If you are going to do a paired test, then each subject can only have 2 time points

On 9/4/19 11:58 AM, miracle ozzoude wrote:

External Email - Use Caution

Hello Freesurfer,

I am trying to run a surface based paired ttest for pet. I started by running mri_vol2surf for each pet longitudinal subject. However, when i use mri_concat with --f, --prune, and --paired-diff i get an error " --paired-xxx specified but there are an odd number of frames". I have 4 subjects, 3 has 3 timepoints and 1 has 2 timepoints (total of 11 frames). Below are my commands and full errors.

How do I solve this problem?

best, Paul

*COMMANDS * mri_vol2surf --mov $in_dir/${session}.gtmpvc.output/mgx.ctxgm.nii.gz --reg $in_dir/${session}.gtmpvc.output/aux/bbpet2anat.lta --hemi lh --projfrac 0.5 \  --o $in_dir/lh.mgx.ctxgm.${session}.fsaverage.sm00.nii.gz --cortex --trgsubject fsaverage

{ echo $in_dir/lh.mgx.ctxgm.${session}.fsaverage.sm00.nii.gz; } >> $lhmgxctx

#concatenate the surface pets mri_concat --f "$lhmgxctxgm" --o ${results_dir}/all.lh.mgx.ctxgm.fsaverage.sm00.nii.gz --prune --paired-diff

session=subject_basline.long.baseid (e.g. 001_baseline.long.001, 001_12mon.long.001 etc)

lhmgxctx = text file containing the pet volumes mapped to the surface for each timepoints. (e.g. /Analysis/001_baseline.long.001/lh.mgx.ctxgm.001_baseline.long.001.fsaverage.sm00.nii.gz /Analysis/001_12mon.long.001/lh.mgx.ctxgm.001_12mon.long.001.fsaverage.sm00.nii.gz etc.)

*ERRORS*

• mri_concat --f Analysis/lh_hemi_ctxgm.txt --o

Analysis/results/all.lh.mgx.ctxgm.fsaverage.sm00.nii.gz --prune --paired-diff-norm1 ninputs = 11 Checking inputs nframestot = 11 ERROR: --paired-xxx specified but there are an odd number of frames

---

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