Peter D. Olcott, Sam Buss, Craig S. Levin, Guillem Pratx,
and Chris K. Sramek
"GRAY: High Energy Photon Ray Tracer for PET Applications
2006 IEEE Nuclear Science Symposition Conference Record, 2006, Volume 4, pp. 2011-2015.
Download manuscript: PDF.
Abstract: GRAY (High Energy Photon Ray Tracer) is a Monte-Carlo ray-driven high energy photon transport engine for mainly PET and SPECT applications that supports complex mesh based primitives for source distributions, phantom shapes, and detector geometries. Monte-Carlo modeling is critical for system design evaluation and image reconstruction development. Ray tracing is a technique used in computer graphics to render scenes with realistic light properties. We adapted an open source ray tracing engine to support the physical properties of high energy photon transport. The main project goal of GRAY is to provide a means to import advanced geometrical mesh primitives from graphical CAD programs to create animated vectorial based phantoms and complex detector geometries while preserving physical accuracy and efficient runtime. These phantoms will be able to model complex moving objects targeted towards developing novel image reconstruction algorithms for cardiac, respiratory, and tracer kinetic modeling. Traditionally, complex geometrical phantoms were simulated using a series of discrete voxelized sources that represent the activity source distributions and attenuation media. By using rejection testing, constructive solid geometry can create primitives that are combined using Boolean operations to create complex phantoms. The high energy photon physics modeling of GRAY has been validated against GATE for both runtime performance and accuracy. GRAY runs an order of magnitude faster than GATE with improved geometric modeling capabilities for detectors and sources. To highlight the capabilities of GRAY, a complex mesh phantom of a rat was filled with uniform activity, placed in a high resolution 1mm3 small animal box PET scanner, simulated with GRAY, and reconstructed using list-mode 3-D OSEM reconstruction.
Back to Sam Buss's publications page.