Multiparametric Estimation of Brain Hemodynamics with MR Fingerprinting ASL

Date

2018-12-28

Authors

Su, Pan

Journal Title

Journal ISSN

Volume Title

Publisher

Abstract

Cerebral perfusion is a process of delivering of blood to tissue capillary beds, supplying sufficient nutrients and oxygen to the brain. It is an important physiology indicator of brain function. The disorder of perfusion such as cerebral vascular disease leads to physiological changes and impaired function of brain. Noninvasive imaging of brain perfusion would contribute significantly to the research of cerebral physiology and clinical application in cerebrovascular diseases. Arterial Spin Labeling (ASL) MRI technique is capable of estimating cerebral blood flow (CBF) without radioactive tracer or contrast agent. However, the signal from ASL is affected by multiple parameters in the kinetic model, causing the complexity in interpreting the underlying physiological mechanism. This thesis develops a novel non-contrast technique for assessing multiple hemodynamic parameters in a single MR scan less than four minutes. The concept of magnetic resonance fingerprinting (MRF) was incorporated into the framework of ASL technique, and a novel sequence, MRF-ASL, was proposed. It is a promising alternative technique in cerebrovascular patients who cannot receive contrast agent based MRI perfusion. This thesis consists of three major components. First, I developed an MR perfusion technique (MRF-ASL) that can provide non-contrast and multi-parametric estimation of hemodynamic markers, in particular measuring timing parameters that are critical for applications in cerebrovascular disease, e.g. bolus arrival time (BAT). Feasibility, reproducibility, comparison to conventional multi-delay ASL and sensitivity were studied. Second, I optimized the key component in MRF-ASL, the TR timing sequence and further increased the spatial coverage of MRF-ASL with multi-slice acquisition. Then with the optimized sequence, I validated the MRF-ASL derived hemodynamic maps with Dynamic Susceptibility Contrast (DSC) MRI. The results showed that these two methods provided visually consistent and quantitatively correlated estimations of CBF and BAT results. Finally, I demonstrated the clinical utility of MRF-ASL by applying our technique to two types of cerebrovascular diseases, ischemic stroke and Moyamoya disease. Results showed that MRF-ASL can detect the prolonged bolus arrival and decreased CBF in these two diseases.

General Notes

Table of Contents

Citation

Related URI