RESUMO
Attenuation correction is widely used in SPECT/CT (Single Photon Emission Computed Tomography) procedures, especially for imaging of the thorax region. Different compensation methods have been developed and introduced into clinical practice. Most of them use attenuation maps obtained using transmission scanning systems. However, this gives extra dose of radiation to the patient. The purpose of this study was to identify when attenuation correction is really important during SPECT/CT procedures.For this purpose, we used Jaszczak phantom and phantom with three line sources, filled with technetium ((99m)-Tc), with scattering materials, like air, water and acrylic, in different detectors configurations. In all images acquired were applied analytic and iterative reconstruction algorithms; the last one with or without attenuation correction. We analyzed parameters such as eccentricity, contrast and spatial resolution in the images.The best reconstruction algorithm on average was iterative, for images with 128 × 128 and 64 × 64 matrixes. The analytical algorithm was effective only to improve eccentricity in 64 × 64 matrix and matrix in contrast 128 × 128 with low statistics. Turning to the clinical routine examinations, on average, for 128 × 128 matrix and low statistics counting, the best algorithm was the iterative, without attenuation correction,improving in 150% the three parameters analyzed and, for the same matrix size, but with high statistical counting, iterative algorithm with attenuation correction was 25% better than that without correction. We can conclude that using the iterative algorithm with attenuation correction in the water, and its extra dose given, is not justified for the procedures of low statistic counting, being relevant only if the intention is to prioritize contrast in acquisitions with high statistic counting.
