Influence of the angle of incidence on the sensitivity of gamma camera based PET

Thicker crystals have been used to increase the detection efficiency of gamma cameras for coincidence imaging. This results in a higher detection probability for oblique incidences than for perpendicular incidences. As the point sensitivity at different radial distances is composed of coincidences with different oblique incidences, the thickness of the crystal will have an effect on the sensitivity profiles. To correct this non-uniform sensitivity, a sensitivity map is needed which can be measured or calculated. For dual- or triple-head gamma camera based positron emission tomography (PET) a calculated sensitivity map is preferable because the radius and the head orientation often change between different acquisitions. First, these sensitivity maps are calculated for 2D acquisitions by assuming a linear relationship between the detection efficiency and the crystal thickness. The 2D approximation is reasonable for gamma cameras with a small axial acceptance angle. The results of the 2D approximation show a good agreement with the results of Monte Carlo simulations of different realistic gamma camera configurations. For dual-head gamma cameras the influence on the sensitivity profile is limited. Greater variation of the sensitivity profile is seen on three-headed gamma cameras and correction of this effect is necessary to obtain uniform reconstruction. To increase the sensitivity of gamma cameras, axial collimators with larger acceptance angles are used. To obtain a correct sensitivity for these cameras a sensitivity calculation in 3D is needed. For a fixed camera position the sensitivity is obtained by integrating the detection efficiency over the solid angle formed by the voxel and the intersection of the first detector with the projection of the second detector on the plane of the first detector. The geometric sensitivity is obtained by averaging this for all camera angles. The values obtained show a good agreement with the Monte Carlo simulations for different points in the field of view. Both 2D and 3D sensitivity profiles show the highest influence of the detector thickness on the radial profiles of the U-shape configuration. Taking the detector thickness into account also has an influence on the axial profiles. This influence is maximal in the centre where more oblique coincidences are present. This method is not limited to gamma camera based PET scanners but can be used to calculate the sensitivity of any PET camera with continuous detector blocks. (author)