Two things. First, I would extract the data straight from the stack used as input (--y) to glmfit instead of mapping the label back to each individual. There are a number of transformation that are hard to make symmetrical going from subject to fsaverage and back. This can particularly affect small ROIs. Second, how did you do the test? What software? Are you totally sure that the design matrix is exactly the same?
On 08/05/2014 02:40 PM, Yunxin Wang wrote:
Hi folks,
I tried to extract the thickness/volume value of the ROI labels transferred from the annot files that're generated automatically by FreeSurfer after a succesful group analysis with corrected significant results. However, the thickness/volume values are no longer correlated with the independent variable that I used in the group analysis model earlier and even the direction of group difference seems to be inconsistent with the group analysis results as well. I have no idea how it went wrong and wondered maybe there's something wrong with my scripts. Have any of you run into such kind of problem?
My script for ROI value extraction goes like this: #annot2label: mri_annotation2label --subject fsaverage --annotation onsetage_thickness --hemi rh --labelbase /expdata/wangyunxin/reconall_lj/fsaverage/labels/onsetage_thickness
for i in sub* do #label2label: mri_label2label --srclabel fsaverage/labels/onsetage_thickness-001.label --srcsubject fsaverage --trglabel ./$i/labels/onsetage_thickness-001.label --trgsubject $i --regmethod surface --hemi rh #label2stats: mris_anatomical_stats -l $i/labels/onsetage_thickness-001.label -t $i/surf/rh.thickness -f $i/stats/rh.onsetage_thickness-001.stats -nsmooth 10 $i rh done
#stats2table aparcstats2table --subjectsfile /expdata/wangyunxin/reconall_lj/music/sublist48 --hemi rh --meas thickness --parc onsetage_thickness-001 --tablefile music/onsetage_thickness-001_rh.txt
Any suggestion would be great! Thanks!
Best, Yunxin
Yunxin Wang Visiting PhD Student Gabrieli Lab McGovern Institute for Brain Research at MIT Office: 46-5107 Tel: 617-955-2809 http://gablab.mit.edu/index.php/gablab/people/14-sample-data-articles/220
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer