SPECT/CT, PET/CT and PET/MR Principles

Abstract

After a short introduction, the general principles, features and limitations of the tomography from projections that form the core of SPECT, PET and CT modalities are reviewed. Mathematical concepts that rule the final image quality, such as ill-posed problem, noise correlation, streak artefacts, iterative algorithm or regularization methods, are explained using simple examples accessible to the non-physicist. Specificities of each modality (CT, MRI, SPECT and PET) are then presented and analyzed. In particular the use of anatomical information provided by CT and MRI is described in the case of hybrid systems.