楔叶、岛叶、额中部内侧、小脑中线两侧及丘脑。阴性激活区有:双侧额前部、顶部及扣带后回。由此推断:以棘慢复合波为表现形式的同步的神经元活动可能反映了丘脑皮层BOLD信号的激活,而失活区域反映了异常放电时的脑功能的静息状态;这类激活在神经元的活动(EEG)与fMRI结果之间有很好的对应关系;EEG-fMRI是研究脑功能状态有效的方式。     

Quantitative Analysis of Asymmetrical Cortical Activity Based on Power Spectrum Changes

The present study intends to quantitatively analyze power changes in blood oxygenation level-dependent (BOLD) signals, and to investigate functional asymmetry of cortical activity in motor areas during sequential finger movements. A power spectrum method was employed, mainly in contrast with the signal magnitude analysis, to investigate functional asymmetry of motor area cortical activity. Six right-handed subjects were included in the functional magnetic resonance imaging (fMRI) experiments. Both bi-handed and single-handed movements were analyzed. The power spectrum method demonstrated that right-handed subjects exhibited a larger power difference in BOLD signals between task and rest states in the right motor area than in the left motor area. These results showed that more nerve cells were evoked in the right motor area of right-handed subjects. In addition, the power spectrum method was confirmed to be a valid quantitative-analysis method for brain asymmetry analyses.     

Generalized Spike and Waves: Effect of Discharge Duration on Brain Networks as Revealed by BOLD fMRI

In the past decade, the possibility of combining recordings of EEG and functional MRI (EEG–fMRI), has brought a new insight into the brain network underlying generalized spike wave discharges (GSWD). Nevertheless, how GSWD duration influences this network is not fully understood. In this study we aim to investigate whether GSWD duration had a threshold (non-linear) and/or a linear effect on the amplitude of the associated BOLD changes in any brain regions. This could help in elucidating if there is an hemodynamic background supporting the differentiation between interictal and ictal events. We studied a population of 42 patients with idiopathic generalized epilepsies (IGE) who underwent resting-state EEG–fMRI recordings in three centres (London, UK; Modena, Italy; Rome, Italy), applying a parametric analysis of the GSWD duration. Patients were classified as having Childhood Absence epilepsy, Juvenile Absence Epilepsy, or Juvenile Myoclonic Epilepsy. At the population level linear GSWD duration-related BOLD signal changes were found in a network of brain regions: mainly BOLD increase in thalami and cerebral ventricles, and BOLD decrease in posterior cingulate, precuneus and bilateral parietal regions. No region of significant BOLD change was found in the group analysis for the non-linear effect of GSWD duration. To explore the possible effect of both the different IGE sub-syndromes and the different protocols and scanning equipment used in the study, a full-factorial ANOVA design was performed revealing no significant differences. These findings support the idea that the amplitude of the BOLD changes is linearly related to the GSWD duration with no universal threshold effect of spike and wave duration on the brain network supporting this activity.     

Blood oxygenation level dependent contrast resting state networks are relevant to functional activity in the neocortical sensorimotor system

The relevance of correlations between blood oxygenation level dependent (BOLD) signal changes across the brain acquired at rest (resting state networks, or RSN) to functional networks was tested using two quantitative criteria: (1) the localisation of major RSN correlation clusters and the task-related maxima defined in BOLD fMRI signal changes from the same subjects; and (2) the relative hemispheric lateralisation (LI) of BOLD fMRI signal changes in sensorimotor cortex. RSN were defined on the basis of signal changes correlated with that of a “seed” voxel in the primary sensorimotor cortex. We found a generally close spatial correspondence between clusters of correlated BOLD signal change in RSN and activation maxima associated with hand movement. Conventional BOLD fMRI during active hand movement showed the expected wide variation in relative hemispheric lateralisation of LI for sensorimotor cortex across the subjects. There was a good correlation between LIs for the active hand movement task and the RSN (r=0.74, p<0.001). The RSN thus define anatomically relevant regions of motor cortex and change with functionally relevant variations in hemispheric lateralisation of sensorimotor cortical interactions with hand movement.     

Aging of human supraspinal locomotor and postural control in fMRI

Standing, walking, and running are sensorimotor tasks that develop during childhood. Thereafter they function automatically as a result of a supraspinal network that controls spinal pattern generators. The present study used functional magnetic resonance imaging (fMRI) to investigate age-dependent changes in the supraspinal locomotor and postural network of normal subjects during mental imagery of locomotion and stance. Sixty healthy subjects (ages: 24-78 years), who had undergone a complete neurological, neuro-ophthalmological, and sensory examination to rule out disorders of balance and gait, were trained for the conditions lying, standing, walking, and running in order to imagine these conditions on command in 20-second sequences with the eyes closed while lying supine in an magnetic resonance imaging (MRI) scanner. The following blood oxygen level-dependent (BOLD) signal changes during locomotion and stance were found to be independent of age: (1) prominent activations in the supplementary motor areas, the caudate nuclei, visual cortical areas, vermal, and paravermal cerebellum; (2) significant deactivations in the multisensory vestibular cortical areas (posterior insula, parietoinsular vestibular gyrus, superior temporal gyrus), and the anterior cingulate during locomotion. The following differences in brain activation during locomotion and stance were age-dependent: relative increases in the cortical BOLD signals in the multisensory vestibular cortices, motion-sensitive visual cortices (MT/V5), and somatosensory cortices (right postcentral gyrus). In advanced age this multisensory activation was most prominent during standing, less during walking, and least during running. In conclusion, the functional activation of the basic locomotor and postural network, which includes the prefrontal cortex, basal ganglia, brainstem, and cerebellar locomotor centers, is preserved in the elderly. Two major age-dependent aspects of brain activation during locomotion and stance were found: the mechanism of cortical inhibitory reciprocal interaction between sensory systems during locomotion and stance declines in advanced age; and consequently, multisensory cortical control of locomotion and stance increases with age. These findings may indicate a more conscious locomotor and postural strategy in the elderly.     

Correlation between near-infrared spectroscopy and magnetic resonance imaging of rat brain oxygenation modulation

We measure the tissue oxygen and haemoglobin concentrations in the rat brain during modulation of inhaled oxygen concentration (FiO2), using non-invasive frequency domain near-infrared oximetry. The rise in oxygenated haemoglobin concentration and the decline in deoxygenated haemoglobin concentration are demonstrated in correspondence with the modulation of FiO2, which is changed from 20% to 100% in increments of 20%. Furthermore, the tissue oxygenation saturation also shows the corresponding trend and changes ranging from approximately 70% to 90%. The relative changes in deoxygenated haemoglobin concentration are compared to the blood-oxygenation-level-dependent (BOLD) MRI signal recorded during a similar FiO2 protocol. A linear relationship with high correlation coefficient between the relative changes in the BOLD MRI signal and the NIRS signal is observed.     

Fuzzy cluster analysis of high-field functional MRI data

Functional magnetic resonance imaging (fMRI) based on blood-oxygen level dependent (BOLD) contrast today is an established brain research method and quickly gains acceptance for complementary clinical diagnosis. However, neither the basic mechanisms like coupling between neuronal activation and haemodynamic response are known exactly, nor can the various artifacts be predicted or controlled. Thus, modeling functional signal changes is non-trivial and exploratory data analysis (EDA) may be rather useful. In particular, identification and separation of artifacts as well as quantification of expected, i.e. stimulus correlated, and novel information on brain activity is important for both, new insights in neuroscience and future developments in functional MRI of the human brain. After an introduction on fuzzy clustering and very high-field fMRI we present several examples where fuzzy cluster analysis (FCA) of fMRI time series helps to identify and locally separate various artifacts. We also present and discuss applications and limitations of fuzzy cluster analysis in very high-field functional MRI: differentiate temporal patterns in MRI using (a) a test object with static and dynamic parts, (b) artifacts due to gross head motion artifacts. Using a synthetic fMRI data set we quantitatively examine the influences of relevant FCA parameters on clustering results in terms of receiver-operator characteristics (ROC) and compare them with a commonly used model-based correlation analysis (CA) approach. The application of FCA in analyzing in vivo fMRI data is shown for (a) a motor paradigm, (b) data from multi-echo imaging, and (c) a fMRI study using mental rotation of three-dimensional cubes. We found that differentiation of true "neural" from false "vascular" activation is possible based on echo time dependence and specific activation levels, as well as based on their signal time-course. Exploratory data analysis methods in general and fuzzy cluster analysis in particular may help to identify artifacts and add novel and unexpected information valuable for interpretation, classification and characterization of functional MRI data which can be used to design new data acquisition schemes, stimulus presentations, neuro(physio)logical paradigms, as well as to improve quantitative biophysical models.