Issues with scouts

Bugs
1

Hi, thank you for reading me.

I have a question about the ROIs and scouts, and I hope you could help me.

I have seen that the FreeSurfer template files has its own atlases. However, when I downsample the cortex file (I prefer to use 5000 vertices), no atlas appears in the list. Is there any way of using atlases with a downsampled cortex?

Thank you in advance.

Kind regards
Victor

1 Like
2

Hi Victor,

This has usually worked for me when I downsample the cortex by using the 'Less vertices ...' option through the GUI. Is that how you did the downsampling?

I would add though that the atlas in the downsampled cortex looks a little different though with gaps between the ROIs. Here is the Desikan-Killiany atlas on the freesurfer template downsampled to 2k vertices:

atlas_downsampleded
atlas_downsampleded701×585 110 KB

So maybe there is a better way?

Best,
Max

3

@VictorRodriguez
I'm not sure what you mean. If I take the FSAverage template, downsample the cortex and load it, I still have all the atlases available:

image
image1021×367 80.8 KB

Please describe better your issue, with screen captures and detailed instructions to reproduce it.

@Nemax
The "gaps" are just a representation issue. A scout is a subset of vertices of the cortex surface, and they are represented with the lines connecting these vertices. All the faces (= triangles) within one region are colored and represented as part of this region, but the faces between two ROIs have no status. And it doesn't matter in Brainstorm because we don't use faces for any computation.

I agree, it would look better if the colored regions were also including parts of the faces, so that we don't see these gaps. But it's not such an easy thing to code... I may work on this some day, but it's not at the top of the list.

image
image790×467 179 KB

1 Like
4

Thanks for the close up. As long as it's just a representation issue it will not affect any analysis.

5

Thank you both!

I'm using ICMB152 template. And when I downsample the cortex (to 5000 vertices, using "Reducepath +Subdivide larger faces") and load it, no atlas appears in the list.

Using the fist option ("Reducepath") there is no problem. However, the second option is suposed to damage the atlases but not removing them. Isn't it?

Thank you again!

6

This option actually deletes the atlases. It's the warning message that is incorrect.
I fixed it: https://github.com/brainstorm-tools/brainstorm3/commit/e0ee2676238d7d3ce52c6742993c700df15ca141#diff-5ba703d036d04f463304a913a21e8293