That does make sense Andrew. But would that be the case even if I weren’t correcting for ICV? In other words, the analysis works when I don’t correct for it. The GLM still looks for variance in group differences regardless if you’re correcting for something or not…right?

 

 

From: freesurfer-bounces@nmr.mgh.harvard.edu [mailto:freesurfer-bounces@nmr.mgh.harvard.edu] On Behalf Of O'Shea,Andrew
Sent: Friday, November 13, 2015 4:07 PM
To: Freesurfer support list
Subject: Re: [Freesurfer] qdec analysis with ICV

 

I believe your issue is you are trying to do a between group analysis but you only have 1 subject in group B. There is no way to estimate the variance in the measure if there is only one subject in the group.

-Andrew

 

From: <freesurfer-bounces@nmr.mgh.harvard.edu> on behalf of "Hirsch, Gabriella" <Gabriella_Hirsch@meei.harvard.edu>
Reply-To: Freesurfer support list <freesurfer@nmr.mgh.harvard.edu>
Date: Friday, November 13, 2015 at 1:58 PM
To: Freesurfer support list <freesurfer@nmr.mgh.harvard.edu>
Subject: Re: [Freesurfer] qdec analysis with ICV

 

Hi Andrew, the values I’m using are already mean centered --  similarly, I’ve never been able to use the full ICV values. Even with David’s suggestion of  reducing the number of digits (effectively making the numbers smaller), it still doesn’t work…

 

 

Gabriella

 

From: freesurfer-bounces@nmr.mgh.harvard.edu [mailto:freesurfer-bounces@nmr.mgh.harvard.edu] On Behalf Of O'Shea,Andrew
Sent: Friday, November 13, 2015 1:55 PM
To: freesurfer@nmr.mgh.harvard.edu; David Vazquez
Subject: Re: [Freesurfer] qdec analysis with ICV

 

Try mean centering your ICV values (I.e. Subtract the mean value from each individual value). This has worked for me to bypass this error. 

-Andrew

 

From: <freesurfer-bounces@nmr.mgh.harvard.edu> on behalf of "Hirsch, Gabriella" <Gabriella_Hirsch@meei.harvard.edu>
Reply-To: "freesurfer@nmr.mgh.harvard.edu" <freesurfer@nmr.mgh.harvard.edu>
Date: Friday, November 13, 2015 at 1:03 PM
To: David Vazquez <dvazq002@ucr.edu>
Cc: "freesurfer@nmr.mgh.harvard.edu" <freesurfer@nmr.mgh.harvard.edu>
Subject: Re: [Freesurfer] qdec analysis with ICV

 

Thanks for the suggestion David. I tried it but I still got the same error :/

 

Gabriella

 

From: David Vazquez [mailto:dvazq002@ucr.edu]
Sent: Friday, November 13, 2015 12:36 PM
To: Hirsch, Gabriella
Subject: Re: [Freesurfer] qdec analysis with ICV

 

Hi Gabriella.

I'm not an expert, but I've had this problem too.  Reduce the number of digits for ICV by converting it be cm^3 instead of mm^3. Also get rid of some decimal places.  It works for us.

Good luck.
-David

 

On Fri, Nov 13, 2015, 8:10 AM Hirsch, Gabriella <Gabriella_Hirsch@meei.harvard.edu> wrote:

Hi Freesurfer Experts,

I am hoping someone can help me solve this error I'm experiencing while conducting a simple qdec analysis. I have 4 subjects (3 vs 1) that I am trying to analyze for volume differences using a demeaned ICV as a nuisance factor (see attached fsgd file - note I changed the .dat extension as my email server didn't like the dat format).

However, I am getting this error:

Normalized matrix condition is 1e+08
Design matrix ------------------
 0.000   1.000   0.000  -28915.199;
 0.000   1.000   0.000   194736.703;
 0.000   1.000   0.000  -184311.000;
 1.000   0.000   18489.000   0.000;
--------------------------------
ERROR: matrix is ill-conditioned or badly scaled, condno = 1e+08
--------------------------------
Possible problem with experimental design:
Check for duplicate entries and/or lack of range of
continuous variables within a class.
If you seek help with this problem, make sure to send:
  1. Your command line:
    mri_glmfit --y /media/seagate_external_/freesurfer_subjects/qdec/Untitled/y.mgh --fsgd /media/seagate_external_/freesurfer_subjects/qdec/Untitled/qdec.fsgd dods --glmdir /media/seagate_external_/freesurfer_subjects/qdec/Untitled --surf fsaverage rh --label /media/seagate_external_/freesurfer_subjects/fsaverage/label/rh.aparc.label --C /media/seagate_external_/freesurfer_subjects/qdec/Untitled/contrasts/rh-Avg-Intercept-volume.mat --C /media/seagate_external_/freesurfer_subjects/qdec/Untitled/contrasts/rh-Avg-volume-ICV-Cor.mat --C /media/seagate_external_/freesurfer_subjects/qdec/Untitled/contrasts/rh-Diff-CVI-PVL-Intercept-volume.mat --C /media/seagate_external_/freesurfer_subjects/qdec/Untitled/contrasts/rh-Diff-CVI-PVL-Cor-volume-ICV.mat
  2. The FSGD file (if using one)
  3. And the design matrix above
Error in Analyze: command failed: mri_glmfit --y /media/seagate_external_/freesurfer_subjects/qdec/Untitled/y.mgh --fsgd /media/seagate_external_/freesurfer_subjects/qdec/Untitled/qdec.fsgd dods --glmdir /media/seagate_external_/freesurfer_subjects/qdec/Untitled --surf fsaverage rh --label /media/seagate_external_/freesurfer_subjects/fsaverage/label/rh.aparc.label --C /media/seagate_external_/freesurfer_subjects/qdec/Untitled/contrasts/rh-Avg-Intercept-volume.mat --C /media/seagate_external_/freesurfer_subjects/qdec/Untitled/contrasts/rh-Avg-volume-ICV-Cor.mat --C /media/seagate_external_/freesurfer_subjects/qdec/Untitled/contrasts/rh-Diff-CVI-PVL-Intercept-volume.mat --C /media/seagate_external_/freesurfer_subjects/qdec/Untitled/contrasts/rh-Diff-CVI-PVL-Cor-volume-ICV.mat

There are no duplicate entries. I have done this same analysis with a higher number of subjects without problems. I have also tried using the inbuilt qdec ICV (which is not demeaned) and got the same error. Could this be a problem with the small subject count? If so, is there any way to remedy it?

Thank you!
Gabriella



• Confidentiality Notice: If you have received this e-mail in error, please immediately notify the sender by replying to this e-mail, keeping the original message intact. Please do not “Reply All.” After replying to the sender, please delete this e-mail from your files. Information contained in this e-mail is only for the use of the individual(s) or entity for whom it is intended, even if addressed incorrectly.

_______________________________________________
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.



• Confidentiality Notice: If you have received this e-mail in error, please immediately notify the sender by replying to this e-mail, keeping the original message intact. Please do not “Reply All.” After replying to the sender, please delete this e-mail from your files. Information contained in this e-mail is only for the use of the individual(s) or entity for whom it is intended, even if addressed incorrectly.



• Confidentiality Notice: If you have received this e-mail in error, please immediately notify the sender by replying to this e-mail, keeping the original message intact. Please do not “Reply All.” After replying to the sender, please delete this e-mail from your files. Information contained in this e-mail is only for the use of the individual(s) or entity for whom it is intended, even if addressed incorrectly.



• Confidentiality Notice: If you have received this e-mail in error, please immediately notify the sender by replying to this e-mail, keeping the original message intact. Please do not “Reply All.” After replying to the sender, please delete this e-mail from your files. Information contained in this e-mail is only for the use of the individual(s) or entity for whom it is intended, even if addressed incorrectly.