[Evaluation of several commercial attenuation correction software programs installed by vendors for data processing: a simulation phantom study].

PURPOSE An attenuation correction is necessary to improve quantitative SPECT evaluation. The aim of this study was to evaluate the performance of several different types of commercial attenuation correction algorithms. METHODS We investigated the performance of four attenuation correction algorithms using a simulation phantom with and without scatter. The attenuation correction was performed on clinical SPECT workstations: GMS5500/PI (Toshiba), GMS7700/A (Toshiba), e.soft (Siemens), and Xeleris (GE). Then, the effect of the attenuation correction was evaluated using the count profile curve and attenuation coefficient that maintained uniformity of the radioisotope distribution in the uniform region of the images. RESULTS In the without-scatter condition, the count profile curve showed a similar shape on all workstations. The attenuation coefficients that maintained uniformity were 0.134, 0.133, 0.133, and 0.135 cm(-1) using GMS5500/PI, GMS7700/A, e.soft, and Xeleris, respectively. There was no significant difference among these workstations. With scatter, the attenuation coefficients differed by 0.109-0.121 cm(-1) with the types of attenuation correction algorithms. CONCLUSION Without scatter, the effects of the attenuation corrections were equivalent. However, with scatter, a few differences were observed in the effects of correction with several types of algorithms. Therefore, our results suggest that careful evaluation should be considered when different types of clinical SPECT workstations are used.