量研学術機関リポジトリ「QST-Repository」は、国立研究開発法人 量子科学技術研究開発機構に所属する職員等が生み出した学術成果（学会誌発表論文、学会発表、研究開発報告書、特許等）を集積しインターネット上で広く公開するサービスです。 Welcome to QST-Repository where we accumulates and discloses the academic research results(Journal Publications, Conference presentation, Research and Development Report, Patent, etc.) of the members of National Institutes for Quantum Science and Technology.
Thank you very much for using our website. On the 11th of March 2019, this site was moved from our own network server to the JAIRO Cloud network server. If you previously bookmarked this site, that bookmark will no longer work. We would be grateful if you could bookmark the website again. Thank you very much for your understanding and cooperation.
[Purpose] Monte Carlo simulation of PET scanners are widely used for performance assessment prior to the development of an actual system as well as detailed analysis of physical processes occurring in the scanners. However, it is very difficult to model application specific PET scanners with more flexible and complex detector arrangement then the conventional cylindrical PET using existing toolkits. In this study, therefore, we developed a software framework that can model arbitrary arranged PET detectors to easily simulate PET data acquisition and image reconstruction for the system with simple configuration files.
[Methods] We developed a Monte Carlo PET simulator and an image reconstruction software, both of which take common configuration file for geometrical arrangement of detectors to model a specific scanner. The detectors can be arranged not only on a cylindrical surface but also as individual detectors with arbitrary positions and orientations. We implemented the Monte Carlo PET simulator using Geant4 toolkit because it is well validated for PET applications. Input data for modelling measurement target are voxel data indicating material distribution and radioactivity distribution. Simulated measurement data are recorded as coincidence list-mode data including random coincidence due to coincidence time window and considering energy resolution, detector dead time. We implemented the image reconstruction software as the ordinary-Poisson list-mode ordered subset expectation maximization algorithm with sensitivity, attenuation, scatter, and random corrections.
[Results] Using the developed framework, we simulated our helmet-neck PET prototype to assess the event component in the measured data and to optimize the reconstruction parameters. As a result, we found that the scatter fraction of the pool phantom used for normalization was 29%, and we could adjust sampling parameters to better fit the scatter distribution in the simulation. The image reconstruction software could reconstruct images with the same configuration file as the simulator.
[Conclusion] We developed an easy-to-use imaging simulation framework with a Monte Carlo simulator and image reconstruction software for PET scanners with arbitrary arranged detectors.
Development of an Imaging Simulation Framework Enabling Modelling of PET Scanners with Arbitrary Detector Arrangement
