-
Notifications
You must be signed in to change notification settings - Fork 29
Examples (PET)
The SIRF distribution comes with a set of demo scripts. Those covering SIRF PET functionality are located in folders
SIRF/examples/Matlab/PET
SIRF/examples/Python/PET
Demos of this group help basic users to get started with SIRF: they show how to use SIRF with minimal user's effort.
osem_reconstruction.m and osem_reconstruction.py
illustrate PET reconstruction by Ordered Subsets Expectation Minimization (OSEM) algorithm. They actually use the OSMAPOSL, an Ordered Subsets (OS) version of the One Step Late algorithm (OSL) from Green et al for Maximum a Posteriori (MAP) maximisation, see http://stir.sourceforge.net/documentation/doxy/html/classstir_1_1OSMAPOSLReconstruction.html. Here it is applied to Maximum Likelihood (ML) objective function, in which case it is equivalent to OSEM.
ossps_reconstruction.py
illustrates PET reconstruction by Ordered Subsets Separable Paraboloidal Surrogate reconstruction algorithm, see
http://stir.sourceforge.net/documentation/doxy/html/classstir_1_1OSSPSReconstruction.html.
Demos of this group are for users that are after greater involvement with the reconstruction process.
steepest_ascent.m and steepest_ascent.py
show how to maximize SIRF PET objective function by user's or third-party optimization algorithm.
user_osmaposl.m and user_osmaposl.py
show how to use SIRF functionality in implementation of user's own reconstruction algorithm, using OSMAPOSL algorithm as an example.
Demos of this group show how to handle basic data objects involved in PET reconstruction.
using_acquisition_data.m and using_acquisition_data.py
show how to handle PET acquisition data using SIRF.
using_acquisition_model.m and using_acquisition_model.py
show how to use PET acquisition models for generating simulated acquisition data and backprojecting it to the image space (see user_osmaposl demos for the illustration on how this functionality is used by the iterative reconstruction algorithms).
(currently only for Python)
These are intended to be run from a Python IDE such as Spyder. Please check https://github.com/CCPPETMR/SIRF/blob/master/examples/Python/PET/interactive/README.md