Echo time dependence of BOLD contrast and susceptibility artifacts

Comparison of fMRI activation as measured with gradient- and spin-echo EPI during visual perception

In this study, we compared fMRI activation measured with gradient- and spin-echo-based fMRI during visual perception of faces, which is mediated by neural activation within a distributed cortical network. With both fMRI techniques, bilateral activation was observed in multiple regions including the inferior occipital gyrus, fusiform gyrus, superior temporal sulcus, amygdala, inferior frontal gyrus, and orbitofrontal cortex. When compared with the gradient-echo sequence, activation measured with the spin-echo sequence was significantly reduced. This decrease was manifested by smaller cluster size, lower statistical significance, smaller amplitude of the fMRI signal, and smaller number of subjects who showed activation in all face-responsive regions. In orbitofrontal cortex, a region prone to susceptibility-related signal dephasing, the spin-echo acquisition considerably restored the signal, but did not reveal stronger activation when compared with the gradient-echo acquisition. Our data indicate that optimized GE sequences that reduce susceptibility artefacts are sufficient to detect activation in regions such as the orbitofrontal cortex.

Sensitivity-encoded (SENSE) echo planar fMRI at 3T in the medial temporal lobe

Parallel imaging techniques are useful for fMRI studies in light of the increasing susceptibility effects at high magnetic field strength. Yet, spatially varying noise amplification constitutes a challenge for the application of these techniques. The medial temporal lobe is particularly vulnerable to susceptibility effect with increasingly strong signal reduction. We present two fMRI studies comparing SENSE single-shot (ssh) echo planar imaging (EPI) at acceleration factors of 2.0, 2.4, 2.7, and 3.0 with conventional sshEPI at TE of 22 and 35 ms. Data were acquired during a learning task which activates the medial temporal lobe bilaterally. Susceptibility related image distortion was markedly reduced with increasing SENSE acceleration. Moreover, in the group results, statistical power increased in the whole brain with SENSE compared to conventional imaging and with a TE of 35 ms compared to 22 ms. Higher SENSE acceleration factors further improved image quality and increased statistical power in the occipital lobe and fusiform gyrus, but not in the medial temporal lobe. We therefore conclude that an sshEPI acquisition protocol with a moderate SENSE acceleration factor of R = 2.0 and TE 35 ms is suitable for the detection of medial temporal activation at 3T.

Spatially selective T2 and T2 * measurement with line-scan echo-planar spectroscopic imaging

Line-scan echo planar spectroscopic imaging (LSEPSI) is applied to quickly measure the T2 and T2* relaxation time constants in pre-selected 2D or 3D regions. Results from brain imaging studies at 3T suggest that the proposed method may prove valuable for both basic research (e.g., quantifying the changes of T2/T2* values in functional MRI with blood oxygenation level-dependent contrast) and clinical studies (e.g., measuring the T2' shortening due to iron deposition). The proposed spatially selective T2 and T2* mapping technique is especially well suited for studies, where T2/T2* quantification needs to be performed dynamically in a pre-selected 2D or 3D region.

EEG-fMRI of focal epileptic spikes: analysis with multiple haemodynamic functions and comparison with gadolinium-enhanced MR angiograms

Combined EEG-fMRI has recently been used to explore the BOLD responses to interictal epileptiform discharges. This study examines whether misspecification of the form of the haemodynamic response function (HRF) results in significant fMRI responses being missed in the statistical analysis. EEG-fMRI data from 31 patients with focal epilepsy were analysed with four HRFs peaking from 3 to 9 sec after each interictal event, in addition to a standard HRF that peaked after 5.4 sec. In four patients, fMRI responses were correlated with gadolinium-enhanced MR angiograms and with EEG data from intracranial electrodes. In an attempt to understand the absence of BOLD responses in a significant group of patients, the degree of signal loss occurring as a result of magnetic field inhomogeneities was compared with the detected fMRI responses in ten patients with temporal lobe spikes. Using multiple HRFs resulted in an increased percentage of data sets with significant fMRI activations, from 45% when using the standard HRF alone, to 62.5%. The standard HRF was good at detecting positive BOLD responses, but less appropriate for negative BOLD responses, the majority of which were more accurately modelled by an HRF that peaked later than the standard. Co-registration of statistical maps with gadolinium-enhanced MRIs suggested that the detected fMRI responses were not in general related to large veins. Signal loss in the temporal lobes seemed to be an important factor in 7 of 12 patients who did not show fMRI activations with any of the HRFs.

Resting fluctuations in arterial carbon dioxide induce significant low frequency variations in BOLD signal

Carbon dioxide is a potent cerebral vasodilator. We have identified a significant source of low-frequency variation in blood oxygen level-dependent (BOLD) magnetic resonance imaging (MRI) signal at 3 T arising from spontaneous fluctuations in arterial carbon dioxide level in volunteers at rest. Fluctuations in the partial pressure of end-tidal carbon dioxide (Pet(CO(2))) of +/-1.1 mm Hg in the frequency range 0-0.05 Hz were observed in a cohort of nine volunteers. Correlating with these fluctuations were significant generalized grey and white matter BOLD signal fluctuations. We observed a mean (+/-standard error) regression coefficient across the group of 0.110 +/- 0.033% BOLD signal change per mm Hg CO(2) for grey matter and 0.049 +/- 0.022% per mm Hg in white matter. Pet(CO(2))-related BOLD signal fluctuations showed regional differences across the grey matter, suggesting variability of the responsiveness to carbon dioxide at rest. Functional magnetic resonance imaging (fMRI) results were corroborated by transcranial Doppler (TCD) ultrasound measurements of the middle cerebral artery (MCA) blood velocity in a cohort of four volunteers. Significant Pet(CO(2))-correlated fluctuations in MCA blood velocity were observed with a lag of 6.3 +/- 1.2 s (mean +/- standard error) with respect to Pet(CO(2)) changes. This haemodynamic lag was adopted in the analysis of the BOLD signal. Doppler ultrasound suggests that a component of low-frequency BOLD signal fluctuations is mediated by CO(2)-induced changes in cerebral blood flow (CBF). These fluctuations are a source of physiological noise and a potentially important confounding factor in fMRI paradigms that modify breathing. However, they can also be used for mapping regional vascular responsiveness to CO(2).

Optimized 3 T EPI of the amygdalae

The optimum parameters for single-shot gradient-recalled (GR) EPI-based fMRI studies of the limbic region are systematically established at 3 T via their ability to mitigate intravoxel dephasing-measured via SNR and T2* in the amygdalae-and their implications for temporal resolution (or brain coverage). Conventional imaging parameters (64 x 64 matrix size and 4-6 mm thick slices) are confirmed to be inadequate for functional studies at 3 T. Measurements of main magnetic field variations across the amygdalae suggest that such variations are equal in the craniocaudal and anterior-posterior directions, and slightly lower in the mediolateral direction, with this and other considerations leading us to conclude an oblique axial orientation to be most suitable. In-plane resolution of approximately 1.7 mm was sufficient to recover signal in the area of the amygdalae. SNR was found to peak at a slice thickness of between 2.0 and 2.5 mm, dependent on the subject. T2* time in the amygdalae was measured with a standard EPI protocol to be 22 +/- 3 ms. Using the optimized (high resolution) EPI protocol proposed here, the measured T2* time increased to 48 +/- 2 ms (compared with 43 +/- 3 ms for a reference FLASH scan), only slightly lower than the cortex (49 +/- 2 ms measured with optimized EPI and 52 +/- 2 ms with FLASH). The FLASH measurement of 43 ms is taken to be a suitable effective echo time (TE(eff)) to achieve maximum BOLD sensitivity in the amygdalae. Time series data acquired with these parameters showed a 60% increase in SNR in the amygdala over that obtained with a standard low-resolution protocol and suggest sufficient SNR and BOLD sensitivity to make functional studies feasible. Arteries, but no substantial draining veins, were found in high-resolution BOLD venograms of the region. Our results indicate that EPI protocols need to be carefully optimized for structures of interest if reliable results from single subjects are to be established in this brain region.

Functional MRI with variable echo time acquisition

A new functional MRI protocol that integrates variable echo time (TE) acquisition and a block-design paradigm is proposed and evaluated with finger-tapping motor task. Simulations and experimental data show that the blood oxygenation level-dependent (BOLD) sensitivity achieved with this approach is comparable to that achieved using a conventional constant-TE protocol. The proposed variable-TE fMRI protocol provides valuable information that cannot be obtained with the constant-TE protocol. First, a field inhomogeneity map can be derived from the multi-TE data and used to correct EPI geometric distortions. Second, changes of T2* values due to the BOLD effect can be quantified. Third, for brain regions with pronounced susceptibility field gradients, the reduced BOLD sensitivity may be compensated for when the acquired multi-TE data are processed appropriately (e.g., with weighted summation). Fourth, large venules and veins may possibly be identified (depending on the vessel orientation and volume fraction) by evaluating the phase values of the multi-TE data. Finally, magnetic field drift over time can be measured from dynamic field maps available with this protocol.

fMRI Activation in Continuous and Spike-triggered EEG-fMRI Studies of Epileptic Spikes

PURPOSE

To evaluate functional magnetic resonance imaging (fMRI) with simultaneous EEG for finding metabolic sources of epileptic spikes. To find the localizing value of activated regions and factors influencing fMRI responses.

METHODS

Patients with focal epilepsy and frequent spikes were subjected to spike-triggered or continuous fMRI with simultaneous EEG. Results were analyzed in terms of fMRI activation, concordance with the location of EEG spiking and anatomic MRI abnormalities, and other EEG and clinical variables. In four patients, results also were compared with those of intracerebral EEG.

RESULTS

Forty-eight studies were performed on 38 patients. Seventeen studies were not analyzed, primarily because no spikes occurred during scanning. Activation was obtained in 39% of 31 studies, with an activation volume of 2.55 +/- 4.84 cc. Activated regions were concordant with EEG localization in almost all studies and confirmed by intracerebral EEG in four patients. Forty percent of patients without an MRI lesion showed activation; 37.5% of patients with a lesion had an activation; the activation was near or inside the lesion. Bursts of spikes were more likely to generate an fMRI response than were isolated spikes (76 vs. 11%; p < 0.05).

CONCLUSIONS

Combining EEG and fMRI in focal epilepsy yields regions of activation that are presumably the source of spiking activity. These regions are highly linked with epileptic foci and epileptogenic lesions in a significant number of patients. Activation also is found in patients with no visible MRI lesion. Intracerebral recordings largely confirm that these activation regions represent epileptogenic areas. It is still unclear why many patients show no activation.

Studying spontaneous EEG activity with fMRI

The multifaceted technological challenge of acquiring simultaneous EEG-correlated fMRI data has now been met and the potential exists for mapping electrophysiological activity with unprecedented spatio-temporal resolution. Work has already begun on studying a host of spontaneous EEG phenomena ranging from alpha rhythm and sleep patterns to epileptiform discharges and seizures, with far reaching clinical implications. However, the transformation of EEG data into linear models suitable for voxel-based statistical hypothesis testing is central to the endeavour. This in turn is predicated upon a number of assumptions regarding the manner in which the generators of EEG phenomena may engender changes in the blood oxygen level dependent (BOLD) signal. Furthermore, important limitations are posed by a set of considerations quite unique to 'paradigmless fMRI'. Here, these issues are assembled and explored to provide an overview of progress made and unresolved questions, with an emphasis on applications in epilepsy.

Selection of voxel size and slice orientation for fMRI in the presence of susceptibility field gradients: application to imaging of the amygdala

The impact of voxel geometry on the blood oxygenation level-dependent (BOLD) signal detectability in the presence of field inhomogeneity is assessed and a quantitative approach to selecting appropriate voxel geometry is developed in this report. Application of the developed technique to BOLD sensitivity improvement of the human amygdala is presented. Field inhomogeneity was measured experimentally at 1.5 T and 3 T and the dominant susceptibility field gradient in the human amygdala was observed approximately along the superior-inferior direction. Based on the field mapping studies, an optimal selection for the slice orientation would be an oblique pseudo-coronal plane with its frequency-encoding direction parallel to the field gradient measured from each subject. Experimentally this was confirmed by comparing the normalized standard deviation of time-series echo-planar imaging signals acquired with different slice orientations, in the absence of a functional stimulus. A further confirmation with a carefully designed functional magnetic resonance imaging study is needed. Although the BOLD sensitivity may generally be improved by a voxel size commensurable with the activation volume, our quantitative analysis shows that the optimal voxel size also depends on the susceptibility field gradient and is usually smaller than the activation volume. The predicted phenomenon is confirmed with a hybrid simulation, in which the functional activation was mathematically added to the experimentally acquired rest-period echo-planar imaging data.

Functional MRI using sensitivity-encoded echo planar imaging (SENSE-EPI)

Parallel imaging methods become increasingly available on clinical MR scanners. To investigate the potential of sensitivity-encoded single-shot EPI (SENSE-EPI) for functional MRI, five imaging protocols at different SENSE reduction factors (R) and matrix sizes were compared with respect to their noise characteristics and their sensitivity toward functional activation in a motor task examination. At constant echo times, SENSE-EPI was either used to shorten the single volume acquisition times (TR(min)) at matrix size 128 x 100 (22 slices) from 3.9 s (no SENSE) to 2.0 s at R = 3, or to increase the matrix size to 192 x 153 (22 slices), resulting in TR(min) = 5.3 s for R = 2 or TR(min) = 3.4 s for R = 3. At the lower resolution, the bisection of echo train length (R = 2) substantially reduced distortions and blurring, while signal-to-noise and statistical power (measured by cluster size and maximum t value per unit time) were hardly reduced. At R = 3 the additional gain in speed and distortion reduction was quite small, while signal-to-noise and statistical power dropped significantly. With enhanced spatial resolution the time course signal-to-noise was better than expected from theory for purely thermal noise because of a reduced contribution of physiological noise, and statistical power almost reached that of the regular, low-resolution single-shot EPI, with a slight drop off toward R = 3. Thus, SENSE-EPI allows to substantially increase speed and spatial resolution in fMRI. At SENSE reduction factors up to R = 2, the potential drawbacks regarding signal-to-noise and statistical power are almost negligible.

Language systems in normal and aphasic human subjects: functional imaging studies and inferences from animal studies

The old neurological model of language, based on the writings of Broca, Wernicke and Lichtheim in the 19th century, is now undergoing major modifications. Observations on the anatomy and physiology of auditory processing in non-human primates are giving strong indicators as to how speech perception is organised in the human brain. In the light of this knowledge, functional activation studies with positron emission tomography (PET) and functional magnetic resonance imaging (fMRI) are achieving a new level of precision in the investigation of language organisation in the human brain, in a manner not possible with observations on patients with aphasic stroke. Although the use of functional imaging to inform methods of improving aphasia rehabilitation remains underdeveloped, there are strong indicators that this methodology will provide the means to research a very imperfectly developed area of therapy.

Functional magnetic resonance imaging technology and traumatic brain injury rehabilitation: guidelines for methodological and conceptual pitfalls

OBJECTIVES

To illuminate the current methodological and conceptual pitfalls inherent in conducting functional magnetic resonance imaging (fMRI) research with individuals who have sustained traumatic brain injury (TBI) and to discuss appropriate remedies. The aim is describe fMRI research, its limitations, and how to best use this technology to examine TBI.

DISCUSSION

The topics discussed in this article include issues regarding signal detection, brain activation measurement, head movement, and sources of signal artifact. Issues surrounding data interpretation and the importance of analyzing the brain as a connected neural network is also discussed. Finally, problems with spatial normalization when examining individuals with TBI are reviewed.

CONCLUSIONS

To date, there is a scarcity of research applying fMRI technology to the study of TBI. However, because it is a noninvasive procedure with high availability in hospital settings across the country, the next decade of TBI research will likely include a proliferation of this form of investigation. At this time, much work is needed to better understand how to optimally use this technology to examine the effects of TBI on behavior. For fMRI to enhance TBI research it will be imperative to establish valid research protocols and reliable methods of data interpretation.

Image distortion correction in fMRI: A quantitative evaluation

A well-recognized problem with the echo-planar imaging (EPI) technique most commonly used for functional magnetic resonance imaging (fMRI) studies is geometric distortion caused by magnetic field inhomogeneity. This makes it difficult to achieve an accurate registration between a functional activation map calculated from an EPI time series and an undistorted, high resolution anatomical image. A correction method based on mapping the spatial distribution of field inhomogeneities can be used to reduce these distortions. This approach is attractive in its simplicity but requires postprocessing to improve the robustness of the acquired field map and reduce any secondary artifacts. Furthermore, the distribution of the internal magnetic field throughout the head is position dependent resulting in an interaction between distortion and head motion. Therefore, a single field map may not be sufficient to correct for the distortions throughout a whole fMRI time series. In this paper we present a quantitative evaluation of image distortion correction for fMRI at 2T. We assess (i) methods for the acquisition and calculation of field maps, (ii) the effect of image distortion correction on the coregistration between anatomical and functional images, and (iii) the interaction between distortion and head motion, assessing the feasibility of using field maps to reduce this effect. We propose that field maps with acceptable noise levels can be generated easily using a dual echo-time EPI sequence and demonstrate the importance of distortion correction for anatomical coregistration, even for small distortions. Using a dual echo-time series to generate a unique field map at each time point, we characterize the interaction between head motion and geometric distortion. However, we suggest that the variance between successively measured field maps introduces additional unwanted variance in the voxel time-series and is therefore not adequate to correct for time-varying distortions.