X-Ray Fluorescence Computed Tomography Imaging of Nanoparticles and Recent Advances in Biomedical X-ray Fluorescence Imaging

Abstract

There are a number of well-established methods in both preclinical and clinical uses of X-rays for non-invasive imaging, including computed tomography and tomographic imaging. Quantitative mapping of various elements in samples of interest is made possible by X-ray fluorescence analysis, while projection radiography offers anatomical information. So far, the technique has mainly found use in the material, archaeological, and environmental sciences for elemental identification and quantification; however, the application in the life sciences has been severely constrained by intrinsic spectral background problems that arise in larger objects. Multiple Compton-scattering events in the target objects provide this background, which severely restricts the minimum detectable marker concentrations that may be achieved. In this article, we take a look back at X-ray fluorescence imaging's (XFI) past, present its current and future promising preclinical applications, and predict its eventual clinical translation, which will be possible by lowering the aforementioned intrinsic background using specialised algorithms and new X-ray sources. Monochromatic synchrotron x-rays are used in conventional x-ray fluorescence computed tomography (XFCT) in order to simultaneously find the concentration and spatial distribution of different elements, such metals, in a sample. It would appear, however, that in a conventional laboratory environment, the synchrotron-based XFCT method is not appropriate for in vivo imaging. To the best of our knowledge, our study is the first to show that XFCT imaging of a small object (about the size of an animal) with low concentrations of gold nanoparticles (GNPs) employing x-rays from the polychromatic diagnostic energy range is possible. To be more precise, we built a polymethyl methacrylate (PMMA) phantom that resembles a small animal's internal organs and tumours by filling two cylindrical columns with saline solution containing 1 and 2 weight percent (wt) GNPs, respectively. After suitable x-ray beam filtering and detector collimation, the phantom was scanned using an XFCT under a pencil beam geometry with a microfocus 110 kVp x-ray beam and a cadmium telluride (CdTe) x-ray detector. With contrast levels inversely proportional to gold concentration, the two GNP-filled columns were clearly visible in the reconstructed images. However, scanning duration is a major reason why the present pencil-beam implementation of XFCT is impractical for commonly used GNPs in in vivo imaging tasks. However, it is still possible to see things smaller than the current phantom size using a combination of several detectors and a small number of projections. The present study proposes various ways to alter the existing XFCT configuration, including using a quasi-monochromatic cone/fan x-ray beam and XFCT-specific spatial filters or pinhole detector collimators, to determine whether a bench-top XFCT system is feasible for GNP-based preclinical molecular imaging systems.