End-to-end acquisition and analysis of whole-slice rosette MRSI for brain metabolic imaging at 3 Tesla

Motivation: Magnetic resonance spectroscopic imaging (MRSI) has strong clinical potential, but also exhibits challenges making it difficult to translate in a clinical context. Goal(s): We aim to develop an end-to-end pipeline with the goal of producing high-quality MRSI spectra in whole brain slices, while also minimizing scan time and work-flow disruptions. Approach: We developed a spin-echo MRSI pulse sequence utilizing rapid rosette k-space trajectories to minimize scan time. Post-processing is achieved using the open-source software FID-A. Results: The spin-echo rosette MRSI sequence, combined with dedicated pre-processing pipelines, produced high-quality MRSI spectra. Overall, this study is a step towards rapid MRSI in a clinical setting. Impact: Magnetic resonance spectroscopic imaging (MRSI) is a powerful tool for quantifying biomarkers in vivo. However, clinical adoption of MRSI has been slow. We propose a complete and coherent pipeline for rapid MRSI acquisitions pre-processing to address this clinical need.