[Gate-users] RE: Reconstruction for Philips Allegro

[Kris Thielemans]

Hi
> I have looked at STIR but I don't believe STIR 
> has profiles for Philips scanners 

People have reported using STIR for the Allegro. However, as far as I know
this was for Philips sinograms, not for listmode data. There's an issue with
the Allegro that it has a large gap between its detectors. Philips handles
this by creating a "geometric" sinogram where 1 LOR corresponds to the gap.
(Siemens does the same). I am not sure how the Philips listmode data works,
or even how you handle this in GATE Root/LMF files. On the other hand, I
believe that Christian Morel's LMF-reader code for STIR takes a similar
viewpoint that it just uses the x,y,z coordinates of the detected events,
and puts those into sinograms. With a bit of luck, this would work for the
Allegro as well.

Of course, even better would be to include the gaps in the reconstruction
model, but STIR doesn't do that yet (not very difficult to do, but it's way
down my own priority list).


> nor does it implement the 
> RAMLA algorithm.  

true, it doesn't, but why are you so interested in it? Do some OSEM, and
then finish of with a bit of MLEM. Next STIR release will have OSSPS (maybe
by the end of the year).

Kris