Good morning Doug,

I am trying to create a contrast matrix (also attached as a text file) for the following conditions -

I have two treatment arms A and B

For treatment arm A, I have subjects 1, 2, 3 imaged at TP1, TP2, TP3

For treatment arm B, I have subjects 4, 5, 6 imaged at TP1, TP2, TP3.

I have used the repeated measures ANOVA tutorial to set up some of the contrasts as follows -

GroupDescriptorFile 1
Title A-Vs-B-Long
Class A-TP1
Class B-TP1
Class A-TP2
Class B-TP2
Class A-TP3
Class B-TP3
Variables	A-TP1-vs-A-TP2	A-TP1-vs-A-TP3	B-TP1-vs-B-TP2	B-TP1-vs-B-TP3	A-TP1-vs-B-TP1	A-TP2-vs-B-TP2	A-TP3-vs-B-TP3
Input sub1-TP1 A-TP1	1	1	0	0
Input sub2-TP1 A-TP1	1	1	0	0
Input sub3-TP1 A-TP1	1	1	0	0
Input sub4-TP1 B-TP1	0	0	1	1
Input sub5-TP1 B-TP1	0	0	1	1
Input sub6-TP1 B-TP1	0	0	1	1
Input sub1-TP2 A-TP2	-1	0	0	0
Input sub2-TP2 A-TP2	-1	0	0	0
Input sub3-TP2 A-TP2	-1	0	0	0
Input sub4-TP2 B-TP2	0	0	-1	0
Input sub5-TP2 B-TP2	0	0	-1	0
Input sub6-TP2 B-TP2	0	0	-1	0
Input sub1-TP3 A-TP3	0	-1	0	0
Input sub2-TP3 A-TP3	0	-1	0	0
Input sub3-TP3 A-TP3	0	-1	0	0
Input sub4-TP3 B-TP3	0	0	0	-1
Input sub5-TP3 B-TP3	0	0	0	-1
Input sub6-TP3 B-TP3	0	0	0	-1

Question 1 - My question is how do I setup the contrasts for the different treatment arms -

A-TP1-vs-B-TP1

A-TP2-vs-B-TP2

A-TP3-vs-B-TP3

Question 2 - Also I have age and cognitive scores for each timepoint. How do I add those to the contrast file ?

Thank you for the help,

Regards

Rito

On Mon, Jul 12, 2021 at 10:51 AM Douglas N. Greve <dgreve@mgh.harvard.edu> wrote:

Yes, it will work on both. If you use a table, then pass the table with --table tablefile instead of --y

On 7/12/2021 7:54 AM, Ritobrato Datta wrote:

External Email - Use Caution

Good morning,

Thanks Doug for the answers. Quick naive question. The analyses I want to perform are on volumetric data parcellated using aseg and not surface data.

Does mri_glmfit work with the output of the asegstats2table or on voxelwise FA maps ?

Many thanks

Rito

On Fri, Jul 9, 2021 at 4:04 PM Douglas N. Greve <dgreve@mgh.harvard.edu> wrote:

On 7/9/2021 11:44 AM, Ritobrato Datta wrote:

External Email - Use Caution

Hi All,

I have the following data –

I have 205 subjects - each subject was imaged at 3 timepoints (baseline, followup 1 and followup 2)

The 205 subjects are split in two treatment arms with 100 subjects in the first one and 105 subjects in the second one.

For each timepoint, I have created FA maps in their native diffusion space.

I have also extracted the mean FA maps for 187 ROIs using mri_segstats.

For each timepoint, I have saved the results as a matrix (FA in 187 ROIs x 205 subjects) in a text file.

So I have three files for the three timepoints.

I have the age and cognitive score for each subject at each timepoint. And their gender.

I want to answer the following questions –

Do the baseline FA correlate with the corresponding cognitive score at baseline ?

This is a straight forward group analysis, so see MailScanner has detected a possible fraud attempt from "secure-web.cisco.com" claiming to be https://surfer.nmr.mgh.harvard.edu/fswiki/FsgdExamples (maybe the Two Groups (1 Factor, Two Levels), One Covariate)

I am interested in testing whether the FA changed significantly across the different timepoints and does that relate to the change in the cognitive score

See MailScanner has detected a possible fraud attempt from "secure-web.cisco.com" claiming to be https://surfer.nmr.mgh.harvard.edu/fswiki/RepeatedMeasuresAnova

Is there an effect of treatment on this change in FA across time ?

Also see MailScanner has detected a possible fraud attempt from "secure-web.cisco.com" claiming to be https://surfer.nmr.mgh.harvard.edu/fswiki/RepeatedMeasuresAnova

Can you please suggest what programs in freesurfer will allow me to test these questions on both voxelwise and ROI wise ?

Many thanks for your help and guidance,

Regards

Rito