 |
 |
FSL experts, I have data where I know that some subjects have considerable motion (>3mm displacement). I use FIX to clean the data, with the -m flag (24 parameter motion regression?) to regress out motion covariates. However, I am considering also running fsl_motion_outliers on the data and regressing out the high motion time points using feat (additional confound EVs). My questions are: 1) Is this a necessary/helpful step for high-motion data? 2) Should this be done before of after clean-up with FIX 3) With a TR of 1000ms, should O also regress out the time points before and after the high-movement time point? If it matters, the data will be used to create voxel-wise functional activity and connectivity metrics (fALFF/ALFF, ReHo, VMHC, etc.)