Dear all,
I'm doing multiple regression on ALFF images. I'd like to perform head motion regression (in nuisance regression step) in preprocessing, and include mean FD Jenkinson as a covariate in 2nd level. Is it reasonable？
It would be great to hear your views and suggestions. Thank you very much!
Best regards,
Feng Zhou
Forums
Hi Feng,
Hi Feng,
As in one of our recent studies (Yan, C.G., Cheung, B., Kelly, C., Colcombe, S., Craddock, R.C., Di Martino, A., Li, Q., Zuo, X.N., Castellanos, F.X., Milham, M.P., 2013. A comprehensive assessment of regional variation in the impact of head micromovements on functional connectomics. Neuroimage 76, 183-201.), ALFF is relatively sensitive to motion (as compared to fALFF). I will suggest you to perform Friston 24 regression, but avoid scrubbing for the current ALFF implementation. At the group level analysis, using mean FD Jenkinson as a covariate is a reasonable way to control motion related false positives (though control at a linear level).
Best,
Chao-Gan
Hi Chao-Gan,
Hi Chao-Gan,
Thanks for your help! I'll try Friston 24 regression.
Happy Chinese New Year of horse!
Feng Zhou
mean of FD Jenksinson for 2 runs, same subject
Hi Dr. Yan,
I am using this procedure as described in your 2013 paper. To apply FD Jenksinson as a group-level covariate when there are 2 runs per subject, is it appropriate to use the arithmetic mean of the FD value for both runs? That is how I am working with the multiple r-to-z FC maps, can you please let me know if there are any problems with these steps?
Thanks in advance,
Felix
You can use mean FD and mean
You can use mean FD and mean FC map for each subject.
thank you
Perfect, thank you for verifying this!