Yarik. I think this is user error. I think that some of the transformations are out of order. But it is hard to see unless we have the original ants commands.
B. On Jul 16, 2010 5:33 PM, "Yaroslav Halchenko" <[email protected]> wrote: Package: ants Version: 1.9+svn532-3 Severity: normal P.S. CCing all possibly interested parties ;) According to docstring of WarpTimeSeriesImageMultiTransform sequence of necessary transformations could be arbitrarily long. Having applied a sequence MNI -> inverse warp -> inverse affine -> inverse affine produces different (and looks like incorrect) result from if I do transformations in two steps, e.g. MNI -> inverse warp -> inverse affine and then result of previous given for the final inverse affine. Here is output of running reproduce.sh. All data, script and results are available from http://www.onerussian.com/Linux/bugs/ants.2steps/ I investigate the result by comparing mni2bold.nii.gz and mni2bold-2.nii.gz overlayed on top of epi_template.nii.gz. mni2bold.nii.gz seems to be offset toward anterior $> ./reproduce.sh I: Lets do directly all the steps at once Running: WarpTimeSeriesImageMultiTransform 4 mni.nii.gz mni2bold.nii.gz -R epi_template.nii.gz anat2mni_InverseWarp.nii.gz -i anat2mni_Affine.txt -i raiders_func2anat_Affine.txt --use-NN FIELD: anat2mni_InverseWarp.nii.gz AFFINE-INV: anat2mni_Affine.txt AFFINE-INV: raiders_func2anat_Affine.txt moving_image_filename: mni.nii.gz output_image_filename: mni2bold.nii.gz reference_image_filename: epi_template.nii.gz [0/3]: FIELD: anat2mni_InverseWarp.nii.gz [1/3]: AFFINE-INV: anat2mni_Affine.txt [2/3]: AFFINE-INV: raiders_func2anat_Affine.txt Four-D image size: [182, 218, 182, 1] 4D-In-Spc [1, 1, 1, 1] 4D-In-Org [-90, 126, -72, 0] 4D-In-Size [182, 218, 182, 1] 4D-In-Dir 1 0 0 0 0 -1 0 0 0 0 1 0 0 0 0 1 ...... 4D-Out-Spc [3, 3, 3, 1] 4D-Out-Org [-70.5, -75.759, 93.1085, 0] 4D-Out-Size [64, 64, 48, 1] 4D-Out-Dir -0 0 1 0 1 -0 0 0 0 -1 -0 0 0 0 0 1 Use Nearest Neighbor interpolation 0 % done ... -0 0 1 1 -0 0 0 -1 -0 100 % complete I: Lets do directly all the steps at once Step 1 mni->anat: WarpTimeSeriesImageMultiTransform 4 mni.nii.gz mni2anat.nii.gz -R brain_avg.nii.gz anat2mni_InverseWarp.nii.gz -i anat2mni_Affine.txt --use-NN FIELD: anat2mni_InverseWarp.nii.gz AFFINE-INV: anat2mni_Affine.txt moving_image_filename: mni.nii.gz output_image_filename: mni2anat.nii.gz reference_image_filename: brain_avg.nii.gz [0/2]: FIELD: anat2mni_InverseWarp.nii.gz [1/2]: AFFINE-INV: anat2mni_Affine.txt Four-D image size: [182, 218, 182, 1] 4D-In-Spc [1, 1, 1, 1] 4D-In-Org [-90, 126, -72, 0] 4D-In-Size [182, 218, 182, 1] 4D-In-Dir 1 0 0 0 0 -1 0 0 0 0 1 0 0 0 0 1 ...... 4D-Out-Spc [1, 1, 1, 1] 4D-Out-Org [-86.8253, -128, 132.723, 0] 4D-Out-Size [256, 256, 176, 1] 4D-Out-Dir -0 0 1 0 1 -0 0 0 0 -1 -0 0 0 0 0 1 Use Nearest Neighbor interpolation 0 % done ... -0 0 1 1 -0 0 0 -1 -0 100 % complete Step 2 anat->mni: WarpTimeSeriesImageMultiTransform 4 mni2anat.nii.gz mni2bold-2.nii.gz -R epi_template.nii.gz -i raiders_func2anat_Affine.txt --use-NN AFFINE-INV: raiders_func2anat_Affine.txt moving_image_filename: mni2anat.nii.gz output_image_filename: mni2bold-2.nii.gz reference_image_filename: epi_template.nii.gz [0/1]: AFFINE-INV: raiders_func2anat_Affine.txt Four-D image size: [256, 256, 176, 1] 4D-In-Spc [1, 1, 1, 1] 4D-In-Org [-86.8253, -128, 132.723, 0] 4D-In-Size [256, 256, 176, 1] 4D-In-Dir -0 0 1 0 1 -0 0 0 0 -1 -0 0 0 0 0 1 ...... 4D-Out-Spc [3, 3, 3, 1] 4D-Out-Org [-70.5, -75.759, 93.1085, 0] 4D-Out-Size [64, 64, 48, 1] 4D-Out-Dir -0 0 1 0 1 -0 0 0 0 -1 -0 0 0 0 0 1 -- System Information: Debian Release: squeeze/sid APT prefers unstable APT policy: (901, 'unstable'), (900, 'testing'), (300, 'experimental') Architecture: amd64 (x86_64) Kernel: Linux 2.6.32-5-amd64 (SMP w/2 CPU cores) Locale: LANG=en_US, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/bash Versions of packages ants depends on: ii libc6 2.11.2-2 Embedded GNU C Library: Shared lib ii libgcc1 1:4.4.4-6 GCC support library ii libinsighttoolkit3.18 3.18.0-3+b1 Image processing toolkit for regis ii libstdc++6 4.4.4-6 The GNU Standard C++ Library v3 ants recommends no packages. Versions of packages ants suggests: ii fsl 4.1.6-1 analysis tools for FMRI, MRI and D pn gridengine-client <none> (no description available) -- no debconf information -- .-. =------------------------------ /v\ ----------------------------= Keep in touch // \\ (yoh@|www.)onerussian.com Yaroslav Halchenko /( )\ ICQ#: 60653192 Linux User ^^-^^ [175555]
