Development of a Phantom for Evaluating Image Quality and Partial-Volume Effects in Hot and Cold Regions in Small-Animal SPECT and PET.

Abstract

The partial-volume effect (PVE) is a major factor affecting quantitative accuracy in small-animal SPECT and PET imaging, leading to underestimation of activity in hot-rod images and overestimation in cold-rod images. However, a phantom capable of assessing image quality, including the PVE in both hot and cold regions, has not been available. This study aimed to determine whether a newly developed phantom can accurately evaluate image quality, including the PVE of hot- and cold-rod images, in small-animal SPECT and PET systems. Methods: The phantom consisted of a cylindric pool and rod sections. The rod section included 6 hot rods and 6 cold rods with diameters of 1-6 mm. The rod and pool sections were filled with solutions containing ^99mTc and ¹⁸F, respectively. In the hot-rod section, the radioactive concentration ratio between rods and background was set to 4:1. Images were acquired for 30 min and reconstructed with a voxel size of 0.8 mm using a pixel-based ordered-subset expectation maximization algorithm. Image contrast, absolute and inverse recovery coefficients (RCs) (indicators of the PVE), and percent coefficient of variation (%CV) and signal-to-noise ratio (indicators of uniformity) were evaluated for both hot and cold rods. Results: The contrast of the hot rod for ^99mTc and ¹⁸F showed lower values with decreasing rod diameters. Furthermore, the ^99mTc image demonstrated a higher contrast than the ¹⁸F image and approached the true contrast. The cold-rod contrasts with ^99mTc and ¹⁸F followed a similar trend. The RCs of ^99mTc and ¹⁸F were lower as rod diameters decreased. The radioactive concentration in the 40-mm pool section was similar to that of the 6-mm hot rod for both radionuclides. The inverse RC was lower with a decreasing cold-rod diameter. Moreover, cold-rod images with ¹⁸F demonstrated lower inverse RCs than with ^99mTc. The %CVs for the ^99mTc and ¹⁸F images were 3.49% ± 0.03% and 6.29% ± 0.07%, respectively, with the ^99mTc image displaying a lower %CV compared with the ¹⁸F image. Conclusion: We developed a phantom that allows the evaluation of physical phenomena in small-animal SPECT and PET images, providing reliable assessment of image contrast, PVE, and uniformity in both hot and cold rods.