 |
 |
Thank you, Gwenaelle, for the tip of saving the time and effort. I will follow your suggestion. On Fri, 7 Nov 2008 14:25:18 +0000, Gwenaëlle DOUAUD <[log in to unmask]> wrote: >Hi, > >as you can't be absolutely sure there is no difference between the two samples (maybe the first one consists of slightly bigger brains for instance), you should redo the analysis merging all the data, sorry. >The good news is that you can skip fslvbm_1 and also the segmentation in fslvbm_2 by slightly changing the script, which will save you a lot of time... Just put all your *_struc.nii.gz *_struc_brain.nii.gz (and *_struc_GM.nii.gz if you've changed the fslvbm_2 script to skip tissue segmentation) in a same, new struc/ directory. > >Hope this helps, >Gwenaelle > > >--- En date de : Ven 7.11.08, X Liu <[log in to unmask]> a écrit : > >> De: X Liu <[log in to unmask]> >> Objet: [FSL] Combining two VBM data sets >> À: [log in to unmask] >> Date: Vendredi 7 Novembre 2008, 14h55 >> I am trying to combine two preprocessed VBM data sets for >> the final >> statistical analysis with "randomise". I have >> processed them before >> separately with the same version of FSL. They are all >> college students >> considered normal adults and I have no reason to believe >> they are >> significantly different samples. Do I need to re-process >> these two data sets >> with a common template or can I just merge the >> pre-processed data sets and >> run "randomise" with the merged data? They are >> all pretty large sample with >> about 40 subjects each. Thank you.