Evaluation of SPECT imaging: a multi-center and multi-vendor phantom study

Abstract Background: Single-photon emission computed tomography (SPECT) is a nuclear medicine (NM) imaging method sensitive for the evaluation and quantification of physiological processes in patients. However, there are different factors that affect qualitatively and quantitatively the SPECT images and, consequently, the reliability of the data. There is a degree of variability underlying the intrinsic performance of systems from the same or different manufacturers. In addition, the different configurations of the gamma cameras (GC) and the acquisition and reconstruction protocols used may lead to heterogeneity of results. This study compared SPECT images acquired in three GC from three NM centers, using the Jaszczak and NEMA IEC Body phantoms, filled with homogeneous solution of technetium-99m. Acquisitions were performed in each GC, following the protocols used in clinical practice of the respective NM center, and the images were analyzed and compared in terms of spatial resolution, sensitivity, noise, contrast, and contrast recovery (CR). Subsequently, the processing and reconstruction parameters were harmonized to assess their impact on the standardization of results. Results: With the clinical protocols, the images acquired with the three GC showed small differences in spatial resolution, noise, and contrast, with the greatest difference being observed in the quantitative accuracy, since the CR has achieved variabilities of 0.17 and 0.28, for different activity concentrations. This intersystem variability was reduced to 0.09 and 0.08 after reconstruction algorithm standardization. However, this change increased the differences in the contrast percentage. Conclusion: GC from different manufacturers showed the greatest heterogeneity of results, however, by eliminating possible sources of variation it is possible to reduce the differences between systems. It was demonstrated the need to include different parameters for the evaluation and comparison of SPECT images, so that procedures can be established between the different NM centers to ensure a trade-off between the various parameters assessed here, as the improvement of some usually leads to the deterioration of others.