[Gate-users] Greater number of delayed coincidences than random coincidences for PET data

[Twyman, Robert]

Dear GATE users,

I have been simulating PET acquisitions using GATE on an XCAT phantom for reconstruction using STIR. The reconstruction processes is iterative and therefore I have need to estimate the contributions due to Random and Scattered events in the measured data. I am estimating the Randoms from the Delayed coincidence data output by GATE using a 200ns Delayed offset. Both the Coincidence and Delayed windows are set to 4.9ns.

Please see my digitiser on GitHub: https://github.com/UCL/STIR-GATE-Connection/blob/master/ExampleScanners/D690/digitiser_D690https://github.com/UCL/STIR-GATE-Connection/blob/master/ExampleScanners/D690/digitiser_D690.mac#L39-L54
.mac#L39-L54

The problem is as follows. I unlist (using STIR) all valid events in the Delayed ROOT file. I unlist all random events (defined as the photons having different eventIDs) from the Coincidence ROOT file. The result is, ~10% more events in the Delayed data, which has knock on effects to my Randoms estimation and into the Scatter estimation…

Does anyone have any idea why I might have more events in my Delayed data than Random events in my Coincidence data?

Regards,
Robert Twyman
Ph.D Student
UCL

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opengatecollaboration.org/pipermail/gate-users/attachments/20210413/e90b52f8/attachment.html>