Dear SPM experts:
We observed an interesting data behavior using spm96
for one of our PET studies, and would like to seek your expert
help on this.
We have 3 groups of subjects: A B C. Group A has 7 subjects, 11 in B and 14 in C.
Each subject in any one of the three groups has two longitudial
FDG-PET scans (roughly two years apart). Scan 1 is the baseline scan
and follow-up scan is scan2.
We are interested in comparing longitudinal glucose uptake reduction
(increase) within each group, and compare the reduction between any
two group (especially between B and C).
Using multi-study, different conditions and proportional scaling design
in spm96, the glucose uptake reduction in group A, B, and
C were tested respectively using
1 -1 0 0 0 0 %for group A
0 0 1 -1 0 0 %for B
0 0 0 0 1 -1 %for C
Glucose uptake increases were tested by changing the signs.
Cross group comparison was made, for example, as
0 0 1 -1 -1 1 (group B glucose uptake reduction is >than the one for C)
etc.
As a post-SPM analysis, we extracted the voxel values (spm-normalized
and corrected) for the
two scans of each subject from
various locations. (i.e, XA(:,i) for XYZ(:,i) location)
For each given location, we have a pair of pixel values for each subject:
p1 for baseline scan and p2 for the follow-up scan.
A scattergraph was made by using matlab commands:
hold;
for j=1:num_subjs;%for subject j
plot([1 2],[p1(j) p2(j)],'o',[1 2],[p1(j) p2(j)]);
%couple of other commands omitted here
end;
What we found is that,for example with group B,
for all the locations that we have examined so far,
highest p1 is always paired with the lowest p2, and
lowest p1 is always paired with the highest p2.
The pairs in the middle follow a similiar trends.
The attached tif file is such an example.
We thought that this was related to the assumption
that sum of subject_diff is zero, but
it may also be:
(1) a true trend of our data, or
(2) caused by any procedure that we might performed in-correctly,
or something else.
Thanks for your help!
Kewei
|