[Gate-users] delayed event

Eunsin Lee eunsin at mail.med.upenn.edu
Wed Jul 7 15:45:55 CEST 2010


Thank you so much!!

Eunsin
******************************************************
Eunsin Lee
Postdoctoral Research Associate
PET Center, Department of Radiology
University of Pennsylvania School of Medicine
eunsin at mail.med.upenn.edu
(215) 349-5751 (office)
(979) 574-7235 (cell)
******************************************************




On Jul 7, 2010, at 7:13 AM, Marcin Balcerzyk wrote:

> Dear Eunsin.
>  
> The best way for me is to make a short (and low event number) simulation and enable ascii output. Users Manual describes columns headings and meaning depending on your geometry (by no means easy numbering scheme). One of the principal characteristics is, as you write, eventID. If for a detected coincidence eventIDs are not the same, and timestamp difference fits within your coincidence window time, that means that they are random – that is they are detected in the same coincidence window time (usually a few ns). For delayed events you have to define new coincidenceSorter. In benchmarkPET in digitizer.mac (Gate 5) there are lines to define it:
>  
> /gate/digitizer/name delay
> /gate/digitizer/insert coincidenceSorter
> /gate/digitizer/delay/setWindow 120. ns
> /gate/digitizer/delay/setOffset 500. ns
> /gate/digitizer/delay/MultiplesPolicy  takeWinnerOfGoods
>  
> Then looking at timestamps of each single forming coincidence (best in ascii output initially), you would see that for eventID different in the case above for more than 500 ns, the coincidences would be delayed. I think setOffset value cannot be less than 100.0 ns.
>  
> To analyze larger files, I would look at benchmarkPET.C and see how random events are counted, and add a few lines of code for root to analyze delays. If you have already run the benchmarkPET.mac, you have the delayed coincidences simulated in benchmarkPET.root fiel and can have a look how many are there. I think that delay coincidence sorter would contain only delayed events.
>  
> Kind regards
>  
> Marcin Balcerzyk
> Universidad of Sevilla, Spain.
>  
>  
> Message: 1
> Date: Tue, 6 Jul 2010 13:53:29 -0400
> From: Eunsin Lee <eunsin at mail.med.upenn.edu>
> Subject: [Gate-users] delayed event
> To: gate-users at lists.opengatecollaboration.org
> Message-ID: <AB460B71-F675-49C3-9CF2-39FC719CE489 at mail.med.upenn.edu>
> Content-Type: text/plain; charset=us-ascii
>  
> Dear GATE users,
>  
> In Gate root output there is a "delay" tree. Are these events are delayed events?
>  
> Also in "Coincidences" tree, there are some events having different eventID numbers, which is defined as random coincidence event in Gate. Aren't these delayed events either?
>  
> Basically I want to tell prompt events from delayed events. How can I do this in GATE?
>  
> Thanks so much in advance,
> Eunsin
>  
>  



More information about the Gate-users mailing list