Hi King-Wai,

Technically you are right. This would be the right way to adjust the scale in area measurements. And eTIV^(1/3) for linear measurements like thickness.

Now practically I am not aware of any work that discusses this (but I did not search for it). I remember that it was consensus in the community that thickness estimates are not adjusted to ICV. This can have two reasons: i) ICV can be a noisy measure and so it can hurt more than it helps (especially when dividing measurements by it as the division amplifies the noise, and ii) it is unclear if thickness is even correlated with head size (or to what extend).

The division by eTIV (mainly for volume estimates) is usually addressed by using it as a confounder variable in the downstream statistics rather than the division. The second assumption (thickness is not correlated with eTIV) can of course be tested. Try to find papers. If it is true, it would mean that larger heads have more brain volume but not necessarily thicker cortices. In that case it is however likely that cortical area increases (more folding for example). Another assumption that should and probably has been tested. If it is true, your normalisation can make sense if you are looking for area increases of regions beyond the effect of head size.

Best, Martin

On 14. Jun 2025, at 04:51, kingwai.chu@gmail.com wrote:

Surface area has the dimension of mm^2 Should it be scaled with something with the same dimension? That is, eTIV^(2/3) which has the dimension of mm^2

From: Douglas N. Greve dgreve@mgh.harvard.edu Sent: Friday, June 13, 2025 11:53 AM To: Freesurfer support list freesurfer@nmr.mgh.harvard.edu Subject: [Freesurfer] Re: Query related to controlling for ICV in the models for surface area and thickness

I'm not sure what you mean. Can you elaborate?

On 6/11/2025 5:09 PM, kingwai.chu@gmail.commailto:kingwai.chu@gmail.com wrote: Hi Doug,

What about the argument of using eTIV^(2/3) makes the scaling dimensionless?

Best,

King-Wai

From: Douglas N. Greve dgreve@mgh.harvard.edumailto:dgreve@mgh.harvard.edu Sent: Wednesday, June 11, 2025 4:09 PM To: freesurfer@nmr.mgh.harvard.edumailto:freesurfer@nmr.mgh.harvard.edu Subject: [Freesurfer] Re: Query related to controlling for ICV in the models for surface area and thickness

Yes, it is recommended. The exact form it should take is in dispute, namely whether you should correct by the eTIV or the eTIV.^(2/3). The second is motivated by the idea of a sphere whose surface area grows with the 2/3 power of the volume. But cortex is not a sphere -- it is folded, so the area can grow faster than 2/3 power. I usually just scale by eTIV. When you run mris_preproc, you can add the --etiv option and it will do the scaling for you. On 6/11/2025 4:05 AM, Muskan Khetan wrote: External Email - Use Caution Dear Team members,

I have a question regarding vertex-wise analysis using FreeSurfer. In my current analysis, I’m looking at the effects of an independent variable on cortical thickness and surface area across the brain. I understand that for ROI-based surface area analyses, it is typically recommended to control for total brain volume (TBV) or intracranial volume (ICV), as surface area is correlated with these global brain size measures. However, I’m unsure whether this recommendation applies to vertex-wise analyses as well or whether any underlying preprocessing steps in FreeSurfer account for brain size differences automatically. Could you please advise on whether TBV or ICV should be included as covariates in vertex-wise surface area models? And more broadly, under what circumstances is it advisable to control for global brain size in vertex-wise analysis? Any insights would be greatly appreciated.

Best, Muskan

_______________________________________________

Freesurfer mailing list -- freesurfer@nmr.mgh.harvard.edumailto:freesurfer@nmr.mgh.harvard.edu

To unsubscribe send an email to freesurfer-leave@nmr.mgh.harvard.edumailto:freesurfer-leave@nmr.mgh.harvard.edu

https://mail.nmr.mgh.harvard.edu/mailman3/lists/freesurfer@nmr.mgh.harvard.e...https://secure-web.cisco.com/1zgqCqj4qntd9bt8wznuUVVVTiFZuyVa-TmLzp-uKKBcf3vcewkEISxSWGva-Ts8eUaJjfAsOXr8ES8n7BAStvHTuIMdtKBdJObfTgS3jtu221BI11OALz_mndTqdO6sM2vKf_9BaQzGhVbpQF5C-yKz74ObFzkPbbS_tf35dZMQ-MAekHIDzeQFrwwG5m0TJGEcg1EtXXkzdclxYpgPTzX4mwjsRWsdkrT6doB6RZO44K4r06Hu310IHnWrfhIqHrcKXSMVC7zAI2rmLoODWf20HcjDOJtE4CZVLbU5g62PmeOv5LybvQb6XFx8qQKC8/https%3A%2F%2Fmail.nmr.mgh.harvard.edu%2Fmailman3%2Flists%2Ffreesurfer%40nmr.mgh.harvard.edu%2F

_______________________________________________ Freesurfer mailing list -- freesurfer@nmr.mgh.harvard.edu To unsubscribe send an email to freesurfer-leave@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman3/lists/freesurfer@nmr.mgh.harvard.e...