| The development of functional MRI technology is rapid since it was reported in the early1990s. Being it can image cerebral cortex functional activation in real time, the advent of functional MRI broadens the clinical application of MRI. Functional MRI has become a significant tool to study brain science and life science with high spatial resolution, temporal resolution, and absolutely non-invasive characteristics.Based on the contrast mechanisms of blood oxygenation level dependent (BOLD), fMRI is usually carried out at high magnetic field strength MRI system. Recently there has been reported that fMRI can also locate activated area of brain based on non-BOLD contrast of proton density weighted image with the deeply research of fMRI, and it can explain the inconsistent of BOLD model. These supply a new method and open a research direction for fMRI. Especially fMRI can be implemented at low field strength using non-BOLD contrast. This is significant for the most users of low field strength MRI at China.The paper research fMRI based on non-BOLD contrast of proton density weighted image. Mainly works include:1) Investigate the methods for fMRI at low field MRI to improve its usability. A method that combines sequence parameters optimized and improved data analysis is proposed.2) Research and develop faster GRASE (Gradient and Spin Echo) sequence of MRI that combines the advantage of TSE (Turbo Spin Echo) and EPI (Echo-planar Imaging) to increase time resolution of non-BOLD fMRI. Thus, applying proton density weighted GRASE for functional study of brain based on non-BOLD contrast improve its implementability further.3) Study the specificity and dependent relationship with TE (Echo Time) of fMRI signal change using proton density-weighted spin-echo EPI. And establish the simple model for fMRI containing the non-BOLD component.The results demonstrate:Using the combination method for fMRI at low field is benefit for its application. By optimizing TSE parameters, decreased its scan time from26s to8s and increased the SNR (Signal to Noise Ratio) above factor2. The method of improved data analysis that combing explicitly mask image with GLM (General linear Model) is very effective for removing pseudo-activated area located on background, skull and scalp of fMRI at low field. To our knowledge, the results acquired from our combination method are better than the similar study result at low field.GRASE sequence decrease scan time to EPI factor parts of TSE scan time by creating multiple short gradient echo trains between successive180pulses. Therefore, considering its advantage on scan speed and image quality, apply proton density weighted GRASE for the brain study of non-BOLD fMRI for the first time. The results demonstrate that time resolution of fMRI is increased and spatial extent of activated area is extended. Comparing with non-BOLD fMRI using TSE, GRASE has more value for fMRI application.fMRI data acquired with high spatial resolution of proton density-weighted SE-EPI consistently demonstrates the activated area can be detected and were matched by areas of activity identified with conventional BOLD fMRI. The spatial distribution of activity observed based on two contrast mechanism is in close proximity. By investigating the characteristics signal change of SE-EPI fMRI with TE, verified the non-BOLD component. There is the linear relationship between signal change with TE when TE>33ms, and positive intercept exist. Considering the non-BOLD component, present a simple mathematical expression for describing fMRI signal change with TE. |
PDF Full Text Request