[Insight-developers] writing volume as a set of image slices

Yaniv, Ziv ZYaniv at childrensnational.org
Tue Jul 12 21:13:59 EDT 2011


Hi all,

I'm sending this to the mailing list before I log a bug report as this is in the gray area between bug and feature.

There seems to be a logical bug/feature in the way itkImageSeriesWriter saves the output images origin. In combination with DICOM writing this results in erroneous values for the "Image Position Patient". The image series writer uses
"double origin[TOutputImage::ImageDimension];" when writing a n-D volume as a set of k-D "slices", n>k  (in my case n==3, k==2) this means that each "slice" is no longer in context. If I look at a slice as a lower dimension subspace and am no longer interested in its relationship with other slices than I'm good. Unfortunately I am interested in retaining the relationships when I write a volume as a set of DICOM slices.

To see what I am talking about run the DicomSeriesReadSeriesWrite.cxx example program on a CT/MR data set ( data available here http://midas.kitware.com/collection/view/90 ). It should just perform copy, but in practice the z coordinate for all slices is set to zero.

Bug (slice loses its location in the higher dimensional space)? Feature (origin must match image dimension)?  

     any ideas suggestions appreciated           
                 Ziv

p.s. "It's on the wiki" didn't work for me this time, although I'm sure it's out there.                 

--
Ziv Yaniv, PhD., 
Principle Investigator,
The Sheikh Zayed Institute for Pediatric Surgical Innovation
Children's National Medical Center
6th Floor Main Hospital, Room M7740
111 Michigan Ave., N.W
Washington, D.C 20010
Phone: 202-476-1288
email: zyaniv at childrensnational.org




-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.itk.org/mailman/private/insight-developers/attachments/20110712/ebf4fd3e/attachment.htm>


More information about the Insight-developers mailing list