Sorry, I meant to say that you cannot use QDEC with more than 2 discrete variables (you have 3). You do not need to extract data from the smoothed thickness files, but the results you get from your ROI analysis may then differ a bit from that you get from the QDEC analysis. doug

krista kelly wrote:

Thanks Doug! Just a few more things:

I only have two groups, so my contrast was wrong. Should it only be 1 -1, even if I have other variables I want to control for? Or should I have a contrast file for each discrete variable?

How would I extract the smoothed data using mri_anatomical_stats?

I realize the importance of smoothing data while doing a vertex-by-vertex analysis in QDec, but if I'm doing an ROI analysis by extracting the average thickness in a certain label (i.e. v1), is it necessary to do the stats on the smoothed data from each participant?

Thanks!

On Tue, Oct 18, 2011 at 2:23 PM, Douglas N Greve <greve@nmr.mgh.harvard.edu mailto:greve@nmr.mgh.harvard.edu> wrote:

krista kelly wrote:

    Hi, just wondering if anyone has any advice on the following
    questions:

    I'm using QDEC to analyze cortical thickness differences
    between groups and I have a few questions:

    1) There are a few variables that I need to control for. First
    variable is scanner - I used two different scanners of
    different field strengths (1.5T GE and 3T Siemens) so I wanted
    to control for scanner. I also wanted to control for age and
    gender. Here's my ideal design setup:

    discrete variables - groups, scanner, gender
    nuisance variable - age

    contrast would be 1 -1 0 0 0 0
    I realize that scanner, gender, and group are all discrete
    variables, and that age would be a continuous one. However, in
    the design tab, I am not able to select all 3 discrete
    variables (groups, scanner, and gender). Should I designate
    scanner or gender as a nuisance variable instead? Would it
    matter which one I use as the nuisance variable? Possible setup:

    discrete vars - groups, scanner
    nuisance vars - age, gender

    contrast - 1 -1 0 0 (Correct?)

Unfortunately, you cannot use QDEC if you have three groups.
You'll have to create an FSGD file and contrasts by hand, then run
mri_glmfit. Do a search for FSGD on our wiki and go to the
examples page. I think there is an example there for what you want
to do.


    2) If I use DODS, should I demean the ages if age is a
    nuisance variable?

You should actually test whether there is an interaction between
age and your group contrast of interest (using the DODS model). If
there is not interaction, then switch to a DOSS where demeaning
does not matter.


    3) If I use a smoothing kernel during the analysis, then
    create an ROI and map that ROI to each individual subject,
    will the resulting .stats file in each subjects stats folder
    reflect the smoothed data or does mapping onto each ROI just
    extract the raw unsmoothed data from each vertex?

It depends on how you do the extraction. If you are using
mris_anatomical_stats on each individual's unsmoothed data, then
it will reflect the unsmoothed.

doug


    Thanks!
    Krista
    ------------------------------------------------------------------------

    _______________________________________________
    Freesurfer mailing list
    Freesurfer@nmr.mgh.harvard.edu
    <mailto:Freesurfer@nmr.mgh.harvard.edu>
    https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer


-- 
Douglas N. Greve, Ph.D.
MGH-NMR Center
greve@nmr.mgh.harvard.edu <mailto:greve@nmr.mgh.harvard.edu>
Phone Number: 617-724-2358 <tel:617-724-2358> Fax: 617-726-7422
<tel:617-726-7422>

Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting
<http://surfer.nmr.mgh.harvard.edu/fswiki/BugReporting>
FileDrop: www.nmr.mgh.harvard.edu/facility/filedrop/index.html
<http://www.nmr.mgh.harvard.edu/facility/filedrop/index.html>



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.