Dear Tugan:
I don't think the inhomogeneity correction does any warping, rather it
increases the signal level in low signal areas and probably reduces signal
in high signal areas so the brain is more even. I would be concerned about
doing this on EPI images unless you're doing it on just a single image for
making the normalization parameter set because it would otherwise change
your underlying signal. I still don't understand why deriving the
parameters from the T1 image doesn't work for them unless the T1 is also
showing a lot of inhomogeneity. Otherwise I think your suggestion is a good
idea.
Darren
At 03:26 PM 11/9/2001 -0800, you wrote:
>Hi, we had the same problem and I'd highly recommend using some
>"inhomogeneity correction" on the EPI scans before normalization.
>Generally this correction nicely wraps the "blown-out" frontal lobe
>back into a reasonable shape.
>
>Good luck,
>
>Tugan
>
>At 9:58 PM +0000 11/9/01, Glenn Wylie wrote:
>>Hello all,
>>
>>We've recently run into a problem using SPM to normalize functional scans
>>acquired on our 3T magnet. Because it's a 3T, we get 'blow-out' of
>>orbital, frontal cortex due to the nasal sinuses. This means that the
>>brains we acquire have a peculiar shape (no orbital cortex). When we go to
>>normalize these, SPM warps them into standard space -- but SPM doesn't know
>>that they have no orbital cortex. This results in distortions: much of
>>the brain is dragged anterior and inferior to where it really should be.
>>
>>The stuctural scans are not affected in the same way, and they normalize
>>correctly. However, applying the parameters determined by normalizing the
>>structurals to the functionals results in much the same problem.
>>
>>Does anyone have any fixes for this sort of problem?
>>
>>Many thanks,
>>
>>Glenn Wylie.
>
>
>--
>****************************************
>
>L. Tugan Muftuler, Ph.D.
>
>Health Sciences,
>Research Imaging Center
>Irvine Hall 164
>University of California
>Irvine, CA, 92697-5020
>
>Tel: (949) 824-6290
>fax: (949) 824-3481
>E-mail: [log in to unmask]
>****************************************
Darren R. Gitelman
Northwestern University, Evanston, IL. USA
[log in to unmask]
|