Print

Print
Hi Alain,

If the percentages of atlasquery do not add up to 100% this implies that there are voxels in your mask outside of the ROIs defined in the atlas. In your case about 43% of the mask appears to lie outside the atlas, so before making any claims about the distribution of your mask I would check where these voxels lie. You can do this in flsview by overlaying your mask with the atlas and check where your mask appears to exceed the atlas ROIs.

Best,

Michiel

On Fri, Jan 30, 2015 at 3:08 PM, Alain Imaging <[log in to unmask]> wrote:
Hi everybody!

I have a rather basic question about the output of atlasquery.

I have performed the shape analysis of the putamen, and I found a cluster of correlation between shape and my behavioral variable.
I would like to know if this cluster fall in the limbic or in the executive part of the putamen, and to get this information I binirize my cluster (fslmaths Puta_tfce_corrp1 -thr 0.95 -bin output) and then I submit this cluster to a query as follow:
> atlasquery -a "Oxford-GSK-Imanova striatal connectivity atlas" -m output

I get the following output:

limbic:2.35577
executive:2.04327
rostral-Motor:4.90385
caudal-Motor:25.4567
parietal:23.0048


Now, what I understand is that this cluster fall mainly in the caudal motor part and in the parietal part, but, since the percentage don't add to 100, I guess that I cannot say that the 25.45% of the cluster is in the caudal-motor and the 23.00% is in the parietal part.

What's the correct way of report the information that I get from the query ?

Thank in advance 

Alain