Search

Thesis (S.M.)--Massachusetts Institute of Technology, Dept. of Electrical Engineering and Computer Science, 2010., Includes bibliographical references (p. 45-48)., Venous oxygen saturation (Yv) in cerebral veins and the cerebral metabolic rate of oxygen (CMRO₂) are important indicators for brain function and disease. Phase-susceptibility measurements in magnetic resonance imaging (MRI) have been used to quantify Yv in candidate cerebral veins. However, currently there is no method to quantify regional CMRO₂ using MRI. Here we propose a novel technique to quantify CMRO₂ from independent MRI estimates of Yv and cerebral blood flow (CBF). Our approach used standard gradient-echo (GRE) and arterial spin labeling (ASL) to make these measurements. Results for in vivo Y, and CMRO₂ estimates on human subjects are presented from application of our technique at 3 Tesla (3T). We also extended our method to high-field human imaging at 7 Tesla (7T), which allows us to take advantage of improved signal-to-noise ratio (SNR) for the same scan duration to achieve higherresolution analysis of vessels of interest. While the higher field strength poses additional challenges, such as increased main field and excitation field inhomogeneities as well as more severe susceptibility artifacts, initial results suggest that substantial benefits can be realized with phase-based regional oxygen metabolism in MRI at high field., by Audrey Peiwen Fan., S.M.

Thesis (Ph. D.)--Harvard-MIT Division of Health Sciences and Technology, 2010., This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections., Cataloged from student submitted PDF version of thesis., Includes bibliographical references (p. 120-128)., Oxygen consumption is an essential process of the functioning brain. The rate at which the brain consumes oxygen is known as the cerebral metabolic rate of oxygen (CMRO₂). CMRO₂ is intimately related to brain health and function, and will change in settings of disease and functional activation. Accurate CMRO₂ measurement will enable detailed investigation of neuropathology and facilitate our understanding of the brain's underlying functional architecture. Despite the importance of CMRO₂ in both clinical and basic neuroscience settings, a robust CMRO₂ mapping technique amenable to functional and clinical MRI has not been established. To address this issue, a novel method called QUantitative Imaging of eXtraction of Oxygen and TIssue Consumption, or QUIXOTIC, is introduced. The key innovation in QUIXOTIC is the use of velocity-selective spin labeling to isolate MR signal exclusively from post-capillary venular blood on a voxel-by-voxel basis. This isolated signal can be related to venular oxygen saturation, oxygen extraction fraction, and ultimately CMRO₂. This thesis first explores fundamental theory behind the QUIXOTIC technique, including design of a novel MRI pulse sequence, explanation of the principal sequence parameters, and results from initial human experiences. A human trial follows, in which QUIXOTIC is used to measure cortical gray matter CMRO₂ in ten healthy volunteers., (cont.) QUIXOTIC-measured CMRO₂ is found to be within the expected physiological range and is comparable to values reported by other techniques. QUIXOTIC is then applied to evaluate CMRO₂ response to carbon-dioxide-induced hypercapnia in awake humans. In this study, CMRO₂ is observed to decrease in response to mild hypercapnia. Finally, pilot studies that show feasibility of QUIXOTIC-based functional MRI (fMRI) and so-called "turbo" QUIXOTIC are presented and discussed., by Divya Sanam Bolar., Ph.D.

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Electrical Engineering and Computer Science, 2008., Includes bibliographical references (p. 219-233)., The hydrodynamics of magnetic fluids, often termed ferrofluids, has been an active area of research since the mid 1960s. However, it is only in the past twenty years that these fluids have begun to be used in magnetic resonance imaging (MRI) where they have found application as contrast agents, improving the contrast to noise ratio of the received MR intensity. The major goal of this work was to assess the feasibility of modulating the signal intensity in MRI in regions near superparamagnetic iron oxide contrast agents by applying a rotating magnetic field. Analytical study and numerical simulations of the effects of such a rotating magnetic field are presented. Simulated variations in fluid magnetization are achieved by changing the frequency of the rotating magnetic field. A linearization of the equilibrium magnetization Langevin relation results in a small signal magnetic susceptibility tensor that can be used to influence fluid flow. The feasibility of the approach under the real-life constraints of MR imaging are assessed. Also examined is the potential for coupling changes in magnetization due to rotating fields with hypothermia treatment of cancerous tissue. In addition to theoretical and simulated analysis, considerable experimental work was undertaken at the MRI systems of the HST Athinoula A. Martinos Center for Biomedical Imaging at Massachusetts General Hospital, Charlestown, Massachusetts. This work evaluated the passive behavior of both commercial ferrofluids and MRI contrast agents. The ability of MRI to serve as a highly accurate indicator of the fluid's physical and magnetic properties is shown. The thesis represents the first investigation into the use of the hydrodynamic properties of magnetic fluid in the presence of a rotating field as a mechanism for contrast in MRI. The results show the possibility of varying MR image intensity under careful selection of field conditions., by Pádraig J. Cantillon-Murphy., Ph.D.

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Electrical Engineering and Computer Science, 2008., This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections., Includes bibliographical references (p. 153-158)., The focus of this dissertation is on the algorithm design, implementation, and validation of parallel transmission technology in Magnetic Resonance Imaging (MRI). Novel algorithms are proposed which yield excellent excitation control, low RF power requirements, methods that extend to non-linear large-flip-angle excitation, as well as a new algorithm for simultaneous spectral and spatial excitation critical to quantification of low-SNR brain metabolites in MR spectroscopic imaging. For testing and validation, these methods were implemented on a newly developed parallel transmission platform on both 3 T and 7 T MRI scanners to demonstrate the ability of these methods for highfidelity B1+ mitigation, first by excitation of phantoms and then by human imaging. Further, spatially tailored RF pulses were demonstrated beyond conventional slice- or slab-selective excitation., by Kawin Setsompop., Ph.D.

Thesis (S.M.)--Massachusetts Institute of Technology, Dept. of Electrical Engineering and Computer Science, 2009., Includes bibliographical references (p. 59-61)., While magnetic resonance imaging (MRI) derives its signal from protons in water, additional and potentially important biochemical compounds are detectable in vivo within the proton spectrum. The detection and mapping of these much weaker signals is known as magnetic resonance spectroscopy or spectroscopic imaging. Among the complicating factors for this modality applied for human clinical imaging are limited chemical-shift dispersion and J-coupling that cause spectral overlap and complicated spectral shapes that limit detection and separation of several brain metabolites using MR spectroscopic imaging. Existing techniques for improved detection include so-called 2D spectroscopy, where additional encoding steps aid in the separation of compounds with overlapping chemical shift. This is achieved by collecting spectral data over a range of timing parameters and introducing an additional frequency axis, fl. Spectral editing techniques attempt to enhance metabolite signal detection along fl in order to resolve specific metabolite signals. While these techniques have been shown to improve signal separation, they carry a penalty in scan time and more complicated reconstruction compared to conventional spectroscopy, and are often prohibitive when combined with spectroscopic imaging. The task of this thesis is to characterize and optimize existing 2D spectroscopy techniques for improved metabolite resolution with reduced number of timing steps through numerical simulation and experimental validation in phantoms., Trina Kok., S.M.

Thesis (S.M.)--Massachusetts Institute of Technology, Dept. of Electrical Engineering and Computer Science, 2007., Includes bibliographical references (leaves 36-38)., Head movements during MEG recordings in children may lead to inaccurate localization of brain activity. In this thesis, we examined the effects of head movements on source estimation in twenty children performing a simple auditory cognitive task. In addition, we tested the ability of a recently introduced spherical harmonic expansion method, signal space separation (SSS), to compensate for the effects of head movements on two source models: equivalent current dipoles (ECDs) and minimum norm estimates (MNE). In the majority of subjects, the goodness-of-fit of ECDs fit to the peak of the auditory N100m response was increased following the SSS correction compared with the averaged forward solution method proposed earlier. The spatial spread of ECDs as determined with a bootstrapping approach was also reduced after SSS correction. In addition, the MNE source estimates were spatially sharpened following SSS application, indicative of an increase in signal to noise ratio. Together these results suggest that SSS is an effective method to compensate for head movements in MEG recordings in children., by Daniel T. Wehner., S.M.

Thesis (M. Eng.)--Massachusetts Institute of Technology, Dept. of Electrical Engineering and Computer Science, 2007., Includes bibliographical references (p. 53)., Special magnetic resonance (MR) scans, such as spiral imaging and echo-planar imaging, require speed and gradient accuracy while putting high demands on the MR gradient system that may cause gradient distortion. Additionally, high field MR scans are prone to inhomogeneities that disturb the gradient system. Regardless of the source, gradient characterization provides a simple tool for distortion correction. An improved method, named the self-encoded slice selection algorithm, of characterizing the gradient system of the magnetic resonance system is proposed. It improves and combines the self-encode method and the direct slice selection method. The new approach is simple and fast, and allows for the measurement of waveform gradients that reach the system's limits. The technique is used to model the gradient system as a linear time-invariant transfer function through frequency-domain analysis and time-domain analysis. A transfer function model of the gradient system on the 3T Siemens Tim Trio scanner is presented here along with the characterization and analysis of common waveform gradients. Possible distortion correction approaches are also suggested., by Joseph Yitan Cheng., M.Eng.

Thesis (S.M.)--Massachusetts Institute of Technology, Dept. of Electrical Engineering and Computer Science, 2006., Includes bibliographical references (p. 71-73)., Conventional Magnetic Resonance Spectroscopic Imaging (MRSI) suffers from both low signal-to-noise (SNR), as well as long acquisition times. The development of high-fidelity gradient coils has opened opportunities for fast k-space encoding schemes that are already used in structural imaging. At the same time, receive-coil arrays using 4 and 23 channels have been developed and reported to produce improved SNR over conventional quadrature detection by single coils. Fast spectroscopic imaging algorithm using spiral k-space trajectories and multiple-channel coil arrays is proposed in order to overcome the long acquisition-time limitations of conventional MRSI., by Borjan Aleksandar Gagoski., S.M.

Thesis (S.M.)--Massachusetts Institute of Technology, Dept. of Electrical Engineering and Computer Science, 2007., Includes bibliographical references (p. 28-29)., Arterial spin labeling (ASL) is the only non-invasive magnetic resonance imaging (MRI) technique that allows absolute quantification of cerebral blood flow (CBF). It involves using radiofrequency pulses designed to invert the spins of water in arterial blood, effectively creating a magnetic bolus. This inverted blood can be considered an endogenous contrast agent; imaging as it traverses the vascular tree allows CBF measurements. Such types of experiments are especially useful for functional neuro-activation studies and in settings of neuropathology. Two flavors of ASL exist: continuous ASL and pulsed ASL. Pulsed ASL has the advantage of not requiring specialized imaging hardware, and can be performed using standard clinical scanners found in most hospitals. Pulsed ASL techniques, however, may yield inaccurate perfusion values and diminished perfusion sensitivity if appropriate labeling parameters are not chosen, particularly during global challenges such as hypercapnia. In this study, the accuracy of QUIPSS II (Quantitative Imaging of Perfusion using a Single Subtraction - second version) ASL for measuring flow changes during a global flow perturbation (hypercapnia) was assessed., (Cont.) Multiple inversion time ASL experiments were performed to examine bolus delivery dynamics under conditions of normocapnia and hypercapnia and at variable inversion band thicknesses. Tag delivery (inflow) curves revealed that typical published parameter values can cause substantial perfusion error during global challenges and render perfusion increases nearly undetectable. Theoretical criteria for choosing optimal QUIPSS II ASL parameter values are explored, and a multiple inversion time method for empirical determination of tag characteristics presented. Single inversion time functional experiments were subsequently performed to show that by using larger inversion band thicknesses and optimized timing parameters, perfusion accuracy and sensitivity can be substantially improved. Activation maps from block design visual cortex activation experiments and normocapnia-hypercapnia experiments support this conclusion., by Divya Sanam Bolar., S.M.