Simulation of a Gamma-Ray Computed Tomography System Using Two Radioisotopes for Structural Inspections: A Preliminary Study

Gamma-ray computed tomography (CT) has widely been used to inspect cracks, obstructions, and fluid flow in pipes. Previous studies have shown that the inspection accuracy is limited when materials in the pipe have a similar density. To overcome this drawback, we investigated a method to discriminate small density differences using a dual-energy gamma-ray CT. In this paper, the performance of a gamma-ray CT system was simulated using Geant4 Application for Tomographic Emission with three types of sources (Am-241, Co-57, and Cs-137). A polyvinyl chloride (PVC) pipe containing oil and sludge was modeled as a phantom. Sparsely acquired data were reconstructed using filtered back-projection and iterative total variation (TV) denoising algorithms. Each material discrimination procedure was performed with a postreconstruction technique. Our results clearly showed that the gamma-ray CT imaging with Am-241 and Co-57 sources and the TV denoising reconstruction algorithm can be used to inspect the internal structures based on discriminated images. The calculated areas of discriminated oil and sludge were within an error margin of 2% (oil: 78.4% and sludge: 21.6%) compared with the simulated phantom (oil: 80% and sludge: 20%). For both gamma sources, the calculated contrast-to-noise ratio (CNR) values with the TV denoising algorithm were the highest values for all structures (Am-241/Co-57: 18.03/28.95 for oil versus sludge, 41.67/31.53 for oil versus PVC pipe, and 23.64/31.53 for sludge versus PVC pipe). The feasibility of our approach and results indicates it usefulness for industrial fields that require structural inspections.