In my opinion, there would probably be benefit to running ICA+FIX on task
fMRI data which regresses out 24 motion parameters together with noise ICA
components. When making the HCP¹s multi-modal parcellation, it was clear
that there were ³activations² that looked very similar to structured noise
you see in resting state fMRI. I am guessing this will make a greater
difference in avoiding false positives, rather than increasing z
statistics.
Peace,
Matt.
On 12/1/16, 5:07 PM, "FSL - FMRIB's Software Library on behalf of Nina de
Lacy" <[log in to unmask] on behalf of [log in to unmask]> wrote:
>Hi there:
>
>I'm setting up a new analysis and was wondering what people are thinking
>these days about regressing out 6 or 12 parameter motion regression at
>the first level (a la Powers) or using as confounds at the first level.
>Would welcome views on this from the FSL team or others on the list.
>
>
>
>Thanks
>Nina
>
>This message and any attached files might contain confidential
>information protected by federal and state law. The information is
>intended only for the use of the individual(s) or entities originally
>named as addressees. The improper disclosure of such information may be
>subject to civil or criminal penalties. If this message reached you in
>error, please contact the sender and destroy this message. Disclosing,
>copying, forwarding, or distributing the information by unauthorized
>individuals or entities is strictly prohibited by law.
|