You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
For the input single-subject brains (on any organ with left-right symmetry), we would like to split along the mid-sagittal plane, so that we can mirror each hemisphere, and thus end up with two symmetric images - one derived from the left and one derived from the right side. This is a necessary step for creating symmetric templates. It will also allow us to use 1/2 hemisphere for template construction, in case the other one is damaged.
This will likely require reorienting the brain so that the midsagittal plane is perfectly vertical and in the middle of the image matrix. We could do this is one of several ways:
manually reorient in napari for one brain, and then rigidly align the other brains to the manually aligned one.
Apply ACPC transform to each brain, for example with 3D slicer. This will require much more manual work and is harder to generalise to non-brain organs.
Some algorithm that automatically finds a plane of symmetry in a 3D image. This, if feasible, could be more general. As input, we could also take a brain mask instead of the image.
The text was updated successfully, but these errors were encountered:
For the input single-subject brains (on any organ with left-right symmetry), we would like to split along the mid-sagittal plane, so that we can mirror each hemisphere, and thus end up with two symmetric images - one derived from the left and one derived from the right side. This is a necessary step for creating symmetric templates. It will also allow us to use 1/2 hemisphere for template construction, in case the other one is damaged.
This will likely require reorienting the brain so that the midsagittal plane is perfectly vertical and in the middle of the image matrix. We could do this is one of several ways:
The text was updated successfully, but these errors were encountered: