Phase imaging is now widely used for anatomical MRI of its own or its derivative forms such as susceptibility-weighted imaging or quantitative susceptibility mapping. The sources of phase signals are differences in the resonance frequency of magnetization in different tissues. Applying off-resonance RF pulses saturate magnetizations of proton spins bound to macromolecules, which are distributed in a wide frequency spectrum and exchange with proton spins in the free water pool, a mechanism called magnetization transfer (MT). Therefore, it is possible that the phase signals can be affected by MT, however, the relationship between MT and phase signals are not explored well.
While phase imaging is often performed with gradient recalled echo, recent preliminary studies showed promising phase signals by using balanced steady state free precession (bSSFP) rather than GRE because of the existence of steep phase change regions called transition bands. However, these transitions bands exist only in limited areas, therefore it is required to develop methods to extend the high phase contrast regions in the whole imaging area.
This dissertation focuses on investigating the relationship between MT and phase signals using the bSSFP sequence. First, multiple phase-cycled bSSFP is proposed to improve the spatial coverage of the bSSFP phase imaging for distortion-free and high phase contrast in the whole brain. Second, to quantitatively understand the relationship between MT and phase signals, a new strategy of quantitative magnetization transfer imaging is suggested using inter-slice bSSFP MT scheme and dictionary-based parameter mapping, which may allow us to overcome the limitations of long scan time and long processing time, respectively. Lastly, the relationship between bSSFP phase signals and MT will be investigated under various conditions. Preliminary trials showed some promising results for the proposed studies.