MATLAB Central Newsreader - 3D CT reconstruction

3D CT reconstruction

maya — Tue, 02 Dec 2008 05:00:44 -0500

I am trying to reconstruct 3D image from 2D CT slices using delaunay triangulation.Can anybody say how to use the delaunay function in Matlab for the above

Thanks
Maya

Re: 3D CT reconstruction

russell.fung@gmail.com — Tue, 02 Dec 2008 14:53:41 -0500

On Dec 1, 11:00=A0pm, "maya" <mayaea...@googlenews.com> wrote:
> I am trying to reconstruct 3D image from 2D CT slices using delaunay tria=
ngulation.Can anybody say how to use the delaunay function in Matlab for th=
e above
>
> Thanks
> Maya

You can probably start by reading the documentation for Delaunay
Triangulation:

help delaunay

Re: 3D CT reconstruction

Gavrilo Bozovic — Tue, 02 Dec 2008 15:00:20 -0500

"maya" <mayaeapen@googlenews.com> wrote in message <gh2fds$b29$1@fred.mathworks.com>...
> I am trying to reconstruct 3D image from 2D CT slices using delaunay triangulation.Can anybody say how to use the delaunay function in Matlab for the above
>
> Thanks
> Maya

Maya, you already asked this same question on August, the 28th. Tens of comments followed, which showed that you did not provide enough information for anyone to give you an answer.

What should be different this time? Try to better formulate your question. You could also try and explain what you have already tried to do to solve your problem (I'm assuming here that you *did* do something in the last three months).

Re: 3D CT reconstruction

Walter Roberson — Tue, 02 Dec 2008 15:24:31 -0500

maya wrote:
> I am trying to reconstruct 3D image from 2D CT slices using delaunay triangulation. Can anybody
> say how to use the delaunay function in Matlab for the above

Based upon the information you supplied in your previous round of postings,
it is not possible to use Delaunay triangulation to do the reconstruction
that you hoped to do.

Delaunay triangulation is for the case where you have points in space, and
you have places where the points are not. That does not match the situation you
described before, where you have a complete matrix of voxels, each of which is
a volume of a different intensity, and you wish to connect together certain
voxels that (if I recall correctly) you have no present algorithm for
firmly differentiating from other voxels. Your segmentation work needs to
start with the aspect of differentiating the voxels of interest from those
that are not of interest: only then should you be concerned about reconstruction.

--
.signature note: I am now avoiding replying to unclear or ambiguous postings.
Please review questions before posting them. Be specific. Use examples of what you mean,
of what you don't mean. Specify boundary conditions, and data classes and value
relationships -- what if we scrambled your data or used -Inf, NaN, or complex(rand,rand)?

Re: 3D CT reconstruction

ImageAnalyst — Tue, 02 Dec 2008 16:23:36 -0500

On Dec 2, 12:00=A0am, "maya" <mayaea...@googlenews.com> wrote:
> I am trying to reconstruct 3D image from 2D CT slices using delaunay tria=
ngulation.Can anybody say how to use the delaunay function in Matlab for th=
e above
>
> Thanks
> Maya

-------------------------------------------------------------
Pretty sloppy terminology. You either reconstruct a 3D image from 2D
projection images (not slices) (using the regular methodology of
filtered back projections, Radon transforms, etc.) or you can make a
surface rendering once you've identified (somehow) which voxels lie on
the surface that you want to render. Which is it?
Regards,
ImageAnalyst