Comparison of prospective and retrospective motion correction for Magnetic Resonance Imaging of the brain - Master's Thesis in Physics

Head motion is one of the most common sources of artefacts for Magnetic Resonance Imaging (MRI) of the brain. Especially children, being intimidated by the dimensions and the noise of the scanner, tend to move considerably during image acquisition. Thus, the use of general anaesthesia or sedation is common practice in clinical routines to avoid motion artefacts. However, general anaesthesia requires additional equipment and personnel and concerns about potential drug-related risks are increasing.With this thesis, I aim to compare different methods for motion correction of MRI as alternative strategies for avoiding motion artefacts during clinical MRI examinations. I evaluate the performance of prospective and retrospective motion correction on a set of six 2D- and 3D-encoded sequences from a clinical pediatric MR protocol. For each sequence, we acquired scans with and without a predefined motion pattern in 22 healthy volunteers. I analyse and statistically compare a set of image qualitymeasures as well as observer quality scores. Furthermore, I quantify the influence of motion correction on motion-related changes in cortical thickness estimates using a general linear model.In this work, I show that for the 3D-encoded sequences, image quality considerably increases from scans without motion correction, over retrospectively to prospectively corrected scans. Selectively reacquiring slices with the highest level of motion additionally improved image quality. Furthermore, cortical thickness estimates from motion corrected scans do not correlate with motion to the same extent as observed for uncorrected scans. For three of the 2D-encoded sequences I observe higher image quality due motion correction and reacquisition compared to the clinical standard, even though the effect size varies between the different sequences. I did not observe an improvement due to motion correction in the T2* sequence, which is known to be highly sensitive to field inhomogeneities. owever, this sequence can be substituted by an accelerated, easily repeatable sequence. My findings confirm that the use of prospective motion correction is feasible for a full clinical children's protocol, reducing motion artefacts and ensuring diagnostic image quality. As a next step in establishing motion correction in clinical routines, the developed and validated set up will be tested on a clinical pediatric population.