NiftyPET

NiftyPET is an open-source, standalone software platform for high-throughput PET image reconstruction and analysis. It focuses on accurately modeling the acquisition processes to deliver high-quality quantitative imaging, particularly for scanners with large axial fields of view. The platform uses parallel computing within the Python package, allowing easy data access, manipulation, and visualization at every processing stage.

The NiftyPET pipeline starts with MR and raw PET input data. It consists of several processing stages, including list-mode data processing, attenuation coefficient map generation, detector normalization, precise forward and back projection between sinogram and image space, reduced-variance random events estimation, fully 3D scatter events estimation, voxel-based partial volume correction, and region- and voxel-level image analysis.

The platform demonstrates its advantages through an amyloid brain scan example, where all processing is performed within a single, uniform Python environment. High-accuracy acquisition modeling uses span-1 ray tracing for true, random, and scatter events. NiftyPET also provides uncertainty estimation for image-derived statistics, enabling robust tracking of subtle physiological changes in longitudinal studies.

NiftyPET also supports the development of new reconstruction and analysis algorithms by allowing users to restrict the axial field of view to any set of rings covering a region of interest, enabling faster fully 3D reconstruction and corrections using real data.