Review of Development and Recent Advances in Biomedical X-ray Fluorescence Imaging

Numerical Study towards In Vivo Tracking of Micro-/Nanoplastic Based on X-ray Fluorescence Imaging

There is a rising awareness of the toxicity of micro- and nanoplastics (MNPs); however, fundamental precise information on MNP-biodistribution in organisms is currently not available. X-ray fluorescence imaging (XFI) is introduced as a promising imaging modality to elucidate the effective MNP bioavailability and is expected to enable exact measurements on the uptake over the physical barriers of the organism and bioaccumulation in different organs. This is possible because of the ability of XFI to perform quantitative studies with a high spatial resolution and the possibility to conduct longitudinal studies. The focus of this work is a numerical study on the detection limits for a selected XFI-marker, here, palladium, to facilitate the design of future preclinical in vivo studies. Based on Monte Carlo simulations using a 3D voxel mouse model, the palladium detection thresholds in different organs under in vivo conditions in a mouse are estimated. The minimal Pd-mass in the scanning position at a reasonable significance level is determined to be <20 ng/mm2 for abdominal organs and <16 μg/mm2 for the brain. MNPs labelled with Pd and homogeneously distributed in the organ would be detectable down to a concentration of <1 μg/mL to <2.5 mg/mL in vivo. Long-term studies with a chronic MNP exposure in low concentrations are therefore possible such that XFI measurements could, in the future, contribute to MNP health risk assessment in small animals and humans.

Enabling Low-Dose In Vivo Benchtop X-ray Fluorescence Computed Tomography through Deep-Learning-Based Denoising

X-ray Fluorescence Computed Tomography (XFCT) is an emerging non-invasive imaging technique providing high-resolution molecular-level data. However, increased sensitivity with current benchtop X-ray sources comes at the cost of high radiation exposure. Artificial Intelligence (AI), particularly deep learning (DL), has revolutionized medical imaging by delivering high-quality images in the presence of noise. In XFCT, traditional methods rely on complex algorithms for background noise reduction, but AI holds promise in addressing high-dose concerns. We present an optimized Swin-Conv-UNet (SCUNet) model for background noise reduction in X-ray fluorescence (XRF) images at low tracer concentrations. Our method's effectiveness is evaluated against higher-dose images, while various denoising techniques exist for X-ray and computed tomography (CT) techniques, only a few address XFCT. The DL model is trained and assessed using augmented data, focusing on background noise reduction. Image quality is measured using peak signal-to-noise ratio (PSNR) and structural similarity index (SSIM), comparing outcomes with 100% X-ray-dose images. Results demonstrate that the proposed algorithm yields high-quality images from low-dose inputs, with maximum PSNR of 39.05 and SSIM of 0.86. The model outperforms block-matching and 3D filtering (BM3D), block-matching and 4D filtering (BM4D), non-local means (NLM), denoising convolutional neural network (DnCNN), and SCUNet in both visual inspection and quantitative analysis, particularly in high-noise scenarios. This indicates the potential of AI, specifically the SCUNet model, in significantly improving XFCT imaging by mitigating the trade-off between sensitivity and radiation exposure.

High-Spatial-Resolution Benchtop X-ray Fluorescence Imaging through Bragg-Diffraction-Based Focusing with Bent Mosaic Graphite Crystals: A Simulation Study

X-ray fluorescence imaging (XFI) can localize diagnostic or theranostic entities utilizing nanoparticle (NP)-based probes at high resolution in vivo, in vitro, and ex vivo. However, small-animal benchtop XFI systems demonstrating high spatial resolution (variable from sub-millimeter to millimeter range) in vivo are still limited to lighter elements (i.e., atomic number Z≤45). This study investigates the feasibility of focusing hard X-rays from solid-target tubes using ellipsoidal lens systems composed of mosaic graphite crystals with the aim of enabling high-resolution in vivo XFI applications with mid-Z (42≤Z≤64) elements. Monte Carlo simulations are performed to characterize the proposed focusing-optics concept and provide quantitative predictions of the XFI sensitivity, in silico tumor-bearing mice models loaded with palladium (Pd) and barium (Ba) NPs. Based on simulation results, the minimum detectable total mass of PdNPs per scan position is expected to be on the order of a few hundred nanograms under in vivo conform conditions. PdNP masses as low as 150 ng to 50 ng could be detectable with a resolution of 600 μm when imaging abdominal tumor lesions across a range of low-dose (0.8 μGy) to high-dose (8 μGy) exposure scenarios. The proposed focusing-optics concept presents a potential step toward realizing XFI with conventional X-ray tubes for high-resolution applications involving interesting NP formulations.

Gold nanoparticle-mediated photothermal therapy guidance with multi-wavelength photomagnetic imaging

Difficulty in heating tumors with high spatial selectivity while protecting surrounding healthy tissues from thermal harm is a challenge for cancer photothermal treatment (PTT). To mitigate this problem, PTT mediated by photothermal agents (PTAs) has been established as a potential therapeutic technique to boost selectivity and reduce damage to surrounding healthy tissues. Various gold nanoparticles (AuNP) have been effectively utilized as PTAs, mainly using strategies to target cancerous tissue and increase selective thermal damage. Meanwhile, imaging can be used in tandem to monitor the AuNP distribution and guide the PTT. Mainly, the parameters impacting the induced temperature can be determined using simulation tools before treatment for effective PTT. However, accurate simulations can only be performed if the amount of AuNPs accumulated in the tumor is known. This study introduces Photo-Magnetic Imaging (PMI), which can appropriately recover the AuNP concentration to guide the PTT. Using multi-wavelength measurements, PMI can provide AuNP concentration based on their distinct absorption spectra. Tissue-simulating phantom studies are conducted to demonstrate the potential of PMI in recovering AuNP concentration for PTT planning. The recovered AuNP concentration is used to model the temperature increase accurately in a small inclusion representing tumor using a multiphysics solver that takes into account the light propagation and heat diffusion in turbid media.

Laboratory Liquid-Jet X-ray Microscopy and X-ray Fluorescence Imaging for Biomedical Applications

Diffraction-limited resolution and low penetration depth are fundamental constraints in optical microscopy and in vivo imaging. Recently, liquid-jet X-ray technology has enabled the generation of X-rays with high-power intensities in laboratory settings. By allowing the observation of cellular processes in their natural state, liquid-jet soft X-ray microscopy (SXM) can provide morphological information on living cells without staining. Furthermore, X-ray fluorescence imaging (XFI) permits the tracking of contrast agents in vivo with high elemental specificity, going beyond attenuation contrast. In this study, we established a methodology to investigate nanoparticle (NP) interactions in vitro and in vivo, solely based on X-ray imaging. We employed soft (0.5 keV) and hard (24 keV) X-rays for cellular studies and preclinical evaluations, respectively. Our results demonstrated the possibility of localizing NPs in the intracellular environment via SXM and evaluating their biodistribution with in vivo multiplexed XFI. We envisage that laboratory liquid-jet X-ray technology will significantly contribute to advancing our understanding of biological systems in the field of nanomedical research.