I'm afraid I havn't been very clear and I have got things all muddled. Here is a very specific outline of my problem, which I should have given you from the start.


The I images are not Dicom, but they are directly from the scanner. This data was taken somewhere else, so it may take me a little while to find exactly what format they are in. For each subject, there is about 100 files name I.001, I.002, etc. The files are unix executable.

The output of the to3d command are two files:
SubjectNumber+orig.BRIK
SubjectNumber+orig.HEAD

3dinfo gives:
aa-devito:040902 head$ 3dinfo 040902+orig.BRIK
++ 3dinfo: AFNI version=AFNI_2008_07_18_1710 (Mar 13 2009) [32-bit]

Dataset File: 040902+orig
Identifier Code: XYZ_DWw7FJbRcUL22DejLUTHqA Creation Date: Wed Mar 18 17:33:11 2009
Dataset Type: MRI Anatomy (-anat)
Byte Order: LSB_FIRST [this CPU native = LSB_FIRST]
Storage Mode: BRIK file
Storage Space: 16252928 (16 million) bytes
Geometry String: "MATRIX(0.8594,0,0,-109.5735,0,0,1.299561,-137.7004,0,-0.8594,0,75.22376):256,256,124"
Data Axes Tilt: Plumb
Data Axes Orientation:
first (x) = Right-to-Left
second (y) = Superior-to-Inferior
third (z) = Anterior-to-Posterior [-orient RSA]
R-to-L extent: -109.574 [R] -to- 109.573 [L] -step- 0.859 mm [256 voxels]
A-to-P extent: -137.700 [A] -to- 22.146 [P] -step- 1.300 mm [124 voxels]
I-to-S extent: -143.923 -to- 75.224 [S] -step- 0.859 mm [256 voxels]
Number of values stored at each pixel = 1
-- At sub-brick #0 '#0' datum type is short: 0 to 13420

----- HISTORY -----
[head@aa-devito.ccn.duke.edu: Wed Mar 18 17:33:11 2009] to3d -anat -session 040902 -prefix 040902 ../rawdata/040902/I.001 ../rawdata/040902/I.002 ../rawdata/040902/I.003 ... ../rawdata/040902/I.123 ../rawdata/040902/I.124
[head@aa-devito.ccn.duke.edu: Tue Mar 24 14:25:00 2009] plug_nudge: -rotate 2.96I 5.00R -0.50A -ashift 0.00S 0.00L 0.00P
[head@aa-devito.ccn.duke.edu: Wed Mar 25 17:49:51 2009] plug_nudge: -rotate -2.96I -5.00R 0.50A -ashift 0.00S 0.00L 0.00P
[head@aa-devito.ccn.duke.edu: Thu Mar 26 18:15:01 2009] plug_nudge: -rotate 2.96I 5.00R -0.50A -ashift 0.00S 0.00L 0.00P



The history is the nudges that I have done. When I use 3dinfo on the raw, pre-nudged afni files, everything is the same except for the nudges. However, both images, nudged and unnudged, when transformed to Nifti and opened in SPM line up. Additionally, the ROI, when transformed to Nifti, lines up in SPM on the correct anatomical region. However, they do not line up with the spoiled grass images that SPM constructed (in analyze format) from the same raw scanner data and on which I have done all of my functional analysis. I need to figure out some way to orient the AFNI drawn ROI onto the SPM spoiled grass image. My original theory is that afni had re-oriented the brain somewhere in the process of transforming the images, but apparently this does not happen. Do you have any suggestions on how to deal with this problem?

Thanks again,
Ian