Hi freesurfer masters, Quick question on the way freesurfer calculates the dice coefficient. I am interested in getting an INTERsubject dice coefficient as a measure of similarity between 2 subjects. My question is, do I need to do an INTERsubject registration step before calculating the dice in freesurfer? In other words, does freesurfer do any kind of INTERsubject registration at the level of the aseg.mgz file? If it does not, when it calculates an intersubject dice coefficient, a low result could be due simply to poor alignment between scans and not poor anatomical correlation. If anyone has details please let me know. Cheers, Jared
Hi Jared,
yes, the talairach.m3z is the intersubject registration we use for generating the aseg, and you could use it to compute a dice.
cheers Bruce
On Tue, 30 Jun 2009, Jared Price wrote:
Hi freesurfer masters, Quick question on the way freesurfer calculates the dice coefficient. I am interested in getting an INTERsubject dice coefficient as a measure of similarity between 2 subjects. My question is, do I need to do an INTERsubject registration step before calculating the dice in freesurfer? In other words, does freesurfer do any kind of INTERsubject registration at the level of the aseg.mgz file? If it does not, when it calculates an intersubject dice coefficient, a low result could be due simply to poor alignment between scans and not poor anatomical correlation. If anyone has details please let me know. Cheers, Jared _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
freesurfer@nmr.mgh.harvard.edu