Quantitative accuracy of slow-rotating dynamic SPECT

We investigated biases in time-activity measurements relevant for quantitative dynamic SPECT/CT imaging when slow-rotating dual-headed gamma camera systems in combination with OSEM-3D (Flash3D) with scatter and attenuation correction are used. The goal was to assess the potential and also the limitations of clinical dual-headed SPECT/CT systems for the quantification of dynamic processes with focus on a renal time-activity function. We used simulations of a SPECT/CT system to estimate absolute quantitation errors in time-activity measurements. We systematically assessed dependencies of these errors on signal to noise ratio and sampling frequencies using a MAG-3 renal time-activity profile. In addition, a physical phantom was developed to measure dynamic processes on a clinical SPECT/CT system. The phantom consisted of a cylindrical chamber placed in a large cylinder phantom and connected to a programmable peristaltic pump. SPECT/CT acquisitions were performed by varying the rotation times of the SPECT system. Absolute activity concentrations were calculated by cross calibrating the imaging system with a well counter and using correction factors derived from simulations. Results from simulations show no significant differences in emission recovery coefficients within the range of 7.5 to 120 seconds per rotation. Phantom experiments using corrections from cross calibration and simulation show average estimation errors of -0.9% and -4.5% for 10 seconds and 60 seconds per rotation, respectively. Conclusion: We showed that quantitation of a renal dynamic process in phantoms using multiple time-contiguous SPECT acquisitions with 3D iterative reconstruction is possible with an accuracy of 4.5%.