W Kuan

Study-specific EPI template improves group analysis in functional MRI of young and older adults

Spatial normalization to a common coordinate space, e.g. via the Montreal Neurological Institute (MNI) brain template, is an essential step of analyzing multi-subject functional MRI (fMRI) datasets. The imperfect compensation for individual regional discrepancies during spatial transformation, which could potentially introduce localization errors of the activation foci and/or reduce the detection sensitivity, may be minimized if a template specifically designed for the subjects of a study is applied. In this fMRI study, we proposed and evaluated the use of a study-specific template (SST) based on the mean of individually normalized echo-planar images for group data analysis. A hand flexion and a word generation tasks were performed on young volunteers in experiment 1. Comparing with the MNI template approach, greater t-values of local maxima and activated voxels were detected within volume-of-interests (VOIs) with the SST approach in both tasks. Moreover, the SST approach reduced Euclidean distances between activation foci of individuals and group by 1.52 mm in motor fMRI and 5.84 mm in language fMRI. Similar results were obtained with or without spatial smoothing of the echo-planar images. Experiment 2 further examined these two approaches in older adults, in which volumetric differences between subjects are of great concerns. With a working memory task, the SST approach showed greater t-values of local maxima and activated voxels within the VOI of prefrontal gyrus. This study demonstrated that the SST resulted in more focused activation patterns and effectively improved the fMRI sensitivity, which suggested potentials of reducing number of subjects required for group analysis.

Breathhold-regulated blood oxygenation level-dependent (BOLD) MRI of human brain at 3 tesla.

To investigate the cerebrovascular response to repeated breathhold challenges using blood oxygenation level‐dependent (BOLD) MRI at 3T and compare the results with previous data at 1.5T.

Inflow effects on hemodynamic responses characterized by event-related fMRI using gradient-echo EPI sequences.

The purpose of this study is to determine whether blood inflow impacts the temporal behavior of BOLD-contrast fMRI signal changes in a typical event-related paradigm. The inflow contributions in the hemodynamic response to repeated single trials of short visual stimulation were assessed with a gradient-echo EPI sequence by altering the flip angle (FA) from 30 degrees to 90 degrees at a repetition time of 1 s. For each FA condition (30 degrees, 60 degrees, and 90 degrees), 30 trials were performed on 15 healthy volunteers on a 3T MRI scanner. Comparing the percent BOLD contrast, prominent inflow effects were found with statistical significance between the 90 degrees- and 30 degrees-FA conditions (0.73 +/- 0.15 versus 0.67 +/- 0.12%, p=0.028). BOLD responses with FA=30 degrees exhibited latencies significantly slower than those with FA=90 degrees (3.69 +/- 0.39 s versus 3.37 +/- 0.28 s, p=0.001). The falling time of the 30 degrees-FA responses was earlier but not statistically different from that of the 90 degrees-FA (8.17 +/- 1.04 s versus 8.03 +/- 1.15 s, p=0.3). Using a voxelwise analysis, the latency variations of the activated visual areas were determined at several contrast-to-noise ratio (CNR) levels (controlled by averaging different numbers of randomly selected trials). The latency variations from the 90 degrees-FA responses were greater at lower CNR but similar at higher CNR levels when comparing to the 30 degrees-FA ones. This study suggests that inflow effects contribute to the BOLD signal, resulting in hemodynamic response with shorter latency.

Variations in BOLD response latency estimated from event-related fMRI at 3T: Comparisons between gradient-echo and Spin-echo

Functional magnetic resonance imaging (fMRI) commonly uses gradient‐recalled echo (GRE) signals to detect regional hemodynamic variations originating from neural activities. While the spatial localization of activation shows promising applications, indexing temporal response remains a poor mechanism for detecting the timing of neural activity. Particularly, the hemodynamic response may fail to resolve sub‐second temporal differences between brain regions because of its signal origin or noise in data, or both. This study aimed at evaluating the performance of latency estimation using different fMRI techniques, with two event‐related experiments at 3T. Experiment I evaluated latency variations within the visual cortex and their relationship with contrast‐to‐noise ratios (CNRs) for GRE, spin echo (SE), and diffusion‐weighted SE (DWSE). Experiment II used delayed visual stimuli between two hemifields (delay time = 0, 250, and 500 ms, respectively) to assess the temporal resolving power of three protocols: GRETR1000, GRETR500, and SETR1000. The results of experiment I showed the earliest latency with DWSE, followed by SE, and then GRE. Latency variations decreased as CNR increased. However, similar variations were found between GRE and SE, when the latter had lower CNR. In experiment II, measured stimulus delays from all conditions were significantly correlated with preset stimulus delays. Inter‐subject variation in the measured delay was found to be greatest with GRETR1000, followed by GRETR500, and the least with SETR1000. Conclusively, blood oxygenation level‐dependent responses obtained from GRE exhibit greater CNR but no compromised latency variations in the visual cortex. SE is potentially capable of improving the performance of latency estimation, especially for group analysis.

SU‐FF‐I‐128: Effects of Slice Orientation and Parallel Acquisition On EPI‐Based PASL Perfusion Imaging in Areas with Susceptibility Artifact

Purpose GE EPI is commonly applied for acquiring perfusion MRI with arterial spin labeling (ASL). A key problem is the long echo train that causes problems in image quality due to the susceptibility artifact. This study aimed to focus on the orbitofrontal cortex (OFC) region and evaluate the effects of slice angle combined with parallel imaging technique on ASL image quality. Methods FAIR combined with a single‐shot GE EPI was applied with TI = 1400 ms, TR/TE/FA = 3000 ms /20 ms /90°, 5mm slice thickness, 7 slices, NEX = 15, FOV = 220 mm and matrix = 64. The slice tilt was set from −45°, 0 to +40° relative to the AC‐PC direction. Images were obtained both with and without applying the SENSE (factor = 2). Signal‐to‐noise ratio (SNR) and contrast was calculated from the nonselective (MNS) and the selective inversion image (MNS). Results The SNR became more stable when the parallel imaging was used (Fig.1). On the contrary, the susceptibility gradient caused a greater effect on the SNR without parallel acquisition with a maximum at −15° slice tilt. At two of the slice orientations the resulted images exhibited greater SNR in the OFC region due to reduced susceptibility artifact. The changes of SNR can be appreciated from the spatially normalized images showed in Fig. 2. Figure 3 demonstrated slight varied contrast at different slice orientation measured with parallel imaging. The highest value is about 1.8% at −15° slice tilt. Conclusion For the purpose of reducing signal losses in OFC, this study found the optimal slice tile without parallel imaging. However, parallel imaging may be preferable because it was less subject to slice orientation. Further studies with more subjects and anatomical areas are required to better understand the changes in inflow contrast as the slice tilt.

SU‐FF‐I‐126: Varied Vasomotor Responses Among Brain Territories in Unilateral ICA Stenosis Patients Studied Using Breath‐Hold BOLD MRI

Introduction: Patients with significant internal carotid artery (ICA) stenosis increase the incidence of ischemic stroke and benefit from carotid interventions, such as carotid angioplasty with stenting (CAS). Hyperperfusion syndrome usually causes disastrous outcome if occurred after carotid interventions. In recent studies, impaired cerebral vasoreactivity (CVR) is considered as one of the predictors for hyperperfusion syndrome. Hypercapnia stresses such as breath holding result in vasodilatation in human subjects. Thus combining breath holding and BOLD MRI may be a useful method to characterize the impaired CVR in patients prone to hyperperfusion after CAS. Most recently, Leoni et. al. found different hemodynamic response among different brain territories in normal subjects during breath holding. The aim of this study was to evaluate the varied hemodynamic responses among different brain territories in unilateral ICA stenosis patients using the same MRI method. Methods: Six patients with unilateral ICA stenosis participated in this study and were asked to hold their breath for 15 sec. Three cycles of the breath holding task were performed. A single‐shot T2* gradient‐echo EPI sequence was used for BOLD measurements at a 1.5T scanner. Twenty axial slices were acquired to cover the whole brain. Echo‐planar images were spatial normalized to the MNI template and averaged BOLD signal time curves were drawn from six regions of interest, i.e. ACA, MCA and PCA on both sides. To show the extent of voxels with significant BOLD signal changes, a correlation analysis (p<0.05, corrected) was applied using the averaged signal time curve from the whole braintissue.Results and Conclusion: In summary, our results showed the impaired CVR in patients with ICA stenosis was territory dependent (Fig. 2). Detailed analysis of the hemodynamic responses measured from breath‐holding BOLD MRI may help predict hyerperfusion syndrome after CAS.

SU‐GG‐I‐120: Improved BOLD Signal Detectability of Short Breath Holding Duration at 3T: A Comparison with 1.5T

Purpose: Hypercapnia stresses such as breath holding (BH) have been showed to result in vasodilatation. Karstrup compared cerebrovascular reactivity between CO 2 inhalation and BH, and found good correlation in BOLD intensity changes. This suggested possibility of using a simple BH task to evaluate abnormal vasomotor functions in clinical. A previous study has focused on the detectability of BOLD signal changes during short BH durations at 1.5 T and found a 10‐s BH could be detected but a 20‐s is suggested for clinical applications. As the 3T scanners recently became more available, we re‐evaluated this issue and aimed to find optimal BH duration for clinical use at 3 T. Method and Materials: A T2* gradient‐echo EPI sequence was used for BOLD measurements at a 3T scanner.Imaging parameters were TR/TE/flip angle= 3000ms/35ms/90°. Five BH duration, from 5 to 30 s, were experimented. To quantitatively examine BOLD responses, a convolution model was used to fit the time curves. Maximum signal change, FWHM and onset time (defined as the time to the first half maximum) were determined from the fitted curve and compared with the previously published results for 1.5 T. To explore the number of voxels with significant BOLD signal changes (p<0.05, corrected), a correlation analysis was applied. Results: Our result shows the fractional activation volume increases with the BH duration and reaches the plateau at 15‐s which is earlier than the data for 1.5T. Conclusion: Our results suggested the BOLD imaging at 3T was more sensitive for detecting BH induced cerebrovacular reactivity than at 1.5 T. The BH duration of 15 s or above is recommended for clinical use at 3 T, which is an improvement from 20‐s BH at 1.5 T.

SU‐GG‐I‐129: Temporal Variations of Hemodynamic Responses of BOLD FMRI at 3T: Spin Echo Vs. Gradient Echo

Purpose: Because of its superior sensitivity, gradient‐echo (GE)‐BOLD signal is currently the most widely used contrast for fMRI. However, several works have suggested that the spin‐echo (SE)‐BOLD can improved spatial localization of neural activity due to its greater weighting to smaller vessels. We hypothesized that the temporal variations of measuredhemodynamic responses (HR), a basic factor of temporal resolution of fMRI, directly related to spatial accuracy of the methods. Therefore this study compared GE‐ to SE‐BOLD in time course and onset time variations of the HR. Method and Materials: Five normal volunteers participated in this study at a 3.0T MRI scanner. The paradigm consisted of 30 trials each with 1‐s visual stimulation and 15‐s fixation. Both the GE and SE experiments used echo‐planar readouts with TR/TE/FA= 1000ms/35ms/64 and TR/TE=1000ms/72ms, respectively. Eight slices with 5‐mm thickness were acquired to cover visual areas. For each activated voxel, the time series were extracted and averaged randomly across 30, 20 and 10 trials, from which the onset times and CNRs were determined with curve fitting to a gamma variate function. Results: Variance of the onset time decreased with CNR increased. At the same CNR levels, we observed significantly smaller onset variances for SE compared to GE. Decreased sensitivities were noted for the SE when comparing the number of activated voxels with the GE results. The mean time courses showed earlier onset for the SE response as compared to the GE.Discussion: We observed earlier onset times with smaller within‐region variances in the SE‐ than in the GE‐BOLD. Since the SE technique gives more weighting to the extravascular contributions around small vessels, we suggest that it could more accurately detect the onset time related to neuronal events. When comparing at the same CNR levels, the smaller latency variations of the SE measurement demonstrated its superior spatiotemporal natures.

SU‐EE‐A4‐05: Model‐Independent Analysis of Cerebrovascular Reactivity MRI During Breath‐Holding

Purpose: Both cerebral tumor and cardiovascular disease may compromise cerebrovascular reactivity due to abnormal angiogenesis or blood supply. Functional MRI during hypercapnia stress, such as breath‐holding (BH), has been used to assess the cerebrovascular reactivity. However, in clinical settings, patients may not be able to hold their breath well, which can reduce the sensitivity for detection when an fMRI model‐based analysis is applied. In view of the global effect induced by breath‐holding, this study proposed a model‐independent analysis, using the whole‐brain‐averaged time curve as a reference function(WBRF), and compared with the results using the canonical hrf(CHRF). Methods: T2*‐weighted images were acquired using a single‐shot GE EPI sequence on 1.5T MR scanner, with TR/TE=3000ms/60ms. Four experiments were performed on two normal volunteers: (1) BH after deep inspiration, (2) after an instruction, BH after the end of the natural expiration, (3) self‐paced BH by button pressing and (4) take in a short breath during the BH duration. For model‐independent analysis, the WBRF was the averaged time curve over the whole braintissue. For model‐dependent analysis, the CHRF was adopted from the SPM2. Voxels with statistically increased signals were determined (p<0.001), and then the fractional activation volume (activated volume/whole volume) was calculated. Results: In all cases, the WBRF was able to show the real breath‐holding paradigm performed by the subjects. From the four experiments, the fractional activation volumes by model‐independent analysis were 62, 53, 57, 54% and 58, 4, 5, 12% by model‐dependent analysis, respectively. Conclusion: The results illustrated model‐independent analysis is a more reliable approach. This is caused by the fact that WBRF better reflects the true beath‐holding situation then CHRF. Due to the impaired patient performance in clinical applications, the method proposed by this study can help increasing the sensitivity of cerebovascular reactivity MRI.

SU‐FF‐I‐89: Effects of EPI Slice Angle, Slice Thickness and Phase Encoding Direction On FMRI Sensitivity in Areas with Susceptibility Artifact

Purpose: A key problem of using EPI for BOLD fMRI is the field inhomogeneities near air/tissue interfaces, or the susceptibility artifact. Therefore, fMRI studies in brain areas such as orbitofrontal cortex (OFC) and temporal lobes (TL) may suffer from signal dropouts and spatial distortions. This study aimed to determine the optimal EPI slice angle, slice thickness and phase‐encoding (PE) direction for the reduction of BOLD sensitivity (BS) losses in TL and OFC. Methods: The study was performed on five healthy volunteers using a 1.5‐T MRIscanner. A single‐shot GE EPI sequence was applied with TR/TE/FA = 3000ms/60ms/900, FOV = 220mm and matrix = 64. The slice tilt was set from −45° to +30° (in 6 steps of 15°) relative to the AC‐PC direction, 35 or 24 slices for the slice thickness of 3 or 6 mm, respectively, and the PE direction either left‐right or anterior‐posterior, which resulted in 24 parameter combinations. BS maps were calculated with corresponding phase maps for each combination and normalized to a brain template using SPM2 for comparison. Results: Significant BS gain (>15%) were observed in several parameter combinations, when comparing to the baseline BS (no slice tilt, slice thickness= 6mm, PE= anterior‐posterior). The optimal parameters that introduced most voxels with such BS gain were +30° slice tile, slice thickness= 3mm and PE = anterior‐posterior in the OFC, and −45° slice tile, slice thickness= 6mm and PE = left‐right in the TL. Conclusion: For the purpose of reducing BS losses in OFC and TL, this study provided the optimal EPI parameters that can be easily adopted in a clinical 1.5T MR scanner. For the PE direction, however, it is better to be kept in the anterior‐posterior direction to avoid peripheral nerve stimulation. Similar studies at 3T are currently under investigation.