The reason I suggested scaling by eTIV is an analogy with a box of paper. The surface area of the paper in the box scales linearly with the volume in the box (twice the volume, twice the paper, twice the area). I vaguely remember doing a test of this, but, if I did, it has been a while. probably using eTIV^(2/3) works well too. I don't think it is so important that it be unitless (though maybe it is easier at review time).

On 5/3/2024 10:56 AM, Fischl, Bruce R.,PHD wrote:

Hi Yunus

 

I think typically you would normalize by eTIV^(2/3), which would result in a dimensionless scaling

 

Cheers

Bruce

 

From: freesurfer-bounces@nmr.mgh.harvard.edu <freesurfer-bounces@nmr.mgh.harvard.edu> On Behalf Of yunus soleymani
Sent: Friday, May 3, 2024 3:31 AM
To: freesurfer@nmr.mgh.harvard.edu
Subject: [Freesurfer] surface and cortical thickness normalization

 

        External Email - Use Caution        

Hello,

I hope you are doing well. Yesterday, I had some questions regarding normalization in Freesurfer. Thanks to Dr. Douglas N. Greve, I received answers for all of them. However, I still have one more question.

When I normalize brain surfaces to eTIV, they appear to be in two different units (mm2 and mm3). Will this be a problem? Additionally, I would like to know your thoughts on normalizing cortical thicknesses. Do you think they should be normalized to eTIV, or would it be better to normalize them to the mean cortical thickness of each subject?

I would greatly appreciate your response.

Thank you,
Yunus

 

 


_______________________________________________
Freesurfer mailing list
Freesurfer@nmr.mgh.harvard.edu
https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer