[Rtk-users] Forward projection rotation matrix from gantry and couch angles

[Simon Rit]

Hi,
I would rather suggest another version of AddProjection
<http://www.openrtk.org/Doxygen/classrtk_1_1ThreeDCircularProjectionGeometry.html#a0fb1475ed76a28cde24fac85eae18e1e>
which uses the source position, detector origin position, u and v vectors.
Be aware that FBP assumes that the rotation is around y, I'm sure that it
won't work for random geometries. If what is important is to have a final
volume in the correct orientation, you can always change its direction and
reset it to identity after reconstruction.
I hope it helps,
Simon

On Fri, Feb 25, 2022 at 1:00 PM pmkclnl at yandex.ru <pmkclnl at yandex.ru> wrote:

> Dear developers,
>
> For DRR image calculation i use a ThreeDCircularProjectionGeometry class
> with AddProjection method. The default rotation direction is around
> Y-axis in LPS coordinates (Z-axis in RAS coordinates in 3D Slicer) as i
> understand from the documentation. Should i calculate rotation matrix
> manually (using gantry, in-plane, out of plane angles) and then use
> AddRotationMatrix method to simulate movement of gantry angle around the
> Y-axis in RAS coordinates (and couch angle rotation around Z-axis in RAS
> coordinates)? Are there any other options to setup a required axis of
> rotation of the gantry?
>
> Best regards,
>
> Mikhail
>
>
> _______________________________________________
> Rtk-users mailing list
> Rtk-users at public.kitware.com
> https://public.kitware.com/mailman/listinfo/rtk-users
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://public.kitware.com/pipermail/rtk-users/attachments/20220225/d8d7c548/attachment.htm>