Dear CASToR users and developers,
I want to know whether the sensitivity image is fixed for a certain PET system and a determined field of view, for example, after I calculate the sensitivity image, and then I reconstruct various data based on this PET, can I directly use the previously calculated sensitivity image? I have always thought that it is only related to the structure of the system, and is a constant. If not, how can I save time when reconstructing different data on the same PET structure? It takes a lot of time to calculate the sensitivity image each time.
Thank you very much in advance for your help.
YuanHang
Hi Yuan Hang,
For list-mode data, the pre-computation of the sensitivity will depend on the system geometry and detector efficiencies, data attenuation, and reconstruction parameters. It will also depend whether a normalization datafile is provided by the user or not (see table 4 and last point of section 2.1 in the documentation ).
If your sensitivity image is generated from the system geometry (default behavior), setting the min angle difference tag to a value greater than 0 should substantially reduce the computation time considering your system is dual-head (sections 5.1/5.2 of the documentation).
Hope this helps,
Thibaut
Hi Thibaut,
Thanks for your reply, I tried setting the angle to a value greater than 0, but since our PET system has a large number of crystal layers, it still takes a lot of time. I checked the documentation for several factors related to the sensitivity image mentioned in your reply, my system structure and crystal detection efficiency should be unchanged, and the reconstruction parameters I have not changed, the only thing that has changed is the simulation data, but the way the data is obtained is the same, only the shape and size of the radioactive source are different, in this case my sensitivity image will change? Regarding “data attenuation”, I don’t quite understand what exactly it means.
Thank you very much in advance for your help.
Sincerely，
YuanHang
Hi Yuan Hang,
I mean attenuation in your object if it is included in your simulation, as attenuation correction factors are integrated in the sensitivity image computation.
If it is not the case, and nothing else changes regarding crystal efficiency and reconstruction parameters, I think you could reuse the same sensitivity image for the reconstruction of other datasets (-sens option for castor-recon).
Best regards,
Thibaut