maximum axial difference for coincidences

General Discussions
1

Dear castor users
I am using castor_v2.1 and at the first I want to calculate the coincidences with regarding the maximum axial difference. Are there any way to apply this option before reconstruction?
Any answer will be highly appreciated!

Best regards

2

Dear Hadi,

As far as I know, and it is valid only if the code has been fully implemented, the maximum axial difference can only be used during the reconstruction to ignore lines of response beyond this provided maximum axial difference.

As far as I remember, it is implemented through a copolar angle in order to be generic, and not a ring difference.

To apply this before reconstruction, you will have to implement your own code to do so.

You can start with one of the toolkits provided, such as castor-dataFileExplorer or castor-dataFileShuffler.

Best

Simon

3

Thank you for your response.

4

Hi Hadi,

This option just affects the computation of the sensitivity image for list-mode, in the case it is computed from the scanner geometry (not from a normalization datafile nor from histogram, as described in documentation 6.1). It is not applied during reconstruction.

What this parameter does is imposing a restriction on the axial distance between the detectors during the sensitivity image computation. The only purpose is to adapt the sensitivity image to the reconstruction of a dataset acquired/simulated with axial restrictions in the system geometry.

Best regards,

5

Hello Everyone,

I was reading about maximum axial difference mm parameter and came across this post. I would appreciate if someone can help to understand some concepts about this parameter.

1- @tmerlin mentioned that this option only affects how sensitivity image is computed. For the actual image what is the maximum axial difference? For the actual image reconstruction, does it by default consider all lines of responses with any ring difference?

2- What if we set a number in mm that is greater that the axial distance between the first and the last ring?

3- Does CASToR place 2D reconstructed images next to one another to create a 3D image (like attached layers)?

Best,
Seyyed

6

Hi Seyyed,

  1. There is no geometrical restrictions/Checks during reconstruction, all LORs are considered. This parameter purpose is to adapt the sensitivity image to the used data (for instance if the data contains only oblique LORs below a certain axial angle) when the sensitivity image is computed from the geometry.

  2. This would have no effect, every LOR passing the restriction.

  3. The reconstructed image remains a 3D volume. What would change is that the projected LORs will be restricted to similar rings for each pair of detectors.

Best,
Thibaut