[ITK] meaning of negative spacing

Dženan Zukić dzenanz at gmail.com
Mon Oct 24 11:16:35 EDT 2016


Hi Zein,

image position patient is the authoritative tag to reconstruct slice
spacing and the vector of Z position/orientation. But I am surprised that
result is negative spacing instead of negative direction cosine
(orientation vector). Can you provide the offending DICOM file?

Regards,
Dženan

On Mon, Oct 24, 2016 at 11:12 AM, Zein Salah <zeinsalah at gmail.com> wrote:

> Hi Dženan,
>
> thanks for the response.
>
> but I still wonder what does this mean. The reason for the question is:
>
> The input image is a multiframe dicom, in which the image position
> patient (0020, 0032)
> is stored in the header for every frame. and these values are having
> decreasing values
> in the z-coordinate. So I am asking myself
>
> 1. if the itkimagereader considers this when it reconstructs the 3D volume?
>
> 2. if this is the reason for negative z-spacing? if this is the case,
> I can this to
> interpret the meaning of the z-spacing.
>
>
>
>
> 2016-10-24 16:55 GMT+02:00 Dženan Zukić <dzenanz at gmail.com>:
> > Hi Zein,
> >
> > I suppose you could invert the 3rd row (Z component) of the
> TransformMatrix
> > and have the spacing positive. It should have the same effect and is more
> > logical.
> >
> > Regards,
> > Dženan
> >
> > On Mon, Oct 24, 2016 at 8:50 AM, Zein Salah <zeinsalah at gmail.com> wrote:
> >>
> >> Hi,
> >>
> >> I have a question regarding reading dicom files with itk,
> >> particularly, multiframe dicom file (all slices in one single file).
> >>
> >> I have read an image with itk 4.10 and write it into an mhd-raw image.
> >> I noticed that the z-spacin is negative.
> >> What does negative spacing mean?
> >>
> >> thanks,
> >> Zein
> >> _______________________________________________
> >> Community mailing list
> >> Community at itk.org
> >> http://public.kitware.com/mailman/listinfo/community
> >
> >
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/community/attachments/20161024/e6371e21/attachment.html>


More information about the Community mailing list