Optimizing contrast agents with respect to reducing beam hardening in nonmedical X-ray computed tomography experiments

Abstract

Iodine is commonly used as a contrast agent in nonmedical science and engineering, for example, to visualize Darcy flow in porous geological media using X-ray computed tomography (CT). Undesirable beam hardening artifacts occur when a polychromatic X-ray source is used, which makes the quantitative analysis of CT images difficult. To optimize the chemistry of a contrast agent in terms of the beam hardening reduction, we performed computer simulations and generated synthetic CT images of a homogeneous cylindrical sand-pack (diameter, 28 or 56 mm; porosity, 39 vol.%) saturated with aqueous suspensions of heavy elements assuming the use of a polychromatic medical CT scanner. The degree of cupping derived from the beam hardening was assessed using the reconstructed CT images to find the chemistry of the suspension that induced the least cupping. The results showed that (i) the degree of cupping depended on the position of the K absorption edge of the heavy element relative to peak of the polychromatic incident X-ray spectrum, (ii) 53 I was not an ideal contrast agent because it causes marked cupping, and (iii) a single element much heavier than 53I (64Gd to 79Au) reduced the cupping artifact significantly, and a four-heavy-element mixture of elements from 64Gd to 79Au reduced the artifact most significantly.

Nondestructive Quantification of Heavy Elements Through the Analysis of Beam Hardening Artifacts Using Double-Exposure X-ray Computed Tomography: A Theoretical Consideration

“…The curves with minimum values are a consequence of the sensitivity of the severity of cupping artifacts to the relative position between the K absorption edge of a heavy element and the spectral peak of an X-ray source [9,16,17]. All data needed for the material differentiation using double-exposure CT are given in Figs.…”

“…The cupping artifact in the reconstructed CT image of the homogeneous sample is characterized by a large LAC value near the sample rim and small LAC value at the sample center ( Fig. 2a) [13][14][15][16][17]. The reconstructed LAC value distribution in the radial direction ( Fig.…”

“…Determination is performed by systematically calculating the effects of sample chemistry and acceleration voltage on the cupping profile. Cupping profiles that are numerically obtained using image-reconstruction computer software [9,13,17] inevitably contain errors that are introduced by the discretization of sample shapes [13,18] (see the noisy fluctuation of the profile in Fig. 2b).…”

Visualization of water drying in porous materials by X‐ray phase contrast imaging

“…This represents a strong limitation in using XACI in combination with water‐mimicking liquids leading to enhanced X‐ray attenuation. An additional limitation consists in the increase of beam hardening artefacts, as thoroughly documented by Nakashima & Nakano (), Nakashima () and Nakashima & Nakano ().…”

Volumetric Assessment of Bone Microstructures by a 3D Local Binary Patterns –Based Method: Bone Changes with Osteoarthritis

“…While applied to bone samples, it allows to visualize the organization of the trabeculae and obtain information related to its density, and eventually on the strength of the bone. Multiple artefacts can affect the µCT scans, the main two for structure analysis being: the partial volume effect (bone not totally within the voxel) [7] and the beam hardening (inconstant brightness within the stack of images) [8]. Thus, the development of methods unaffected by these issues is a challenging topic to be addressed in bone structural analysis.…”