Re: [Freesurfer] Flipping left-right hemispheres

You have uncovered a bug. I don't think anyone has tried to run xhemi on a longitudinal data set before. There is a work-around. First run

xhemireg --s subject --no-tal

Then run

surfreg --s subj1_Tp1.long.subj1_base --t fsaverage_sym --rh --xhemi

I think this should work

doug

On 03/14/2014 03:39 PM, Peggy Skelly wrote:

Hi All,

Although the question of flipping hemispheres has come up before, I still struggling with it. I'm working on a longitudinal study with stroke subjects (half with lh lesions, half with rh lesions). We want to compare only the non-lesioned, healthy, hemispheres at 2 timepoints (pre- and post-therapy).

Using the tutorial (http://freesurfer.net/fswiki/FsTutorial/LongitudinalTutorial_freeview) as a guide, I've completed the longitudinal analysis (cross-base-long steps) for both lh- and rh-lesioned subjects.

I've done the thickness data qcache for the right/healthy hemi for the lh-lesioned group: long_mris_slopes --qdec long_qdec_lh_lesion_subjects_table.dat --meas thickness --hemi rh --do-avg --do-rate --do-pc1 --do-spc --do-stack --do-label --time my_time_var --qcache fsaverage_sym --sd $SUBJECTS_DIR

Convert the long_qdec_lh_lesion_subjects_table.dat into a cross.qdec.table.dat to use in qdec. This works well---I can look for clusters where athe average thickness-rate is different from 0.

How can I combine the rh-lesioned group with the first group?

I thought I could flip hemispheres of the results at each subject's longitudinal output timepoint with: surfreg --s subj1_Tp1.long.subj1_base --t fsaverage_sym --rh --xhemi

but this terminated with an error:

...

recon-all -sb s1311_15T_post.long.s1311_15T_base/xhemi -talairach

ERROR: Are you trying to run or re-run a longitudinal time point? If so, please specify the following parameters:

   \' -long <tpNid> <templateid> \'

   where <tpNid> is the time point id (SAME as cross sectional
   ID) and <templateid> is the ID created in the -base run.
   The directory <tpNid>.long.<templateid> will be created
   automatically or used for output, if it already exists.

I know it's recommended to flip hemispheres after processing (http://www.mail-archive.com/freesurfer%40nmr.mgh.harvard.edu/msg34040.html). When (and how) is it best done?

Thanks, Peggy

---

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