[MINC-development] minc 2.0 file format definition

romain minc-development@bic.mni.mcgill.ca
Tue, 10 May 2005 18:17:22 +0200


/Hello/


/I have carefully read the MINC 2.0 File Format document written
by Robert D Vincent, and I have a few question, in order to
fully understand how the orientation is handle./


/First, I wonder if the formula (4) from section 2.1 describing
the relation between voxel space and world space is not false ?
I would have put Ox, Oy, Oz in the last colon rather than in the last line./


/Then I have a problem with the definition of xspace, yspace and zspace : /

in the definition of direction_cosines it is say :
"Numeric vector with 3 elements giving the direction cosines of the axes.
Although axes are labeled x, y and z, they may in fact have a significantly
different orientation - this attribute allows the direction relative to
the true axes to be specified exactly. "


The definition of the true axes is clear : this is the patient 
coordinate system,
 but then the precise definition of the axes (that is xspace, yspace and 
zspace)
is not clear to me.


In the dicom format the equivalent definition is clear. The first given
cosines direction is the direction of the first row in the patient 
coordinate
system. (and the first row is the faster varying index of the stored data)


If you use the same definition then the name is really confusing... if not
cant someone precise it  ?


If someone could give me an example for the definition of those 
dimension variables,
in the case of coronal acquired volume (ie, where the data is stored in 
the order,
x axis, z axis and y axis.)


Thanks for the help

Romain