I'm attempting to draw scouts on a heavily damaged brain from stroke, and use the lesion drawing inform my scout drawing. I'm wondering whether my approach of projecting the lesion (as a scout) is appropriate. Could you please commenting on the below steps?
I have a lesion drawing and a T1.
First, I ran FreeSurfer on the T1 to segment the brain. I imported FreeSurfer results into Brainstorm.
After this, outside of Brainstorm, I normalized the lesion drawing (and T1) to MNI space. Now, I have lesion drawing in MNI space. I used this script: (spmScripts/nii_enat_norm.m at master · rordenlab/spmScripts · GitHub).
Finally, back in Brainstorm with FreeSurfer results, I computed MNI normalization of the imported anatomy, then imported the normalized lesion drawing as a "scout" and overlaid it to the cortex. My steps for this second part were as follows: atlas tab --> add scouts to atlas --> selected the normalized lesion mask under the settings volume mask or atlas (dilated, MNI space). When I got the warning (screenshot below), I said Yes, just to see what it would look like. Then, I applied the MNI transformation as prompted (since I computed it in Brainstorm already, screenshot below), which I believe matches up the lesion drawing and Brainstorm anatomy to MNI space?
The Brainstorm results are below in green, with the MRIcroGL viewer on the right side of the screenshot for comparison. Note: only the "lesion scout" is showing, with no other scouts present.
Is this the correct approach to overlaying the lesion drawing onto cortex in Brainstorm? The results seem reasonable to me, but wanted to obtain your thoughts, especially about the warning I received. I also hope this may be helpful to others who are trying to complete similar tasks.
In this approach, you need to ensure that the MNI normalization that is performed in Brainstorm uses the Segment method. As this is the method that is used in the nii_enat_norm.m script.
Moreover, you can import the normalized lesion volume, with right-click on the Subject, then Import MRI and the option Volume atlas (MNI space). By doing so, it will be possible to verify the alignment between the volumes in Brainstorm.
While using a volume mask to create cortical scouts is possible, results must be verified, as the warning points, the results are not optimal. This happens as to be assigned to a scout a cortical vertex must be INSIDE the volume parcellation, which is not always the case as the cortex vertices are often at the border of the the volume parcellation. See these examples, of cases where it goes wrong:
