Impact of scatter correction in the kinetic analysis of a D2 receptor ligand SPECT study

Abstract

Scatter correction (SC) has been shown to be essential for the quantification of brain perfusion in SPECT. However, in the case using the tracer showing a high activity accumulation in a small region with only a small activity in other areas of brain, the contribution of SC has yet to be studied. In this work, the impact of SC on the quantitative kinetic parameters derived from SPECT studies using a dopamine D2 receptor ligand was investigated. SPECT data were acquired dynamically with serial arterial blood sampling following intravenous injection of I-123 iodobezofuran (IBF). SC was performed with the previously validated transmission dependent convolution subtraction (TDCS) technique. Images were reconstructed by OSEM including the attenuation correction (AC), and FBP with post AC by Chang's method. Time activity curves (TACs) were generated from the striatum (STR) and occipital lobe (OCC) regions, with and without SC. Quantitative values of binding potential (BP) were then compared for four kinetic models, including the 3 compartment model (3COM), linear graphic plot (GRP), reference tissue method (RTM), and multi-regression method (MLR). SC changed both TACs of STR and OCC, and introduced significant changes in kinetic parameters, which induced the significantly increased BP in 4 kinetic models, compared to those without SC. The change in BP by SC was greater with 3COM model, than the other models. SC caused significant change in the shape of TACs in both STR and OCC, resulting in increased BP with all four kinetic methods. Likewise for the brain perfusion study, SC is required for the dopamine D2 receptor ligand studies in SPECT.