[Insight-developers] Re: [Fwd: Re: itkImage]

[Leila Baghdadi]

Hey Petit frere,

1. I do not believe I can claim anything about DICOM simply because I
have hardly used it, so I agree with whatever the DICOM experts say!

2. By the same token, I know enough about MINC to know that the origin
is infact rotated by the direction cosines, believe me I have checked
various minc tools and programs!

3. I think till last week I had the wrong assumption 

There is a generalized definition for voxel-world conversion using the
direction cosines for different image formats.

I am now convinced that the above is simply NOT TRUE!

4. I "HUMBLY" agreed to change my code reflecting the changes to the
origin.

Thanks again for everyone's time and effort

Leila

p.s I will send an e-mail to the list once I have fixed everything at my
end!

On Mon, 2006-13-03 at 11:55 -0500, Mathieu Malaterre wrote:
> > Hello Master Luis,
> > 
> > That is exactly what I was talking about,
> > 
> > If I recall Bill was the one who agreed with me and said that the origin
> > should be rotated! (check e-mail archives),
> 
> Leila,
> 
> 	If you follow that exact same thread, you'll see that Bill also agree 
> on the fact that this is wrong. Gordon Kindlmann pointed out in the 
> DICOM reference that the Direction Cosines should *not* be applied to 
> the origin.
> 
> 	For more info, see PS3.3. C.7.6.2.1.1 Image Position and Image 
> Orientation. (~p275 in 2004 edition).
> 
> 
> >>>OK, I will investigate this further with a DICOM expert,
> >>>
> >>>but it would be awful if different image formats had different methods
> >>>of calculating voxel -- world based on direction cosines!
> 
> This I don't know.
> 
> HTH,
> Mathieu