SS_Warper Standardizing Error

AFNI Message Board
1

I downloaded a directory of anatomical scans from an online dataset. I’m having an issue when I try to standardize these images from native space to MNI space. I have successfully used SSwarper in the past so I would like to know if my issue is more afni-related or with the scans I’m trying to process.
The script I used:

@SSwarper -input 1_MPRAGE+orig -base $File_Path/MNI152_2009_template_SSW.nii.gz -subid $subj_number

There’s a cascade of errors in the process that result from junk.SSwarper.2ovuXrwwb3N_TAL5.nii not being output. I get a series of "can’t open dataset ./junk.SSwarper.2ovuXrwwb3N_TAL5.nii " fatal errors. I tried double checking the terminal log with one of my successful uses of SSwarper and the main difference I see is during the 3dQwarp section. I get the following warning/errors in my failed unsuccessful attempts:

++ 3dQwarp: replacing source dataset with 3dAllineate result ./junk.SSwarper.2ovuXrwwb3N_TAL5_Allin.nii

  • 3dAllineate output files have been renamed

mat44 3dAllineate matrix:

  0.868885      0.003196     -0.021110    -246.061996

 -0.022162      0.892608      0.201009    -270.204987

  0.016392     -0.186738      0.811656     540.333008

++ Dataset zero-pad: xbot=549 xtop=548 ybot=543 ytop=552 zbot=549 ztop=543 voxels

*+ WARNING: At least one padding is more than 50% of dataset grid size!

*+ WARNING: Alignment of dataset grid centers might help a LOT: @Align_Centers

** ERROR: EDIT_volpad: Can’t malloc space for new array

** ERROR: EDIT_volpad: Can’t malloc space for new array

Fatal Signal 11 (SIGSEGV) received

3dQwarp

Bottom of Debug Stack

** AFNI version = AFNI_19.1.09 Compile date = May 17 2019

** [[Precompiled binary macos_10.12_local: May 17 2019]]

** Program Death **

I appreciate any help you could provide!

2

Hello,

The first error message shows:

** ERROR: EDIT_volpad: Can’t malloc space for new array

It suggests you do not have enough RAM available to run the command. How much RAM is there, and do you have other computers to use?

  • rick
3

My original attempt was on my desktop with 16 GBs of RAM. I tried repeating my analysis on my lab’s cluster which holds 128 GBs of RAM and the same error occurred. I’m wondering how memory could be a factor.

It seems like AFNI is trying to utilize as much memory available since for some reason it’s struggling to warp the anatomical images in this dataset to MNI space. Could the issue have more to do with my anatomical data or the way I’m using SSwarper?

4

What is the output of:

A)


3dinfo -ad3 -o3 -orient -prefix 1_MPRAGE+orig $File_Path/MNI152_2009_template_SSW.nii.gz

and B)


3dCM 1_MPRAGE+orig $File_Path/MNI152_2009_template_SSW.nii.gz

?

–pt

5

The script output suggests running @Align_Centers, so that’s what I would do.

*+ WARNING: Alignment of dataset grid centers might help a LOT: @Align_Centers

The nonlinear scripting step zeropads the volumes to include the original and template. That can use a lot of memory if they are far apart.

6

Input:

3dinfo -ad3 -o3 -orient -prefix 1_MPRAGE+orig $File_Path/MNI152_2009_template_SSW.nii.gz


Output:

1.200000 1.000000 1.000000 -130.783707 -137.438400
419.236206 LPI                       1_MPRAGE

1.000000 1.000000 1.000000 96.000000 132.000000
-78.000000 LPI MNI152_2009_template_SSW.nii.gz






Input:

3dCM 1_MPRAGE+orig $File_Path/MNI152_2009_template_SSW.nii.gz


Output:

-226.882  -237.64  549.06

0.107468  22.5192  9.14651

Then after using @Align_Centers to align the anatomical to the MNI template, here are my results with the same two lines.

Input:

3dinfo -ad3 -o3 -orient -prefix 1_MPRAGE+orig $File_Path/MNI152_2009_template_SSW.nii.gz


Output:

1.200000 1.000000 1.000000 95.400009 137.500000
-109.500000 LPI                       1_MPRAGE

1.000000 1.000000 1.000000 96.000000 132.000000
-78.000000 LPI MNI152_2009_template_SSW.nii.gz






Input:

3dCM 1_MPRAGE+orig $File_Path/MNI152_2009_template_SSW.nii.gz


Output:

-0.698524  37.2983  20.3242

0.107468  22.5192  9.14651

After using @Align_Centers I tried running @SSwarper again but it resulted in the same error.