 |
 |
Dear Jesper, thanks for you quick response! I am using a 1mm template because my T1 is a 1mm one, so I thought that would be better. But maybe that's not the case, since the functional images are not? The refmask you are talking about would be the MNI152_T1_1mm_brain_mask_dil.nii, correct? So, basically, I would fnirt the images, using my previously generated mats etc by FLIRT (I have already run 1st level analyses), using the modified config file, and afterward applywarp, right? since I don't want to mess up anything in my directories (I want to later try to proceed with 2nd and 3rd level analyses using both the flirted and fnirted images, separately, for comparison): what exactly does the fnirting then generate? does it make its own "reg" directory, or does it just put additional stuff into the already existing one? I'm assuming it doesnt overwrite anything. Thanks a lot for your help, I appreciate it! Kind regards, Hannah Dear Hannah, > I have yet another question about FNIRT, this time about the cnf file. I would like to use one of the configuration files you provide. However, since I need a 1mm instead of 2 mm template brain, I was wondering, if I could just change that part in the configure file, or if then, all the other set parameters would not work anymore? why do you need a 1mm template? That is a quite high resolution to perform your 2nd level analysis in given that the resolution of your fMRI data is typically ~3mm or so. The, slightly sad, truth is that changing to a 1mm template does change some other things. Luckily the changes are quite subtle and you would get away with just changing the line --subsamp=4,4,2,2,1,1 to --subsamp=8,8,4,4,2,2 You will then also need to specify your own mask in the space of your template by including --refmask=your_mask_goes_here where the mask should be in the space of your template and have one everywhere where there is brain and zero elsewhere. Good Luck Jesper