Development of a Method to Determine Electron Density and Effective Atomic Number of High Atomic Number Solid Materials Using Dual-Energy Computed Tomography

Development of Cu application using dual-energy computed tomography for detecting medication-related osteonecrosis of the jaw

INTRODUCTION

The present study developed an application using dual-energy computed tomography (DECT) focused on Cu for detecting medication-related osteonecrosis of the jaw (MRONJ).

MATERIALS AND METHODS

First, we performed two types of phantom studies using a Cu wire syringe and pig mandible with Cu wire to detect Cu on DECT. Second, DECT examinations of 44 patients with MRONJ were performed to compare lesion and normal bone sites using single-energy CT, DECT-virtual non-calcium (VNCa), and DECT-Cu applications. Quantitative analyses of VNCa CT and CT values were performed, and a cut-off value was calculated using receiver operating characteristic analysis. Third, we compared the Cu content in the MRONJ and normal bone groups using inductively coupled plasma atomic emission spectroscopy (ICP-AES).

RESULTS

The material-specific differences in attenuation between the two different energies enabled the accurate separation of Cu from Ca in phantom studies. The sensitivity and specificity for single-energy CT, DECT-VNCa, and DECT-Cu applications were 97.7% and 2.3%, 86.4% and 81.8%, and 88.6% and 97.7%, respectively. Thus, VNCa CT values obtained on DECT-Cu application images showed the highest area under the curve value and maximal diagnostic efficacy in differentiating lesion sites from normal bone sites. On ICP-AES analyses, the Cu content was significantly higher in the MRONJ group than in the normal bone group.

CONCLUSION

DECT-Cu application demonstrated better diagnostic performance in detecting MRONJ compared with single-energy CT or DECT-VNCa.

Non-Invasive characterisation of renal stones using dual energy CT: A method to differentiate calcium stones

BACKGROUND

Non-invasive DECT based characterization of renal stones using their effective atomic number (Z_eff) and the electron density (ρe) in patients.

AIM

This paper aims to develop a method for in-vivo characterization of renal stone. Differentiation of renal stones in-vivo especially sub types of calcium stones have very important advantage for better judgement of treatment modality.

MATERIALS AND METHODS

50 extracted renal stones were scanned ex-vivo using dual energy CT scanner. A method was developed to characterize these renal stones using effective atomic number and electron density obtained from dual energy CT data. The method and formulation developed in ex-vivo experiments was applied in in-vivo study of 50 randomly selected patients of renal stones who underwent dual energy CT scan.

RESULTS

The developed method was able to characterize Calcium Oxalate Monohydrate (COM) and the combination of COM and Calcium Oxalate Dihydrate (COD) stones non-invasively in patients with a sensitivity of 81% and 83%respectively. The method was also capable of differentiating Uric, Cystine and mixed stones with the sensitivity of 100, 100 and 85.71% respectively.

CONCLUSION

The developed dual energy CT based method was capable of differentiating sub types of calcium stones which is not differentiable on single energy or dual energy CT images.

Dual-Energy Computed Tomography For Differentiation Between Osteoblastic Metastases and Bone Islands

Objective

The objective of our study was to evaluate the utility of Rho/Z on dual-energy computed tomography (DECT) for the differentiation of osteoblastic metastases (OBMs) from bone islands (BIs).

Methods

DECT images of 110 patients with malignancies were collected. The effective atomic number (Z), electron density (Rho), dual energy index (DEI), and regular CT (rCT) values were measured by two observers. Independent-sample t-test was used to compare these values between OBMs and BIs. The diagnostic performance was assessed by receiver operating characteristic (ROC) analysis and the cutoff values were evaluated according to ROC curves.

Results

A total of 205 OBMs and 120 BIs were included. The mean values of Z, Rho, DEI, and rCT of OBMs were significantly lower than those of BIs, whereas the standard deviation values were higher than those of BIs (all p ≤ 0.05). ROC analysis showed that 11.86 was the optimal cutoff value for Z, rendering an area under the ROC curve (AUC) of 0.91, with a sensitivity of 91.2% and a specificity of 82.5%.

Conclusion

DECT can provide quantitative values of Z, Rho, and DEI and has good performance in differentiating between OBMs and BIs.

Monochromatic image reconstruction via machine learning

X-ray computed tomography (CT) is a nondestructive imaging technique to reconstruct cross-sectional images of an object using x-ray measurements taken from different view angles for medical diagnosis, therapeutic planning, security screening, and other applications. In clinical practice, the x-ray tube emits polychromatic x-rays, and the x-ray detector array operates in the energy-integrating mode to acquire energy intensity. This physical process of x-ray imaging is accurately described by an energy-dependent non-linear integral equation on the basis of the Beer–Lambert law. However, the non-linear model is not invertible using a computationally efficient solution and is often approximated as a linear integral model in the form of the Radon transform, which basically loses energy-dependent information. This approximate model produces an inaccurate quantification of attenuation images, suffering from beam-hardening effects. In this paper, a machine learning-based approach is proposed to correct the model mismatch to achieve quantitative CT imaging. Specifically, a one-dimensional network model is proposed to learn a non-linear transform from a training dataset to map a polychromatic CT image to its monochromatic sinogram at a pre-specified energy level, realizing virtual monochromatic (VM) imaging effectively and efficiently. Our results show that the proposed method recovers high-quality monochromatic projections with an average relative error of less than 2%. The resultant x-ray VM imaging can be applied for beam-hardening correction, material differentiation and tissue characterization, and proton therapy treatment planning.

Virtual Monoenergetic CT Imaging via Deep Learning

Conventional single-spectrum computed tomography (CT) reconstructs a spectrally integrated attenuation image and reveals tissues morphology without any information about the elemental composition of the tissues. Dual-energy CT (DECT) acquires two spectrally distinct datasets and reconstructs energy-selective (virtual monoenergetic [VM]) and material-selective (material decomposition) images. However, DECT increases system complexity and radiation dose compared with single-spectrum CT. In this paper, a deep learning approach is presented to produce VM images from single-spectrum CT images. Specifically, a modified residual neural network (ResNet) model is developed to map single-spectrum CT images to VM images at pre-specified energy levels. This network is trained on clinical DECT data and shows excellent convergence behavior and image accuracy compared with VM images produced by DECT. The trained model produces high-quality approximations of VM images with a relative error of less than 2%. This method enables multi-material decomposition into three tissue classes, with accuracy comparable with DECT.