Author: Sang Hyun Cho, Sandun Y. Jayarathna, Neerajan Nepal 👨🔬
Affiliation: The University of Texas MD Anderson Cancer Center 🌍
Purpose: To develop/validate a Monte Carlo (MC) model of an experimental benchtop x-ray fluorescence computed tomography (XFCT) system with a pixelated CdTe detector that can be used for MC simulations of detection/imaging of gold nanoparticles (GNPs) present within a small animal-sized object under various irradiation conditions.
Methods: The Geant4 MC toolkit was used to model an experimental benchtop XFCT system adopting a pixelated CdTe detector (128×128-pixels, 250µm-pixel-pitch) coupled with a 5cm-thick 2D parallel-hole collimator (11×11 array of 2mm-diameter apertures) made of stainless steel. The developed MC model was validated by performing a simulation mirroring a typical experimental task where a 1.8mm-Sn-filtered 125 kVp x-ray cone-beam was used to irradiate a 3cm-diameter PMMA phantom with three 2mm-diameter cylindrical holes filled with various concentrations of GNPs, depositing 1.77 and 3.55 cGy per projection. Additional MC simulations were performed with hypothetical monoenergetic beams (85, 90, and 95 keV). After obtaining x-ray fluorescence (XRF)/scatter spectra from MC simulations, background-subtracted XRF counts were corrected for attenuation of XRF photons and the incident x-ray beam. The GNP detection limits were then determined from a calibration curve (corrected XRF counts vs. GNP concentrations).
Results: The MC simulation with a filtered 125 kVp beam suggested the GNP detection limit of 300 ppm achievable with 1.77 cGy. The same result was also observed from the reconstructed XFCT image analysis. Among the tested monoenergetic x-ray beams, the 85 keV beam showed the best performance, lowering the GNP detection limit to 150 ppm (at 1.77 cGy) and 100 ppm (at 3.55 cGy).
Conclusion: The developed MC model was capable of producing the XFCT image and the GNP detection limit that matched our previous experimental results. It also enabled other investigations difficult to perform experimentally, such as quantitative assessment regarding the impact of an incident beam spectrum on the GNP detection limit.