SIRF-Contribs

Users contributions to SIRF.

This space is for contributions that are not quite ready yet to be incorporated into SIRF's main repository. We welcome the sharing of any contributions that might help other researchers. We hope that widely used features can then be finalised, and then added to the main SIRF repository.

Please add your contributions as a pull request.

• Jupyter notebooks can be added to ./src/notebooks.
• Python code should be added to ./src/Python/sirf/contrib with an __init__.py file. In this fashion, once a user has checked out this repository, they can use it in their code with, for example:

from sirf.contrib import kcl

Installation

pip install git+https://github.com/SyneRBI/SIRF-Contribs

Current contents

• Dynamic PET simulation - batch generation of simulated dynamic PET data using SIRF/STIR.
• KCL priors - de Pierro, Bowsher, etc. priors
• brainweb - preliminary script to create brainweb data with some extra features.
• MCIR - scripts for the MR and PET MCIR reconstructions.
• Grappa_and_CIL - notebook demonstrating CIL integration with SIRF, with the use case of a GRAPPA reconstruction of MR data.
• example functions and notebooks for MAP optimisation with PET (or SPECT)
  • BSREM implementation and notebook using a preconditioned gradient ascent algorithm.
  • LBFGS-B-PC implementation and notebook using a preconditioned LBFGS-B algorithm (calling SciPy).
  • MaGeZ ALG1 implementation, the winner of the first PETRIC