<div dir="ltr"><div><div><div><div><div>Dear FSL experts,<br><br>I have a diffusion data set with
 3 groups, 2 time points each (pre&amp;post training), and want to see, 
for each group, where there are changes in FA from pre 
to post, and also where these changes differ between the groups (i.e. 
time x group interaction), while controlling for age and gender.<br><br></div>I know that for VBM, the image fed to randomise is created by subtracting the 4D merged&amp;registered image corresponding to the pre scans from that of the post scans, and then manually smoothing it. However, I haven&#39;t found a similar guide for TBSS on the wiki and in the list archives. <br>

<br>I am wondering:<br><br>1)  How is the difference image created exactly?Is the subtraction made for the FA values themselves (i.e. are the individual images are the dti_FA images) or for the skeleton (e.g. all_FA_skeletonised)? <br>

</div>2) At which stage should this subtraction be performed, e.g. before bedpostx, before tbss_1_preproc etc ?<br></div>3) Does the TBSS difference image need to be manually smoothed in the same way (e.g. sigma=2 if expecting a small effect) as for VBM?<br>

</div>4) I also want to do tractography to quantify the strength of several homotopic tracs (e.g. left DLPFC to right DLPFC) and extract FA values from those specific tracts. Again, at which point in this operation would the above subtraction intervene?<br>

<br></div><div>Thanks very much for your help!<br><br>Tudor<br></div>


</div>