Motivation: Analyze the origin of "cloudy ghost" artifacts observed during quality assurance in temporal-fluctuation-noise-images from 10-minute Echo-planar imaging, to understand and improve consistency of fMRI-based neuroimaging. Goal(s): Determine gradient behavior differences between two 3T MRI systems and link them to the observed "cloudy ghost" artifacts. Approach: We measured the gradient trajectory of a standard Echo-planar Imaging sequence with a dynamic field camera at two MRI systems and simulated artifacts based on individual components of the measured trajectory. Results: A simulation routine based on Pulseq and the MRzero framework was developed, that can partially reproduce the observed artifacts from measured zero and first-order trajectory deviations. Impact: Enhance understanding of gradient related artifact sources in fMRI neuroimaging studies. Our open-source simulation routine allows to partially reproduce artifacts from dynamic zero and first-order trajectory deviations. The improved understanding lays the groundwork for development of more reliable future MRI-systems.
Wehkamp et al. (Tue,) studied this question.