Magnetic resonance imaging (MRI) is a powerful and flexible medical imaging modality that can provide excellent morphological and functional information non-invasively. However, MRI is inherently signal-limited. There is a trade-off between the various desired goals of high signal-to-noise ratio (SNR), high contrast-to-noise ratio (CNR), broad spatial field-of-view (FOV), fine spatial resolution, and short scan tune. This dissertation presents efficient acquisition methods for MRI based on a particular concentric rings data sampling trajectory, which can achieve a more flexible trade-off between the many imaging objectives.

The concentric rings non-Cartesian readout trajectory samples data on a polar grid in spatial frequency k-space with a set of concentric circles. The unique circular sampling property of concentric rings enables a reduction in scan time compared to conventional Cartesian encoding while maintaining a high degree of robustness to imperfections that limit the application of other non-Cartesian trajectories.

A detailed account of concentric rings implementation procedures is presented to support both three-dimensional spatial and one-dimensional spectral encoding. Simultaneous encoding of spatial and spectral dimensions is possible for the concentric rings by using an efficient retracing design that takes advantage of its circular symmetry. Trade-offs between spatial/spectral encoding are discussed and important properties are analyzed.

Off-resonance effects pose a major limitation to the application of non-Cartesian trajectories for MRI. Since the concentric rings retracing design can efficiently encode spatial/spectral information, off-resonance effects due to field inhomogeneity and chemical shift can be characterized and accounted for. Experimental results from in vivo fat/water imaging of the knee, head, larynx, and calf show that uniform and reliable fat/water separation can be achieved with the concentric rings. The concentric rings are also applied to the general case of MR spectroscopic imaging.

In addition to enabling efficient spatial/spectral encoding, the concentric rings trajectory can be used for efficient magnetization-prepared imaging. Concentric rings are inherently centric-ordered, provide smooth weighting in k-space, and enable shorter scan times—all of which are vital to magnetization-prepared imaging. Experimental results from in vivo volumetric brain imaging demonstrate that the concentric rings can achieve higher SNR and CNR within a shorter scan duration than conventional Cartesian-encoding methods.

In all of these imaging scenarios and for different anatomical regions, the concentric rings prove to be a robust readout trajectory that enables flexible trade-offs. The concentric rings are easy to implement and can be used to enhance a wide variety of clinical MRI applications.