[Rtk-users] Reconstruction of CBCT using FDK algorithm

[Simon Rit]

Hi,
Please keep the mailing list in the loop.
Are the distances in pixel units? If not, you need to set the spacing of
the projections according to the pixel spacing of your projections.
Simon

On Wed, Jun 19, 2024 at 8:56 AM yx deng <yxd15227813801 at gmail.com> wrote:

> Hi,Simon
> I had red the doc of  *RTK 3D circular projection geometry*   carefully,
>   GantryAngle is the rotation around y, and the code of setting geometry
> as follows:
> using GeometryType = rtk::ThreeDCircularProjectionGeometry;
>   GeometryType::Pointer geometry = GeometryType::New();
>   unsigned int          numberOfProjections = 1201;
>   double                firstAngle = 0;
>   double                angularArc = 360;
>   unsigned int          sid = 66.;// source to isocenter distance
>   unsigned int          sdd = 199.;  // source to detector distance
>   for (unsigned int noProj = 0; noProj < numberOfProjections; noProj++)
>   {
>
>     double angle = firstAngle + noProj * angularArc / numberOfProjections;
>     double ProjOffsetX = 0.;
>     double ProjOffsetY = 0.;
>     double OutoffPlanAngle = 0;
>     double InPlanAngle = 0;
>     double SourceOffSetX = 0.;
>     double SourceOffSetY = 0.;
>     geometry->AddProjection(
>       sid, sdd, angle, ProjOffsetX, ProjOffsetY, OutoffPlanAngle,
> InPlanAngle, SourceOffSetX, SourceOffSetY);
>   }
> I also tried to change the value of  OutoffPlanAngle and InPlanAngle,
> but  it didn't work.Looking forward to your answer.
>
> Simon Rit <simon.rit at creatis.insa-lyon.fr> 于2024年6月18日周二 20:24写道：
>
>> Hi,
>> It seems to be a geometry problem. I would suggest to check if your
>> geometry object is adequately set, see doc here
>> <http://www.openrtk.org/Doxygen/DocGeo3D.html>.
>> Simon
>>
>> On Tue, Jun 18, 2024 at 1:32 PM yx deng <yxd15227813801 at gmail.com> wrote:
>>
>>> *Hi Simon，*
>>>
>>> *I'm a image processing engineer doing project in image** reconstruction using rtk.I met a question that the shape of the region corresponding to the reconstructed target changes from a rectangle to a diamond in the slice.
>>> As the pictures show:The result should be a circle, not a diamond.Please give me some advices to correct the reconstruction ,thanks a lot.
>>> *
>>>
>>> [image: 企业微信截图_1718708067845.png] [image: 企业微信截图_17187080755468.png] [image:
>>> 企业微信截图_17187080843559.png]
>>>
>>> the projection images'size is 972*768*1200,
>>> origin is (-485.5,-383.5,0),spacing is (1.0,1.0,1.0)
>>>
>>> the volume's size is 768*768*128
>>>
>>> origin is (-383.5,-383.5,-63.5)，
>>>
>>> spacing is (1.0,1.0,1.0).
>>>
>>> the class of CudaFDKConeBeamReconstructionFilter was applied.
>>>
>>>
>>>
>>>
>>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.creatis.insa-lyon.fr/pipermail/rtk-users/attachments/20240619/a7657fc2/attachment-0001.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: ??????_1718708067845.png
Type: image/png
Size: 8097 bytes
Desc: not available
URL: <http://www.creatis.insa-lyon.fr/pipermail/rtk-users/attachments/20240619/a7657fc2/attachment-0003.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: ??????_17187080755468.png
Type: image/png
Size: 19115 bytes
Desc: not available
URL: <http://www.creatis.insa-lyon.fr/pipermail/rtk-users/attachments/20240619/a7657fc2/attachment-0004.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: ??????_17187080843559.png
Type: image/png
Size: 39411 bytes
Desc: not available
URL: <http://www.creatis.insa-lyon.fr/pipermail/rtk-users/attachments/20240619/a7657fc2/attachment-0005.png>