[Gate-users] order of events in a coincidence

[Marcin Balcerzyk]

Dear Gaters.

I looked at first and second event of coincidence and wonder how the event
is determined as first or second. It looks that it is the same as in singles
sequence, i.e. the first single forming a coincidence is a first event of
coincidence, second single, forms second event of coincidence. 

However I cannot find an order for singles. Definitely they are listed in
ASCII files by the order of eventID. But then what makes an order? For true
coincidences, eventID is the same. I checked timestamp, but the order is not
by that, albeit there are more events of the correct timestamp sequence in
singles file then reversed ones. It is neither energy deposited, nor
volumeID, nor scattering events. Definitely it is not random (or symmetric),
as the graphs show.

It seems secondary, but I discovered that when the coincidences are 511 vs.
secondary prompt event of higher energy (other gamma), the asymmetry becomes
much more pronounced, and for me disturbing.  For whatever reason in a
symmetric PET scanner and source precisely in the center, the second event
in coincidence carries more energy on average. There is about twice the
amount of events, where the second part of coincidence carries more energy
than the first one. It looks the effect does not have anything to do with
positronium. My digitizer module is perfectly symmetric.

If you do not know, maybe you can point at C++ file which is responsible for
ordering singles? Or maybe it is some GEANT4 artifact? I remember a
discussion on this forum about the energy spectra dependency on Gate
version.

I use Gate 3.0.0 and corresponding software.

Marcin Balcerzyk
Instituto Pluridisciplinar
Universidad Compultense de Madrid
Paseo Juan XXIII n. 1
28040 Madrid, Spain
tel. +34 91 3943261
fax. +34 91 3943266
movil +34 697 322 126
Skype: balcerzm