Mismatch in Actual ROI locations vs Suma representation when using 3dundump + 3dNetCorr

sandra · May 10, 2024, 10:43am

AFNI version info (afni -ver):

uname -a
Linux DESKTOP-RJJ49UB 4.4.0-19041-Microsoft #3996-Microsoft Thu Jan 18 16:36:00 PST 2024 x86_64 x86_64 x86_64 GNU/Linux

afni -ver
Precompiled binary linux_ubuntu_16_64: Apr 5 2024 (Version AFNI_24.1.01 'Publius Septimius Geta')

Paul,

I used the tutorial scripts, modified with our cleaned epi and anatomy data in the $subject.results directory (created by afni_proc.py which was run after sswarper using the nmi cost function). Our subject is masters02.

set vepi = all_runs.masters02-baseline+tlrc
set vanat = anatSS.masters02-baseline+orig

Our ROI should be on both the left and right cortices, so should essentially be mirror images (ish). Our ROI don’t look like they should at all.

Is this possibly a misregistration between XYZ and IJK coordinates of ROIs, or might there somehow be a scale difference? Notice in the SUMA representation of Netcorr’s ROIs that are being processed, the ROIs are all on one side and not mirror images like they should be.

IJK coordinates of ROIs
19 37 14
46 34 14
39 22 18
22 24 17
42 28 21
38 50 13

Suma representation:

Suma-representation-xyz-vx-ijk-mismatch

Actual ROI locations:

As always, thank you so much for your time and all your help. You are greatly appreciated.

Sandra

ptaylor · May 13, 2024, 2:34pm

Hi, Sandra-

The vepi data are in standard space, which I can tell because of the +tlrc in the BRIK/HEAD file name suffix. I guess you used a template to define the final space, which is certainly fine and common to do.

However, the selected anatomical vanat is not in standard space---it has a +orig part of its file name, and also I recognize that as the output from @SSwarper or sswarper2 as being the skullstripped anatomical in native/original space.

The GUI will not overlay/underlay data from different spaces. I believe that was an original design choice to help prevent people from mistakenly overlay/underlay data that weren't in the same space and therefore should not be displayed together (so, a guard rail on making a mistake). The GUI will change the underlay to be some dataset that matches with your overlay spacename when you select it. It probably puts some text to the effect of either

*+ WARNING: Forced switch from 'Original View' to 'Talairach View' [#1]

or

*+ WARNING: Forced switch from 'Talairach View' to 'Original View' [#2]

in the terminal, depending on which one you are over/underlaying first.

So, in this case, indeed, a different anatomical should be paired with the final, standard space EPI. I would guess that you have the subject anatomical in standard space, called something like anat_final.*+tlrc.HEAD in your results directory? That would be an appropriate thing to be vanat in your code.

--pt

sandra · May 14, 2024, 11:36pm

Hey Paul,

I implemented your suggestions, and am still having the same issues with ROI locations on the Suma image. On the upside, at least its generating montage images. When I right-click on all the ROI in the SUMA representation it gives more info which I’m including here along with the command line info. Let me know if you'd also like to see output from all the programs do_0?_* Any ideas on what more I can try to fix our ROI spacing issue? Thanks a lot for your time and your help.

Variable settings

set ilist = 00_list_of_all_roi_centers.txt

set vepi = all_runs.masters02-baseline+tlrc

set vanat = anat_final.masters02-baseline+tlrc.HEAD

Output from right-clicking the SUMA ROI

sandra@DESKTOP-RJJ49UB:~/data/Data/automation-test/newcases/masters02.rois/afni_tutorial_rois_sswarper_fixed_ROIs$ ./do_

00_setup_example.tcsh

++ 3dUndump: AFNI version=AFNI_24.1.01 (Apr 5 2024) [64-bit]

++ Starting to fill via -xyz coordinates

++ Total number of voxels filled = 7

++ Wrote out dataset ./roi_mask_001.nii.gz

++ 3dUndump: AFNI version=AFNI_24.1.01 (Apr 5 2024) [64-bit]

++ Starting to fill via -xyz coordinates

++ Total number of voxels filled = 7

++ Wrote out dataset ./roi_mask_002.nii.gz

++ 3dUndump: AFNI version=AFNI_24.1.01 (Apr 5 2024) [64-bit]

++ Starting to fill via -xyz coordinates

++ Total number of voxels filled = 7

++ Wrote out dataset ./roi_mask_003.nii.gz

++ 3dUndump: AFNI version=AFNI_24.1.01 (Apr 5 2024) [64-bit]

++ Starting to fill via -xyz coordinates

++ Total number of voxels filled = 7

++ Wrote out dataset ./roi_mask_004.nii.gz

++ 3dUndump: AFNI version=AFNI_24.1.01 (Apr 5 2024) [64-bit]

++ Starting to fill via -xyz coordinates

++ Total number of voxels filled = 7

++ Wrote out dataset ./roi_mask_005.nii.gz

++ 3dUndump: AFNI version=AFNI_24.1.01 (Apr 5 2024) [64-bit]

++ Starting to fill via -xyz coordinates

++ Total number of voxels filled = 7

++ Wrote out dataset ./roi_mask_006.nii.gz

++ Skull-stripping the anat vol.

-- Error SUMA_BrainWrap_ParseInput (SUMA_3dSkullStrip.c:1296):

Output dset ./all_runs_mprage_ss.nii.gz+tlrc exists, will not overwrite

++ DONE!

sandra@DESKTOP-RJJ49UB:~/data/Data/automation-test/newcases/masters02.rois/afni_tutorial_rois_sswarper_fixed_ROIs$ rm mprage

sandra@DESKTOP-RJJ49UB:~/data/Data/automation-test/newcases/masters02.rois/afni_tutorial_rois_sswarper_fixed_ROIs$ ./do_00_setup_example.tcsh

++ 3dUndump: AFNI version=AFNI_24.1.01 (Apr 5 2024) [64-bit]

++ Starting to fill via -xyz coordinates

++ Total number of voxels filled = 7

++ Wrote out dataset ./roi_mask_001.nii.gz

++ 3dUndump: AFNI version=AFNI_24.1.01 (Apr 5 2024) [64-bit]

++ Starting to fill via -xyz coordinates

++ Total number of voxels filled = 7

++ Wrote out dataset ./roi_mask_002.nii.gz

++ 3dUndump: AFNI version=AFNI_24.1.01 (Apr 5 2024) [64-bit]

++ Starting to fill via -xyz coordinates

++ Total number of voxels filled = 7

++ Wrote out dataset ./roi_mask_003.nii.gz

++ 3dUndump: AFNI version=AFNI_24.1.01 (Apr 5 2024) [64-bit]

++ Starting to fill via -xyz coordinates

++ Total number of voxels filled = 7

++ Wrote out dataset ./roi_mask_004.nii.gz

++ 3dUndump: AFNI version=AFNI_24.1.01 (Apr 5 2024) [64-bit]

++ Starting to fill via -xyz coordinates

++ Total number of voxels filled = 7

++ Wrote out dataset ./roi_mask_005.nii.gz

++ 3dUndump: AFNI version=AFNI_24.1.01 (Apr 5 2024) [64-bit]

++ Starting to fill via -xyz coordinates

++ Total number of voxels filled = 7

++ Wrote out dataset ./roi_mask_006.nii.gz

++ Skull-stripping the anat vol.

++ DONE!

sandra@DESKTOP-RJJ49UB:~/data/Data/automation-test/newcases/masters02.rois/afni_tutorial_rois_sswarper_fixed_ROIs$ ./do_01_make_single_roi_map.tcsh

++ 3dMean: AFNI version=AFNI_24.1.01 (Apr 5 2024) [64-bit]

++ OK: ROI masks don't appear to overlap

++ 3dTcat: AFNI version=AFNI_24.1.01 (Apr 5 2024) [64-bit]

*+ WARNING: Set TR of output dataset to 1.0 s

++ elapsed time = 0.0 s

++ 3dTstat: AFNI version=AFNI_24.1.01 (Apr 5 2024) [64-bit]

++ Authored by: KR Hammett & RW Cox

++ Output dataset ./final_roi_map.nii.gz

++ 3drefit: AFNI version=AFNI_24.1.01 (Apr 5 2024) [64-bit]

++ Authored by: RW Cox

++ Processing AFNI dataset final_roi_map.nii.gz

setting labeltable
loading and re-writing dataset final_roi_map.nii.gz (/home/sandra/data/Data/automation-test/newcases/masters02.rois/afni_tutorial_rois_sswarper_fixed_ROIs/final_roi_map.nii.gz in NIFTI storage)

++ 3drefit processed 1 datasets

++ My command:

@chauffeur_afni -ulay anat_final.masters02-baseline+tlrc.HEAD -olay final_roi_map.nii.gz -box_focus_slices all_runs_mprage_ss.nii.gz -cbar ROI_i32 -func_range 32 -pbar_posonly -opacity 9 -blowup 1 -save_ftype JPEG -prefix final_roi_map -montx 6 -monty 3 -set_xhairs OFF -label_mode 1 -label_size 3 -do_clean

++ Using AFNI ver : AFNI_24.1.01

++ chauffeur ver : 6.7

------------------ start of optionizing ------------------

++ Found input file: anat_final.masters02-baseline+tlrc.HEAD

++ Found input file: final_roi_map.nii.gz

++ Found focus refbox file: all_runs_mprage_ss.nii.gz

++ Using blowup factor: 1

++ Using opacity: 9

++ Making temporary work directory to copy vis files: ./__tmp_chauf_mPHvlOA9Isk

++ Converted 0 to labels

++ Final subbrick indices: -1 -1 -1

++ Copy ulay to visualize (volumetric) within user's range:

++ Ulay to be visualized within user range:

[0%, 98%] -> [0.000000, 1080.322021]

++ 3dcalc: AFNI version=AFNI_24.1.01 (Apr 5 2024) [64-bit]

++ Authored by: A cast of thousands

++ Output dataset ././__tmp_chauf_mPHvlOA9Isk/tmp_ulay.nii

++ 3dcalc: AFNI version=AFNI_24.1.01 (Apr 5 2024) [64-bit]

++ Authored by: A cast of thousands

++ Output dataset ././__tmp_chauf_mPHvlOA9Isk/tmp_olay.nii

++ User-entered function range value value (32)

++ Dimensions (xyzt): 161 191 151 1

++ (initial) Slice spacing ordered (x,y,z) is: 8 10 8

++ 3dAutobox: AFNI version=AFNI_24.1.01 (Apr 5 2024) [64-bit]

++ Auto bbox: x=11..147 y=12..181 z=9..139

++ 3dAutobox: output dataset = ./__tmp_chauf_mPHvlOA9Isk/ulay_box_0.nii

++ 3dmaskdump: AFNI version=AFNI_24.1.01 (Apr 5 2024) [64-bit]

++ Executing 3dcalc()

argv[0]=3dcalc argv[1]=-a argv[2]=./__tmp_chauf_mPHvlOA9Isk/ulay_box_0.nii[0] argv[3]=-expr argv[4]=a argv[5]=-byte argv[6]=-session argv[7]=/tmp argv[8]=-prefix argv[9]=3dcalc_AFN_XJUCcnl_9Zm8k-eFg3UiSQ argv[10]=-verbose

++ 3dcalc: AFNI version=AFNI_24.1.01 (Apr 5 2024) [64-bit]

++ Authored by: A cast of thousands

++ Computing sub-brick 0

++ Scaling output to type byte brick(s)

++ Sub-brick 0 scale factor = 0.182275

++ Computing output statistics

++ Output dataset /tmp/3dcalc_AFN_XJUCcnl_9Zm8k-eFg3UiSQ+tlrc.BRIK

++ 3050990 voxels in the entire dataset (no mask)

++ 1 voxels in the boxes and/or balls

++ Using only the boxes+balls mask

++ How many coors? 3

++ Will have the ref box central coors : SET_DICOM_XYZ -4 21 13

++ Will have the ref box central gapord: 7 9 7

------------------- end of optionizing -------------------

-- trying to start Xvfb :858

[1] 19355

_XSERVTransmkdir: Owner of /tmp/.X11-unix should be set to root

+++ Command Echo:

afni -q -no1D -noplugins -no_detach -com SWITCH_UNDERLAY tmp_ulay.nii -com SWITCH_OVERLAY tmp_olay.nii -com SEE_OVERLAY + -com OPEN_WINDOW sagittalimage opacity=9 mont=6x3:7:0:black crop=0:0,0:0 -com OPEN_WINDOW coronalimage opacity=9 mont=6x3:9:0:black crop=0:0,0:0 -com OPEN_WINDOW axialimage opacity=9 mont=6x3:7:0:black crop=0:0,0:0 -com SET_PBAR_ALL +99 1 ROI_i32 -com DO_NOTHING -com SET_SUBBRICKS -1 -1 -1 -com SET_ULAY_RANGE A.all 0.000000 1080.322021 -com SET_FUNC_RANGE 32 -com SET_THRESHNEW 0 * -com SET_FUNC_ALPHA No -com SET_FUNC_BOXED No -com SET_FUNC_RESAM NN.NN -com SET_XHAIRS OFF -com SET_XHAIR_GAP -1 -com SET_DICOM_XYZ -4 21 13 -com SAVE_JPEG axialimage ./final_roi_map.axi blowup=1 -com SAVE_JPEG sagittalimage ./final_roi_map.sag blowup=1 -com SAVE_JPEG coronalimage ./final_roi_map.cor blowup=1 -com QUITT ./__tmp_chauf_mPHvlOA9Isk

+/home/sandra/data/Data/automation-test/newcases/masters02.rois/afni_tutorial_rois_sswarper_fixed_ROIs/__tmp_chauf_mPHvlOA9Isk++ Writing one 966x573 image to filter '/home/sandra/abin/cjpeg -quality 95 > ./final_roi_map.axi.jpg'

++ Writing one 1146x453 image to filter '/home/sandra/abin/cjpeg -quality 95 > ./final_roi_map.sag.jpg'

++ Writing one 966x453 image to filter '/home/sandra/abin/cjpeg -quality 95 > ./final_roi_map.cor.jpg'

AFNI QUITTs!

+* Removing temporary image directory './__tmp_chauf_mPHvlOA9Isk'.

[1] + Done Xvfb :858 -screen 0 1024x768x24

++ DONE (good exit)

see: ./final_roi_map*

++ DONE!

sandra@DESKTOP-RJJ49UB:~/data/Data/automation-test/newcases/masters02.rois/afni_tutorial_rois_sswarper_fixed_ROIs$ ./do_02_surfaceize_rois.tcsh

Notice SUMA_AddNodeIndexColumn:

Assuming node indexing

is explicit.

1st row is for node 0

2nd is for node 1, etc.

Notice SUMA_AddNodeIndexColumn:

Assuming node indexing

is explicit.

1st row is for node 0

2nd is for node 1, etc.

Notice SUMA_AddNodeIndexColumn:

Assuming node indexing

is explicit.

1st row is for node 0

2nd is for node 1, etc.

Notice SUMA_AddNodeIndexColumn:

Assuming node indexing

is explicit.

1st row is for node 0

2nd is for node 1, etc.

Notice SUMA_AddNodeIndexColumn:

Assuming node indexing

is explicit.

1st row is for node 0

2nd is for node 1, etc.

Notice SUMA_AddNodeIndexColumn:

Assuming node indexing

is explicit.

1st row is for node 0

2nd is for node 1, etc.

vvvvvvvvvvvvvvvvvvvvvvvvvvvvSurface #1/6(Local Domain Parent), loading ...

++ Notice SUMA_AutoLoad_SO_Dsets (SUMA_Load_Surface_Object.c:1619 @21:53:24):

Auto Loading ./final_roi_map.SURF.left_au.k1.niml.dset onto ./final_roi_map.SURF.left_au.k1.gii

++ Notice SUMA_LoadDsetOntoSO_eng (SUMA_Color.c:10444 @21:53:24):

dset has a mesh parent, Checking relationship

oo Warning SUMA_Contralateral_SO (SUMA_DOmanip.c:2885):

Surface sides are not clearly defined. If this is in error, consider adding

Hemisphere = R (or L or B) in the spec file

to make sure surfaces sides are correctly labeled.

Similar warnings will be muted

vvvvvvvvvvvvvvvvvvvvvvvvvvvvSurface #2/6(Local Domain Parent), loading ...

++ Notice SUMA_AutoLoad_SO_Dsets (SUMA_Load_Surface_Object.c:1619 @21:53:24):

Auto Loading ./final_roi_map.SURF.left_bro.k4.niml.dset onto ./final_roi_map.SURF.left_bro.k4.gii

++ Notice SUMA_LoadDsetOntoSO_eng (SUMA_Color.c:10444 @21:53:24):

dset has a mesh parent, Checking relationship

vvvvvvvvvvvvvvvvvvvvvvvvvvvvSurface #3/6(Local Domain Parent), loading ...

++ Notice SUMA_AutoLoad_SO_Dsets (SUMA_Load_Surface_Object.c:1619 @21:53:24):

Auto Loading ./final_roi_map.SURF.right_au.k2.niml.dset onto ./final_roi_map.SURF.right_au.k2.gii

++ Notice SUMA_LoadDsetOntoSO_eng (SUMA_Color.c:10444 @21:53:24):

dset has a mesh parent, Checking relationship

vvvvvvvvvvvvvvvvvvvvvvvvvvvvSurface #4/6(Local Domain Parent), loading ...

++ Notice SUMA_AutoLoad_SO_Dsets (SUMA_Load_Surface_Object.c:1619 @21:53:24):

Auto Loading ./final_roi_map.SURF.right_bro.k3.niml.dset onto ./final_roi_map.SURF.right_bro.k3.gii

++ Notice SUMA_LoadDsetOntoSO_eng (SUMA_Color.c:10444 @21:53:24):

dset has a mesh parent, Checking relationship

vvvvvvvvvvvvvvvvvvvvvvvvvvvvSurface #5/6(Local Domain Parent), loading ...

++ Notice SUMA_AutoLoad_SO_Dsets (SUMA_Load_Surface_Object.c:1619 @21:53:24):

Auto Loading ./final_roi_map.SURF.right_motor.k5.niml.dset onto ./final_roi_map.SURF.right_motor.k5.gii

++ Notice SUMA_LoadDsetOntoSO_eng (SUMA_Color.c:10444 @21:53:24):

dset has a mesh parent, Checking relationship

vvvvvvvvvvvvvvvvvvvvvvvvvvvvSurface #6/6(Local Domain Parent), loading ...

++ Notice SUMA_AutoLoad_SO_Dsets (SUMA_Load_Surface_Object.c:1619 @21:53:24):

Auto Loading ./final_roi_map.SURF.right_occ.k6.niml.dset onto ./final_roi_map.SURF.right_occ.k6.gii

++ Notice SUMA_LoadDsetOntoSO_eng (SUMA_Color.c:10444 @21:53:24):

dset has a mesh parent, Checking relationship

SUMA_Engine: Starting to listen ...

vvvvvvvvvvvvvvvvvvvvvvvvvvvv

Selected surface final_roi_map.SURF.left_au.k1.gii (Focus_DO_ID # 1).

FaceSet 37, Closest Node 13

Nodes forming closest FaceSet:

25, 13, 14

Coordinates of Nodes forming closest FaceSet:

19.745371, 34.311199, 16.704662

19.695551, 32.293709, 14.874831

22.178751, 34.313274, 13.408155

^^^^^^^^^^^^^^^^^^^^^^^^^^^^

vvvvvvvvvvvvvvvvvvvvvvvvvvvv

Selected surface final_roi_map.SURF.left_bro.k4.gii (Focus_DO_ID # 2).

FaceSet 34, Closest Node 13

Nodes forming closest FaceSet:

13, 25, 10

Coordinates of Nodes forming closest FaceSet:

23.195551, 21.793713, 18.374832

23.245371, 23.811197, 20.204660

21.119585, 22.364252, 17.490644

^^^^^^^^^^^^^^^^^^^^^^^^^^^^

vvvvvvvvvvvvvvvvvvvvvvvvvvvv

Selected surface final_roi_map.SURF.right_bro.k3.gii (Focus_DO_ID # 4).

FaceSet 28, Closest Node 10

Nodes forming closest FaceSet:

13, 10, 9

Coordinates of Nodes forming closest FaceSet:

40.695549, 18.293713, 18.374832

38.619583, 18.864252, 17.490644

40.279606, 17.234304, 17.066328

^^^^^^^^^^^^^^^^^^^^^^^^^^^^

vvvvvvvvvvvvvvvvvvvvvvvvvvvv

Selected surface final_roi_map.SURF.right_motor.k5.gii (Focus_DO_ID # 5).

FaceSet 34, Closest Node 10

Nodes forming closest FaceSet:

13, 25, 10

Coordinates of Nodes forming closest FaceSet:

40.695549, 25.293713, 21.874832

40.745373, 27.311197, 23.704660

38.619583, 25.864252, 20.990644

^^^^^^^^^^^^^^^^^^^^^^^^^^^^

vvvvvvvvvvvvvvvvvvvvvvvvvvvv

Selected surface final_roi_map.SURF.right_occ.k6.gii (Focus_DO_ID # 6).

FaceSet 33, Closest Node 25

Nodes forming closest FaceSet:

25, 26, 10

Coordinates of Nodes forming closest FaceSet:

37.245373, 48.311199, 16.704662

34.848953, 49.659824, 16.506144

35.119583, 46.864254, 13.990644

^^^^^^^^^^^^^^^^^^^^^^^^^^^^

vvvvvvvvvvvvvvvvvvvvvvvvvvvv

Selected surface final_roi_map.SURF.right_au.k2.gii (Focus_DO_ID # 3).

FaceSet 29, Closest Node 13

Nodes forming closest FaceSet:

12, 13, 9

Coordinates of Nodes forming closest FaceSet:

48.565605, 28.822611, 13.133023

47.695549, 28.793711, 14.874831

47.279606, 27.734304, 13.566326

dglen · May 15, 2024, 3:57pm

Does your 3dUndump contain the option -ijk ? The output shows it's using xyz coordinates in mm rather than ijk in indices. That would explain why all the coordinates are one side (positive x, y and z).