Print

Print


Hi Congying,
I've been a little separated from the coding on SIRF over the last few months, but I don't believe the MHKEM has been incorporated into SIRF. The development required would not be particularly huge (I think?). Although it would probably require some C++ level changes in SIRF to support the vector of anatomical images.
However, it is all up and running in the STIR engine itself. I'd suggest you have a go with STIR first.

Depending how familiar you are with STIR, I'd suggest first making sure you are able and comfortable to do a regular OSEM reconstruction. Once you're happy there, it is very simple to try [M]HKEM, see the example parameter file Daniel wrote here:
https://github.com/UCL/STIR/blob/master/examples/samples/KOSMAPOSL_mhkem.par

If you're not so familiar with STIR, the User Guide is best to start. (Kris - the version on sourceforge is old, is there an updated version? I've attached my most recently compiled version, which is a few weeks old just before 4.0)
You could use some of the publicly available information from the private Short Course run at MIC in 2018:
http://stir.sourceforge.net/MIC2018ShortCourse/main.htm
https://github.com/UCL/STIR-exercises

Converting your own data from the scanner can be a bit tricky, here's a starting point:
https://github.com/SyneRBI/SIRF/wiki/PET-raw-data

Cheers,
Ash

--

Ashley Gillman
Postdoctoral Research Fellow | CSIRO AEHRC
E [log in to unmask]<mailto:[log in to unmask]> | T +617 4753 8531 | M +61 404 085 204

From: SyneRBI Users list <[log in to unmask]> On Behalf Of Cy Chu
Sent: Monday, 11 May 2020 8:19 PM
To: [log in to unmask]
Subject: About the MHKEM

Dear SIRF Team,

Recently, I have noticed that the HKEM method available in SIRF seems to be updated to MHKEM which could work with multiple anatomical priors simultaneously (https://github.com/UCL/STIR/pull/424/). I am wondering whether this MHKEM approach has become available in STIR or SIRF? If the new approach is available now, is it possible to find an example of using it? I really want to try this kernel-based method with our mMR PET data.

Thank you very much.

With best regards,
Congying Chu



------------------------------------------------------------------------------------------------
------------------------------------------------------------------------------------------------
Forschungszentrum Juelich GmbH
52425 Juelich
Sitz der Gesellschaft: Juelich
Eingetragen im Handelsregister des Amtsgerichts Dueren Nr. HR B 3498
Vorsitzender des Aufsichtsrats: MinDir Volker Rieke
Geschaeftsfuehrung: Prof. Dr.-Ing. Wolfgang Marquardt (Vorsitzender),
Karsten Beneke (stellv. Vorsitzender), Prof. Dr.-Ing. Harald Bolt
------------------------------------------------------------------------------------------------
------------------------------------------------------------------------------------------------

________________________________

To unsubscribe from the SYNERBI-USERS list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=SYNERBI-USERS&A=1

########################################################################

To unsubscribe from the SYNERBI-USERS list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=SYNERBI-USERS&A=1