Michael Ljungberg

Images created using a standard collimator-based scintillation camera are essentially 2D images, lacking information regarding the source depth, since the value in a particular pixel in the image represents detection of photons along a line determined by the collimator. However, it is possible to reconstruct a set of 2D images that together form a 3D image of the underlying activity distribution from data acquired of the same source distribution at different projection angles around the object. This chapter will describe the way in which these data can be used to reconstruct transversal images by the filtered back-projection (FBP) method as well as by iterative algorithms, and also how noise regularization can be implemented. Various physical factors that affect the reconstructed images will also be discussed. If we combine a SPECT system with a CT system and display the images from both as a single hybrid image, additional useful information can be obtained. The CT information can also be used to correct for non-homogeneous attenuation, scatter, and partial-volume effects. Finally, the chapter will discuss how the combination of quantitative SPECT images registered to CT images can be used for dosimetry calculations.