Improved combined T2*-weighted MR Imaging of the Brain and Cervical Spinal Cord

Abstract

Functional magnetic resonance imaging (fMRI) based on the blood oxygen level-dependent contrast is used to study the function of the central nervous system non-invasively in vivo and has become an important tool in neuroscience and biomedical research. It has been well-established in the human brain for decades and has more recently been increasingly investigated in the human spinal cord. Combined brain and spinal cord fMRI, which can be used to investigate the interaction between these two regions of the central nervous system, is still challenging due to the different geometries, timings, and shim settings that typically would be used for the individual regions. The shimming procedure was cumbersome, time-consuming, and user-dependent; the result was not optimal and sometimes failed. Additionally, the volume coverage was limited. Therefore, the combined brain and spinal cord fMRI was not feasible for more clinical applications.

In this thesis, these issues are addressed. Not only the performance of combined brain and spinal cord T2* weighted MR Imaging has been improved, as is demonstrated in phantom and in vivo experiments, but also its applicability in research and clinical studies has been facilitated considerably. It is shown that a dedicated region-wise shim algorithm increases field homogeneity, which reduces the image artifacts of echo planar imaging, like geometric distortions and signal losses. The procedure to determine the optimum z-shim values has been automatized, which is less time-consuming and not user-dependent, without affecting the image quality. The implementation of the simultaneous multi-slice acquisition method for the brain volume improves the volume coverage and/or temporal resolution for the brain and spinal cord combined T2* weighted image. The image reconstruction has been extended to support on-the-fly image calculation with full functionality without the need for retrospective reconstruction as in previous applications. First steps towards a navigator-based motion detection of the spine have been performed by a navigator echo implemented in the echo planar image sequence, which shows that with the anatomic character of the vertebrae and intervertebral disc, the displacement of the spine along the slice direction can be detected. With these improvements, the performance and applicability of brain and spinal cord combined acquisitions have been improved considerably, not only facilitating its usage in basic research but also paving its way into clinical applications.