Reconstruction of PET data acquired with the BrainPET using STIR

The Software for Tomographic Image Reconstruction (STIR) is an open source framework, which has several implemented reconstruction algorithms [1] and it is often considered a standard for reconstruction. A special challenge for reconstructing images with STIR occurs in the case of data acquired in positron emission tomography (PET) systems with gaps. The BrainPET is the PET component of the hybrid scanner 3TMR-BrainPET developed by Siemens for human brain MR-PET [2], [3]. The BrainPET has axial as well as transaxial gaps between its detector modules. This paper describes how data acquired with the BrainPET was reconstructed with STIR. The method to handle the gaps in the STIR reconstruction framework is described. Results of reconstructed images with acquired data are shown.