[gate-users] STIR code for handling LMF data

[Kris Thielemans]

Hi Long
>
> I remember that we have disscussed lmf file. You'd like to do ML
> reconstruction
> directly from List Mode data, isn't it?
>
sure. the code almost works.

> While I need a Interfile interface. So I moved to GATE to build
> the interfile
> interface between STIR and Gate (GATE has LMF lib support at that
> time), and has
> more information about the geometry. And hence comes out the
> (logical ring,
> detector number) and span, mashing, michelogram and interfile,
> and we do not
> conflict at all.
>
> It seems that you won't like to convert the lmf into interfile,
> so please give me
> further information on how to convert it, I am very curiouse
> toward this. :-)

hang on.

STIR has lm_to_projdata, which converts list mode data to sinograms, which
are output at present as interfile, but that will change once we have the
relevant code for switching output file formats for sinograms.

lm_to_projdata works for any data format that can be read by
stir::CListModeData. So, what I sent you were some classes to get this to
work with the LMF format. (You could support root data if you wanted of
course).

GATE might have more info about the geometry of course, but then again, it
probably writes that in the lmf data anyway (or in root). If not, it's lost.

So, as far as writing code that reads GATE data and outputs Interfile, it's
largely a matter of choice/convenience. However, if you want STIR list mode
reconstructions to work on GATE data, then there is no choice: you need to
work via stir::CListModeData.

I hope this clears this up.

Kris