Comparative studies of collimator performance in DaTscan (Ioflupane I-123) striatal SPECT

Purpose: To determine optimal collimator and gamma camera combination for Ioflupane I-123 (DaTscan) striatal SPECT. Methods: Anthropomorphic basal ganglia phantom was used. The striatal chambers (caudate and putamen chambers) and the large chamber simulating nonspecific background activity in the remainder of the brain were filled with I-123 with the specific activity ratio 7.7. SPECT data were acquired using triple-head gamma camera (THGC) with fan-beam low-energy ultra high-resolution (LEUR) collimators, with dual-head gamma camera (DHGC) with parallel-beam lowenergy high-resolution (LEHR) collimators and medium-energy general-purpose (MEGP) collimators. Data were acquired at 159 keV with a 20% window, with I-123 and Tc-99m flood table for THGC and DHGC, respectively. The images were reconstructed using the OSEM algorithm with resolution modeling and uniform attenuation correction, and Butterworth postfilter, 5th order and 0.64, 0.78 and 1.0 Ny for LEUR, LEHR, and MEGP, respectively. The filter parameters were chosen to optimize the balance between image noise and spatial resolution. Results: The best image quality in terms of spatial resolution and contrast was obtained with fan-beam LEUR/THGC. The MEGP/DHBC produced images with better contrast-to-noise ratio than LEHR/DHGC. The measured ratio of mean activity in striatal chambers to the remainder of the brain was comparable for all three collimator/camera combinations. Conclusions: Based on phantom DaTscan striatal SPECT, the THGC with fan-beam LEUR collimators is preferable. If DHGC is used MEGP collimators provide better image quality, as compared to LEHR. More studies including patient studies are needed to confirm best collimator/camera combination.