Hi Nicolai, your first column and second column are redundant. You cannot have both. doug ps. Please remember to send these things to the freesurfer list instead of directly to me. thanks!

On 07/03/2012 02:55 PM, Nicolai Grüner-Hegge wrote:

Hi Doug,

I have been trying to run qdec but have been having an error message pop up. The error message that popped up is provided below:

Design matrix ------------------ 1.000 18.000 2.485; 1.000 18.000 2.526; 1.000 18.000 2.531; 1.000 18.000 2.492; 1.000 18.000 2.456; 1.000 18.000 2.444; 1.000 18.000 2.538;

1.000 18.000 2.529;

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 

/Users/christine/Desktop/TEST_NGH/Reproducibility_FS/RECON/qdec/Untitled/y.mgh --fsgd /Users/christine/Desktop/TEST_NGH/Reproducibility_FS/RECON/qdec/Untitled/qdec.fsgd dods --glmdir /Users/christine/Desktop/TEST_NGH/Reproducibility_FS/RECON/qdec/Untitled --surf fsaverage lh --C /Users/christine/Desktop/TEST_NGH/Reproducibility_FS/RECON/qdec/Untitled/contrasts/lh-Avg-Intercept-thickness.mat --C /Users/christine/Desktop/TEST_NGH/Reproducibility_FS/RECON/qdec/Untitled/contrasts/lh-Avg-thickness-Age-Cor.mat

2. The FSGD file (if using one)
3. And the design matrix above

Error in Analyze: command failed: mri_glmfit --y /Users/christine/Desktop/TEST_NGH/Reproducibility_FS/RECON/qdec/Untitled/y.mgh --fsgd /Users/christine/Desktop/TEST_NGH/Reproducibility_FS/RECON/qdec/Untitled/qdec.fsgd dods --glmdir /Users/christine/Desktop/TEST_NGH/Reproducibility_FS/RECON/qdec/Untitled --surf fsaverage lh --C /Users/christine/Desktop/TEST_NGH/Reproducibility_FS/RECON/qdec/Untitled/contrasts/lh-Avg-Intercept-thickness.mat --C /Users/christine/Desktop/TEST_NGH/Reproducibility_FS/RECON/qdec/Untitled/contrasts/lh-Avg-thickness-Age-Cor.mat

I do not know what an FSGD file is so I guess I am not using one. My measurement was thickness with a smoothing of 10 and the left hemisphere. I have tried it for other parts of the brain and the same message is displayed. Thank you for your time!

Best Regards, Nicolai