Development of MRI techniques for measuring cerebral blood volume, blood flow, and blood oxygenation within a single scan.

摘要

Functional MRI (fMRI) is commonly performed using the blood-oxygenation-level-dependent (BOLD) approach, which is sensitive to ensemble changes in cerebral blood volume (CBV), cerebral blood flow (CBF), and cerebral metabolic rate of oxygen (CMRO2). In order to understand and quantify the BOLD fMRI signal, it is essential to design multi-modal fMRI approaches that are sensitized to these individual hemodynamic parameters, and to further determine the oxygen metabolism. This dissertation aims to develop and improve current quantitative fMRI techniques to detect relaxation times T2*, cerebral blood volume (CBV), cerebral blood flow (CBF), blood oxygenation level hemodynamics, oxygen extraction fraction (OEF), and CMRO2 during neuronal activation in a time efficient manner.;Total and extravascular R2* values in the parenchyma in human visual cortex are measured by combining multi-echo BOLD and vascular-space-occupancy (VASO) fMRI with visual stimulation at 7T. The VASO method is expected to suppress the intravascular signals in the microvessels. Both the absolute parenchymal extravascular R2* and R2 * changes (DeltaR2*) upon activation are determined, and the ratio of extravascular DeltaR2* to total DeltaR2* is calculated, confirming a predominant contribution from the extravascular component of the BOLD effect at 7T. Parenchymal OEF during stimulation is also estimated based on these measurements, the value of which is consistent with those reported at lower field strengths.;While normally in most of the quantitative fMRI approaches, BOLD, CBV, and CBF measurements are separately performed to estimate CMRO2 dynamics, the ability to acquire these physiological parameters simultaneously would be very useful to improve image acquisition efficiency, and more importantly reduce the sensitivity to temporal variations due to factors such as subject head motion, task performance, and physiologic fluctuations between the fMRI experiments. A large portion of this dissertation is devoted to design single-scan approaches to detect changes in the multi-modal hemodynamic signals. First, a novel 3D whole-brain MRI pulse sequence, dubbed 3D VASO-FAIR, is proposed to detect CBV and CBF responses in a single scan. Second, a new 3D acquisition strategy that extends VASO-FAIR and incorporates a 3D T2-preparation gradient-echo (GRE) BOLD method is implemented to simultaneously measure BOLD, CBV, and CBF reactivity during functional stimulation. Compared to individually performed multi-modal fMRI scans, similar image quality, activation patterns, relative signal changes (?S/S), tSNRs and CNRs can be achieved using the proposed combined sequences. Finally, based on the BOLD, CBV, and CBF responses obtained from the combined sequence, the oxygen metabolism alterations (OEF and CMRO2) are quantified.