Hi John, it has been a long time since I looked at version 4.3, so I don't know if it can or not. One thing you can try is to use the version 5.x mri_segstats (the program which computes the aseg.stats table). You could run this outside of recon-all and generate a file with a different name. That way you won't have to worry about mixing version. doug
On 11/30/2012 03:00 PM, John Drozd wrote:
Hello,
My freesurfer version is freesurfer-Linux-centos4_x86_64-stable-pub-v4.3.0
I noticed that the aparc+aseg.mgz label map is resampled to 1mm x 1mm x 1mm voxels, eventhough the .mgz grayscale mri image file remains as per the original voxel size and is not 1x1x1. mri_convert .nii to .mgz translates the grayscale image down the y direction in the horizontal plane transverse/axial slices.
I noticed that the aseg.stats files list the number of voxels and the volume, and these numbers match because the voxel size is 1 x 1 x 1.
My question is whether freesurfer 4.3 can be invoked to perform partial volume error corrections, and if so how do you invoke this feature?
My interests lie in segmentations of the lateral ventricles including their posterior and inferior horns.
thank you, john _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer