Functional and perfusion magnetic resonance imaging at 3 tesla

This thesis deals with the development and optimization of fast magnetic resonance imaging (MRI) methods for non-invasive functional studies of the human brain and perfusion imaging on a 3 Tesla (T) whole body NMR system. The functional MRI (fMRI) experiments performed showed that single-shot multi-echo EPI and spiral imaging techniques provide fast tools to obtain information about T2* distributions during functional activation in the human brain. Both sequences were found to be useful in the separation of different sources contributing to the functional MR signal like inflow or susceptibility effects in the various vascular environments. An fMRI study dealing with the involvement of prefrontal brain regions in movement preparation lead to inconsistent results. It could not be clarified if these were caused by problems during a spatial normalization process of the individual brains or if the functional paradigm, using very short inter-stimulus intervals, was not suited for the problem investigated. Blood flow velocity measurements in the human finger showed that the use of a strong, small-bore gradient system permits short echo times that reduce flow artefacts and allows high spatial resolution in order to keep systematic errors due to partial volume effects small. With regard to the perfusion investigations an inversion recovery snapshot-FLASH sequence was implemented, which allowed the acquisition of T1 parameter maps of the human brain within a few seconds. The accuracy of this method was demonstrated in test objects. The perfusion investigations with FAIR showed good qualitative results, whereas the quantitative analysis did not yield reproducible findings. A reason for the poor results could be the low signal-to-noise ratio (SNR) of the FAIR images or an incomplete global inversion of the magnetization due to the transmission characteristics of the radio-frequency coil. The BASE sequence that did not require a global inversion yielded quantitative perfusion changes due to brain activity in a functional study with visual stimulation. Most of the performed experiments took advantage of the higher SNR and functional contrast at the high magnetic field strength of 3 T, compared to in clinical routine used field strengths of 1.5 T and below. On the other hand, single-shot imaging techniques showed increased susceptibility artefacts. In conclusion, the various experiments performed within this thesis highlight the potential of MRI to obtain a plurality of information by choosing appropriate imaging sequences that may be used in future clinical studies. (author)