Performance characterization of a prototype dual-layer cone-beam computed tomography system.

PURPOSE: Conventional cone-beam computed tomography CT (CBCT) provides limited discrimination between low-contrast tissues. Furthermore, it is limited to full-spectrum energy integration. A dual-energy CBCT system could be used to separate photon energy spectra with the potential to increase the visibility of clinically relevant features and acquire additional information relevant in a multitude of clinical imaging applications. In this work, the performance of a novel dual-layer dual-energy CBCT (DL-DE-CBCT) C-arm system is characterized for the first time. METHODS: A prototype dual-layer detector was fitted into a commercial interventional C-arm CBCT system to enable DL-DE-CBCT acquisitions. DL-DE reconstructions were derived from material-decomposed Compton scatter and photoelectric base functions. The modulation transfer function (MTF) of the prototype DL-DE-CBCT was compared to that of a commercial CBCT. Noise and uniformity characteristics were evaluated using a cylindrical water phantom. Effective atomic numbers and electron densities were estimated in clinically relevant tissue substitutes. Iodine quantification was performed (for 0.5-15 mg/ml concentrations) and virtual noncontrast (VNC) images were evaluated. Finally, contrast-to-noise ratios (CNR) and CT number accuracies were estimated. RESULTS: The prototype and commercial CBCT showed similar spatial resolution, with a mean 10% MTF of 5.98 cycles/cm and 6.28 cycles/cm, respectively, using a commercial standard reconstruction. The lowest noise was seen in the 80 keV virtual monoenergetic images (VMI) (7.40 HU) and the most uniform images were seen at VMI 60 keV (4.74 HU) or VMI 80 keV (1.98 HU), depending on the uniformity measure used. For all the tissue substitutes measured, the mean accuracy in effective atomic number was 98.2% (SD 1.2%) and the mean accuracy in electron density was 100.3% (SD 0.9%). Iodine quantification images showed a mean difference of -0.1 (SD 0.5) mg/ml compared to the true iodine concentration for all blood and iodine-containing objects. For VNC images, all blood substitutes containing iodine averaged a CT number of 43.2 HU, whereas a blood-only substitute measured 44.8 HU. All water-containing iodine substitutes measured a mean CT number of 2.6 in the VNC images. A noise-suppressed dataset showed a CNR peak at VMI 40 keV and low at VMI 120 keV. In the same dataset without noise suppression applied, a peak in CNR was obtained at VMI 70 keV and a low at VMI 120 keV. The estimated CT numbers of various clinically relevant objects were generally very close to the calculated CT number. CONCLUSIONS: The performance of a prototype dual-layer dual-energy C-arm CBCT system was characterized. Spatial resolution and noise were comparable with a commercially available C-arm CBCT system, while offering dual-energy capability. Iodine quantifications, effective atomic numbers, and electron densities were in good agreement with expected values, indicating that the system can be used to reliably evaluate the material composition of clinically relevant tissues. The VNC and monoenergetic images indicate a consistent ability to separate clinically relevant tissues. The results presented indicate that the system could find utility in diagnostic, interventional, and radiotherapy planning settings.

Spectral Photon-counting CT: Initial Experience with Dual-Contrast Agent K-Edge Colonography.

Purpose To investigate the feasibility of using spectral photon-counting computed tomography (CT) to differentiate between gadolinium-based and nonionic iodine-based contrast material in a colon phantom by using the characteristic k edge of gadolinium. Materials and Methods A custom-made colon phantom was filled with nonionic iodine-based contrast material, and a gadolinium-filled capsule representing a contrast material-enhanced polyp was positioned on the colon wall. The colon phantom was scanned with a preclinical spectral photon-counting CT system to obtain spectral and conventional data. By fully using the multibin spectral information, material decomposition was performed to generate iodine and gadolinium maps. Quantitative measurements were performed within the lumen and polyp to quantitatively determine the absolute content of iodine and gadolinium. Results In a conventional CT section, absorption values of both contrast agents were similar at approximately 110 HU. Contrast material maps clearly differentiated the distributions, with gadolinium solely in the polyp and iodine in the lumen of the colon. Quantitative measurements of contrast material concentrations in the colon and polyp matched well with those of actual prepared mixtures. Conclusion Dual-contrast spectral photon-counting CT colonography with iodine-filled lumen and gadolinium-tagged polyps may enable ready differentiation between polyps and tagged fecal material. © RSNA, 2016.

Quantitative spectral K-edge imaging in preclinical photon-counting x-ray computed tomography.

OBJECTIVES: The objective of this study was to investigate the feasibility and the accuracy of spectral computed tomography (spectral CT) to determine the tissue concentrations and localization of high-attenuation, iodine-based contrast agents in mice. Iodine tissue concentrations determined with spectral CT are compared with concentrations measured with single-photon emission computed tomography (SPECT) and inductively coupled plasma mass spectrometry (ICP-MS). MATERIALS AND METHODS: All animal procedures were performed according to the US National Institutes of Health principles of laboratory animal care and were approved by the ethical review committee of Maastricht, The Netherlands. Healthy Swiss mice (n = 4) were injected with an iodinated emulsion radiolabeled with indium as multimodal contrast agent for CT and SPECT. The CT and SPECT scans were acquired using a dedicated small-animal SPECT/CT system. Subsequently, scans were performed with a preclinical spectral CT scanner equipped with a photon-counting detector and 6 energy threshold levels. Quantitative data analysis of SPECT and spectral CT scans were obtained using 3-dimensional volumes-of-interest drawing methods. The ICP-MS on dissected organs was performed to determine iodine uptake per organ and was compared with the amounts determined from spectral CT and SPECT. RESULTS: Iodine concentrations obtained with image-processed spectral CT data correlated well with data obtained either with noninvasive SPECT imaging (slope = 0.96, r = 0.75) or with ICP-MS (slope = 0.99, r = 0.89) in tissue samples. CONCLUSIONS: This preclinical proof-of-concept study shows the in vivo quantification of iodine concentrations in tissues using spectral CT. Our multimodal imaging approach with spectral CT and SPECT using radiolabeled iodinated emulsions together with ICP-based quantification allows a direct comparison of all methods. Benchmarked against ICP-MS data, spectral CT in the present implementation shows a slight underestimation of organ iodine concentrations compared with SPECT but with a more narrow distribution. This slight deviation is most likely caused by experimental rather than technical issues.

Gold nanocrystal labeling allows low-density lipoprotein imaging from the subcellular to macroscopic level.

Low-density lipoprotein (LDL) plays a critical role in cholesterol transport and is closely linked to the progression of several diseases. This motivates the development of methods to study LDL behavior from the microscopic to whole-body level. We have developed an approach to efficiently load LDL with a range of diagnostically active nanocrystals or hydrophobic agents. We performed focused experiments on LDL labeled with gold nanocrystals (Au-LDL). The labeling procedure had minimal effect on LDL size, morphology, or composition. Biological function was found to be maintained from both in vitro and in vivo experiments. Tumor-bearing mice were injected intravenously with LDL, DiR-LDL, Au-LDL, or a gold-loaded nanoemulsion. LDL accumulation in the tumors was detected with whole-body imaging methods, such as computed tomography (CT), spectral CT, and fluorescence imaging. Cellular localization was studied with transmission electron microscopy and fluorescence techniques. This LDL labeling procedure should permit the study of lipoprotein biointeractions in unprecedented detail.

Statistical reconstruction of material decomposed data in spectral CT.

Photon-counting detector technology has enabled the first experimental investigations of energy-resolved computed tomography (CT) imaging and the potential use for K-edge imaging. However, limitations in regards to detecter technology have been imposing a limit to effective count rates. As a consequence, this has resulted in high noise levels in the obtained images given scan time limitations in CT imaging applications. It has been well recognized in the area of low-dose imaging with conventional CT that iterative image reconstruction provides a superior signal to noise ratio compared to traditional filtered backprojection techniques. Furthermore, iterative reconstruction methods also allow for incorporation of a roughness penalty function in order to make a trade-off between noise and spatial resolution in the reconstructed images. In this work, we investigate statistically-principled iterative image reconstruction from material-decomposed sinograms in spectral CT. The proposed reconstruction algorithm seeks to minimize a penalized likelihood-based cost functional, where the parameters of the likelihood function are estimated by computing the Fisher information matrix associated with the material decomposition step. The performance of the proposed reconstruction method is quantitatively investigated by use of computer-simulated and experimental phantom data. The potential for improved K-edge imaging is also demonstrated in an animal experiment.

Second Generation Gold Nanobeacons for Robust K-Edge Imaging with Multi-Energy CT.

Spectral CT is the newest advancement in CT imaging technology, which enhances traditional CT images with the capability to image and quantify certain elements based on their distinctive K-edge energies. K-edge imaging feature recognizes high accumulations of targeted elements and presents them as colorized voxels against the normal grayscale X-ray background offering promise to overcome the relatively low inherent contrast within soft tissue and distinguish the high attenuation of calcium from contrast enhanced targets. Towards this aim, second generation gold nanobeacons (GNB(2)), which incorporate at least five times more metal than the previous generation was developed. The particles were synthesized as lipid-encapsulated, vascularly constrained (>120 nm) nanoparticle incorporating tiny gold nanoparticles (2-4 nm) within a polysorbate core. The choice of core material dictated to achieve a higher metal loading. The particles were thoroughly characterized by physicochemical techniques. This study reports one of the earlier examples of spectral CT imaging with gold nanoparticles demonstrating the potential for targeted in vitro and in vivo imaging and eliminates calcium interference with CT. The use of statistical image reconstruction shows high SNR may allow dose reduction and/or faster scan times.

Effect of computed tomography scanning parameters on gold nanoparticle and iodine contrast.

PURPOSE: Gold nanoparticles (gold-NPs) have lately been proposed as alternative contrast agents to iodine-based contrast agents (iodine-CA) for computed tomography (CT) angiography. The aims of this study were to confirm an appropriate environment in which to evaluate such novel contrast agents, to investigate the comparative contrast of iodine-CA versus gold-NP, and to determine optimal scanning parameters for gold-NP. MATERIALS AND METHODS: Three different clinical scanners were used to acquire CT images. A range of concentrations (10 mM to 1.5 M) of gold-NP and iodine-CA were scanned with varying x-ray tube voltages and currents, reconstruction kernels, protocols, and scanner models. The different environments investigated were air, water, and water with a bone simulant (Ca3(PO4)2). Regression coefficients were derived from the attenuation values plotted against concentration and compared for statistical significance using t values. RESULTS: As expected, contrast was linearly related to concentrations up to 500 to 1000 mM, depending on the conditions used, whereupon a plateau of 3000 Hounsfield units was reached. Attenuation was significantly different depending on the environment used (air, water, or water and bone simulant). Contrast is dependent on the x-ray tube voltage used, with the contrast produced from iodine-CA sharply declining with increasing voltage, whereas the contrast of gold-NP varied less with tube voltage but was maximal at 120 kV in water with bone simulant. Current, reconstruction kernels, protocols, and scanner model had less effect on contrast. CONCLUSION: Water with a bone simulant is a preferable environment for evaluating novel cardiac CT contrast agents. Relative iodine-CA versus gold-NP contrast is dependent on the scanning conditions used. Optimal scanning conditions for gold-NP will likely use an x-ray tube voltage of 120 kV.

An early investigation of ytterbium nanocolloids for selective and quantitative "multicolor" spectral CT imaging.

We report a novel molecular imaging agent based on ytterbium designed for use with spectral "multicolor" computed tomography (CT). Spectral CT or multicolored CT provides all of the benefits of traditional CT, such as rapid tomographic X-ray imaging, but in addition, it simultaneously discriminates metal-rich contrast agents based on the element's unique X-ray K-edge energy signature. Our synthetic approach involved the use of organically soluble Yb(III) complex to produce nanocolloids of Yb of noncrystalline nature incorporating a high density of Yb (>500K/nanoparticle) into a stable metal particle. The resultant particles are constrained to vasculature (∼200 nm) and are highly selective for binding fibrin in the ruptured atherosclerotic plaque. Nanoparticles exhibited excellent signal sensitivity, and the spectral CT technique uniquely discriminates the K-edge signal (60 keV) of Yb from calcium (bones). Bioelimination and preliminary biodistribution reflected the overall safety and defined clearance of these particles in a rodent model.

Sensitivity of photon-counting based K-edge imaging in X-ray computed tomography.

The feasibility of K-edge imaging using energy-resolved, photon-counting transmission measurements in X-ray computed tomography (CT) has been demonstrated by simulations and experiments. The method is based on probing the discontinuities of the attenuation coefficient of heavy elements above and below the K-edge energy by using energy-sensitive, photon counting X-ray detectors. In this paper, we investigate the dependence of the sensitivity of K-edge imaging on the atomic number Z of the contrast material, on the object diameter D , on the spectral response of the X-ray detector and on the X-ray tube voltage. We assume a photon-counting detector equipped with six adjustable energy thresholds. Physical effects leading to a degradation of the energy resolution of the detector are taken into account using the concept of a spectral response function R(E,U) for which we assume four different models. As a validation of our analytical considerations and in order to investigate the influence of elliptically shaped phantoms, we provide CT simulations of an anthropomorphic Forbild-Abdomen phantom containing a gold-contrast agent. The dependence on the values of the energy thresholds is taken into account by optimizing the achievable signal-to-noise ratios (SNR) with respect to the threshold values. We find that for a given X-ray spectrum and object size the SNR in the heavy element's basis material image peaks for a certain atomic number Z. The dependence of the SNR in the high- Z basis-material image on the object diameter is the natural, exponential decrease with particularly deteriorating effects in the case where the attenuation from the object itself causes a total signal loss below the K-edge. The influence of the energy-response of the detector is very important. We observed that the optimal SNR values obtained with an ideal detector and with a CdTe pixel detector whose response, showing significant tailing, has been determined at a synchrotron differ by factors of about two to three. The potentially very important impact of scattered X-ray radiation and pulse pile-up occurring at high photon rates on the sensitivity of the technique is qualitatively discussed.

Atherosclerotic plaque composition: analysis with multicolor CT and targeted gold nanoparticles.

PURPOSE: To investigate the potential of spectral computed tomography (CT) (popularly referred to as multicolor CT), used in combination with a gold high-density lipoprotein nanoparticle contrast agent (Au-HDL), for characterization of macrophage burden, calcification, and stenosis of atherosclerotic plaques. MATERIALS AND METHODS: The local animal care committee approved all animal experiments. A preclinical spectral CT system in which incident x-rays are divided into six different energy bins was used for multicolor imaging. Au-HDL, an iodine-based contrast agent, and calcium phosphate were imaged in a variety of phantoms. Apolipoprotein E knockout (apo E-KO) mice were used as the model for atherosclerosis. Gold nanoparticles targeted to atherosclerosis (Au-HDL) were intravenously injected at a dose of 500 mg per kilogram of body weight. Iodine-based contrast material was injected 24 hours later, after which the mice were imaged. Wild-type mice were used as controls. Macrophage targeting by Au-HDL was further evaluated by using transmission electron microscopy and confocal microscopy of aorta sections. RESULTS: Multicolor CT enabled differentiation of Au-HDL, iodine-based contrast material, and calcium phosphate in the phantoms. Accumulations of Au-HDL were detected in the aortas of the apo E-KO mice, while the iodine-based contrast agent and the calcium-rich tissue could also be detected and thus facilitated visualization of the vasculature and bones (skeleton), respectively, during a single scanning examination. Microscopy revealed Au-HDL to be primarily localized in the macrophages on the aorta sections; hence, the multicolor CT images provided information about the macrophage burden. CONCLUSION: Spectral CT used with carefully chosen contrast agents may yield valuable information about atherosclerotic plaque composition.