> I have two sets of EPI (=perfusion) images, which are test-retest images
> for each subject and I scanned 13 subjects with 3 different acquisition
> methods. Those EPI images are not fMRI images. I mean those are not
> time-course. Only one image is for each slice. Those EPI images were
> coregistered to T1 high resolution images. The T1 high resolution images
> were normalized to T1 template and segmented for gray (seg1=i1),
> white(seg2=i2), and CSF(seg3=i3). I was also extract brain (brain=i4). To
> obtain better segmentation result, I did Imagray=i1./(i1+i2+i3+eps).*i4 for
> gray matter and
>
> Imawhite=i2./(i1+i2+i3+eps).*i4 for white matter.
>
> 1) I want to segment EPI images according to T1 segmented images.
> I choose EPI(i1) and Imagray (i2) for gray matter, and then i1.*i2 using
> ImaCal and EPI(i1) and Imawhite (i2) for white matter, and then i1.*i2
> using ImaCal
>
> Am I right, or any mistake?
>
> I will obtain mean signal intensity and standard deviation for each subject
> of gray matter, white matter, or whole brain (gray+white) volume for
> calculating intra-class correlation coefficient to drive which one method
> is the best for measuring reliability of perfusion.
Try the following piece of code for computing means and variances...
P=spm_get(2,'*.img','Select Perfusion image and Grey/White image');
V=spm_vol(P);
s1=0;
s2=0;
for i=1:V(1).dim(1:2),
M = spm_matrix([0 0 i]);
y = spm_slice_vol(V(1),M,V(1).dim(1:2),0);
p = spm_slice_vol(V(2),M,V(1).dim(1:2),0);
s1 = s1 + sum(p(:));
s2 = s2 + sum(p(:).*y(:));
end;
mn = s2/s1;
s3=0;
for i=1:V(1).dim(1:2),
M = spm_matrix([0 0 i]);
y = spm_slice_vol(V(1),M,V(1).dim(1:2),0);
p = spm_slice_vol(V(2),M,V(1).dim(1:2),0);
s3 = s3 + sum(p(:).*(y(:)-mn).^2);
end;
vr = s3/s1;
fprintf('Mean=%g Variance=%g\n',mn,vr);
> 2) I also want to use SPM for statistical analysis of segmented test-retest
> EPI data of gray matter, white matter, or whole brain. Would you recommend
> me the best way to analysis those images using SPM? I am very new for SPM
> statistical analysis.
I would need to know more about what you plan to do in order to comment.
Best regards,
-John
--
Dr John Ashburner.
Functional Imaging Lab., 12 Queen Square, London WC1N 3BG, UK.
tel: +44 (0)20 78337491 or +44 (0)20 78373611 x4381
fax: +44 (0)20 78131420 http://www.fil.ion.ucl.ac.uk/~john
|