Abstract
The problem of using 360' or 180' data sampling techniques in transaxial SPECT imaging is still to be solved. A theoretical point source study for an ideal response detector has shown, for objects which are close enough to the origin of the reconstructed area, that there are significant differences between sections obtained by different sampling methods. A computer simulation study of line sources in a homogeneous attenuated medium has confirmed the results of clinical studies in which significant image distortion has been observed in 180' sections but not in 360' reconstructed images.