[Castor-users] Siddon multiray returns prematurely (I think)

[Andreas Bolke]

Dear Castor Developers,

first of all, thanks for making Castor open source and maintaining the 
code in the first place. It has already helped me
to find Bugs in my own reconstruction when I was comparing my results to 
the castor result.

Well I guess I get right to the point then.
I use very simple GATE data to make PET reconstructions and compare them 
to my own reconstruction.
If I understand multiray correctly, then I think in 
iProjectorIncrementalSiddonMulti::ProjectWithoutTOF() the line 282:

/ if( alphaMax <= alphaMin ) return 0;/

will cause the function to return prematurely without having iterated 
over all lines (I mean multiray lines, not oProjectionLine objects).
This can cause for example low sensitivity values on the edges (I added 
a screenshot to the attachment).
When for example you use 16 lines, but because of the
random position of each line, line no. 2 might not hit a voxel, the 
function will return and stop iterating over
the remaining 14 lines.

Or am I missing something? If this is the case, I apologize in advance.


Best regards,
Andreas
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.castor-project.org/pipermail/castor-users/attachments/20210224/a73dbfbc/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: castor_multiray_sens_comparisson.png
Type: image/png
Size: 115406 bytes
Desc: not available
URL: <http://lists.castor-project.org/pipermail/castor-users/attachments/20210224/a73dbfbc/attachment-0001.png>