Acquisition and processing techniques for rapid, robust, and patient-specific measurement of cerebral diffusion and perfusion using Magnetic Resonance Imaging

Magnetic resonance imaging (MRI) is a powerful technique that enables accurate medical diagnoses. Not only does it allow the visualization of static tissue contrasts, but it also provides the ability to measure the function and physiology behind important life processes. This dissertation presents two methods for imaging physiological effects, namely diffusion and perfusion.

For diffusion-weighted imaging, but not limited to this application, a technique called simultaneous multi-contrast (SMC) imaging is introduced to acquire multiple image contrasts from different scan types within a single measurement. This technique offers new opportunities in clinical protocols where examination time is a critical factor and multiple image contrasts need to be acquired. The method was tested in a phantom and healthy subjects.
The results showed that simultaneous acquisition of two contrasts (here diffusion-weighted imaging and T2*-weighting) with SMC imaging is possible with robust contrast separation and minimal impact on image quality. Thereby, the simultaneous acquisition of multiple contrasts reduces the overall examination time and there is inherent registration between contrasts.

For cerebral perfusion-weighted imaging, the goal of this work was to increase the robustness of time-encoded arterial spin labeling (ASL) to image artifacts by adjusting the timing of imaging during active scanning individually for each subject. One of the major challenges in ASL imaging is the high variability of arterial transit times (ATT), which leads to arterial transit delay (ATD) artifacts. In particular, in patients with pathologic changes, these artifacts occur when post-labeling delay (PLD) and bolus duration are not optimally matched to the individual, resulting in difficult quantification of cerebral blood flow (CBF) and ATT. This is also true for the free-lunch (FL) approach in Hadamard-encoded pseudocontinuous ASL (H-pCASL). The method was tested in healthy subjects, with FL-bolus timing individualized by the developed and implemented adaptive H-pCASL sequence in combination with an automatic feedback algorithm. Automatic subject-specific adjustment of the FL-bolus PLD during an ongoing measurement may reduce uncertainties in perfusion quantification and reduce the number of cases in which measurements have to be repeated because the recommended timing does not fit the patient.