I acquired list mode data through simulation using GATE (geant4 application tomography emission). However, although scatter correction (-sc) supported by CASToR was performed, ring artifacts were generated as shown in the image below, and the proper image did not come out.
I would appreciate it if you could let me know if there are any precautions or essential elements regarding the GATE simulation process or the use of CASToR.
Regarding the -sc option, it does not really provide a scatter correction estimation, but the rates as computed from the Monte-Carlo simulated data. Currently there is only the possibility to ignore scatter/random events in the resulting datafile using the -t option, otherwise the scatter correction factors estimation must be preformed and implemented by the user.
Did you try to reconstruct your data without scatter ? Ring artifacts can be caused by various issues. Also did you include attenuation correction in your data ?
The problem is that the images of items 1, 2, and 4 are all normally reconstructed, but the image of item 3 does not work normally, just like the picture in the text.
All four types of attenuation were applied using the -atn function provided by castor using the attenuation map.
Could there be other causes of ring artifacts even after attenuation correction?
Hi,
The use of -sc should give you a similar output than trues only, really. I am not sure what causes the issue except maybe if the acquisition duration information are not correctly recovered, this can lead to over/underestimation of scatter as these data are stored as rates (check if the values of the Start time (s) and Duration (s) fields are correct in your datafile header).
In any case, -sc does not provide a proper scatter correction. You can have a look at the following thread for an alternative approach to imitate scatter-corrected data:
