I have a two part question regarding subcortical segmentation:

1) To visually inspect if segmentation accurately captures structure boundaries (e.g., amygdala) I am using the norm.mgz file and overlaying it with the aseg.mgz file in Freeview. 

Is this correct? Or should I be using brainmask.mgz (instead of norm.mgz)?

Also, Freeview gives me some options - "Resample to standard RAS space" and "Apply registration file". Where can I find information about whether or not to use these options? 

2) If I manually edit the boundaries of the amygdala (either by erasing voxels or using the color brush to add voxels), what command should I use to re-run such that the corrections are applied to my aseg.stats output? 

The command I found on the wiki page says to use recon-all -autorecon-2-noaseg -subjid <Subj_ID>. My understanding was that the -noaseg flag will skip all the subcortical segmentation steps and that is not what I want. What I want is for FS to re-run subcortical segmentation such that it reflects the voxel editing I did in Freeview.

This is the first time I am doing something like this so any help you can provide or pointing me to the right resource would be great!

Thank you!

Reema

--