Motion correction of parametric fMRI data from multi-slice single-shot multi-echo acquisitions

Fast parametric imaging using multi-echo techniques has been proven to yiel d quantitative parameter maps with high stability for functional MRI (fMRI) . Due to the different contrasts and signal-to-noise ratios (SNRs) in the v arious images, motion correction of the echo images or the resulting parame ter maps is not a straightforward process. 3D motion correction of parametr ic imaging data has not yet been examined thoroughly. However, motion corre ction is an essential step in fMR1 data processing. In this study several p ossible motion detection methods were tested and compared. Motion parameter s can be estimated from the different echo images as well as from the param eter maps. The accuracy of the different methods was examined in simulation s and in in vivo experiments. Motion parameters should be estimated from th e I-0-parameter maps and subsequently applied to the T-2(*)-parameter maps. (C) 2001 Wiley-Liss, Inc.