Breast tissue decomposition with spectral distortion correction: a postmortem study.

PURPOSE: To investigate the feasibility of an accurate measurement of water, lipid, and protein composition of breast tissue using a photon-counting spectral computed tomography (CT) with spectral distortion corrections. METHODS: Thirty-eight postmortem breasts were imaged with a cadmium-zinc-telluride-based photon-counting spectral CT system at 100 kV. The energy-resolving capability of the photon-counting detector was used to separate photons into low and high energy bins with a splitting energy of 42 keV. The estimated mean glandular dose for each breast ranged from 1.8 to 2.2 mGy. Two spectral distortion correction techniques were implemented, respectively, on the raw images to correct the nonlinear detector response due to pulse pileup and charge-sharing artifacts. Dual energy decomposition was then used to characterize each breast in terms of water, lipid, and protein content. In the meantime, the breasts were chemically decomposed into their respective water, lipid, and protein components to provide a gold standard for comparison with dual energy decomposition results. RESULTS: The accuracy of the tissue compositional measurement with spectral CT was determined by comparing to the reference standard from chemical analysis. The averaged root-mean-square error in percentage composition was reduced from 15.5% to 2.8% after spectral distortion corrections. CONCLUSIONS: The results indicate that spectral CT can be used to quantify the water, lipid, and protein content in breast tissue. The accuracy of the compositional analysis depends on the applied spectral distortion correction technique.

Optimization of K-edge imaging for vulnerable plaques using gold nanoparticles and energy resolved photon counting detectors: a simulation study.

We investigated the effect of different imaging parameters, such as dose, beam energy, energy resolution and the number of energy bins, on the image quality of K-edge spectral computed tomography (CT) of gold nanoparticles (GNP) accumulated in an atherosclerotic plaque. A maximum likelihood technique was employed to estimate the concentration of GNP, which served as a targeted intravenous contrast material intended to detect the degree of the plaque's inflammation. The simulation studies used a single-slice parallel beam CT geometry with an x-ray beam energy ranging between 50 and 140 kVp. The synthetic phantoms included small (3 cm in diameter) cylinder and chest (33 × 24 cm(2)) phantoms, where both phantoms contained tissue, calcium and gold. In the simulation studies, GNP quantification and background (calcium and tissue) suppression tasks were pursued. The x-ray detection sensor was represented by an energy resolved photon counting detector (e.g., CdZnTe) with adjustable energy bins. Both ideal and more realistic (12% full width at half maximum (FWHM) energy resolution) implementations of the photon counting detector were simulated. The simulations were performed for the CdZnTe detector with a pixel pitch of 0.5-1 mm, which corresponds to a performance without significant charge sharing and cross-talk effects. The Rose model was employed to estimate the minimum detectable concentration of GNPs. A figure of merit (FOM) was used to optimize the x-ray beam energy (kVp) to achieve the highest signal-to-noise ratio with respect to the patient dose. As a result, the successful identification of gold and background suppression was demonstrated. The highest FOM was observed at the 125 kVp x-ray beam energy. The minimum detectable GNP concentration was determined to be approximately 1.06 µmol mL(-1) (0.21 mg mL(-1)) for an ideal detector and about 2.5 µmol mL(-1) (0.49 mg mL(-1)) for a more realistic (12% FWHM) detector. The studies show the optimal imaging parameters at the lowest patient dose using an energy resolved photon counting detector to image GNP in an atherosclerotic plaque.

Spectral CT imaging of vulnerable plaque with two independent biomarkers.

The purpose of this paper is to investigate the feasibility of a novel four-material decomposition technique for assessing the vulnerability of plaque with two contrast materials spectral computer tomography (CT) using two independent markers: plaque's inflammation and spotty calcification. A simulation study was conducted using an energy-sensitive photon-counting detector for k-edge imaging of the coronary arteries. In addition to detecting the inflammation status, which is known as a biological marker of a plaque's vulnerability, we use spotty calcium concentration as an independent marker to test a plaque's vulnerability. We have introduced a new method for detecting and quantifying calcium concentrations in the presence of two contrast materials (iodine and gold), calcium and soft tissue background. In this method, four-material decomposition was performed on a pixel-by-pixel basis, assuming there was an arbitrary mixture of materials in the voxel. The concentrations of iodine and gold were determined by the k-edge material decomposition based on the maximum likelihood method. The calibration curves of the attenuation coefficients, with respect to the concentrations of different materials, were used to separate the calcium signal from both contrast materials and different soft tissues in the mixtures. Three different materials (muscle, blood and lipid) were independently used as soft tissue. The simulations included both ideal and more realistic energy resolving detectors to measure the polychromatic photon spectrum in single slice parallel beam geometry. The ideal detector was used together with a 3 cm diameter digital phantom to demonstrate the decomposition method while a more realistic detector and a 33 × 24 cm(2) digital chest phantom were simulated to validate the vulnerability assessment technique. A 120 kVp spectrum was generated to produce photon flux sufficient for detecting contrast materials above the k-edges of iodine (33.2 keV) and gold (80.7 keV). By performing simulations on a 3 cm diameter digital phantom, we successfully identified four materials that were simultaneously present in the mixture at different proportions and in multiple locations on the phantom. Quantitative analysis with a chest digital phantom showed that the results for iodine, gold and calcium were highly correlated with the known concentrations. The analysis revealed a potentially powerful technique for assessing a plaque's vulnerability with two independent markers. High correlation and low relative errors between calculated and known materials' concentrations showed that the method is feasible. This technique can potentially have a high clinical impact.