Preview

PDF, English Download (34MB) | Terms of use

Abstract

Conventional magnetic resonance imaging (MRI) utilizes the signal of water protons for non-invasive imaging of anatomical structures, while depending on contrast agents for reliable tumor detection. In chemical exchange saturation transfer (CEST) MRI, selective saturation of chemically exchanging protons leads to a transfer and accumulation of saturation in the water proton magnetization, effectively resulting in signal enhancement. Thus CEST forms a spatially highly resolved imaging technique providing information on endogenous low concentrated biomolecules and their chemical environment. However, in living tissue, only qualitative information can be obtained with conventional evaluation methods, as CEST effects overlap each other and are biased by the signal of macromolecular cellular structures and water relaxation parameters. The aim of this thesis therefore was the isolation of unbiased CEST effects in vivo for quantitative CEST MRI of brain tumors. The separation of effects was possible by applying a fit model to highly resolved data obtained at ultra-high magnetic fields. This required the development of a method to correct for apparent transmit field inhomogeneities. In a patient study, the vanishing of tumor contrast upon relaxation-compensation was observed. Precise experiments with tissue samples identified an additional, previously disregarded contribution, allowing the progress towards a novel CEST contrast. As it highlights similar structures as gadolinium contrast-enhanced MR images, it may allow tumor detection, and possibly staging, without the injection of contrast agents.