Preview

PDF, English Download (10MB) | Terms of use

Abstract

In this thesis, a novel technique for the volumetric quantification of the longitudinal relaxation time T1 by ultra high field (UHF) magnetic resonance imaging (MRI) is introduced. It is based upon the prediction of the MR signal disturbances, due to static magnetic and RF field inhomogeneities as well as readout effects by the imaging process itself, and the corrections resulting hereof. For this reason, the mathematics of the magnetization’s equation of motion and the Bloch equations are implemented into a new simulation framework and regarded for by the evaluation algorithms. Furthermore, different MR signal simulation strategies additionally considering the k-space filters and various correction approaches are investigated. The introduced SIMBA IR and SIMBA DESPOT1-HIFI methods are capable of quantifying T1 with respective maximum deviations to the nominal values of (-0.42±1.23)% and (1.99±1.58)% within a T1 range of 1100 ms to 3300 ms. A minimization of the repetition time TR within the SIMBA IR experiments shortens the measurement time by up to 50% and further improves the accuracy. The use of a non-adiabatic preparation reduces the SAR exposure by up to 70% and allows examinations near organs of risk. Eventually, the even faster SIMBA DESPOT1-HIFI method was applied on a volume of 256×256×176mm3 with an isotropic resolution of 1mm within less than 30 min. A study of the whole human brain revealed a clearly differentiated soft tissue contrast and T1 values of (1917±95) ms for the gray and (1246±56) ms for the white matter. In a study on the human calf muscle, T1 was quantified to a value of (1877±92) ms. All T1 values are in a strong agreement with literature values.