multiple functional overlays

Questions about Anatomist manipulation

Moderators: denghien, riviere

Post Reply
tobias
Posts: 2
Joined: Wed Oct 06, 2004 3:11 pm
Location: Columbia University, NY

multiple functional overlays

Post by tobias »

Dear all,

I would like to overlay several functional volumes (SPM contrast images) onto a single inflated white matter mesh using Anatomist 3D Fusion. I can only see a way to overlay a single image, does anybody know how to have several overlays, ideally colour-coded differently (e.g. red, blue, green)?

many thanks,

Tobias.
Guest

multiple functional overlays

Post by Guest »

As an addition to my previous question: I tried loading a mesh and two functional images into Anatomist and then fusion them together, the result being that my mesh turns black (with the activation clusters of the functional images being grey/white). I have tried to solve this problem by inverting the overlay image values, but the results are the same. Does anybody know how to solve this problem?
Another suggestion for multiple overlays was to turn the clusters into meshes; how do you do that?

Many thanks and best wishes,

Tobias.
User avatar
riviere
Site Admin
Posts: 1361
Joined: Tue Jan 06, 2004 12:21 pm
Location: CEA NeuroSpin, Saint Aubin, France
Contact:

Post by riviere »

Hi,

I have checked that fusioning several volumes with a mesh desn't work in anatomist 1.30: only one volume is taken into account. It will work in the next version of Anatomist.
The black colors may be due to the colormap you are using for the volume. Try a colormap with a "fusion" extension, which have the first color white. Then you can play with the threshold by changing the min color value slider.
If your activation images don't overlap (if no voxel is involved in several clusters), you can build a "label volume" with all clusters and a specific value for each cluster.
To build it, you have to play with commandlines: threshold and binarize each activation map using AimsThreshold:
AimsThreshold -i image -o thresholded.ima -m ge -t <your_threshold> -b
AimsLinearComb -i thresholded.ima -o thresholded.ima -b 32767
(this is to obtain value 1 rather than 32767 on the thresholded image)
Then merge them two by two using coefficients that will lead to different values: if your binarized images are I1, I2... In, you can apply:
AimsLinearComb -i I1 -j I2 -c 2 -o tempimage.ima
(this will do: I1 + 2*I2 so voxels of clusters from I1 will have the value 1 and clusters from I2 will have the value 2)
then:
AimsLinearComb -i tempimage.ima -j I3 -c 3 -o tempimage.ima
and so on.
Then you can use this image and merge it with the mesh using an adequate colormap.

Alternately you can make meshes of the different clusters (after you have binarized images) using AimsMesh but it's a bit more complex and you will probably end up with many different meshes if you have many activation patches.

There is a last way to deal with such data but up to now only "expert" users were able to use it: make a "clusters graph" from the labels volume, a compound object which holds several sub-objects (including meshes), one for each cluster or activation patch. This is done using AimsClusterArg.

Denis
Post Reply