Optimization of Incident X-Ray Source Spectrum Through Filtration for a Benchtop X-Ray Fluorescence Computed Tomography (XFCT) System
N Manohar1,3*, B Jones2, S Cho1,3, (1) Georgia Tech, Atlanta, GA, (2) Univ. of Colorado School of Medicine, Aurora, CO, (3) The University of Texas MD Anderson Cancer Center, Houston, TXTH-A-141-6 Thursday 8:00AM - 9:55AM Room: 141
Purpose: To optimize acquisition of fluorescence photon signal from gold nanoparticles (GNPs) and reduce the effect of Compton scatter in benchtop x-ray fluorescence computed tomography (XFCT).
Methods: In polychromatic XFCT, fluorescence signal is obscured by Compton scatter of source photons, and thus the choice of source spectrum is crucial. An experimental XFCT system for irradiation of a GNP-containing small animal-sized phantom was accurately modeled using the MCNP5 code for Monte Carlo simulations. The model used either a lead (Pb) or tin (Sn) filter to harden the 105 kVp source spectrum, and was validated with fluorescence/scatter profile measurements. Spectra (Pb: 1, 2, and 3 mm; Sn: 0.9, 1, 2, and 3 mm) were input into the model and their effect on fluorescence production/detection was investigated. Simulations were also run with hypothetical quasi-monochromatic x-ray sources (81, 85, 90, 95, and 100 keV) to determine the ideal energy for discriminating gold K-shell fluorescence peaks from Compton scatter background. Fluorescence signal-to-dose ratio (FSDR) and relative scan time (RST) were used to assess the degree of optimization during simulations.
Results: For both materials, increasing filtration hardened source spectra and significantly decreased fluence available for producing fluorescence photons, while increasing FSDR. The RST increased at a much higher rate for a given increase in FSDR. Compared to Pb, Sn produced generally higher overall fluence, more prominent fluorescence peaks, favorably harder scatter profiles, increased FSDR, and shorter RST. Simulations using hypothetical quasi-monochromatic spectra showed that increasing source energy, between 81 and 100 keV, increased signal-to-background ratio and FSDR.
Conclusion: Judicious choice of filter material/thickness to tailor polychromatic x-ray spectrum, considering trade-offs between FSDR and RST, can dramatically improve the performance of benchtop XFCT. This work will foster the further development of a benchtop XFCT system for routine pre-clinical imaging applications with GNPs.
Supported by NIH/NCI grant R01CA155446
Funding Support, Disclosures, and Conflict of Interest: NIH/NCI grant R01CA155446