Abstract
Accurate, absolute quantification of activity in SPECT depends on the counting sensitivity of the gamma camera collimator system and on the contribution of the tail of the point spread function (PSF) that contains both scattered radiation and septal penetration. These factors were studied for the radionuclides 99mTc, 131I and 123I on different cameras and collimators. The ability of the geometric mean line source scatter function subtraction technique to correct for the effect of the tail of the PSF was investigated in a computer simulation study with 99mTc and a physical phantom study with 123I. The sensitivities of 99mTc collimators were relatively constant with distance in air whereas those of 123I and 131I decreased considerably. For these radionuclides the contribution of the PSF tail was also increased as compared to 99mTc. The geometric mean line source scatter function subtraction technique was confirmed as being able to correct effectively for this factor. Collimator design affects both the fall off of sensitivity in air and the tail of the PSF. Attention to correction of these factors enables accurate quantification.