Hi all,
I using dual regression to examine differences in within network and between network connectivity between groups of participants. Some participants have a different number of timepoints for their scans because a few high motion volumes were removed at the beginning or end of some scans. When I performed group ICA (temporal concatenation) and dual regression, I did not get any errors about the different number of timepoints. But, I have read in the archives that these analyses require the same number of timepoints across participants. Is that still true? Do I need to redo my analyses?
Second, I am using FSLnets to examine between network connectivity. When I load the time series from the dual regression, it throws an error saying Participants have different number of timepoints. It seems that regardless of whether it is necessary for dual regression, I will need to make all the scan lengths the same to use FSLNets (unless there is an updated version that can handle different scan lengths). Is this correct?
Thank you,
Dina Dajani
|