Research on system response matrix modelling method of rectangular PET scanner based on Monte Carlo simulation

• Published in: Journal of instrumentation Vol. 16; no. 1; p. P01009
• Main Authors: Deng, Z.Z., Xu, S.K., Chen, Z., Zhou, K., Hong, W.Y.
• Format: Journal Article
• Language: English
• Published: Bristol IOP Publishing 01.01.2021
• Subjects: Coefficient of variation; Image contrast; Image quality; Image reconstruction; Monte Carlo simulation; Positron emission; Quality assessment; Ray tracing; Simulation; Spatial resolution; Symmetry; Tomography
• ISSN: 1748-0221; 1748-0221
• DOI: 10.1088/1748-0221/16/01/P01009

In positron emission tomography (PET) image reconstruction, the accurate system response matrix (SRM) is one of the key factors for reconstructing high-quality images. To reduce the influence of various factors on the spatial resolution, this paper focuses on calculating the SRM based on Monte Carlo (MC) simulation and proposes a method based on system geometric symmetry to reduce the simulation time. We split the symmetry into two types: intra-ring symmetry and inter-axis symmetry, a total of 208 times the symmetry is used. According to the standards formulated by the American National Electrical Manufacturers Association (NEMA), the spatial resolution and image quality are assessed by simulation experiments. Results of point source measurement show that the full width at half maximum (FWHM) range in each direction is about 1.1–1.3 mm, the average value is about 1.2 mm. In Derenzo phantom imaging, the MC method can clearly distinguish the 1.2 mm rods. To assess image quality, we calculated the uniformity of the image, contrast recovery coefficients (CRC), and coefficient of variation (COV). At the same time, we compared with the ray tracing (RT) and pixel drive (PD) models. In general, the MC method achieves better results in simulation experiments than RT and PD methods.