Re: [Freesurfer] Fwd: Questions aobut using monte carlo sim in Qdec

On 04/01/2014 05:40 PM, Ashley Shurick wrote:

Hi all,

What is the default cluster-wise threshold in qdec for monte carlo simulations? My output says the area threshold is 0mm^2, which doesn't make sense (see attachment).

the threshold is .05. The minimum area threshold is ignored in favor of a the cluster p-value threshold (which is a different way to set the minimum area).

Additionally, I have two follow-up questions to an email posted Sat :

1. How do you explicitly change the CW P threshold in the qdec gui?

I'm attaching a pic of my qdec gui and the corresponding terminal output. According to a previous answer the vertex-wide threshold is .01 while the CW Pvalue threshold is .05 (see terminal output) and the visualization threshold is 2 (p < .01; please let me know if I'm wrong with any of this info). How do I go about changing the CW Pvalue threshold in the qdec GUI?

Sorry, this was an oversite in the GUI design. It is hardcoded at .05. If you want to change it, you have to run mri_glmfit-sim from the command-line

2. why does the gui only display one cluster when I hit "find clusters

and go to max" when the simulation outputs 2 clusters? According to previous posts I believe I should be setting my vertex-wide threshold to the same thing used in the simulation, which is 2 (p< .01), yet I'm getting different results. Can you please explain the discrepancy?

You may need to change the threshold min to 1.3 (p<.05) doug

Thank you for your assistance,

Ashley

On Sun, Mar 30, 2014 at 11:47 AM, Douglas Greve <greve@nmr.mgh.harvard.edu mailto:greve@nmr.mgh.harvard.edu> wrote:

On 3/29/14 3:09 PM, Ashley Shurick wrote:

...

Hi,

I have some questions about using monte carlo sim in qdec:

First of all, I just want to verify it's ok to use the simulation
with a whole-brain analysis and not just an ROI analysis.

yes

...

Second, I'm a bit confused about how to calculate the p-threshold
for my max clusters. Looking here:
http://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/GroupAnalysisClusterSummary

...

tells me how to calculate the p-value of my clusters, but
according to the page, max = -log(base10) (pvalue). The example
given is the 13th cluster where max = -5.981, and the stated p
value is p < .006. I can't figure out how you came up with this
p-value, can you please clarify your calculations?

This is not the cluster-wise p-value, this the most significant
vertex inside the cluster. This p-value is computed from the GLM
and contrast that you supply. The cluster p-value is in a
different column in the table.

...

Finally, in qdec when I change my "threshold" from say 1.3 (.05)
to 2.0 (.01) I notice the CW Pvalue threshold doesn't change in
the terminal output, and remains at .05. Can you please explain
this? What exactly am I changing when I change the "threshold" in
qdec?

That is the vertex-wise threshold. You need to explicitly change
the CWP threshold in the qdec GUI
doug

...

Thank you in advance for your assistance,

Ashley


-- 
Ashley A. Shurick
Ph.D. Candidate
Department of Psychology
Stanford University


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

_______________________________________________
Freesurfer mailing list
Freesurfer@nmr.mgh.harvard.edu <mailto: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.

-- Ashley A. Shurick Ph.D. Candidate Department of Psychology Stanford University

-- Ashley A. Shurick Ph.D. Candidate Department of Psychology Stanford University

---

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