Castor mCT PET/CT recon error

Hello. CASToR users,

Step 1/2: Computing scatter and/or random histogram…
***** castor-GATERootToCastor :: Issue detected during the conversion!
***** detector ID larger than the max nb of detectors: 32448
***** please email to the castor-users mailing list with this log and the mac file
***** rsectorID1:25
***** moduleID1:0
***** submoduleID1:0
***** crystalID1:43
***** layerID1:-1
***** computed castorID:1349765887
***** Exit function called. Abort with code 1.

I have a question for reconstruction process above the error message.
I simulated the mCT PET/CT scanner by Siemens

I introduce the brief specification.

#siemens mct spec
#crystal: LSO(4 x 4 x 20 mm)
#13 x 13 crystals/detector block
#48 module (4 detector blocks/module)
#Detector diameter: 84.9 cm (dia. 42.45 cm)
#4 rings

If i want to know problem, what information I should provide.

In addition, I use vGATE v.9.0.

Thank you.

Best Regards
PCR

Greetings,

This error feels familiar. In short, layerID type can cause problems due to the C++ type casting or something similar. A patch was provided and merged in the develop branch but it is not currently in a released version, I believe.

So, you can either get the develop branch of CASToR in the gitlab, which I do not know if it is in a working state, or look at the modification made in the merge request named “Changed layer variable type to fit with layerId” .

Bests,
Maxime Toussaint

Thank you so much for your answer.

I’m still a beginner and don’t understand everything,

If possible, could you please tell me specifically how I can solve it?

Thanks again for your help.

Best Regards
PCR

Where do you need help?

  • Compiling CASToR?
  • Getting access to the gitlab?
  • Viewing the modifications proposed in a merge request?

We have access to CASToR code, so I believe it is worth it to learn how to use these tools.

Dear Maxime_Toussaint

Thank you again for your kind response.

As you mentioned,

I accessed to gitLab (Castor), but I can’t find where and what the patch is.

Also, can I get the merge request proposed?

Could you let me know what are the methods and tools to access the CASToR code?

Thank you so much.

Best Regards
PCR

Once you are in CASToR’s gitlab, look in the left column/menu for “Code”. In that section, you will see “Merge requests”, click on it. The one I talk about right now was merged, so click on “Merged” (around middle of the screen). The merged request is named “Changed layer variable type to fit with layerId”. Click on it. There should four options “Overview”, “Commits”, “Pipeles” and “Changes”. “Changes” shows what the merge request modified. From this, you will see that only one word was changed!

Best of luck!

Dear Maxime_Toussaint

Thank you so much.

You are helping me a lot with my research.

Again, thank you so much.

I entered the CASToR gitlab (link below) you mentioned, but I can’t find “changed layer variable type to fit with layerid”.

(castor gitlab - Google Search)

Is gitlab paid by any chance?

I’m a beginner, so I’m really sorry.

If you don’t mind, could you please send me the “changed layer variable type to fit with layerid” file and contents

Could you please send it to “tigeaglepcr@gmail.com”?

Thank you very much.

Best Regards
PCR

Creating a gitlab account is free. The paid part is when you create a project or something. To access CASToR gitlab, you only need a gitlab account.

It is fine being a beginner, it means you can learn! Which is also why I tried to show you how to fish at the place of giving you the fish. After all, “Give a man a fish, and you feed him for a day. Teach a man to fish, and you feed him for a lifetime.”.

See the discussion in Problem in crystal index to find two possible solutions.

Good luck!

If I am not mistaken, the repository on Gitlab is invite-only, so creating an account is not enough, one also has to request access.

If you don’t mind @Maxime_Toussaint I’m sharing your fix here for future reference: just change the type of the variable layer defined in src/management/gDataConversionUtilities.cc at line 828 from uint8_t to int32_t.

Thank you.

I try to do again, following your advice.