On Mon, 5 Apr 1999, ning ma; ENEE wrote:
> Dear SPMers,
>
> I have some questions about realignment in SPM96.
> According to John Ashberner's suggestions given in last June (please see
> below), when I realign PET images in SPM96, I do it twice. But I am a bit
> confused by *.mat files generated in realignment procedure.
>
> When I do the 1st realignment, I choose:
> 1) Coregister and reslice
> 2) Sinc interpolation
> 3) Mean images only
> 4) Mask the images
>
> The results I get are: mean*01.img(hdr) and a set of *.mat files.
These .mat files contain transformation parameters that
reflect realignment to the 1st image.
> Then I go to the 2nd realignment by using mean*01.img(hdr) and the
> original pet images *.img(hdr):
> 1) Coregister and reslice
> 2) Sinc interpolation
> 3) All images + Mean image
> 4) Mask the images
>
> This time the results I get are: meanmean*01.img(hdr), a set of
> r*.img(hdr) and a new set of *.mat files.
>
What you get here, depends on which image was entered first. If it is the
mean*01.img, then all images are realigned to this image, and the results
are stored in the new set of .mat files. (That is what Dr.
Ashburner recommended). However, you should also have a
set of r*.mat files - is this what you mean by the "new set of .mat
files"? I would imagine that the new .mat files will overwrite the ones
from the 1st realignment, and the r*.mat files will have the same
information as the new .mat files since a .mat file becomes connected to
the .img file in the same way as .hdr file is.
If, however, your first entered image was *01.img ("baseline"), then you
realign everything to the 1st image of the series using parameters (in
.mat files of step 1) that reflect realignment to the 1st image -
repeating step 1.
> For some reason, for some data set, before I do the 2nd realignment, I
> delete the '*.mat' files obtained from the 1st realignment, would this
> action affect the resulting r*.img(hdr) files and meanmean*01.img(hdr)??
> And how much is it do you think?
In this case, SPM would calculate starting estimates from the first image
entered. - same as step 1. If this image is mean*01.img, it should not
matter (in theory) since the starting estimates would be produced from the
image to which all images were realigned in step 1 - and these estimates
should be the same as ones stored in first set of .mat files.
If first image is image #1 from the series, the starting estimates would
come from that - you will be realigning to the 1st image without using
the mean.
> The generation of '*.mat' files is a new feature in SPM96. From SPM96 help
> I know that it is not necessary to get r*.img(hdr) images before the
> spatial normalisation stage. Could you tell me how to use '*.mat' files
> during the spatial normalisation? What should I pay attention to when I do
> this?
>
Assuming that you figured out what procedure you want to use for
realignment, you choose "coregister only" then you do your last pass - it
won't write r*img files, but instead will create a new set of .mat files
replacing the old ones (I think). When you do spatial normalization,
simply enter a .img file from which you want SPM to determine
normalization parameters - it will probably be mean*01.img if you follow
Dr. Ashburner's procedure. Then enter all images you want to normalize
(.img files and each has an associated .mat file, whose contents will be
taken into account during spatial normalization).
I hope this helps -
--Vitaly Furman
Brain Imaging Lab
Columbia-Presbyterian Medical Center
New York, NY
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|