External Email - Use Caution
Dear Thomas,
Thanks for your response.
I have one more question here: I was looking at the Schaefer parcellation paper where authors have 100, 200, 300, 500...1000 areas parcellation. Is there a way to choose among these parcellations (i.e., what's the basic difference/justification/significance of each of these parcellations over one another) for cortical structure analysis?
Thanks!
On Thu, Jun 24, 2021 at 7:24 PM Thomas Yeo ythomas@csail.mit.edu wrote:
External Email - Use CautionHi Martin,
With regards to adolescents versus adults, there's obviously differences between the two age groups. In our experience, when we re-estimate parcellations from this study (*MailScanner has detected a possible fraud attempt from "secure-web.cisco.com" claiming to be* https://secure-web.cisco.com/1IudlKj4RIkQs10Z1MVq-qyzXMK1KFK6wvN63xEvKi24e5r... https://secure-web.cisco.com/1TF7OmC5fKijUUx5uZkr1n5IUfkGr1RFX7MO-EmdsI3fo1_9zP0Jp-ETRw5NAjxS8BJDa6kKcSMMpJ2sdiX-ot-j-5QFedPJw2Pv_w_Lx09S2qip068nmS_qYJ7jPemY60cjnwP8Sv5YsnioL9TmWNC25owkxNqJp4Z23siRPlxWMF6odSKRD90PbJNM9pQK60dgMUgk07qeKbxs2PfIzD7evz-QcmOcS6NgyV6QttV9sTK7p2YtU6VOPRTXrkqV29tPjdaVkUQg-0QMGTS-45w/https%3A%2F%2Fdoi.org%2F10.1093%2Fcercor%2Fbhab101) in kids (ABCD), the parcellations are not better based on various metrics we computed, so I suspect it won't matter much. But the kids results are not in our paper, so I am not sure the reviewer will buy this argument. For this issue, you could reapply the Yeo2011 algorithm (*MailScanner has detected a possible fraud attempt from "secure-web.cisco.com" claiming to be* https://secure-web.cisco.com/1STyuALGZ52-UjmmBJfV1CdFGDZwA6RF_hF3zsYX0GWpLCk... https://secure-web.cisco.com/1Z2srwkLUjJ_Dnco-Ccq_78ppkQhV3eTVmdR3bkvuvg64MLymHCpblVH7s3BbaYqCgmBhjfn2fLmvOxE52eTv_pNlqqJU66mrQ_kyS9r5xo7vnpF49mnl1CurnSQ-g0lBHQrOu-r1VaMu4bdeod4wOunBUn8z0Ilju3bxxxEVIBwddrah-4RY6d0MV-0qlYnFyxyw5vpq66r2twE7Xzv--mIsL85psoPahHst1Vto9_q39hxhT94w3fNtgOpITnrC9jSZStJGUBEPMuLDb7fWgQ/https%3A%2F%2Fgithub.com%2FThomasYeoLab%2FCBIG%2Ftree%2Fmaster%2Fstable_projects%2Fbrain_parcellation%2FYeo2011_fcMRI_clustering) to your sample and estimate cohort-specific networks and repeat your analysis. We have also developed individual-specific versions of the Yeo2011 networks (*MailScanner has detected a possible fraud attempt from "secure-web.cisco.com" claiming to be* https://secure-web.cisco.com/1uV4DGwqj8a1p5Y1x2Tqm4bmH1OC1GsJDtJUxazbhQpCeX1... https://secure-web.cisco.com/1ja6ZcGwaTlynHGpv68CpBe5wYiryA6Cd9PoPXavhiQcOfiz9Dqk182Jm1NOGknuS5V4LN5APd8rzUErhPpJ6N8fmbPbU1ofhucuPYirEkxbiN-f7KwiyDUiW26HQYY_APS_F3PptKP35VM14zsTm7TwwBSzRZcp1x2GUabhSkC2Uwedw4JQXh_Wned1PegQitYecAZLY0yG_RxpEISUp9zo-MvBV5DdGwVNYsVx8cRr1imnAjiZgQ241c3ZHSKJsuA3WlEi0QZmaYA9NYNL2Ng/https%3A%2F%2Fgithub.com%2FThomasYeoLab%2FCBIG%2Ftree%2Fmaster%2Fstable_projects%2Fbrain_parcellation%2FKong2019_MSHBM), which you might find helpful, although it's probably not necessary based on the reviewer's comments.
The other issue is whether a functional atlas might be appropriate for structural data. I think that really depends on your question. For example, if your question is whether default network cortical thickness might be different between adolescents and adults, I think it's reasonable to use a functional atlas where the default network is well defined. If you use a structural atlas like the Desikan-Kiliany atlas, then you will still need to use a functional atlas to roughly assign the Desikan-Kiliany parcels to the default network, which does not seem that great either.
Regards, Thomas
On Fri, Jun 25, 2021 at 12:52 AM Martin Juneja mj70481@gmail.com wrote:
External Email - Use CautionHi all,
I am reposting my concerns below. I would really appreciate any help.
---------- Forwarded message --------- From: Martin Juneja mj70481@gmail.com Date: Fri, Jun 18, 2021 at 6:16 PM Subject: Yeo atlas for children! To: Freesurfer support list freesurfer@nmr.mgh.harvard.edu
Dear experts,
I used Yeo's 17-network parcellation to study the association between network-wise cortical volume and a clinical variable in adolescents. I received the following two questions from the reviewers regarding the use of Yeo's atlas. I would really appreciate if someone could help us in addressing these questions:
*1) Could the author justify their choice to use a functional parcellation on structural data? How is using this parcellation helpful in understanding the brain? All in all, the authors need to justify the superiority of using Yeo's 17-area parcellations on structural data over other methods of parcellation/analysis. * *2) One issue with Yeo's parcellation also stems from the fact that this parcellation was derived from adult connectomes. Could the authors also address this issue and provide examples that justify their approach.*
Thanks. _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu *MailScanner has detected a possible fraud attempt from "secure-web.cisco.com" claiming to be* https://secure-web.cisco.com/1fMLZlGIypxEQELpYv3-VDWcZPrSydwqaavTCiC-CauyEg8... https://secure-web.cisco.com/1vYwj4u8tnnloAFkq4GNee0aO8bp4AJUBz-f6_rsP7s37_gWii9gsMI3T4i80MIoKCmaAbgSDavMI9Bra-Qk7FqwHAR__FBeRuuC3XCpet50brzAn5-A8K5-0JOTdZXjrfA1qCqFs04yCTLQy1mTe7wOXzmWFuHfI7dyoNfQpgcuJM0neywH2WO4quKfZx9e2C_pm4ronht7RD98Axtw5iLCfeiE4ofV57OGHobrbOveEjP3csgYMC7ZS_59V7Kxia-b8F9Ct_3ip4R_Lpq0PyA/https%3A%2F%2Fmail.nmr.mgh.harvard.edu%2Fmailman%2Flistinfo%2Ffreesurfer
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://secure-web.cisco.com/1fMLZlGIypxEQELpYv3-VDWcZPrSydwqaavTCiC-CauyEg8...