Dear Lilla,

Thank you very much for your answer and your help. I followed all the recommended steps (I introduced a white matter mask that will ensure that I am  only considering diffusion values in the white matter and then also I did corrected for multiple comparison running mri_glmfit-sim with the option grf as Doug recommend me).

My question is how can I conclude if there is an increase and/or decrease in the FA values in one group (for example Patients) versus another group (example controls).

In other words can we conclude on the FA values based on the GLM results/clusters  from the http://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/Diffusion (I don't see anymore the image with red and blue clusters posted on this website. Is something going on?)

Thank you very much! Antonella

________________________________ From: Lilla Zollei lzollei@nmr.mgh.harvard.edu To: Antonella Kis atorok9@yahoo.com Cc: "freesurfer@nmr.mgh.harvard.edu" freesurfer@nmr.mgh.harvard.edu Sent: Tuesday, December 13, 2011 12:02 PM Subject: Re: [Freesurfer] FA interpretation from cvs registration for group analysis

Hi Antonella,

I will appreciate a lot if you can help me understand what is the meaning of the results after running after running mri_cvs_register and the GLM group analysis.

More exactly, I followed all the instructions from the webpage:

http://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/Diffusion

and I wonder if based on the cluster's colours I can say something about my FA values in one group related to the other one.

More exactly I am looking to understand if I have an increase in the FA value for the controls versus patients or the FA in patients is larger than in the control group.

My contrast matrix is C = [+1.00000 -1.00000 ] and I have 30 controls and 32 patients.

I am not sure how to comment on your outcomes as I don't know anything about your data. However, the diffusion tutorial that we put together is a very top-level roadmap on how one might go about voxel-based analysis of their diffusion data. It is not complete in a sense that you would want to make sure that there is some further processing done on your data sets before you start interpreting the results. One is that you would want to introduce a white matter mask that will ensure that you are only considering diffusion values in the white matter and then also you would want to correct for multiple comparison running mri_glmfit-sim. We are planning on updating this tutorial further for our next course to reflect all these additional steps.

What is the age range for the cvs_avg35 space? I wonder if is the best space to register my subjects if they are children with age between 6-14.

The atlas was constructed from 35 adults, consisting of males and females ages 18 to 93.

We have been successul registering pediatric images also in that space, but the outcome will also depend on the quality of your images.

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

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.

Hi Antonella,

I am doing a DTI group analysis study more exactly I am interested to see any changes in the FA values within two groups: controls versus patients (children with epilepsy). I introduced a white matter mask that will ensure that I am  only considering diffusion values in the white matter and then also I did corrected for multiple comparison running mri_glmfit-sim with the option grf (as Doug recommend me).

My questions are:

1)How can I conclude if there is an increase and/or decrease in the FA values in one group (for example patients) versus another group (example controls). In other words what I can conclude regarding the FA values based on the GLM results: blue versus red clusters  from the http://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/Diffusion (I don't see anymore the image with red and blue clusters posted on this website!)

See my previous email.

2. If my fa-mask was defined using the wmparc2diff.mgz so is a segmentation volume, in order to resample this volume (wmparc or aseg) in CVS space, do I need  to use the nearest neighbor (nearest) interpolation option instead of --interp trilin? What will be the

difference and is the difference visible in the intensity or in another parameter such as a smoother registration?

Yes, with any of the segmentation / parcellation file you would want to use the nearest neighbor interpolation option. The difference (if you used trilinear instead) would be visible. You would see new and uninterpretable segmentation lables in the resampled volume which are not valid if you want to interpret them as anatomical references. (The interpolation does not change the registration results though, it just affects the way your resampling gets executed.)

--Lilla

Hi Antonella,

Thank you very much for your answer and your help. I followed all the recommended steps (I introduced a white matter mask that will ensure that I am  only considering diffusion values in the white matter and then also I did corrected for multiple comparison running mri_glmfit-sim with the option grf as Doug recommend me).

My question is how can I conclude if there is an increase and/or decrease in the FA values in one group (for example Patients) versus another group (example controls).

In other words can we conclude on the FA values based on the GLM results/clusters  from the http://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/Diffusion (I don't see anymore the image with red and blue clusters posted on this website. Is something going on?)

You will need ot look at your corrected significance map to conclude whether the hypothesis that you set up holds or not.

And the reason why some part of the diffusion group analysis is missing now is that we are trying to make the tutorial more complete and comprehensive and while we are doing that we decided not to display the older version.

--Lilla

---

From: Lilla Zollei lzollei@nmr.mgh.harvard.edu To: Antonella Kis atorok9@yahoo.com Cc: "freesurfer@nmr.mgh.harvard.edu" freesurfer@nmr.mgh.harvard.edu Sent: Tuesday, December 13, 2011 12:02 PM Subject: Re: [Freesurfer] FA interpretation from cvs registration for group analysis

Hi Antonella,

...

I will appreciate a lot if you can help me understand what is the meaning of the results after running after running mri_cvs_register and the GLM group analysis.

More exactly, I followed all the instructions from the webpage:

http://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/Diffusion

and I wonder if based on the cluster's colours I can say something about my FA values in one group related to the other one.

More exactly I am looking to understand if I have an increase in the FA value for the controls versus patients or the FA in patients is larger than in the control group.

My contrast matrix is C = [+1.00000 -1.00000 ] and I have 30 controls and 32 patients.

I am not sure how to comment on your outcomes as I don't know anything about your data. However, the diffusion tutorial that we put together is a very top-level roadmap on how one might go about voxel-based analysis of their diffusion data. It is not complete in a sense that you would want to make sure that there is some further processing done on your data sets before you start interpreting the results. One is that you would want to introduce a white matter mask that will ensure that you are only considering diffusion values in the white matter and then also you would want to correct for multiple comparison running mri_glmfit-sim. We are planning on updating this tutorial further for our next course to reflect all these additional steps.

...

What is the age range for the cvs_avg35 space? I wonder if is the best space to register my subjects if they are children with age between 6-14.

The atlas was constructed from 35 adults, consisting of males and females ages 18 to 93.

We have been successul registering pediatric images also in that space, but the outcome will also depend on the quality of your images.

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

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.