[MINC-users] icbm_template stx sampling lattice definitions

[Alex Zijdenbos]

Hello all,

I realized that the way the icbm_sampling lattices are defined, is not
ideal for certain applications. Specifically, all the
icbm_template_X.XXmm.mnc lattice definitions, all have the same start
values, and I believe that the world coordinate origin always falls in the
center of a voxel.

This gets a bit annoying when dealing with issues of symmetry. E.g., if you
are looking to define L/R in these spaces, there is a x=0 plane of voxels
for which it is not clear whether it belongs to L or R. By the same token,
the bounding boxes of these spaces do not line up, and the voxel dimensions
do not follow a regular series, where twice the resolution means twice the
number of voxels. I am fairly sure that this inherent asymmetry actually
affects the behaviour of some tools as well.

I've always thought it would be better to define the standard sampling
lattice with respect to the world coordinate origin which would be *between*
voxels; such that the lattices, albeit with different start values, would
line up with respect to the origin.

I don't fully recall why we (Peter?) developed these lattice templates like
this back then. Obviously I can define new lattices to suit my needs; but
standardization has its benefits :-) Was curious if anybody had any
thoughts about this.

-- A