As a general rule, could I advise list members to please go easy with very
large attachments. These are received by about 2000 people, and some
recipients don't have very large mailboxes. If possible, it is better to
make data available by having an ftp site, or a web site where people can
download from.
I've looked at the data, and it doesn't seem too strange. There are only 7
slices in the original images, so the interpolated data will not appear too
good. Did the masked perfusion images look OK when displayed in SPM?
Best regards,
-John
> Hello, I am working with data from an AD population and have used Christian
> Gaser's create template and optimized VBM procedures in order to generate
> the template and priors and also the normalization parameters (from VBM) to
> apply to perfusion MRI images. The perfusion images are 7 slice and have
> been masked for gray matter and partial volume corrected. I used the
> default [2 2 2] dimensions when I applied the normalization parameters to
> the perfusion image. I also wrote the normalized image with [1 1 1], since
> this is how the parameters were generated in the first place in VBM. The
> normalized perfusion image looks strange to me (with both [1 1 1] and [2 2
> 2]) and I am unsure what the best procedure for normalization might be in
> my case. I have attached the normalized perfusion image [2 2 2]. Thanks!
> Audrey
>
> BTW: the dimensions of the perfusion image before normalization were 128 x
> 128 x 7 with 2.34 x 2.34 x 10 voxels. The warped T1 image from VBM was 157
> x 189 x 156. My T1 template and priors looked very good.
|