A prototype coded aperture detector for small animal SPECT

S. R. Meikle, Royal Prince Alfred Hospital, Sydney
P. Kench, University of Sydney
A. G. Weisenberger, Thomas Jefferson National Accelerator Facility, Virginia, USA
R. Wojcik, Thomas Jefferson National Accelerator Facility, Virginia, USA
M. F. Smith, Thomas Jefferson National Accelerator Facility, Virginia, USA
S. Majewski, Thomas Jefferson National Accelerator Facility, Virginia, USA
S. Eberl, Royal Prince Alfred Hospital, Sydney
R. R. Fulton, Royal Prince Alfred Hospital, Sydney
A. Rosenfeld, University of Wollongong
M. J. Fullham, Royal Prince Alfred Hospital, Sydney

Article comments

This article was originally published as: Meikle, SR, Kench, P, Weisenberger, AG, Wojcik, R, Smith, MF, Majewski, S, Eberl, S, Fulton, RR, Rosenfeld, AB & Fulham, MJ, A prototype coded aperture detector for small animal SPECT, IEEE Transactions on Nuclear Science, October 2002, 49(5)1, 2167-2171. Copyright IEEE 2002.

Abstract

In a previous simulation study, we demonstrated the feasibility of using coded apertures together with pixelated detectors tors for small animal SPECT. In this paper; we further explore the potential of this approach with a prototype detector and simulated multipinhole apertures. We also investigated the effect of multiplexing due to overlapped projections on convergence properties, image signal-to-noise ratio (SNR) and spatial resolution. The detector comprises a 48/spl times/44 array of NaI(Tl) crystals, each 1 mm/spl times/1 mm/spl times/5 mm on a 1.25-mm pitch. The crystal array is directly coupled to a Hamamatsu R3941 8 cm position sensitive photomultiplier tube. Multipinhole apertures were simulated by performing repeated SPECT acquisitions of the same object with a single tungsten pinhole translated to different positions in the aperture plane. Image reconstruction is based on a three-dimensional ray driven projector which is an extension of a method described for single pinhole SPECT with a displaced center of rotation. Image estimates are updated using the maximum likelihood expectation maximization (ML-EM) algorithm. The effect of multiplexing was to slow convergence and reduce the achievable SNR by approximately 15% compared with nonmultiplexed data (but the result may be achieved in a fraction of the time). The reconstructed resolution obtained with a resolution phantom was 1.5-mm full width at half maximum and there was no appreciable difference between the resolution of multiplexed and nonmultiplexed data. These results encourage us to develop a prototype coded aperture system for high sensitivity, high resolution small animal SPECT.