Black results from CASToR reconstruction

General Discussions
1

Hi everyone,

I am having problems with the resulting reconstruction images from CASToR.

I am working with a Pet scanner simulation with a NEMA IEC phantom in GATE. I previously submitted a question in the forum related to the normalization computation file in CASToR. In the meantime, I was able to voxelize the NEMA IEC phantom I had defined to provide it to the castor-norm as the true image.

I successfully converted the root file from the simulation to a castor histogram file. I provided every necessary file to castor-norm. However, I obtain the following message:

sScannerManager::Initialize() → From scanner GED690
oProjectorManager::Initialize() → Initialize projectors and projection lines
castor-norm() → Compute normalization factors for 95544576 events
castor-norm() → 2438 out of 95544576 (2.551689e-03%) normalization factors correctly computed.
Please be aware that a low percentage may result in poor reconstruction quality!

I decided to try anyway to do the reconstruction providing the resulting normalization file and attenuation file but I obtained only black images.

Then, I decided to do just the reconstruction the normal way without the normalization or attenuation files. I used:

castor-recon -df SIM_NEMA_df.Cdh -it 8:8 -dim 102,79,74 -vox 3,3,3 -fout ReconstructedImageNema -vb 2

But still, I obtain a black image… I confirmed with a GATE dose test (without association with the scanner) if I had placed the source image in the right position, and the resulting image from this test confirmed that everything was placed correctly. I also checked that the activity was doing its thing right with the visualization qt.

I really don’t understand why when reconstructing the root file from the pet scanner simulation, CASToR can’t reconstruct it right. I uploaded here the voxelized image from NEMA IEC phantom I am using.

attenuationedited.hdr (1.0 KB)
iec_3mmteste2.raw (582.3 KB)
iec_3mmteste2edited.hdr (971 Bytes)

I would really appreciate if anyone could help me understanding what could be the cause for the no results with CASToR.

Thank you!

Greetings,
Beatriz

2

Hi Beatriz,

Indeed you should get some results even without normalization.

If you could send me the root/mac files of the simulation, I can have a look at the end of next week.

Best,
Thibaut

3

Hi Thibault,

I would really appreciate it if you could have a look at the root file. I checked it and it seemed fine to me, but something may be missing that I don’t know. I’ll send everything by email.

Thank you.

Greetings,
Beatriz

4

Hi Beatriz,
The piece of data is mostly composed of random coincidences, you can see it from CASToR conversion script log (below), or by checking EventIDs of photons directly from the ROOT datafile. You’ll see most of them are different.

image
image779×307 30.7 KB

I believe something must be wrong in the simulation.

Best,
Thibaut

5

Hi,

Thank you for checking!

Do you perhaps know where the problem could come from in the simulation? Could it be related with the phantom itself or the physics?

6

Hi,
Nothing looks wrong in your system from a first (quick) look.

I would advice you try some basic simulations first (e.g point source) to check if the problem remains, so that you could know if this comes from the system/settings or the source/phantom.

1 Like
7

Okay, Thank you so much for all the help! :slight_smile: