 |
 |
Hi, It is worth pointing out here that we don't recommend this approach to registration (using an intermediate T2) and I would advise you to just go from the epi to the T1 to standard. If you turn on fieldmap correction, FSL currently assumes that the "initial highres" is a wholebrain EPI, and applies the fieldmap distortion correction to it. At the moment this behaviour cannot be turned off in the GUI and if you really want to use both an intermediate T2 image _and_ fieldmaps, then you will have to run it yourself outside of the GUI. You can always do this and then copy the resulting registration into the FEAT directory and run "updatefeatreg" to incorporate these results into the FEAT pipeline. I hope this helps. All the best, Mark On 18 Apr 2013, at 02:05, David Grayson <[log in to unmask]> wrote: > Dear FSL experts, > > I am trying to use the feat command to apply the following registration procedure: > > epi --> initial T2 structural (dof 6) > initial T2 structural --> main T1 structural (dof 6) > main T1 structural --> standard atlas (dof 12) > > fieldmap unwarping is on. I have made these settings in the design.fsf file. However, when I go through the log files where it calls the 'mainfeatreg' command, I see that it has overwritten the dof to 'BBR' for the initial T2 structural --> highres T1 stage. > > Here is the command as it is called (notice that there are 2 -w options, the first is what I set, the one at the end is apparently what overwrites): > > /usr/share/fsl/5.0/bin/mainfeatreg -d /home/david/Projects/Amyg_Monkeys/Analyses/unknown/processed_data/d_446/d_44 > 6_func_1_.feat -l /home/david/Projects/Amyg_Monkeys/Analyses/unknown/processed_data/d_446/d_446_func_1_.feat/logs/ > feat5_reg -R /home/david/Projects/Amyg_Monkeys/Analyses/unknown/processed_data/d_446/d_446_func_1_.feat/report_unw > arp.html -r /home/david/Projects/Amyg_Monkeys/Analyses/unknown/processed_data/d_446/d_446_func_1_.feat/report_reg. > html -i /home/david/Projects/Amyg_Monkeys/Analyses/unknown/processed_data/d_446/d_446_func_1_.feat/example_func.n > ii.gz -n 10 -j /home/david/Projects/Amyg_Monkeys/Analyses/unknown/processed_data/d_446/T2_late2epi_brain -u 6 -v > 90 -h /home/david/Projects/Amyg_Monkeys/Analyses/unknown/processed_data/d_446/d_446_MPRAGE_fslreorient_strip -w 6 > -x 90 -s /home/david/Projects/Primate_Atlases/inia19/inia19_t1_brain -y 12 -z 180 -a /home/david/Projects/Amyg_Mo > nkeys/Analyses/unknown/processed_data/d_446/uphase_rad -b /home/david/Projects/Amyg_Monkeys/Analyses/unknown/proce > ssed_data/d_446/d_446_MAG_fslreorient_strip -e 22 -f 10 -g 0.94 -p y- -w BBR > > This leads to the fieldmap unwarping being applied to the T2 image, in addition to unwarping being applied to the example_func image, which is clearly not what I want. Is this a bug or what can I do about this? P.S. I am using a design file optimized for use with a previous version of fsl, before bbr was incorporated. In case that matters... > > Thanks very much for reading, > > David