We've been having some problems with our MNI registration of the fMRI data. 
After going through the PreFS, FS, PostFS, and the GenericfMRIbatch scripts, we 
get a 4D dataset: subjectID/MNINonLinear/Results/taskname/taskname.nii.gz

I'm assuming those 4D images should be registered to the MNI template?

We're finding that when we run stats on those 4D files, the resulting 
activations are pulled back from the edge of the brain. It seems that the 
registration of the fMRI series is not quite good enough on some part of every 
subject, and cumulatively the activation is cut off around the edge of the 
brain. We were using SPM to do the stats, see the attached picture.

If I do FSL stats, and register those 4D images to the MNI template, the 
resulting activations look fine, but I'm not sure this is the best thing to do, 
as they are then being registered to the MNI template twice.

I'm also going to assume that this is because of poor quality structural data. 
I've emailed this list about various problems related to our MPRAGE sequence, 
and I suspect this may be another problem :)

_________________________________________________
Gregory Book
Senior Technology Manager
Olin Neuropsychiatry Research Center, Institute of Living, Hartford Hospital
200 Retreat Avenue
Hartford, CT 06106
Tel: 860-545-7267 Fax: 860-545-7797
[email protected]<mailto:[email protected]>
http://nidb.sourceforge.net


This e-mail message, including any attachments, is for the sole use of the 
intended recipient(s) and may contain confidential and privileged information. 
Any unauthorized review, use, disclosure, or distribution is prohibited. If you 
are not the intended recipient, or an employee or agent responsible for 
delivering the message to the intended recipient, please contact the sender by 
reply e-mail and destroy all copies of the original message, including any 
attachments.

_______________________________________________
HCP-Users mailing list
[email protected]
http://lists.humanconnectome.org/mailman/listinfo/hcp-users

Reply via email to