On Wed, 20 Jan 2010 09:57:20 -0500, Julie McEntee <[log in to unmask]>
wrote:
>Hi all,
>
>I am preprocessing fMRI data using SPM5. In the process of running Realign:
>Estimate & Reslice with the following parameters:
>
>Estimation options:
>Quality -> 1
>Separation -> 4 Smoothing (FWHM) -> 5 Num Passes -> Register to mean
>Interpretation -> 7th Degree B-Spline
>Wrapping -> No wrap Weighting -> <None>
>Reslice options:
>Resliced Images -> Mean Image Only
>Interpolation -> 7th Degree B-Spline
>Wrapping -> No wrap Masking -> Mask images
>
>some subjects realignment output graphs show ³spikes²- one image here or
>there that is greatly out of alignment compared to rest of images (see
>attached example). Should I/is there any way to exclude a particular scan
>from remainder of processing/analysis and should it be done before
>normalization/smoothing or during model specification/estimation?
I think you should also view the images that refuse to align well, in case
there's something odd going on in the dataset.
Also, if you do slice timing correction, you might want to consider getting rid
of the images before you do that. Not by removing them, but by replacing
them by the averages of the neighboring timepoints. That's in addition to
removing them at the model specification stage, not in lieu of.
>
>Thank-you for your help,
>
>Julie
>
>Julie E. McEntee, MA, CCRP
>Senior Research Program Coordinator
>Department of Psychiatry- Neuroimaging
>Johns Hopkins University School of Medicine
>600 N. Wolfe St./Phipps 300
>Baltimore, MD 21287
>Phone: 410-502-0468
>Fax: 410-614-3676
>
|