Hi,
in order to understand our results I looked into the AAL output more deeply.
On the first glimps it looks really simple but there seems to be a factor somehow in the numbers. And I cannot make sense out of it.
Here ist a small example
Label % Cluster Nb Vx Cluster % Label Nb Vx Label
-----------------------------------------------------------------------------------------
ParaHippocampal_R 71.83 71 1.90 1132
Fusiform_R 25.35 71 0.30 2518
Hipocampus_R 1.41 71 0.04 946
OUTSIDE 1.41 71 0.00 0
When I look at the first row, I can see that 71.83% of my current cluster belongs to the ParaHipp_R. Since the cluster consists of 71 voxel that would be about 51 voxel for this label.
These 51 voxel are 1.9% of the ParaHipp_R, which would mean that the ParaHipp_R consists of 51/0.019 = 2684 voxels.
And this is in contradiction to the 1132 voxels in the table.
Ratio: 2684/1132=2.37
When I do the same calculations for row 2 and 3 I always get this ratio.
In a next step I looked into the code of the m-file: gin_clusters_plabels.m
and found this code (line 368-374):
voiCluster = M(1,1)*M(2,2)*M(3,3);
voiROI = MNID.mat(1,1)*MNID.mat(2,2)*MNID.mat(3,3);
if voiCluster ~= voiROI
fact = abs(voiROI/voiCluster);
else
fact = 1;
end
A test run were I printed out the value of the factor revealed that this is indeed the ratio I found in the output.
My questions are now what this factor stands for and how I can interprete the results.
Thank you for taking the time to look at this.
Best regards,
Katharina
|