Correction of severe beam‐hardening artifacts via a high‐order linearization function using a prior‐image‐based parameter selection method

Abstract

The prior-image-based linearization method exhibited better correction performance than conventional methods. Because the proposed method did not require time-consuming iterative reconstruction processes to obtain the optimal correction function, it can expedite the correction procedure and incorporate more high-order terms in the linearization correction function in comparison to the conventional methods.

A new dental CBCT metal artifact reduction method based on a dual-domain processing framework

“…Many researchers have done effect on these problems but most of them consider the artifacts as an image problem without taking the physical imaging process into account. Due to the different x-ray absorption features of the human tissues, Oh et al (2018) proposed a sinogram linearization correction method to correct the projections before reconstruction. But the metal x-ray absorption features are quite different from some human tissues.…”

“…The commonly used projection inpainting based metal artifact reduction methods, such as LI-MAR and NMAR (Meyer et al 2010), cannot fully remove the metal artifacts and may involve new artifacts during the processing because the assumption of the algorithm does not match the physical situation. Some researchers (Kyriakou et al 2010, Oh et al 2018, Lee et al 2019 pointed out that the main cause of metal artifacts is the nonlinear relationship between the polychromatic projection measurements and the line integral of the scanned objects, as described in section 2.1. The solution is to use a polynomial fitting to correct the nonlinear attenuated polychromatic projection to the equivalent linear attenuated monochromatic projection, which is more consistent with the actual physical situation.…”

“…The solution is to use a polynomial fitting to correct the nonlinear attenuated polychromatic projection to the equivalent linear attenuated monochromatic projection, which is more consistent with the actual physical situation. Therefore, we integrated a beam-hardening artifact reduction method (Oh et al 2018) based on a high-order linearization function into our framework. The correction function is shown in equation (5).…”

“…The experimental results of the three typical simulated cases (figure 5) are shown in figures 6-8 respectively. In each case, the original constructed image, the Ground Truth, the result of the Projection-correction (Oh et al 2018), LI-MAR (Kalender et al 1987), FS-MAR (Meyer et al 2012), CNN+NMAR (Kim et al 2022) and the proposed method are illustrated to show the comparative results. The ROI area in rectangle is shown in a higher resolution at the down corner of each image to better observe the difference of these results.…”

“…Meyer et al (2010) used a prior image to preserve the tissue features and normalize the projection with a forward projection of the prior image before interpolation, but it is extremely dependent on the quality of the prior image. Projection linearization correction method (Kyriakou et al 2010, Verburg and Seco 2012, Schuller et al 2015, Oh et al 2018, Lee et al 2019 does not consider the existence of the metal in projection data as the cause of metal artifacts. It considers the nonlinear relationship between the polychromatic projection measurements and the line integral of the scanned objects as the cause of metal artifacts.…”