Hi,
I've previously posted comments about the orient header field of the Analyze
7.5 format.
After discussion with [log in to unmask], I've confirmed an error in
my matlab code and some associated comments related to handling the flipped
codes for this field. I had assumed the flips were in the slice dimension,
so axial flipped was Z flipped, coronal flipped was Y flipped and sagittal
flipped was X flipped. However, this is incorrect. The flipping is only
required for raster graphics movie animations and the flips are done in the
row dimensions so the orthoviews are drawn from top to bottom as quickly as
possible. Hence, axial flipped is a Y flip (drawn from anterior to
posterior), coronal flipped is a Z flip and so too is sagittal flipped a Z
flip (drawn from superior to inferior). Thus, the flipped orient volumes
are NOT RECOMMENDED for general volume storage and handling, according to
the Analyze 7.5 standard.
Furthermore, this means that the Analyze 7.5 standard does not "officially"
support neurological orientations. The default Analyze 7.5 orientation is
the transverse unflipped volume, which accords with the radiological
convention, with +X left, +Y anterior and +Z superior, and the origin (voxel
1,1,1) is always at the bottom left of the ortho-display. I had assumed
that the sagittal flipped volume provides the neurological convention, with
only a flip in X dimension, giving +X *right*, +Y anterior, +Z superior.
This is not the case. Hence, SPM and associated software must discard the
standard orientation of the Analyze format to handle volumes with
neurological convention (eg, Talairach normalized volumes). The adoption of
a .mat file to hold a 3D transformation matrix is a reasonable extension of
the Analyze 7.5 format. The Analyze 3D coordinate system and associated
orient header field
then becomes irrelevant.
In any case, for anyone interested, some revised matlab code to handle the
orient field, according to the "official" methods, is available in the
mri_toolbox at http://sourceforge.net/project/showfiles.php?group_id=50847,
along with associated documentation. A new function called avw_flip
provides for any ortho-flip you like, although it will most likely
invalidate the "official" Analyze 7.5 orientation system. This code does
not handle the SPM .mat file, sorry. The avw_view function has been
revised, with brightness controls and an easy left/right flip to assist with
viewing in neurological or radiological orientation (only in the viewer, the
data in the workspace and on disk are assumed to be radiological).
I hope this makes sense.
Kind regards, Darren
--
Darren Weber, PhD Student
Cognitive Neuroscience, School of Psychology
Flinders University of SA, GPO Box 2100, Adelaide, SA 5001, Aust.
Ph: (61 8) 8201 3998, Fax: (61 8) 8201 3877
http://brain.mhri.edu.au/~dlw/homepages/index.html
|