Hi experts,
I ran fMRIPrep on a pediatric dataset using the MNIPediatricAsym_cohort-2_res-2 template with fieldmap-based distortion correction. I was further planning to analyze confounds and exclude participants with more than 20% motion outliers, based on the default thresholds (FD > 0.5 and std_dvars > 1.5).
I noticed that one participant has 28% motion outliers (42 volumes), flagged based on std_dvars , but not FD (mean FD = 0.08; only 2 volumes exceeded the 0.5 threshold). According to MRIQC , this participant shows very high overall signal quality (TSNR = 81).
The std_dvars values are consistently elevated, without large scan-to-scan variability (range = 1.18-2.5; sorry, I guess I can’t upload a “confounds.tsv” file as a new user).
If the elevated std_dvars values are not due to movement (low FD) or residual distortion (addressed by fieldmap correction), what might be the cause? Do you have any recommendations for how to handle this participant in the analysis?