Hello everone,
I am using brainstorm for MEG data and studying the spatial topography maps created by 20 component ICA.
I am not loading any MRI scan or any other anatomy.
I am trying to build a classifier using these images. The problem I have is that the background of the topography maps (ear, nose) are not constant between subjects. Sometimes the nose moves up, and the head moves down. Furthermore, the shape of the head also changes significantly. I have attached 2 sample figures to illustrate what I am trying to say. (Why is there this change?)
According to this link, what I need is a â2D discâ topography, which creates a more uniform circle.
However, I do not know how to create such images for ICA active projectors. The default here seems to be â2D sensor capâ and there is no option of changing it.
Alternatively, is there a way for me to just work with the actual topography, ie, without any background elements (nose, ears)?
Thank you for your help in this manner,
Sincerely,
drgarg