hi doug

Thanks for checking my code and the hint with the smoothing rate. I appreciate your time on my issue. I’ve ran a series of different smoothing rates (from 10-30mm) and 25mm showed the best results for our data in terms of cluster size but still prominent effects in summary statistics.

vivian

Am 06.11.2013 um 16:43 schrieb Douglas N Greve greve@nmr.mgh.harvard.edu:

Yes, that all looks correct. Though 25mm is a lot of smoothing. I usually use 10-15mm. doug

On 11/06/2013 10:34 AM, Vivian R. Steiger wrote:

...

Hi Doug

Thanks for your answer on my issue "glmfit for paired-diff on 2 groups" in the mailinglist

Commands:

mris_preproc --target 1_average_30VPs --fsgd responder.txt --hemi lh --meas thickness --out respnon_lh_thickness.mgh —paired-diff-norm

mri_surf2surf --hemi lh --s 1_average_30VPs --fwhm 25 --sval respnon_lh_thickness.mgh --cortex --tval respnon_lh_thickness.sm25.mgh

mri_glmfit --y respnon_lh_thickness.sm25.mgh --fsgd responder_paired.txt doss --C sad_groups.txt --surf 1_average_30VPs lh --cortex --glmdir respnon_lh_thickness

FSGD 1: (responder.txt)

GroupDescriptorFile 1 Title Resonder_NonResponder Class resp Class nonresp Input 1Pre_VP02.long.VP02_Base resp Input 2Post_VP02.long.VP02_Base resp Input 1Pre_VP03.long.VP03_Base nonresp Input 2Post_VP03.long.VP03_Base nonresp Input 1Pre_VP04.long.VP04_Base resp Input 2Post_VP04.long.VP04_Base resp Input 1Pre_VP05.long.VP05_Base nonresp Input 2Post_VP05.long.VP05_Base nonresp Input 1Pre_VP06.long.VP06_Base resp Input 2Post_VP06.long.VP06_Base resp Input 1Pre_VP07.long.VP07_Base resp Input 2Post_VP07.long.VP07_Base resp Input 1Pre_VP08.long.VP08_Base resp Input 2Post_VP08.long.VP08_Base resp Input 1Pre_VP09.long.VP09_Base resp Input 2Post_VP09.long.VP09_Base resp Input 1Pre_VP10.long.VP10_Base nonresp Input 2Post_VP10.long.VP10_Base nonresp . . .

FSGD 2: (responder_paired.txt)

GroupDescriptorFile 1 Class resp Class nonresp Input VP02pair resp Input VP03pair nonresp Input VP04pair resp Input VP05pair nonresp Input VP06pair resp Input VP07pair resp Input VP08pair resp Input VP09pair resp Input VP10pair nonresp . . .

Best, Vivian

---

On 11/04/2013 03:34 PM, Douglas N Greve wrote:

...

Yes, that looks correct, though I could be more confident if you send the command lines used and your fsgd file(s). doug

---

On 10/31/2013 07:32 AM, Vivian R. Steiger wrote:

...

Dear FS-Experts,

I would like to compare 2 Groups (Responder (N=19) vs. Non-Responder (N=11)) with longitudinal data. Participants from both groups had only 2 scan sessions in each case 10 weeks apart.

So far is used the glmfit-variant with --paired-diff option in the first place. (based on

http://surfer.nmr.mgh.harvard.edu/fswiki/PairedAnalysis )

...

The paired-fsgd contained then 2 classes (Responder, Non-Responder) which is used in glm-fit for comparison with the contrast (1 -1) for group diffs.

This approach confirmed our hypothesized effects but I'm not sure whether I used the right strategy.

Could you might help me with this issue?

Thank in advance

Best,

Vivian

-- Douglas N. Greve, Ph.D. MGH-NMR Center greve@nmr.mgh.harvard.edu Phone Number: 617-724-2358 Fax: 617-726-7422

Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting FileDrop: https://gate.nmr.mgh.harvard.edu/filedrop2 www.nmr.mgh.harvard.edu/facility/filedrop/index.html Outgoing: ftp://surfer.nmr.mgh.harvard.edu/transfer/outgoing/flat/greve/

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.