Hi Mike
nice to be right once in a while :). You shouldn't have to register to an atlas brain to build the folding pattern template though - you can pick one of your own dataset to start with.
cheers Bruce On Thu, 9 Jan 2014, Michael Chee wrote:
Hi Bruce and Linda,
Yes, this is my work email. Pacific internet shut down email operations in Singapore in Dec 2012…
We built an ‘Asian Only’ template with around 30 Chinese participants but the use of this template did not generate a result that differed from that obtained using the stock MGH template. We were quite disbelieving of Bruce’s prediction that this would be so but he proved to be correct. The caveat (and its not a small one) is that to BUILD the template, one has to register to an atlas brain which is presumably Caucasian weighted.
The paper comparing predominantly Caucasian and Chinese brains using FS was published in 2011 in J Cog Neurosci and can be found on my lab website. We did not mention building the custom atlas in that work because of the lack of a difference in using this vs. the standard template.
Zheng Hui (copied) did all the hard work on this.
Michael Chee MBBS, FRCP(Edin) Professor Cognitive Neuroscience Laboratory Neuroscience and Behavioral Disorders Program Duke-NUS Graduate Medical School 8 College Rd, #06-18 Singapore 169857.
email: michael.chee@duke-nus.edu.sg office: (65) 65164916 CNL Home page: http://www.cogneuro-lab.org
From: Bruce Fischl fischl@nmr.mgh.harvard.edu Date: Thursday, 9 January, 2014 5:24 am To: Linda Zhang lzhang87@gmail.com Cc: Freesurfer freesurfer@nmr.mgh.harvard.edu, Michael Chee michael.chee@duke-nus.edu.sg Subject: Re: Freesurfer templates for cross-ethnicity comparisons
Hi Linda
I had an old address for Mike in the first email. I think this one is correct. Building a folding atlas is pretty quick, depending on how many subjects you have. You'll need to pick one subject, run mris_make_template to build a single-subject template (for each hemi), register everyone to it, rerun mris_make_template to build a multi-subject atlas, rerun mris_register with the new atlas as the target and iterate this procedure until it converges (atlas building and reregistering). Usually it does after only one additional round.
cheers Bruce
On Wed, 8 Jan 2014, Linda Zhang wrote:
Thanks Bruce! How long would it take to build an atlas?What are your thoughts Mike?
On 7 January 2014 14:38, Bruce Fischl fischl@nmr.mgh.harvard.edu wrote: Hi Linda
I actually don't know the ethnicity of the subjects in our atlas. I wouldn't expect it to be an issue for folding patterns, but it is possible. You could build your own atlas and redo the analyses if you want to see if that changes things. I'll cc Mike Chee who I think has looked at this question as well.
cheers Bruce
On Tue, 7 Jan 2014, Linda Zhang wrote:
Hi Bruce,I've been using Freesurfer to compare subcortical volumes and cortical thickness between Caucasian and Chinese controls. There are some significant findings, but the question of whether or not these may be biased due to the template has been raised (the assumption being that the majority of subjects used in the template are Caucasian). Is it possible to reduce the weight of the atlas during normalisation (or some other method to address the issue)? I would greatly appreciate your thoughts on this.
Cheers, Linda
The information in this e-mail is intended only for the person to whom it is addressed. If you believe this e-mail was sent to you in error and the e-mail contains patient information, please contact the Partners Compliance HelpLine at http://www.partners.org/complianceline . If the e-mail was sent to you in error but does not contain patient information, please contact the sender and properly dispose of the e-mail.