부산대학교 도서관

As a preprocessing to the reconstruction of the Compton image, it is important to determine the interaction sequence of the incident gamma ray in a detection system. In the conventional Compton cameras consisted of scattering and absorption detectors, the interaction sequence can be determined based on timing resolution is significantly larger than the flight time of a gamma-ray between the events, and hence, the sequence order cannot be determined by the timing information. Various sequence tracking methods have been studied depending on the number of interactions. The sequence tracking methods were introduced Simple comparison method and Deterministic method for two interactions, and Minimum squared difference (MSD) method and Three compton method for three interactions. In this study, two and three interactions sequence methods were compared for our Compton cameras consisted of single and double layered CdZnTe arrays module. The single array module structure consisted of 6 × 6 mosaic array of virtual Frisch-grid CZT crystals with dimensions of 5 × 5 × 12 mm3 and 2.1 mm gap between the crystals. The double layered array modules comprised two identical CZT arrays in parallel and the widths between two array modules were set 3 cm and 10 cm. The three geometry of CZT compton cameras simulated by Monte Carlo N-Particle Transport code (MCNPX). In the simulations, a point 137Cs (662 keV) was positioned 15 cm apart from the CZT array modules. Also, source having angular offset was processed by 0°, 20°, 40°. The source image was reconstructed by simple back-projection (SBP) method and weighted list-mode maximum likelihood expectation maximization (MLEM). also, The results were quantitatively evaluated by full width at half maximum (FWHM) and signal to noise ratio (SNR) for optimizing multiple sequence method in our Compton camera system.