Optimization of image reconstruction and performance evaluation for a small-animal SPECT and PET imaging

Abstract Objective : In the field of small animal imaging, Versatile emission computed tomography/computed tomography (VECTor/CT) allows for fully simultaneous acquisition with SPECT and PET tracer using a clustered multi-pinhole collimator. We determined optimal image reconstruction parameter settings and assessed performance characteristics for 99mTc and 18F imaging. Methods : A line source phantom, a cylindrical phantom and a hot-rod phantom containing either 99mTc or 18F were used to perform quantitative analyses including the FWHM, uniformity, contrast and contrast-to-noise ratio and visual analysis. The number of subsets was set as 4, 8, 16 and 32 and the number of iterations was varied from 1 to 30. After the determination of the optimal reconstruction parameters, sensitivity and activity recovery coefficients was measured. Results : The optimal reconstruction parameters were 32 subsets and 8 iterations for 99mTc and 32 subsets and 17 iterations for 18F. Using optimal parameters, 1.1 mm and 1.3 mm hot-rod phantoms can be distinguished for 99mTc and 18F, respectively. In addition, sensitivity was 3606 cps/MBq for 99mTc and 2052 cps/MBq for 18F. The activity recovery coefficients of 99mTc and 18F were maintained above 1.0 for more than 2 mm diameter rod and more than 3 mm diameter rod, respectively. Conclusions : We conducted a performance evaluation with 3 phantoms while changing reconstruction parameters to determine optimal settings for 99mTc and 18F. When the optimal reconstruction parameters we found were used, it was able to visually distinguish small hot-rods diameter, compared with generally used parameter settings and then we could evaluate performance characteristics of the VECTor/CT equipped with the collimator newly designed for the size of a rat.