Dear Yunjin,
this difference is due to the fact that the very first step of dual regression is “creating common mask”. The brain masking is setup such that a common mask is estimated: it is the intersection of the masks across all subjects (keeps only voxels with nonzero values for all subjects). Hence this step could cause slight differences in (the masking of) any single subject's maps for different runs when you include different subjects.
This is because the expected thing to do with dualreg outputs is to feed them into randomise (stage 3) - which therefore requires a common mask to have been applied (because randomise can't handle missing data).
Hope this helps,
Ludovica
> On 21 Jan 2019, at 6:33 am, Yunjin <[log in to unmask]> wrote:
>
> Hi all,
>
> I performed dual regression for patients group and control group.
>
> However, after excluding 6 people from patients group and conducting dual regression again,
>
> I found same participant has different time courses (1st stage) for the same ICA component when compared to the first result.
>
> I expected the same time course value since I used the same group spatial map before and after excluding patients.
>
> How could it be explained?
>
> Thanks,
> Yunjin
>
> ########################################################################
>
> To unsubscribe from the FSL list, click the following link:
> https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1
########################################################################
To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1
|