I don't think there is one "right" way to do it. The choice of normalization changes what you are testing, so you have to figure out whether it makes sense based on your hypothesis of what is going on.
On 8/11/2020 1:24 PM, Haewon Roh wrote:
External Email - Use Caution
Hi. team!
Recently, I submitted a paper dealing with volumes of hippocampus and its subfields in epilepsy patients who have significantly reduced hippocampal volume (compared with normal controls) and, in that paper, I used the normalization method using ICV, which was suggested by Jack CR in 1989: normalized volume = volume (observed) – b (ICV – mean ICV), where b is the slope of regression for each ICV.
But, one of the reviewers pointed out that I should use the method based on each patient's hippocampus rather than total intracranial volume. Since I've seen that a lot of papers, which dealt about these kinds of issues about hippocampus and subfields, used the same method with me, I am really doubtful that it is more appropriate if I should use the method using individual volume of hippocampus (rather than ICV) in order to normalize its volumes of substructures in the comparison analysis with those of controls.
# And, whether this is true or not, could you please recommend to me several literatures that can show which method of normalizing volumes of the hippocampus and its subfields is more appropriate for this kind of issue?
# Really sorry for bothering you. Thank you always for your kind and delicate solutions ! : )
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer