Back to PLS Helpsave BS image to .img file
Posted on 01/04/08 16:54:33
Number of posts: 41
Hello again ~
I am stuck on how to efficiently extract bootstrap ratios to an external image. Using the 'extract to .img file' fromt the main menu works - but the file produced loses information about the image.
e.g., To run PLS/create datamats I use fully pre-processed and normalized (spatially) files either within MNI or Talairach space. Thus xyz mm in PLS represent this space. When I save the bootstrap ratios using the GUi - the .img image (presumably nifti format but perhaps not?) loses these space coordinates and thus, when loaded into AFNI (for instance) it's mismatched to other files and in its own coordinate system.
Thus, I typically use raw coordinates from PLS and create images manually to view the results outside of the PLSGUI. This is cumbersome, however, and so I am wondering how to best use the internal functions to generate an external image of the result. I typically use AFNi, not SPM, for analysis - thus nifti files are preferred.
Any thoughts?
Best
Agatha
I am stuck on how to efficiently extract bootstrap ratios to an external image. Using the 'extract to .img file' fromt the main menu works - but the file produced loses information about the image.
e.g., To run PLS/create datamats I use fully pre-processed and normalized (spatially) files either within MNI or Talairach space. Thus xyz mm in PLS represent this space. When I save the bootstrap ratios using the GUi - the .img image (presumably nifti format but perhaps not?) loses these space coordinates and thus, when loaded into AFNI (for instance) it's mismatched to other files and in its own coordinate system.
Thus, I typically use raw coordinates from PLS and create images manually to view the results outside of the PLSGUI. This is cumbersome, however, and so I am wondering how to best use the internal functions to generate an external image of the result. I typically use AFNi, not SPM, for analysis - thus nifti files are preferred.
Any thoughts?
Best
Agatha
Replies:
Untitled Post
I'm Online
Posted on 01/04/08 17:10:48
Number of posts: 291
> I am stuck on how to efficiently extract bootstrap ratios to an
> external image. Using the 'extract to .img file' fromt the main
> menu works - but the file produced loses information about the image.
>
> e.g., To run PLS/create datamats I use fully pre-processed and
> normalized (spatially) files either within MNI or Talairach
> space. Thus xyz mm in PLS represent this space. When I save the
> bootstrap ratios using the GUi - the .img image (presumably nifti
> format but perhaps not?) loses these space coordinates
It's ANALYZE 7.5. Any NIfTI supported program should have not problem to load it. In addition, it keeps the space coordinates, including space dimension, voxel size, and originator.
and thus,
> when loaded into AFNI (for instance) it's mismatched to other
> files and in its own coordinate system.
>
> Thus, I typically use raw coordinates from PLS and create images
> manually to view the results outside of the PLSGUI. This is
> cumbersome, however, and so I am wondering how to best use the
> internal functions to generate an external image of the result. I
> typically use AFNi, not SPM, for analysis - thus nifti files are
> preferred.
step?
Posted on 01/04/08 17:53:27
Number of posts: 41
Jimmy ~ Yes, that's why I am perplexed.
It seems, however, that the dimensions of the image are distored. If anyone knows the answer to this let me know... I'll try to figure it out in the meantime. The steps I use to get to AFNI:
- save BS values as .img file
- 3dcalc -a BI_OUTPUT.hdr -prefix BI_OUTPUT -expr 'a' (TO CHANGE TO BRIK - JUST TO BE SURE)
- 3drefit -fim -view 'tlrc' BI_OUTPUT (TO MATCH MY VIEW SETTINGS)
Best
Agatha
It seems, however, that the dimensions of the image are distored. If anyone knows the answer to this let me know... I'll try to figure it out in the meantime. The steps I use to get to AFNI:
- save BS values as .img file
- 3dcalc -a BI_OUTPUT.hdr -prefix BI_OUTPUT -expr 'a' (TO CHANGE TO BRIK - JUST TO BE SURE)
- 3drefit -fim -view 'tlrc' BI_OUTPUT (TO MATCH MY VIEW SETTINGS)
Best
Agatha
details...
Posted on 01/04/08 18:04:32
Number of posts: 41
fyi: here are the dimensions of the file - below it are the dimensions this file SHOULD have (i.e., this is what went INTO making PLS datamats!
R-to-L extent: -158.000 [R] -to- -2.000 [R] -step- 4.000 mm [ 40 voxels]
A-to-P extent: -190.000 [A] -to- -2.000 [A] -step- 4.000 mm [ 48 voxels]
I-to-S extent: 2.000 [S] -to- 150.000 [S] -step- 4.000 mm [ 38 voxels]
PROPER FILE
R-to-L extent: -78.000 [R] -to- 78.000 [L] -step- 4.000 mm [ 40 voxels]
A-to-P extent: -79.000 [A] -to- 109.000 [P] -step- 4.000 mm [ 48 voxels]
I-to-S extent: -64.000 [I] -to- 84.000 [S] -step- 4.000 mm [ 38 voxels]
R-to-L extent: -158.000 [R] -to- -2.000 [R] -step- 4.000 mm [ 40 voxels]
A-to-P extent: -190.000 [A] -to- -2.000 [A] -step- 4.000 mm [ 48 voxels]
I-to-S extent: 2.000 [S] -to- 150.000 [S] -step- 4.000 mm [ 38 voxels]
PROPER FILE
R-to-L extent: -78.000 [R] -to- 78.000 [L] -step- 4.000 mm [ 40 voxels]
A-to-P extent: -79.000 [A] -to- 109.000 [P] -step- 4.000 mm [ 48 voxels]
I-to-S extent: -64.000 [I] -to- 84.000 [S] -step- 4.000 mm [ 38 voxels]
got it!
Posted on 01/04/08 18:29:11
Number of posts: 41
got it!
FYI - in case any one ever has this problem - you need to change the origin manually - for some reason AFNI loses it when reading in the nifti files. the program 3drefit can be used to correct the origin and recenter the brain.
best
agatha
FYI - in case any one ever has this problem - you need to change the origin manually - for some reason AFNI loses it when reading in the nifti files. the program 3drefit can be used to correct the origin and recenter the brain.
best
agatha
Calculating Jacobian
Posted on 01/16/08 19:24:23
Number of posts: 1
Hello,
I have obtained deformation fields after registring a subject to the reference image. Is it possible to calculate the Jacobian from these deformation fields using the PLS software?
Thank you
I have obtained deformation fields after registring a subject to the reference image. Is it possible to calculate the Jacobian from these deformation fields using the PLS software?
Thank you
Untitled Post
Posted on 01/16/08 19:38:13
Number of posts: 394
quote:
Hello,
I have obtained deformation fields after registring a subject to the reference image. Is it possible to calculate the Jacobian from these deformation fields using the PLS software?
Thank you
It is not possible to calculate the Jacobians with the PLS software. You can certainly analyse them with PLS, but our package has very limited tools for image processing. There is a paper published that has used Jacobians to compare brain structure between mouse strains: Hello,
I have obtained deformation fields after registring a subject to the reference image. Is it possible to calculate the Jacobian from these deformation fields using the PLS software?
Thank you
Neuroanatomical differences between mouse strains as shown by high-resolution 3D MRI
Chen, Kovacevic, Lobaugh, N.J., Sled, J.G. Henkelman, R.M. Henderson, J.T. NeuroImage Volume 29, Issue 1, 1 January 2006, Pages 99-105
Login to reply to this topic.
