Novel methods for steady-state neuroimaging

Magnetic Resonance Imaging (MRI) has had a major impact on neuroscience due to the flexibility of MRI contrast. In particular, functional MRI (FMRI) and diffusion-weighted imaging (DWI) have provided neuroscientists with unique and powerful methods for studying brain function and structure. However, neither of these methods has the signal-to-noise ratio (SNR), spatial resolution or image quality achieved by many other MRI methods. This work presents new techniques for FMRI and DWI that are based on steady-state free precession (SSFP) imaging. SSFP achieves unusually high SNR efficiency by allowing the magnetization to persist from one RF excitation to the next. In addition, SSFP methods tend to have unusual signal properties that can be used to manipulate image contrast.; The standard method for FMRI, using the Blood Oxygenation Level Dependent (BOLD) effect, has significant limitations that result from the coupling of functional contrast to sources of image artifact. We have developed a method that uses the SSFP phase profile to invert the signal in deoxygenated blood relative to oxygenated blood. The resulting Blood Oxygenation Sensitive Steady-state (BOSS) signal decouples functional contrast from imaging, enabling significantly better image quality than BOLD FMRI. BOSS FMRI is very SNR-efficient, achieves strong functional contrast and is relatively immune to susceptibility gradients.; Standard DWI methods suffer from low SNR, poor image resolution and severe image artifacts. Although it has long been known that SSFP DWI is one of the most efficient DWI pulse sequences, this method has traditionally suffered from intense sensitivity to brain motion. Motion in DWI causes phase offsets in the magnetization that create image artifacts in multi-shot acquisitions. This work has developed a method of removing phase artifacts from SSFP DWI using navigator echoes. Navigated SSFP DWI produces high-resolution diffusion images with high SNR and no visible artifact.; A major component of this work was the development of a novel navigator correction that removes phase artifacts caused by non-rigid motion based on an approximation to the least-squares optimal reconstruction. This method, referred to as a refocusing reconstruction, produces significantly better images than standard rigid-body corrections and should be applicable to other navigated DWI sequences.