Note: You must be registered in order to post a reply. To register, click here. Registration is FREE!
T O P I C R E V I E W
Jeff_Browndyke
Posted - Oct 14 2013 : 2:45:35 PM I've been working with FSL basic TBSS scripts with relative success, but being a FSL novice I've been using the processing recipe here: http://fsl.fmrib.ox.ac.uk/fsl/fslwiki/tbss/UserGuide
This guide, however, only provides help on constructing design.mat and design.con files for basic 2-sample t-tests. I'm trying to figure out how to construct a repeated measures x 2 sample ANOVA (i.e., group x time) for 12 patients and 12 controls to examine differences in FA between baseline and 6wk post-op and their interaction. Does anyone know how to code this sort of model in FSL and the required contrasts? I looked at the code in the design.mat and design.con files for the 2-sample and that code seems pretty simple, but I'm not certain what to do about a repeated measures ANOVA where within- and between-variance needs to be computed.
BTW - I'm processing all four timepoints together to obtain an overall study specific mean FA skeleton and have named the files alphabetically to indicate the groups by timepoints (e.g., Controls Tp1 = A0...n; Patients Tp1 = B0...n; Controls Tp2 = C0...n; Patients Tp2 = D0...n).
Posted - Oct 14 2013 : 3:35:15 PM no, you keep using randomise .. Set-up your glm with Glm, but now you'll need design.mat, design.con and design.fts
So add this flag to your randomise line:
-f <design.fts> f contrasts file
Jeff_Browndyke
Posted - Oct 14 2013 : 3:29:38 PM Thanks, Chris! I'm assuming that I cannot use "randomize?" The TBSS script line I've been using for the 2-sample procedure with TFCE correction and 500 permutations is:
