[gate-users] RE: STIR reconstruction software (GATE simualtions)

[Claude Comtat]

Dear Vandana and Kris,

just to confirm that Kris is correct.

> 
> My understanding is that as long as you are simulating a CTI/Siemens PET
> scanner that actually exists and is supported by STIR (which essentially
> means one that outputs ECAT7 data, so no PET/CT), then you don't have to do
> anything specific. 
> 
> One of the problems with ECAT7 output is that it doesn't contain all scanner
> dimensions. Instead, it contains a system_type field. If GATE fills this in
> correctly, then STIR can figure out the dimensions and off you go.

You need the following command in GATE to specify our system ID:

/gate/output/ecat7/system XXX


> An alternative way would be to let GATE write Interfile headers that STIR
> can understand. This shouldn't be incredibly difficult, but I don't know of
> anybody who has plans in that direction.

For now, the sinogram outputs (raw or ecat7) are only defined for ECAT
systems as the LOR-to-SinogramBIN conversion follows the ECAT
specifications. It means, in particular, that the sinogram radial
sampling is not uniform (arc effect) and requires reconstruction
routines that handle this specific feature. If we can keep these ECAT
specifications, it is quite straightforward to convert an ECAT7 sinogram
to an Interfile sinogram (remove the binary headers and directory
matrices and write a separate ascii header file, plus byte swapping for
the raw data if needed).

An alternative, valid for all PET systems, would be to have an "exact"
rebining from the LOR to the sinogram (or projection ) space (calculate
for each LOR the four projection coordinates --- radial and axial
positions ; azimuthal and co-polar angles --- and then fill the
corresponding 4D sinogram or projection bin). As they are different ways
to define the projections coordinates (co-polar angle or its tangential
?), the reconstruction routines should follows the same definition ...

Kris, do you think that such an output would be useful for STIR ? If
yes, a correct geometrical normalization of the data will also be needed.

All the best,


Claude


-- 
Claude Comtat					comtat at ieee.org
Image Acquisition and Processing group		tel: +33 1 69 86 78 01
Frédéric Joliot Hospital Facility		fax: +33 1 69 86 77 49
French Atomic Energy Commission 		www-dsv.cea.fr/shfj