Surfsmooth error

AFNI Message Board
1

Hi,

I´ve been trying to run a smooth on a subject surface with this command in a script:

SurfSmooth -spec $SUMA_path/mesh140.${subj}“$h”.spec
-surf_A mesh140.“$h”.smoothwm.asc
-met HEAT_07 -input $data_path/“$h”${subj}decon$condition.TENT.ave_.niml.dset
-target_fwhm 6 -Niter -1
-output $data_path/“$h”${subj}decon$conds.TENT.ave.blur6.niml.dset

but get this error message and I do not know why the SyrfSmooth can get a mean fwhm:

– Error SurfSmooth (SUMA_SurfSmooth.c:2128):
Failed to get mean fwhm
– Error SurfSmooth (SUMA_SurfSmooth.c:2128):
Failed to get mean fwhm
– Error SurfSmooth (SUMA_SurfSmooth.c:2128):
Failed to get mean fwhm
– Error SurfSmooth (SUMA_SurfSmooth.c:2128):
Failed to get mean fwhm
– Error SurfSmooth (SUMA_SurfSmooth.c:2128):
Failed to get mean fwhm
– Error SurfSmooth (SUMA_SurfSmooth.c:2128):
Failed to get mean fwhm
– Error SurfSmooth (SUMA_SurfSmooth.c:2128):
Failed to get mean fwhm
– Error SurfSmooth (SUMA_SurfSmooth.c:2128):
Failed to get mean fwhm
– Error SurfSmooth (SUMA_SurfSmooth.c:2128):
Failed to get mean fwhm
– Error SurfSmooth (SUMA_SurfSmooth.c:2128):
Failed to get mean fwhm
– Error SurfSmooth (SUMA_SurfSmooth.c:2128):
Failed to get mean fwhm
– Error SurfSmooth (SUMA_SurfSmooth.c:2128):
Failed to get mean fwhm

I would appreciate so much to receive some guidance because couldn’t find a response to an error of this type.

2

Hi-

Hmm, I am not sure. I just tried running this, which is a similar command to what you are running (just different file names), using the data in the AFNI Bootcamp directory AFNI_data6/FT_analysis/FT/SUMA, and it ran fine:


SurfSmooth \
  -spec std.141.FT_lh.spec \
  -surf_A std.141.lh.smoothwm.gii \
  -met HEAT_07 \
  -input std.141.lh.sulc.niml.dset \
  -target_fwhm 6 \
  -Niter -1 \
  -output TEST.niml.dset

What is your “$SUMA_path/mesh140.${subj}_”$h".spec" dset, and how was it generated (I don’t think it was created by @SUMA_Make_Spec_FS)? I am not familiar with what that is; I assume that $h is the hemisphere (“lh” or “rh”).

–pt

3

Hi,

Thank you for your answer, yes I have another data from other experiments before and it ran well with those (the only difference is, apart from the task and subjetcs, that these data is from a 1.5 T machine, using a TR=3.5s, while the other is from a 3 T machine, using TR=1.5 s). Yes, I have the paths like this in the script:

set top_path = /Users/neurociencia/Desktop/Semantic_Priming/

set data_path = “${top_path}/DICOM/DICOM-${subj}/Deconvolve”

set SUMA_path = “${top_path}/freesurfer_files/${subj}/SUMA”

Yes I used @SUMA_Make_Spec_FS to create the specs and they seem fine, and you’re right $h is for both hemispheres lh and rh.

I have varied the target_fwhm value up and down with the same results and as I do not use cmask, I do not have the same problem as the other surfsmooth entrance of this message board.

Any help with be greatly appreciated.

4

Hi-

If you run FS’s recon-all and then AFNI’s @SUMA_Make_Spec_FS, you generally get the same set of files created each time. To my knowledge, there is no file whose name starts with “mesh140”. So, how did you create these files (please paste commands)?

–pt

5

Hi Taylor,

Thank you very much for your help. This is what I ran in the script with these subjects and previous data.

After FS recon-all:

@SUMA_Make_Spec_FS -sid ${subj} -debug 1

Create meshes

    foreach h (lh rh)

        MapIcosahedron -spec ${subj}_"$h".spec -ld 140 -prefix mesh140. -NN_dset_map $h.aparc.a2009s.annot.1D.roi -all_surfs_spec -verb

    end

Do you think I should use Surfsmooth better with the lh and rh specs instead of meshes?.

Thank you very much.

6

Howdy-

I wonder if one issue is that for the option -NN_dset_map you have entered not a full dset (.niml.dset), but an ROI map (.roi). Though, note that there is already a set of *annot.niml.dset files for the aparcs:
std.{60,141}.lh.aparc.a2009s.annot.niml.dset
std.{60,141}.rh.aparc.a2009s.annot.niml.dset
Can I ask what your larger goal is with this command? Is using “-ld 140” really necessary, when there is already a std mesh with essentially “-ld 141”?

One final note: when running @SUMA_Make_Spec_FS, we generally recommend include one of the -GNIFTI/-GIFTI/-IFTI options; when @SUMA_Make_Spec finishes and one of those options has not been used, there is a message displayed:


WARNING!
  You didn't use the '-NIFTI' or '-GIFTI' or similarly-named
  option flag! Are you *sure* you didn't want to?  Learning about
  this option changed my life-- it might just change yours, too.
  Please read the helpfile entry for '-NIFTI' if you don't know
  what I'm talking about here.

–pt

7

Hi,

Thank you for your response. We were not using -GNIFTI options due to the note of problems in SUMA_Make_Spec:

** Note: from 22 Feb 2013 through 20 Mar 2017, use of -NIFTI
would distort standard mesh surfaces. To evaluate
effects of this, consider: MapIcosahedron -write_dist.

And that is why we were using MapIcosahedron to build a mesh in this script.

But let me try to use the std 141 meshes and try to smooth them with Surfsmooth instead of the built with MapIcosahedron. I´ll get back here.

8

Hi-

Are you using a version of AFNI from between 22 Feb 2013 through 20 Mar 2017?? Because if you are not, then this is not a problem-- that bug was fixed long ago. (And if you are reading that help message, then I don’t think you have that old version of AFNI!)

You don’t need to use MapIcosahedron to do this on your own-- I think you should just use “-NIFTI” or “-GNIFTI”.

–pt

9

Hi again,

I have run the @SUMA_Make_Spec_FS using -GNIFTI flag. It gave me the meshes std.141 as gii, and used them in my script of SurfSmooth instead of the std.141.lh.smoothwm.asc as shown:

foreach conds (“Nopal” “Norel” “Rel” “Xnopal” “Xnorel” “Xrel”)

foreach h (lh rh)

SurfSmooth -spec $SUMA_path/std.141.${subj}_"$h".spec \
-surf_A std.141."$h".smoothwm.gii \
-met HEAT_07 -input $data_path/"$h"_${subj}_decon_$condition.TENT.ave_.niml.dset \
-target_fwhm 6 -Niter -1 \
-output $data_path/"$h"_${subj}_decon_$conds.TENT.ave_.blur6.niml.dset

end

end

But the result was the same as with the previous meshes:

– Error SurfSmooth (SUMA_SurfSmooth.c:2128):
Failed to get mean fwhm
– Error SurfSmooth (SUMA_SurfSmooth.c:2128):
Failed to get mean fwhm
– Error SurfSmooth (SUMA_SurfSmooth.c:2128):
Failed to get mean fwhm
– Error SurfSmooth (SUMA_SurfSmooth.c:2128):
Failed to get mean fwhm
– Error SurfSmooth (SUMA_SurfSmooth.c:2128):
Failed to get mean fwhm
– Error SurfSmooth (SUMA_SurfSmooth.c:2128):
Failed to get mean fwhm
– Error SurfSmooth (SUMA_SurfSmooth.c:2128):
Failed to get mean fwhm
– Error SurfSmooth (SUMA_SurfSmooth.c:2128):
Failed to get mean fwhm
– Error SurfSmooth (SUMA_SurfSmooth.c:2128):
Failed to get mean fwhm
– Error SurfSmooth (SUMA_SurfSmooth.c:2128):
Failed to get mean fwhm
– Error SurfSmooth (SUMA_SurfSmooth.c:2128):
Failed to get mean fwhm
– Error SurfSmooth (SUMA_SurfSmooth.c:2128):
Failed to get mean fwhm

A failure for each condition, so I´m wondering about my data. I attach how it looks the data over the meshes without the smoothing, with the activation bar at 0.25. It looks fine to me, but…

rel.jpg
rel.jpg1099×744 94.8 KB

10

Hi again,

I have run the @SUMA_Make_Spec_FS using -GNIFTI flag. It gave me the meshes std.141 as gii, and used them in my script of SurfSmooth instead of the std.141.lh.smoothwm.asc as shown:

foreach conds (“Nopal” “Norel” “Rel” “Xnopal” “Xnorel” “Xrel”)

foreach h (lh rh)

SurfSmooth -spec $SUMA_path/std.141.${subj}“$h”.spec
-surf_A std.141.“$h”.smoothwm.gii
-met HEAT_07 -input $data_path/“$h”${subj}decon$condition.TENT.ave_.niml.dset
-target_fwhm 6 -Niter -1
-output $data_path/“$h”${subj}decon$conds.TENT.ave.blur6.niml.dset

end

end

But I have the same error:

– Error SurfSmooth (SUMA_SurfSmooth.c:2128):
Failed to get mean fwhm
– Error SurfSmooth (SUMA_SurfSmooth.c:2128):
Failed to get mean fwhm
– Error SurfSmooth (SUMA_SurfSmooth.c:2128):
Failed to get mean fwhm
– Error SurfSmooth (SUMA_SurfSmooth.c:2128):
Failed to get mean fwhm
– Error SurfSmooth (SUMA_SurfSmooth.c:2128):
Failed to get mean fwhm
– Error SurfSmooth (SUMA_SurfSmooth.c:2128):
Failed to get mean fwhm
– Error SurfSmooth (SUMA_SurfSmooth.c:2128):
Failed to get mean fwhm
– Error SurfSmooth (SUMA_SurfSmooth.c:2128):
Failed to get mean fwhm
– Error SurfSmooth (SUMA_SurfSmooth.c:2128):
Failed to get mean fwhm
– Error SurfSmooth (SUMA_SurfSmooth.c:2128):
Failed to get mean fwhm
– Error SurfSmooth (SUMA_SurfSmooth.c:2128):
Failed to get mean fwhm
– Error SurfSmooth (SUMA_SurfSmooth.c:2128):
Failed to get mean fwhm

Can you help me?