Regarding the mcflirt, you need to do mcflirt on a 4D image, not on 290 3D
images. In other words, you need to use fslmerge first to create one single
4D image, and then use mcflirt on that 4D image.
This will merge the 290 volumes along the time dimension, in the order that
they are listed by "ls":
fslmerge -t merged.nii swapped_*.nii
Then run mcflirt:
mcflirt -in merged.nii -stats -plots -report
You might be better off using motion correction in the pre-stats tab in the
Feat GUI. That makes it nice and easy to use motion estimates as covariates
in your GLM if you want.
--Greg
____________________________________________________________________
Greg Burgess, Ph.D.
Research Associate, Institute of Cognitive Science
University of Colorado - Boulder
Phone: 303-735-5802
Email: [log in to unmask]
University of Colorado - Boulder
UCB 594
Boulder, CO 80309-0594
On Wed, 20 Feb 2008 16:18:57 -0500, Emily T Stoneham <[log in to unmask]> wrote:
>Thanks!
>Emily
>
>----- Original Message -----
>From: Andreas Bartsch <[log in to unmask]>
>Date: Wednesday, February 20, 2008 3:49 pm
>Subject: [FSL] AW: [FSL] McFLIRT basic use question
>
>> Hi,
>> >a) do we use the Motion Corrected images for our processing?
>> you can but in my experience mcflirt is more robust.
>> >b) do we then do McFLIRT on those images?
>> No: if you use the Moco images from the scanner you should NOT run
>> mcflirt on top (simply because it will degrade your images).
>> Cheers-
>> Andreas
>>
>>
>> ________________________________
>>
>> Von: FSL - FMRIB's Software Library im Auftrag von Emily Stoneham
>> Gesendet: Mi 20.02.2008 18:40
>> An: [log in to unmask]
>> Betreff: [FSL] McFLIRT basic use question
>>
>>
>>
>> Good Afternoon!
>>
>> This is a basic question regarding use of McFIRT.
>> Our Scanner puts out both a file of the DICOM images per run
>> (290 volumes, 30 slices per
>> image) and a file that is motion corrected by the scanner (same
>> image numbers as above). I then
>> convert these to .nii using MRIConvert, and reorient the images
>> using a script provided to me by
>> Mark Jenkinson. I would like to know:
>>
>> a) do we use the Motion Corrected images for our processing?
>> b) do we then do McFLIRT on those images?
>> Assuming we do a McFLIRT prior to doing Slice Timing:
>> c) I have been having a hard time figuring out how to batch those
>> 290 images so that it does a
>> McFLIRT on each one. I would like to stay within FSL to do this (I
>> had been advised to do this in
>> another program like AFNI), as it would greatly simplify things. I
>> have a batch file for reorienting
>> images (works amazingly well- thank you Mark!):
>>
>> for fn in *.nii ; do fslswapdim $fn -x y -z swapped_$fn ; done
>>
>> ... BUT I can't extrapolate how to do this kind of thing for
>> McFLIRT - i.e., something that I can also
>> apply to slice timing and 4D conversion (no, I don't know Unix
>> other than basic commands). Can
>> someone give me a hint as to the basics of batch filing (if that is
>> what it is called) in FSL?
>>
>> Cheers!
>> Emily
>>
>
>begin:vcard
>n:Stoneham;Emily T
>fn:Emily T Stoneham
>tel;cell:240-577-4542
>version:2.1
>email;internet:[log in to unmask]
>end:vcard
>
|