Objectives: We designed a multi-pinhole (MPH) collimator for clinical brain SPECT imaging, which could potentially be used with the recently reported Tau-SPECT tracers (H. Watanabe et al, 2016). In this design two identical MPH collimators replace the existing collimators on a general-purpose-dual-head SPECT system. The pinholes are arranged at different axial positions to improve sampling. In addition, the collimator on one head is rotated 180 degree (about the axis normal to the collimator plane) with respect to the other one to further increase the axial sampling. Method: We determined the geometry of the system (e.g., number of pinholes, aperture size, aperture positions, aperture-center-of-field-of view distance, and aperture-detector distance) by maximizing system sensitivity for a target ellipsoidal (principal semi-axes, a:10.5, b:10.5, c:9 cm) volume-of-interest (VOI) covering 99-percentile head size and a target system resolution of 7 mm at the focal point on the axis-of-rotation. Constraints in optimization were fitting around the 99-percentile shoulder width of patients during rotation, minimum radius of rotation for brain, and <30% of multiplexing. We performed analytic and GATE Monte Carlo simulations of this system for uniform activity within the VOI, and clinical distributions within the XCAT brain phantom.

Results: Our preliminary results showed that 6-pinhole version offers the highest sensitivity. However, better axial sampling is obtained with the 7-pinhole version, which offers the second highest sensitivity (0.008%) and slightly lower multiplexing (24%). We altered the pinhole positions on the 7-pinhole version to further improve the axial sampling, which also resulted in further reduction in multiplexing (18%). We will present the results for the optimization algorithm over a range of numbers of pinholes and simulation results obtained from selected collimator designs. Conclusions: We designed and simulated an MPH system for general brain SPECT imaging. In future work we will assess through simulations the performance of the selected designs for quantitative measures such as striatal binding ratio (SBR) for Parkinson’s Disease imaging and cortical to cerebellum ratio (CCR) in Tau imaging for Alzheimer’s Disease. Acknowledgment This work was supported by the National Institute of Biomedical Imaging and Bioengineering (NIBIB) grant R01-EB022092. The contents are solely the responsibility of the authors and do not represent the official views of the NIBIB.