Premium
TH‐A‐141‐03: High‐Sensitivity L‐Shell X‐Ray Fluorescence CT Imaging of Cisplatin
Author(s) -
Bazalova M,
Pratx G,
Xing L
Publication year - 2013
Publication title -
medical physics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.473
H-Index - 180
eISSN - 2473-4209
pISSN - 0094-2405
DOI - 10.1118/1.4815708
Subject(s) - imaging phantom , attenuation , physics , x ray fluorescence , electron shell , nuclear medicine , materials science , fluorescence , optics , ion , medicine , quantum mechanics , ionization
Purpose: To increase the sensitivity of x‐ray fluorescence computed tomography (XFCT) imaging by excitation of low‐energy L‐shell x‐rays with higher interaction cross‐section compared to K‐shell x‐rays. Methods: XFCT images of a 2‐cm diameter acrylic phantom with 0.1–0.4% (or 1–4 mg/mL) Cisplatin solutions were simulated in a modified version of the EGSnrc/DOSXYZnrc Monte Carlo code. XFCT images were reconstructed based on platinum fluorescence x‐rays excited by 15, 30, and 80 keV x‐rays. Based on L‐edge and K‐edge energies of platinum, 15 and 30 keV XFCT images were reconstructed using L‐shell x‐rays (∼10 keV) and 80 keV images were reconstructed with K‐shell x‐rays (∼67 keV). XFCT images were reconstructed with simple filtered back‐projection (FBP), maximum‐likelihood expectation maximization (ML‐EM) without and with attenuation correction to account for the low energy 15 and 30 keV excitation x‐rays and L‐shell fluorescence x‐rays. Results: For a 2.5 mGy imaging dose, K‐shell XFCT images reconstructed with simple FBP resulted in Cisplatin imaging sensitivity limit of 740 μg/mL, which was defined by contrast‐to‐noise ratio (CNR) value of 4. For the same dose, 15 keV L‐shell XFCT imaging sensitivity increased by a factor of 3.8 for objects at the periphery of the phantom. However, contrast in L‐shell images decreased drastically towards the center of the phantom due to attenuation of the excitation and fluorescence x‐rays. Our ML‐EM reconstruction algorithm with attenuation correction applied on the 15 keV L‐shell XFCT image 1) significantly increased CNR and resulted in 2) a uniform signal across the phantom, and 3) accurate quantification of Cisplatin concentration. Conclusion: We have shown that L‐shell XFCT imaging in connection with ML‐EM and attenuation correction reconstruction algorithm resulted high imaging sensitivity and accurate quantification of imaging contrast. A clinically relevant concentration of 6 μg/mL of Cisplatin can be accurately detected with a low imaging dose of 18 mGy.