Orthohull.py

From MEG Core
Revision as of 09:25, 18 March 2019 by Nugenta (talk | contribs)
Jump to navigation Jump to search

Return to Source Localization - SAM

Description

The program orthohull.py takes a T1 weighted anatomical MRI, with tags placed for the three fiducial points (nasion, and left and right preauricular), and creates a headmodel for use with the SAM software. This program will not function if afni is not installed.

Usage

orthohull.py anat+orig

Other options:

 -p surfacename: :::specifies the surface to use to create the headmodel. The default is the brainhull.ply produced by AFNI's 
                 3dSkullStrip routine.  Other options include innerskull.ply or outerskull.ply
 -x prefix:      Used to specify a prefix for all output files, this defaults to no prefix. 
 -i <inf>:       Used to specify the amount by which the brainhull.ply surface is inflated for the final headmodel.  Defaults
                 to 0.005, which inflates the surface by 0.5cm, preventing clipping. use -i0 for no inflation.
 -t              Do the Talairach transform as part of the preprocessing (recommended). This will be performed using the AFNI
                 program @auto_tlrc. The Talairach transform will be stored in the header of the ortho+tlrc image.
 -m              Same as -t, but instead of TT_N27+tlrc, use MNI_caez_N27+tlrc as the template. 
 -o              By default, the Talairach transform is performed on the ortho/MEG space image, which has been rotated such that 
                 all three fiducial points lie in a horizontal plane. Sometimes the Talairach transform will fail, because of 
                 the large initial mismatch between the ortho image and the template. This option requests that the Talairach
                 transform be derived using the original space skull stripped image. In this case, the transform will be stored
                 in the file ORTHO_to_TLRC.1D, and can be applied to MEG images using 3dWarp -matvec_in2out.
 -q              Forces the output of the hull to be convex (only strictly necessary for problem surfaces)