# Project 2 - Synergistic PET and SPECT reconstruction Lead: Sam Porter Co-lead: Gemma Fardell ### Description This group will focus on synergistic reconstruction using PET/SPECT acquisition models. We will work through some basic notebooks on how to set up PET/SPECT reconstruction with SIRF before moving on to synergistic PET/SPECT reconstruction using CIL algorithms. ### Preparation #### Recommended videos: * [Fully3D 2021 - SIRF/CIL training school - 06 Optimisation based reconstruction with CIL](https://youtu.be/n2py8X_WdVE), **Jakob Sauer Jorgensen** * [Fully3D 2021 - SIRF/CIL training school - 08 Iterative image reconstruction - PET specific](https://youtu.be/an9Yjbusv58), **Andrew Reader** * [Fully3D 2021 - SIRF/CIL training school - 11 Synergistic reconstruction - approaches](https://youtu.be/Xicdg0WyekM), **Kris Thielemans** #### Recommended papers: * https://iopscience.iop.org/article/10.1088/0266-5611/31/1/015001/pdf (section 3.2) ### Notebooks to be tackled during the conference: * SPECT/SPECT_acquisition_model * SPECT/SPECT_OSEM * SPECT/SPECT_OSEM_measured_data * Synergistic/cil_joint_tv_PET * Synergistic/cil_joint_tv_PET_SPECT ### Misc: We have a 32 core CPU machine availavble. Please use __Training School for SIRF and CIL no GPU__ In order to make use of multi-threading please add the line ```acq_model_matrix.set_keep_all_views_in_cache(True)``` before calling ```acq_model = spect.AcquisitionModelUsingMatrix(acq_model_matrix) acq_model.set_up(templ_sino,uMap)``` ### Q&A # Back to main [Main page](https://hackmd.io/oNneaDUOQNK6XdDRmooITw?view#SIRF-training-school-PSMR-TBP-2022---Main)