[MINC-users] Dicom to Minc converter for Phillips ?

[Jonathan HARLAP]

It seems that your dcm files have the ImageOrientationPatient
(0x20,0x37) field filled in, but not the ImagePositionPatient
(0x20,0x32).  dicom3_to_minc uses both these fields in order to
determine the slice position in the out of slice dimension...  There
is an option (-ignore_positions) in dicom3_to_minc which lets you tell
it to use the image number instead of the slice position, which I had
added specifically to handle cases where slice position could not be
calculated properly, with the caveat that your step in the out of
slice dimension will be 1mm (assuming consecutive image numbers)
instead of whatever it should really be, and the start coordinate in
that dimension will be wrong.  However, a simple header hack could fix
that.

There also seems to be a field SliceLocation (0x20,0x1041) which your
data contains, but the values are not unique across the range of
slices, so it would be difficult to use that, as dicom3_to_minc would
complain of duplicate slices at a given position, and proceed to
ignore all but one slice at each position.

I'm open to suggestions if you can think of a sensible way to handle
this kind of data...

Cheers,
Jon


On Fri, Mar 11, 2005 at 09:52:48PM +0900, Hyun-Pil Kim wrote:
> Hi Jon,
> 
> I can send you the files only if you remove them after using testing
> purpose. Just let me know where to upload them.
> 
> Thanks,
> Phil
> |-----Original Message-----
> |From: minc-users-admin@bic.mni.mcgill.ca [mailto:minc-users-
> |admin@bic.mni.mcgill.ca] On Behalf Of Jonathan HARLAP
> |Sent: Friday, March 11, 2005 9:37 PM
> |To: minc-users@bic.mni.mcgill.ca
> |Subject: Re: [MINC-users] Dicom to Minc converter for Phillips ?
> |
> |Hi,
> |
> |If you have a sample Philips dcm file you can send me, I can look to
> |see what kind of fix to dicom3_to_minc would make it work for you...
> |It's also always useful for us to have "problem" data for testing new
> |versions of the converters.
> |
> |Cheers,
> |Jon
> |
> |
> |
> |On Tue, Mar 08, 2005 at 08:35:08PM +0900, Hyun-Pil Kim wrote:
> |>
> |>    Is there any working version of dicom to minc converter for Phillips
> |>    dcm files?
> |>
> |>    The most recent version of dicom3_to_minc utility didn’t work
> |>    for me.
> |>
> |>    After some googling, I found phil_to_minc but somehow the link is not
> |>    present.
> |>
> |>    [1]http://www.cmr.uq.edu.au/~rotor/minc/
> |>
> |>
> |>
> |>    Thanks,
> |>
> |>    Phil.
> |>
> |>    *------------------------------------------------*
> |>    Hyun-Pil Kim,
> |>    Computational Neuroimage Analysis Lab.,
> |>    Dept. of Biomedical Engineering, Hanyang Univ.
> |>    Sungdong P.O. Box 55, Seoul, 133-605, KOREA
> |>    Tel: +82-2-2290-0697; FAX: +82-2-2296-5943
> |>    Mobile: +82-19-286-0485
> |>    E-mail: hpkim at ihanyang.ac.kr; aimtop77 at empal.com
> |>    *------------------------------------------------*
> |>
> |> References
> |>
> |>    1. http://www.cmr.uq.edu.au/~rotor/minc/
> |_______________________________________________
> |MINC-users@bic.mni.mcgill.ca
> |http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users
> 
> _______________________________________________
> MINC-users@bic.mni.mcgill.ca
> http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users