PT - JOURNAL ARTICLE AU - Arda Konik AU - Navid Zeraatkar AU - Kesava Kalluri AU - Benjamin Auer AU - Lars R. Furenlid AU - Michael A. King TI - Improved Quantitative DAT Imaging with Multiplexing in a Multi-pinhole SPECT System DP - 2019 May 01 TA - Journal of Nuclear Medicine PG - 524--524 VI - 60 IP - supplement 1 4099 - http://jnm.snmjournals.org/content/60/supplement_1/524.short 4100 - http://jnm.snmjournals.org/content/60/supplement_1/524.full SO - J Nucl Med2019 May 01; 60 AB - 524Objectives: we have designed a multi-pinhole (MPH) collimator to improve the performance of the dopamine transporter (DAT) imaging by replacing one of the existing collimators on a dual-head SPECT system. MPH collimator focuses on the striatum and hence offers a better sensitivity/resolution trade-off than the parallel or fan-beam collimators within this clinically most relevant region. Our original MPH design consisted of 9 pinholes (PH) with minimal multiplexing [1, 2]. In this simulation study, we aim to obtain further improvements in the performance of the MPH by increasing the number of pinholes, hence enhancing the angular/axial sampling and sensitivity at the expense of increased multiplexing. Methods: we performed analytic simulations of 9, 13, and 16 PH configurations for aperture diameters of 4, 5, and 6 mm using modified XCAT brain phantoms [3] with standard and 10% larger size (in each dimension) striata on a “lumpy” background (to approximate the variation in uptake) modeling DAT imaging. The standard and large size striata represent the 50th and 99th percentile of the population [4], respectively. MPH configurations studied here were narrowed down based on our previous studies [5, 6]. The simulations modeled the object attenuation (no scatter), effective collimator resolution/sensitivity, and intrinsic resolution of 3 mm. For all cases, we avoided Str/Str multiplexing. While 9-PH design had only a minimal Bkg/Bkg multiplexing (~3%), 13- and 16-PH ones had a Bkg/Str multiplexing of 50% and 30%, respectively. The respective sensitivities obtained from the attenuated projections were 1:1.4:1.6 for the MPH collimators with 9, 13 and 16 PHs. Five noise realizations were obtained for each case and reconstructed with attenuation correction. The normalized root mean square error (NRMSE) was calculated within a spherical volume of 10 cm diameter including the striatum. Results: our current results are based on the projections that were acquired from 8 projection views over 360º and using a Str:Bkg contrast ratio of 8:1. At this count level (1.6 M counts for 9PH-4mm case), both 13 and 16 PH provided significantly better quantitative accuracy than the original 9 PH design. Also, 4 mm apertures resulted in highest NRMSE for all cases. The lowest NRMSE was obtained using 16-ph configuration using 5 mm aperture size. Conclusions: We determined through a series of simulations that additional pinholes can further improve the performance of our MPH system despite increased multiplexing in the projections. Based on our current results, 16 pinholes with 5 mm diameter aperture size appears to perform the best in terms of NRMSE. We will continue our investigation for different contrast ratios and different number of projection views (for equal time acquisition). We will also investigate altering pinhole positions slightly to improve the angular and axial sampling, which is especially more important when we use limited number of projection views. Finally, we will compare the results of these studies with the conventional systems.