Faculty Bibliography

Parallel transmission has emerged as an efficient means for implementing multidimensional spatially selective radiofrequency excitation pulses. To date, most theoretical and experimental work on parallel transmission radiofrequency (RF) pulse design is based on the small-tip-angle approximation to the Bloch equation. The small-tip-angle, while mathematically compact, is not an exact solution and leads to significant errors when large-tip-angle pulses are designed. Methods have been proposed to overcome the limitations of the small-tip-angle using regularized least-square optimization or optimal control algorithms. These methods, however, are based on further approximations to the Bloch equation or require the use of general purpose algorithms that do not capitalize fully on the dynamics of the physical model at hand. In this article, a novel algorithm for large-tip-angle parallel transmission pulse design is proposed. The algorithm relies on a perturbation analysis of the Bloch equation and it depicts the relationship between the excited magnetization, its deviation from the target pattern and the desired pulses. Simulations and experiments are used to validate the proposed method on a 7 T 8-channel transmit array. The results demonstrate that the perturbation analysis algorithm provides a fast and accurate approach for multidimensional large-tip-angle pulse design, especially when large acceleration factors and/or echo-planar trajectories are used.

The aims of the present study were as follows: (i) to perform the first (87)Rb MRI in live rats with focal ischemic stroke; and (ii) to test the hypothesis that K(+) egress from the brain in this model is quantifiable in individual animals by high-field (7-T) K/Rb substitution MRI. Rats preloaded with dietary Rb(+) (resulting in Rb/(K + Rb) replacement ratios of 0.1-0.2 in the brain) were subjected to permanent occlusion of the middle cerebral artery, and (87)Rb MRI was implemented with 13-min temporal resolution using a dedicated RF coil and a spiral ultrashort-TE sequence (TR/TE = 3/0.07 ms). The ischemic core was localized by apparent diffusion coefficient mapping, by microtubule-associated protein-2 immunohistochemistry and by changes in surface reflectivity. [K], [Na] and [Rb] were determined independently in the micropunched samples by post-mortem flame photometry. Both techniques were generally in agreement in the nonischemic cortex; however, the MRI-assessed [K(+) + Rb(+)] drop in ischemic brain was less pronounced (average efflux rate of 4.8 +/- 0.2 nEq/mm(3) /h versus 10 +/- 1 nEq/mm(3)/h by flame photometry; p < 0.0001). The use of higher field gradients for better spatial resolution, and hence more accurate quantification, is suggested.

It is unclear whether ethnic differences exist in adipose tissue (AT) and skeletal muscle (SM) distribution in black and white youth. Investigation into the pattern of AT and SM distribution in black versus white youth may provide insight into the previously reported health disparities between these ethnicities. Therefore, we examined total and regional AT and SM in overweight black and white boys. The study sample included overweight black (n = 19) and white (n = 21) boys (11-18 yr, BMI >/= 85th) whose body composition was evaluated using whole-body MRI. Despite similar age, Tanner stage, and BMI, black boys had significantly (P < .05) less visceral AT than white boys and more (P < .05) total and lower-body subcutaneous AT (SAT) in both absolute (kg) and relative (%) terms. There was a main effect (P < .05) of ethnicity on the relationship between total and regional AT, such that for a given amount of total body AT (kg), black boys had a greater (P < .05) lower-body SAT and less visceral AT than their white peers. For a given amount of total SM, black boys had more (P < .05) SM in the thigh. Compared with overweight white boys, overweight black boys have less visceral fat, more subcutaneous fat, and more thigh skeletal muscle.

The authors have applied high-definition fiber tracking (HDFT) to the resection of an intraparenchymal dermoid cyst by using a minimally invasive endoscopic port. The lesion was located within the mesial frontal lobe, septal area, hypothalamus, and suprasellar recess. Using high-dimensional (256 directions) diffusion imaging, more than 250,000 fiber tracts were imaged before and after surgery. Trajectory planning using HDFT in a computer model was used to facilitate cannulation of the cyst with the endoscopic port. Analysis of the proposed initial surgical route was overlaid onto the fiber tracts and was predicted to produce substantial disruption to prefrontal projection fibers (anterior limb of the internal capsule) and the cingulum. Adjustment of the cannulation entry point 1 cm medially was predicted to cross the corpus callosum instead of the anterior limb of the internal capsule or the cingulum. Following cyst resection performed using endoscopic port surgery, postoperative imaging demonstrated accurate cannulation of the lesion, with improved quantitative signal from both the anterior limb of the internal capsule and the cingulum. The observed fiber preservation from the cingulum and the anterior limb of the internal capsule, with minor injury to the corpus callosum, was in close agreement with preoperative trajectory modeling. Comparison of pre- and postoperative HDFT data facilitated quantification of the benefits and costs of the surgical trajectory. Future studies will help to determine whether HDFT combined with endoscopic port surgery facilitates anatomical and functional preservation in such challenging cases.

Disorganization of collagen fibers is a sign of early-stage cartilage degeneration in osteoarthritic knees. Water molecules trapped within well-organized collagen fibrils would be sensitive to collagen alterations. Multicomponent effective transverse relaxation (T2*) mapping with ultrashort echo time acquisitions is here proposed to probe short T(2) relaxations in those trapped water molecules. Six human tibial plateau explants were scanned on a 3T MRI scanner using a home-developed ultrashort echo time sequence with echo times optimized via Monte Carlo simulations. Time constants and component intensities of T2* decays were calculated at individual pixels, using the nonnegative least squares algorithm. Four T2*-decay types were found: 99% of cartilage pixels having mono-, bi-, or nonexponential decay, and 1% showing triexponential decay. Short T2* was mainly in 1-6 ms, while long T2* was approximately 22 ms. A map of decay types presented spatial distribution of these T2* decays. These results showed the technical feasibility of multicomponent T2* mapping on human knee cartilage explants.

Hypertension, diabetes, obesity, and dyslipidemia are risk factors that characterize metabolic syndrome (MetS), which increases the risk for stroke by 40%. In a preliminary study, our aim was to evaluate cerebrovascular reactivity and oxygen metabolism in subjects free of vascular disease but with one or more of these risk factors. Volunteers (n=15) 59+/-15 (mean+/-SD)years of age clear of cerebrovascular disease by magnetic resonance angiography but with one or more risk factors were studied by quantitative positron emission tomography for measure ment of cerebral blood flow, oxygen consumption, oxygen extraction fraction (OEF), and acetazolamide cerebrovascular reactivity. Eight of ten subjects with MetS risk factors had OEF >50%. None of the five without risk factors had OEF >50%. The presence of MetS risk factors was highly correlated with OEF >50% by Fisher's exact test (p<0.007). The increase in OEF was significantly (P<0.001) correlated with cerebral metabolic rate for oxygen. Increased OEF was not associated with compromised acetazolamide cerebrovascular reactivity. Subjects with one or more MetS risk factors are characterized by increased cerebral oxygen consumption and ischemic stress, which may be related to increased risk of cerebrovascular disease and stroke.

High-resolution (approximately 0.22 mm) images are preferably acquired on whole-body 7T scanners to visualize minianatomic structures in human brain. They usually need long acquisition time ( approximately 12 min) in three-dimensional scans, even with both parallel imaging and partial Fourier samplings. The combined use of both fast imaging techniques, however, leads to occasionally visible undersampling artifacts. Spiral imaging has an advantage in acquisition efficiency over rectangular sampling, but its implementations are limited due to image blurring caused by a strong off-resonance effect at 7T. This study proposes a solution for minimizing image blurring while keeping spiral efficient. Image blurring at 7T was, first, quantitatively investigated using computer simulations and point-spread functions. A combined use of multishot spirals and ultrashort echo time acquisitions was then employed to minimize off-resonance-induced image blurring. Experiments on phantoms and healthy subjects were performed on a whole-body 7T scanner to show the performance of the proposed method. The three-dimensional brain images of human subjects were obtained at echo time = 1.18 ms, resolution = 0.22 mm (field of view = 220 mm, matrix size = 1024), and in-plane spiral shots = 128, using a home-developed ultrashort echo time sequence (acquisition-weighted stack of spirals). The total acquisition time for 60 partitions at pulse repetition time = 100 ms was 12.8 min without use of parallel imaging and partial Fourier sampling. The blurring in these spiral images was minimized to a level comparable to that in gradient-echo images with rectangular acquisitions, while the spiral acquisition efficiency was maintained at eight. These images showed that spiral imaging at 7T was feasible.

The delineation of early infarction in large gyrencephalic brain cannot be accomplished with triphenyl-tetrazolium chloride (TTC) due to its limitations in the early phase, nor can it be identified with microtubule-associated protein 2 (MAP2) immunohistochemistry, due to the fragility of large thin sections. We hypothesize that MAP2 immunostaining of thick brain sections can accurately identify early ischemia in the entire monkey brain. Using ischemic brains of one rat and three monkeys, a thick-section MAP2 immunostaining protocol was developed to outline the infarct region over the entire non-human primate brain. Comparison of adjacent thick and thin sections in a rat brain indicated complete correspondence between ischemic regions (100.4mm(3)+/-1.2%, n=7, p=0.44). Thick sections in monkey brain possessed the increased structural stability necessary for the extensive MAP2 immunostaining procedure permitting quantification of the ischemic region as a percent of total monkey brain, giving infarct volumes of 11.4, 16.3, and 19.0% of total brain. Stacked 2D images of the intact thick brain tissue sections provided a 3D representation for comparison to MRI images. The infarct volume of 16.1cm(3) from the MAP2 sections registered with MRI images agreed well with the volume calculated directly from the stained sections of 16.6 cm(3). Thick brain tissue section MAP2 immunostaining provides a new method for determining infarct volume over the entire brain at early time points in a non-human primate model of ischemic stroke

PURPOSE: To demonstrate the use of sodium MRI for measuring the time course of tissue sodium concentration (TSC) in a nonhuman primate model of reversible focal brain ischemia. MATERIALS AND METHODS: Reversible endovascular focal brain ischemia was induced in nonhuman primates (n = 4), and sodium MRI was performed on a 3 Tesla scanner for monitoring changes in TSC during both the middle cerebral artery (MCA) occlusion and MCA reperfusion portions of the experiment. RESULTS: The TSC increased linearly in the ischemic tissue during MCA occlusion (ranging from a mean TSC increase of 5.44%/h to 7.15%/h across the four subjects), and then there was a statistically significant change from a positive TSC slope during MCA occlusion to a TSC slope after MCA reperfusion that was not statistically different from zero. The linear increase in sodium MRI during brain ischemia was used to estimate the stroke onset time to within 0.45 h in each of the four subjects (with a maximum 95% confidence interval of +/- 1.147 h). CONCLUSION: The data indicate that sodium MRI increases linearly during brain ischemia, and that this increase is stopped by tissue reperfusion within 5.4 h after stroke onset