Improving Quantitative Cerebral Perfusion MR Imaging based on the Bookend Technique

Quantitative cerebral perfusion is a fundamental physiologic parameter that reflects the severity and progression of a broad range of pathologies. Quantitative cerebral blood flow (qCBF) measurements provide an ischemic threshold that distinguishes tissue destined for infarct from salvageable tissue in ischemic stroke patients. Quantitative cerebral perfusion has also been shown effective at monitoring diseases such as brain tumors, cerebrovascular occlusive disease, Alzheimer's disease, among others. Unfortunately, there is no widely available method for qCBF measurement.;Dynamic susceptibility contrast (DSC) magnetic resonance imaging (MRI), using a paramagnetic contrast agent, has shown potential for determining absolute quantification of flow. The Bookend method is a DSC-MRI technique allowing the computation of qCBF, quantitative cerebral blood volume (qCBV) and mean transit time (MTT). However, the implementation of the Bookend technique in a clinical setting is cumbersome, since it consists of several scanning steps and requires offline image post-processing with user input and time-consuming post-processing steps. In addition, Bookend perfusion maps suffer from artifacts and inaccuracies that are due to either image acquisition strategy or to patient physiology.;This thesis discusses the various technical improvements that facilitate the implementation of the Bookend technique in a clinical setting. The body of the thesis is divided into four main parts. The first part describes the improvements to two major perfusion postprocessing steps: (1) automation of the selection of the arterial input function (AIF) to eliminate user input; and (2) acceleration of the T1 mapping procedure. The second part presents two clinical implementation studies of the Bookend technique, one in normative posterior fossa brain regions, and another one in patients with multiple sclerosis, i.e. diseased white matter tissue. The third part describes the automation of Bookend image acquisition through the development of a single, self-calibrating MRI pulse sequence. The fourth part presents a correction method for arterial-tissue delay and dispersion effects that greatly affect the accuracy of DSC-based, quantitative cerebral perfusion measurements. Finally, future directions of the work are discussed and suggested.;The long term objective of this thesis is to improve the diagnosis and treatment of cerebrovascular disease.