failed alignment

[LuisaBoh]

Hi AFNI!

I am running afni_proc but I am having an issue aligning my EPI to my T1. My datasets are very close that’s why I change the option -giant_move for big move and the results are better but still not good. I also tried with LPC+ZZ and is not different than the picture attached. Any suggestions? Here is my command

afni_proc.py
-subj_id 198_061814
-copy_anat /rcc/stor1/projects/GerDep/Luisa/Grief_V1/afni_proc/198_061814/anat/anatSS.ssWarp.nii
-anat_has_skull no
-dsets /rcc/stor1/projects/GerDep/Luisa/Grief_V1/afni_proc/198_061814/func/EPI_retroicor_fixed.nii
-blocks tshift align tlrc volreg blur mask scale regress
-radial_correlate_blocks tcat volreg
-tcat_remove_first_trs 5
-align_opts_aea -cost lpc -big_move -check_flip
-tlrc_base /rcc/stor1/projects/GerDep/Luisa/Grief_V1/afni_proc/MNI152_2009_template_SSW.nii.gz
-tlrc_NL_warp
-tlrc_NL_warped_dsets /rcc/stor1/projects/GerDep/Luisa/Grief_V1/afni_proc/198_061814/anat/anatQQ.ssWarp.nii
/rcc/stor1/projects/GerDep/Luisa/Grief_V1/afni_proc/198_061814/anat/anatQQ.ssWarp.aff12.1D
/rcc/stor1/projects/GerDep/Luisa/Grief_V1/afni_proc/198_061814/anat/anatQQ.ssWarp_WARP.nii
-volreg_align_to MIN_OUTLIER
-volreg_align_e2a
-volreg_tlrc_warp
-mask_epi_anat yes
-blur_size 4.0
-regress_stim_times
/rcc/stor1/projects/GerDep/Luisa/Grief_V1/afni_proc/198_061814/func/newTask_1_Face_IdealHRF.1D
/rcc/stor1/projects/GerDep/Luisa/Grief_V1/afni_proc/198_061814/func/newTask_Shape_IdealHRF.1D
-regress_stim_labels Face Shape
-regress_opts_3dD -jobs 2 -gltsym ‘SYM: Face -Shape’
-glt_label 1 F-S
-regress_motion_per_run
-regress_censor_motion 0.3
-regress_censor_outliers 0.05
-regress_3dD_stop
-regress_reml_exec
-regress_compute_fitts
-regress_make_ideal_sum sum_ideal.1D
-regress_est_blur_epits
-regress_est_blur_errts
-regress_run_clustsim no
-html_review_style pythonic
~

[dglen]

There’s not much spatial contrast in your EPI data. Try changing the cost function to lpa, lpa+ZZ or nmi.

[LuisaBoh]

Daniel thank you so much! it improves A LOT but still not my favorite what do you think?

Here is what I used

-align_opts_aea -cost lpa+ZZ -big_move -check_flip

[dglen]

If you have pre-steady state data, then those might have better contrast, and you could use that as an external volreg_base_dset. Aside from that, you could try limiting the alignment to rigid body by adding “-rigid_body” to your “-align_opts_aea” options.

[LuisaBoh]

Daniel thank you so much for your help!

Yes I have pre-steady data, when you say “use that as an external volreg_base_dset” you are referring to change the epi_base to 1? (sorry if this is a dumb question)

================================= align ==================================

for e2a: compute anat alignment transformation to EPI registration base

(new anat will be current anatSS.ssWarp+orig)

align_epi_anat.py -anat2epi -anat anatSS.ssWarp+orig
-suffix _al_junk
-epi vr_base_min_outlier+orig -epi_base 1
-epi_strip 3dAutomask
-anat_has_skull no
-cost lpa+ZZ -rigid_body -big_move -check_flip
-volreg off -tshift off

[dglen]

I mean that option as an input to afni_proc.py, so if you wanted to use the zeroth volume (AFNI starts counting with 0), then you could give afni_proc.py this option along with all your other options:
-volreg_base_dset myepidset.nii.gz’[0]’

In that case, as with most cases, you would let afni_proc.py fill in the epi_base itself.