CT metal artifact reduction method correcting for beam hardening and missing projections

CT metal artefact reduction for hip and shoulder implants using novel algorithms and machine learning: A systematic review with pairwise and network meta-analyses

INTRODUCTION

Many tools have been developed to reduce metal artefacts in computed tomography (CT) images resulting from metallic prosthesis; however, their relative effectiveness in preserving image quality is poorly understood. This paper reviews the literature on novel metal artefact reduction (MAR) methods targeting large metal artefacts in fan-beam CT to examine their effectiveness in reducing metal artefacts and effect on image quality.

METHODS

The PRISMA checklist was used to search for articles in five electronic databases (MEDLINE, Scopus, Web of Science, IEEE, EMBASE). Studies that assessed the effectiveness of recently developed MAR method on fan-beam CT images of hip and shoulder implants were reviewed. Study quality was assessed using the National Institute of Health (NIH) tool. Meta-analyses were conducted in R, and results that could not be meta-analysed were synthesised narratively.

RESULTS

Thirty-six studies were reviewed. Of these, 20 studies proposed statistical algorithms and 16 used machine learning (ML), and there were 19 novel comparators. Network meta-analysis of 19 studies showed that Recurrent Neural Network MAR (RNN-MAR) is more effective in reducing noise (LogOR 20.7; 95 % CI 12.6-28.9) without compromising image quality (LogOR 4.4; 95 % CI -13.8-22.5). The network meta-analysis and narrative synthesis showed novel MAR methods reduce noise more effectively than baseline algorithms, with five out of 23 ML methods significantly more effective than Filtered Back Projection (FBP) (p < 0.05). Computation time varied, but ML methods were faster than statistical algorithms.

CONCLUSION

ML tools are more effective in reducing metal artefacts without compromising image quality and are computationally faster than statistical algorithms. Overall, novel MAR methods were also more effective in reducing noise than the baseline reconstructions.

IMPLICATIONS FOR PRACTICE

Implementation research is needed to establish the clinical suitability of ML MAR in practice.

PND-Net: Physics-Inspired Non-Local Dual-Domain Network for Metal Artifact Reduction

Metal artifacts caused by the presence of metallic implants tremendously degrade the quality of reconstructed computed tomography (CT) images and therefore affect the clinical diagnosis or reduce the accuracy of organ delineation and dose calculation in radiotherapy. Although various deep learning methods have been proposed for metal artifact reduction (MAR), most of them aim to restore the corrupted sinogram within the metal trace, which removes beam hardening artifacts but ignores other components of metal artifacts. In this paper, based on the physical property of metal artifacts which is verified via Monte Carlo (MC) simulation, we propose a novel physics-inspired non-local dual-domain network (PND-Net) for MAR in CT imaging. Specifically, we design a novel non-local sinogram decomposition network (NSD-Net) to acquire the weighted artifact component and develop an image restoration network (IR-Net) to reduce the residual and secondary artifacts in the image domain. To facilitate the generalization and robustness of our method on clinical CT images, we employ a trainable fusion network (F-Net) in the artifact synthesis path to achieve unpaired learning. Furthermore, we design an internal consistency loss to ensure the data fidelity of anatomical structures in the image domain and introduce the linear interpolation sinogram as prior knowledge to guide sinogram decomposition. NSD-Net, IR-Net, and F-Net are jointly trained so that they can benefit from one another. Extensive experiments on simulation and clinical data demonstrate that our method outperforms state-of-the-art MAR methods.

An image-based metal artifact reduction technique utilizing forward projection in computed tomography

The projection data generated via the forward projection of a computed tomography (CT) image (FP-data) have useful potentials in cases where only image data are available. However, there is a question of whether the FP-data generated from an image severely corrupted by metal artifacts can be used for the metal artifact reduction (MAR). The aim of this study was to investigate the feasibility of a MAR technique using FP-data by comparing its performance with that of a conventional robust MAR using projection data normalization (NMARconv). The NMARconv was modified to make use of FP-data (FPNMAR). A graphics processing unit was used to reduce the time required to generate FP-data and subsequent processes. The performances of FPNMAR and NMARconv were quantitatively compared using a normalized artifact index (AIn) for two cases each of hip prosthesis and dental fillings. Several clinical CT images with metal artifacts were processed by FPNMAR. The AIn values of FPNMAR and NMARconv were not significantly different from each other, showing almost the same performance between these two techniques. For all the clinical cases tested, FPNMAR significantly reduced the metal artifacts; thereby, the images of the soft tissues and bones obscured by the artifacts were notably recovered. The computation time per image was ~ 56 ms. FPNMAR, which can be applied to CT images without accessing the projection data, exhibited almost the same performance as that of NMARconv, while consuming significantly shorter processing time. This capability testifies the potential of FPNMAR for wider use in clinical settings.

Spectral CT image reconstruction using a constrained optimization approach-An algorithm for AAPM 2022 spectral CT grand challenge and beyond

BACKGROUND

CT reconstruction is of essential importance in medical imaging. In 2022, the American Association of Physicists in Medicine (AAPM) sponsored a Grand Challenge to investigate the challenging inverse problem of spectral CT reconstruction, with the aim of achieving the most accurate reconstruction results. The authors of this paper participated in the challenge and won as a runner-up team.

PURPOSE

This paper reports details of our PROSPECT algorithm (Prior-based Restricted-variable Optimization for SPEctral CT) and follow-up studies regarding the algorithm's accuracy and enhancement of its convergence speed.

METHODS

We formulated the reconstruction task as an optimization problem. PROSPECT employed a one-step backward iterative scheme to solve this optimization problem by allowing estimation of and correction for the difference between the actual polychromatic projection model and the monochromatic model used in the optimization problem. PROSPECT incorporated various forms of prior information derived by analyzing training data provided by the Grand Challenge to reduce the number of unknown variables. We investigated the impact of projection data precision on the resulting solution accuracy and improved convergence speed of the PROSPECT algorithm by incorporating a beam-hardening correction (BHC) step in the iterative process. We also studied the algorithm's performance under noisy projection data.

RESULTS

Prior knowledge allowed a reduction of the number of unknown variables by85.9 % $85.9\%$ . PROSPECT algorithm achieved the average root of mean square error (RMSE) of3.3 × 10 - 6 $3.3\,\times \,10^{-6}$ in the test data set provided by the Grand Challenge. Performing the reconstruction with the same algorithm but using double-precision projection data reduced RMSE to1.2 × 10 - 11 $1.2\,\times \,10^{-11}$ . Including the BHC step in the PROSPECT algorithm accelerated the iteration process with a 40% reduction in computation time.

CONCLUSIONS

PROSPECT algorithm achieved a high degree of accuracy and computational efficiency.

Quad-Net: Quad-Domain Network for CT Metal Artifact Reduction

Metal implants and other high-density objects in patients introduce severe streaking artifacts in CT images, compromising image quality and diagnostic performance. Although various methods were developed for CT metal artifact reduction over the past decades, including the latest dual-domain deep networks, remaining metal artifacts are still clinically challenging in many cases. Here we extend the state-of-the-art dual-domain deep network approach into a quad-domain counterpart so that all the features in the sinogram, image, and their corresponding Fourier domains are synergized to eliminate metal artifacts optimally without compromising structural subtleties. Our proposed quad-domain network for MAR, referred to as Quad-Net, takes little additional computational cost since the Fourier transform is highly efficient, and works across the four receptive fields to learn both global and local features as well as their relations. Specifically, we first design a Sinogram-Fourier Restoration Network (SFR-Net) in the sinogram domain and its Fourier space to faithfully inpaint metal-corrupted traces. Then, we couple SFR-Net with an Image-Fourier Refinement Network (IFR-Net) which takes both an image and its Fourier spectrum to improve a CT image reconstructed from the SFR-Net output using cross-domain contextual information. Quad-Net is trained on clinical datasets to minimize a composite loss function. Quad-Net does not require precise metal masks, which is of great importance in clinical practice. Our experimental results demonstrate the superiority of Quad-Net over the state-of-the-art MAR methods quantitatively, visually, and statistically. The Quad-Net code is publicly available at https://github.com/longzilicart/Quad-Net.

Quality improvement of images with metal artifact reduction using a noise recovery technique in computed tomography

In metal artifact reduction (MAR) in computed tomography (CT) based on projection data inpainting, X-ray photon noise has not been considered in the inpainting process. This study aims to assess the effectiveness of a MAR technique incorporating noise recovery in such projection data regions, compared with existing MAR techniques based on projection data normalization (NMAR), including one with frequency splitting (FSNMAR). Phantoms simulating hip prostheses and dental fillings were scanned using a 64-row multi slice CT scanner. The projection data was processed by NMAR and NMAR with noise recovery (NRNMAR); the processed data was sent back to the CT system for reconstruction. For the phantoms and clinical cases with hip prostheses and dental fillings, images were reconstructed without MAR, and with NMAR, NRNMAR, and FSNMAR (incorporated in the CT system). To validate the efficacy of noise recovery, noise power spectra (NPSs) were measured from the images of the hip prosthesis phantom with and without metals. The artifact index (AI) was compared between NRNMAR and FSNMAR. The resultant NPSs of NRNMAR were very similar to those of phantom images with no metals, endorsing the efficacy of noise recovery. The NMAR images had unnatural noise textures and FSNMAR caused additional streaks. NRNMAR exhibited some significant improvements in these respects: It reduced the AI by as much as 66.2-88.6% compared to FSNMAR, except for the case of a unilateral prosthesis. In conclusion, NRNMAR, which simply adds white noise to the projection data, would be effective in improving the quality of CT images with metal artifacts reduction.

Semi-supervised segmentation of metal-artifact contaminated industrial CT images using improved CycleGAN

Accurate segmentation of industrial CT images is of great significance in industrial fields such as quality inspection and defect analysis. However, reconstruction of industrial CT images often suffers from typical metal artifacts caused by factors like beam hardening, scattering, statistical noise, and partial volume effects. Traditional segmentation methods are difficult to achieve precise segmentation of CT images mainly due to the presence of these metal artifacts. Furthermore, acquiring paired CT image data required by fully supervised networks proves to be extremely challenging. To address these issues, this paper introduces an improved CycleGAN approach for achieving semi-supervised segmentation of industrial CT images. This method not only eliminates the need for removing metal artifacts and noise, but also enables the direct conversion of metal artifact-contaminated images into segmented images without the requirement of paired data. The average values of quantitative assessment of image segmentation performance can reach 0.96645 for Dice Similarity Coefficient(Dice) and 0.93718 for Intersection over Union(IoU). In comparison to traditional segmentation methods, it presents significant improvements in both quantitative metrics and visual quality, provides valuable insights for further research.

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

Objective.Cone beam computed tomography (CBCT) has been wildly used in clinical treatment of dental diseases. However, patients often have metallic implants in mouth, which will lead to severe metal artifacts in the reconstructed images. To reduce metal artifacts in dental CBCT images, which have a larger amount of data and a limited field of view compared to computed tomography images, a new dental CBCT metal artifact reduction method based on a projection correction and a convolutional neural network (CNN) based image post-processing model is proposed in this paper. Approach.The proposed method consists of three stages: (1) volume reconstruction and metal segmentation in the image domain, using the forward projection to get the metal masks in the projection domain; (2) linear interpolation in the projection domain and reconstruction to build a linear interpolation (LI) corrected volume; (3) take the LI corrected volume as prior and perform the prior based beam hardening correction in the projection domain, and (4) combine the constructed projection corrected volume and LI-volume slice-by-slice in the image domain by two concatenated U-Net based models (CNN1 and CNN2). Simulated and clinical dental CBCT cases are used to evaluate the proposed method. The normalized root means square difference (NRMSD) and the structural similarity index (SSIM) are used for the quantitative evaluation of the method.Main results.The proposed method outperforms the frequency domain fusion method (FS-MAR) and a state-of-art CNN based method on the simulated dataset and yields the best NRMSD and SSIM of 4.0196 and 0.9924, respectively. Visual results on both simulated and clinical images also illustrate that the proposed method can effectively reduce metal artifacts.Significance. This study demonstrated that the proposed dual-domain processing framework is suitable for metal artifact reduction in dental CBCT images.

The effect of common dental fixtures on treatment planning and delivery for head and neck intensity modulated proton therapy

PURPOSE

Proton treatment plan perturbation by common dental fixtures such as amalgams (Am) and porcelain-fused-to-metal (PFM) crowns has, to date, been uncharacterized. Previous studies have been conducted to determine the physical effect of these materials within the beam path for single spots, but their effects on complex treatment plans and clinical anatomy have not yet been quantified. The present manuscript aims to study the effect of Am and PFM fixtures on proton treatment planning in a clinical setting.

METHODS

An anthropomorphic phantom with removable tongue, maxilla, and mandible modules was simulated on a clinical computed tomography (CT) scanner. Spare maxilla modules were modified to include either a 1.5 mm depth central groove occlusal amalgam (Am) or a porcelain-fused-to-metal (PFM) crown, implanted on the first right molar. Modified tongue modules were 3D printed to accommodate several axial or sagittal oriented pieces of EBT-3 film. Clinically representative spot-scanning proton plans were generated in Eclipse v.15.6 using the proton convolution superposition (PCS) algorithm v.15.6.06 using a multi-field optimization (MFO) technique with the goal of delivering a uniform 54 Gy dose to a clinical target volume (CTV) typical of a base-of-tongue (BoT) treatment. A typical geometric beam arrangement of two anterior oblique (AO) beams and a posterior beam was employed. Plans optimized without any material overrides were delivered to the phantom A) without implants; B) with Am fixture; or C) with PFM crown. Plans were also reoptimized and delivered with inclusion of material overrides to equate relative stopping power of the fixture with that of a previously measured result.

RESULTS

Plans exhibit slightly greater dose weight towards AO beams. The optimizer accounted for inclusion of fixture overrides by increasing beam weights to the beam closest to the implant. Film measurements exhibited cold spots directly within the beam path through the fixture in plans with and without overridden materials. Cold spots were somewhat mitigated in plans including overridden materials in the structure set but were not entirely eliminated. Cold spots associated with Am and PFM fixtures were quantified at 17% and 14% for plans without overrides, respectively, and 11% and 9% with using Monte Carlo simulation. Compared with film measurements and Monte Carlo simulation, the treatment planning system underestimates the dose shadowing effect in plans including material overrides.

CONCLUSIONS

Dental fixtures create a dose shadowing effect directly in line with the beam path through the material. This cold spot is partially mitigated by overriding the material to measured relative stopping powers. Due to uncertainties in modeling perturbation through the fixture, the magnitude of the cold spot is underestimated using the institutional TPS when compared to measurement and MC simulation.

Physics-informed sinogram completion for metal artifact reduction in CT imaging

Objective.Metal artifacts in the computed tomography (CT) imaging are unavoidably adverse to the clinical diagnosis and treatment outcomes. Most metal artifact reduction (MAR) methods easily result in the over-smoothing problem and loss of structure details near the metal implants, especially for these metal implants with irregular elongated shapes. To address this problem, we present the physics-informed sinogram completion (PISC) method for MAR in CT imaging, to reduce metal artifacts and recover more structural textures.Approach.Specifically, the original uncorrected sinogram is firstly completed by the normalized linear interpolation algorithm to reduce metal artifacts. Simultaneously, the uncorrected sinogram is also corrected based on the beam-hardening correction physical model, to recover the latent structure information in metal trajectory region by leveraging the attenuation characteristics of different materials. Both corrected sinograms are fused with the pixel-wise adaptive weights, which are manually designed according to the shape and material information of metal implants. To furtherly reduce artifacts and improve the CT image quality, a post-processing frequency split algorithm is adopted to yield the final corrected CT image after reconstructing the fused sinogram.Main results.We qualitatively and quantitatively evaluated the presented PISC method on two simulated datasets and three real datasets. All results demonstrate that the presented PISC method can effectively correct the metal implants with various shapes and materials, in terms of artifact suppression and structure preservation.Significance.We proposed a sinogram-domain MAR method to compensate for the over-smoothing problem existing in most MAR methods by taking advantage of the physical prior knowledge, which has the potential to improve the performance of the deep learning based MAR approaches.

Metal Artifact Reduction in Dental CBCT Images Using Direct Sinogram Correction Combined with Metal Path-Length Weighting

Metal artifacts in dental computed tomography (CT) images, caused by highly X-ray absorbing objects, such as dental implants or crowns, often more severely compromise image readability than in medical CT images. Since lower tube voltages are used for dental CTs in spite of the more frequent presence of metallic objects in the patient, metal artifacts appear more severely in dental CT images, and the artifacts often persist even after metal artifact correction. The direct sinogram correction (DSC) method, which directly corrects the sinogram using the mapping function derived by minimizing the sinogram inconsistency, works well in the case of mild metal artifacts, but it often fails to correct severe metal artifacts. We propose a modified DSC method to reduce severe metal artifacts, and we have tested it on human dental images. We first segment the metallic objects in the CT image, and then we forward-project the segmented metal mask to identify the metal traces in the projection data with computing the metal path length for the rays penetrating the metal mask. In the sinogram correction with the DSC mapping function, we apply the weighting proportional to the metal path length. We have applied the proposed method to the phantom and patient images taken at the X-ray tube voltage of 90 kVp. We observed that the proposed method outperforms the original DSC method when metal artifacts were severe. However, we need further extensive studies to verify the proposed method for various CT scan conditions with many more patient images.

DuDoUFNet: Dual-Domain Under-to-Fully-Complete Progressive Restoration Network for Simultaneous Metal Artifact Reduction and Low-Dose CT Reconstruction

To reduce the potential risk of radiation to the patient, low-dose computed tomography (LDCT) has been widely adopted in clinical practice for reconstructing cross-sectional images using sinograms with reduced x-ray flux. The LDCT image quality is often degraded by different levels of noise depending on the low-dose protocols. The image quality will be further degraded when the patient has metallic implants, where the image suffers from additional streak artifacts along with further amplified noise levels, thus affecting the medical diagnosis and other CT-related applications. Previous studies mainly focused either on denoising LDCT without considering metallic implants or full-dose CT metal artifact reduction (MAR). Directly applying previous LDCT or MAR approaches to the issue of simultaneous metal artifact reduction and low-dose CT (MARLD) may yield sub-optimal reconstruction results. In this work, we develop a dual-domain under-to-fully-complete progressive restoration network, called DuDoUFNet, for MARLD. Our DuDoUFNet aims to reconstruct images with substantially reduced noise and artifact by progressive sinogram to image domain restoration with a two-stage progressive restoration network design. Our experimental results demonstrate that our method can provide high-quality reconstruction, superior to previous LDCT and MAR methods under various low-dose and metal settings.

Low-dimensional Manifold Constrained Disentanglement Network for Metal Artifact Reduction

Deep neural network based methods have achieved promising results for CT metal artifact reduction (MAR), most of which use many synthesized paired images for supervised learning. As synthesized metal artifacts in CT images may not accurately reflect the clinical counterparts, an artifact disentanglement network (ADN) was proposed with unpaired clinical images directly, producing promising results on clinical datasets. However, as the discriminator can only judge if large regions semantically look artifact-free or artifact-affected, it is difficult for ADN to recover small structural details of artifact-affected CT images based on adversarial losses only without sufficient constraints. To overcome the illposedness of this problem, here we propose a low-dimensional manifold (LDM) constrained disentanglement network (DN), leveraging the image characteristics that the patch manifold of CT images is generally low-dimensional. Specifically, we design an LDM-DN learning algorithm to empower the disentanglement network through optimizing the synergistic loss functions used in ADN while constraining the recovered images to be on a low-dimensional patch manifold. Moreover, learning from both paired and unpaired data, an efficient hybrid optimization scheme is proposed to further improve the MAR performance on clinical datasets. Extensive experiments demonstrate that the proposed LDM-DN approach can consistently improve the MAR performance in paired and/or unpaired learning settings, outperforming competing methods on synthesized and clinical datasets.

[An adaptive CT metal artifact reduction algorithm that combines projection interpolation and physical correction]

OBJECTIVE

To propose an adaptive weighted CT metal artifact reduce algorithm that combines projection interpolation and physical correction.

METHODS

A normalized metal projection interpolation algorithm was used to obtain the initial corrected projection data. A metal physical correction model was then introduced to obtain the physically corrected projection data. To verify the effectiveness of the method, we conducted experiments using simulation data and clinical data. For the simulation data, the quantitative indicators PSNR and SSIM were used for evaluation, while for the clinical data, the resultant images were evaluated by imaging experts to compare the artifact-reducing performance of different methods.

RESULTS

For the simulation data, the proposed method improved the PSNR value by at least 0.2 dB and resulted in the highest SSIM value among the methods for comparison. The experiment with the clinical data showed that the imaging experts gave the highest scores of 3.616±0.338 (in a 5-point scale) to the images processed using the proposed method, which had significant better artifact-reducing performance than the other methods (P < 0.001).

CONCLUSION

The metal artifact reduction algorithm proposed herein can effectively reduce metal artifacts while preserving the tissue structure information and reducing the generation of new artifacts.

DICDNet: Deep Interpretable Convolutional Dictionary Network for Metal Artifact Reduction in CT Images

Computed tomography (CT) images are often impaired by unfavorable artifacts caused by metallic implants within patients, which would adversely affect the subsequent clinical diagnosis and treatment. Although the existing deep-learning-based approaches have achieved promising success on metal artifact reduction (MAR) for CT images, most of them treated the task as a general image restoration problem and utilized off-the-shelf network modules for image quality enhancement. Hence, such frameworks always suffer from lack of sufficient model interpretability for the specific task. Besides, the existing MAR techniques largely neglect the intrinsic prior knowledge underlying metal-corrupted CT images which is beneficial for the MAR performance improvement. In this paper, we specifically propose a deep interpretable convolutional dictionary network (DICDNet) for the MAR task. Particularly, we first explore that the metal artifacts always present non-local streaking and star-shape patterns in CT images. Based on such observations, a convolutional dictionary model is deployed to encode the metal artifacts. To solve the model, we propose a novel optimization algorithm based on the proximal gradient technique. With only simple operators, the iterative steps of the proposed algorithm can be easily unfolded into corresponding network modules with specific physical meanings. Comprehensive experiments on synthesized and clinical datasets substantiate the effectiveness of the proposed DICDNet as well as its superior interpretability, compared to current state-of-the-art MAR methods. Code is available at https://github.com/hongwang01/DICDNet.

3D Printing of Tooth Impressions Based on Multi-Detector Computed Tomography Images Combined with Beam Hardening Artifact Reduction in Metal Structures

We investigated the role of metal artifact reduction by taking 3D print impressions using 3D data of Computed Tomography (CT) images based on the algorithm applied. We manufactured a phantom of a human mandible tooth made of gypsum and nickel alloy to measure the metal artifacts. CT images were obtained by changing the phantom tube voltage and tube current. The signal intensity of the image generated by the metal artifacts before and after the iterative metal artifact reduction algorithm (iMAR) was measured. A 3D printing process was performed after converting the images, before and after iMAR application, into STL files using InVesalius version 3.1.1 by selecting the conditions that minimized the effect of the artifact. Regarding metal artifacts, the Hounsfield unit (HU) value showed low as the tube voltage increased. The iMAR-applied images acquired under the same conditions showed a significantly lower HU. The artifacts, in the form of flashes, persisted in the 3D-printed product of the image not subjected to iMAR, but were largely removed in the 3D-printed product following iMAR application. In this study, the application of iMAR and data acquired using high tube voltage eliminated a significant portion of the metal artifacts, resulting in an impression shape that was consistent with the human body.

Physical characterization of therapeutic proton delivery through common dental materials

PURPOSE

Dental fixtures are commonplace in an aging, radiation treatment population. The current, local standard of practice in particle therapy is to employ treatment geometries to avoid delivery through implanted dental fixtures. The present study aims to observe the physical effect of delivering therapeutic proton beams through common dental fixture materials as prelude to an eventual goal of assessing the feasibility of using treatment geometries not specified for avoidance of oral implants. A sampling of common dental materials was selected based on prosthodontic consult and was evaluated in terms of relative stopping power and three-dimensional (3D) dose perturbation.

METHODS

Amalgams, porcelain-fused-to-metal (PFM) crowns consisting of zirconia and non-noble base metals, and lithium disilicate implants were chosen for analysis. Theoretical stopping power (S) and mass stopping power (S/ρ) were calculated using the Stopping and Range of Ions in Matter (SRIM) application, basing stoichiometric compositions of each fixture on published materials data. S and S/ρ were calculated for a range of historically available compositions of amalgams from 1900 until the current era. The perturbance of S and S/ρ as a function of clinically relevant ranges of amalgam compositions for the modern era was analyzed. Water equivalent thickness (WET) and relative stopping power (S_rel ) of each material was measured for a clinical spot-scanning proton beam with monoenergies of 159.9 and 228.8 MeV with a multi-layer ionization chamber (MLIC). Subsequently, 3D dose perturbation was assessed by delivering proton beams through a custom phantom designed to simulate both en-face and on-edge treatment geometries through the selected materials. A treatment plan mimicking the experimental delivery was constructed in the institutional treatment planning system and calculated using TOPAS based Monte Carlo Simulation (MCS). Experimental results were used to validate the MCS. Finally, TPS outputs were compared to MCS to determine the accuracy of the dose calculation model.

RESULTS

Historical compositions of amalgams ranged in S from 44.8 to 42.9 MeV/cm, with the greatest deviation being observed for the 1900-1959 era. Deviation as a function of amalgam composition from the modern era was most sensitive to proportion of Hg, accounting for deviations up to -4.2% at the greatest clinically relevant concentration. S/ρ was not found to vary greatly between each porcelain and metal alloy material for porcelain-fused-to-metal (PFM) type crowns. Relative stopping powers ranged between 1.3 and 5.4 for all studied materials, suggesting substantial changes in proton range with respect to water. Film measurements of pristine spots confirm dose perturbance and shortening of proton range, with an upstream shift of each Bragg peak being observed directly behind the installed fixture. At high energies, cold spots were found in all cases directly behind each material feature with a medial fill-in of dose occurring distally. Qualitative agreement of spot perturbance was confirmed between film measurements and MCS. Finally, when comparing integrated depth doses (IDD) by summing over all axial directions, good agreement is observed between TPS and MCS.

CONCLUSIONS

All dental materials studied substantially perturbed the dosimetry of pristine proton spots both in terms of WET/S_rel as well as the spatial distribution of dose. Proton range was quantifiably shortened, and each dental material affected a cold spot directly behind the object with medial dose back-filling was observed distally. Monte Carlo simulations and Eclipse dose calculations exhibited good agreement with measurements, suggesting that treatment planning without employing avoidance strategies may be possible with further investigation. This article is protected by copyright. All rights reserved.

Inner-ear augmented metal artifact reduction with simulation-based 3D generative adversarial networks

Metal Artifacts creates often difficulties for a high quality visual assessment of post-operative imaging in computed tomography (CT). A vast body of methods have been proposed to tackle this issue, but these methods were designed for regular CT scans and their performance is usually insufficient when imaging tiny implants. In the context of post-operative high-resolution CT imaging, we propose a 3D metal artifact reduction algorithm based on a generative adversarial neural network. It is based on the simulation of physically realistic CT metal artifacts created by cochlea implant electrodes on preoperative images. The generated images serve to train a 3D generative adversarial networks for artifacts reduction. The proposed approach was assessed qualitatively and quantitatively on clinical conventional and cone beam CT of cochlear implant postoperative images. These experiments show that the proposed method outperforms other general metal artifact reduction approaches.

Relationship Between Glenoid Component Shift and Osteolysis After Anatomic Total Shoulder Arthroplasty: Three-Dimensional Computed Tomography Analysis

BACKGROUND

The purpose of this study was to evaluate glenoid component position and radiolucency following anatomic total shoulder arthroplasty (TSA) using sequential 3-dimensional computed tomography (3D CT) analysis.

METHODS

In a series of 152 patients (42 Walch A1, 16 A2, 7 B1, 49 B2, 29 B3, 3 C1, 3 C2, and 3 D glenoids) undergoing anatomic TSA with a polyethylene glenoid component, sequential 3D CT analysis was performed preoperatively (CT1), early postoperatively (CT2), and at a minimum 2-year follow-up (CT3). Glenoid component shift was defined as a change in component version or inclination of ≥3° from CT2 to CT3. Glenoid component central anchor peg osteolysis (CPO) was assessed at CT3. Factors associated with glenoid component shift and CPO were evaluated.

RESULTS

Glenoid component shift occurred from CT2 to CT3 in 78 (51%) of the 152 patients. CPO was seen at CT3 in 19 (13%) of the 152 patients, including 15 (19%) of the 78 with component shift. Walch B2 glenoids with a standard component and glenoids with higher preoperative retroversion were associated with a higher rate of shift, but not of CPO. B3 glenoids with an augmented component and glenoids with greater preoperative joint-line medialization were associated with CPO, but not with shift. More glenoid component joint-line medialization from CT2 to CT3 was associated with higher rates of shift and CPO. A greater absolute change in glenoid component inclination from CT2 to CT3 and a combined absolute glenoid component version and inclination change from CT2 to CT3 were associated with CPO. Neither glenoid component shift nor CPO was associated with worse clinical outcomes.

CONCLUSIONS

Postoperative 3D CT analysis demonstrated that glenoid component shift commonly occurs following anatomic TSA, with increased inclination the most common direction. Most (81%) of the patients with glenoid component shift did not develop CPO. Longer follow-up is needed to determine the relationships of glenoid component shift and CPO with loosening over time.

LEVEL OF EVIDENCE

Therapeutic Level IV. See Instructions for Authors for a complete description of levels of evidence.

Dosimetric impact of commercial CT metal artifact reduction algorithms and a novel in-house algorithm for proton therapy of head and neck cancer

PURPOSE

To compare the dosimetric impact of all major commercial vendors' metal artifact reduction (MAR) algorithms to one another, as well as to a novel in-house technique (AMPP) using an anthropomorphic head phantom.

MATERIALS AND METHODS

The phantom was an Alderson phantom, modified to allow for artifact-filled and baseline (no artifacts) computed tomography (CT) scans using teeth capsules made with metal amalgams or bone-equivalent materials. It also included a cylindrical insert that was accessible from the bottom of the neck and designed to introduce soft tissue features into the phantom that were used in the analysis. The phantom was scanned with the metal teeth in place using each respective vendor's MAR algorithm: OMAR (Philips), iMAR (Siemens), SEMAR (Canon), and SmartMAR (GE); the AMPP algorithm was designed in-house. Uncorrected and baseline (bone-equivalent teeth) image sets were also acquired using a Siemens scanner. Proton spot scanning treatment plans were designed on the baseline image set for five targets in the phantom. Once optimized, the proton beams were copied onto the different artifact-corrected image sets, with no reoptimization of the beams' parameters, to evaluate dose distribution differences in the different MAR-corrected and -uncorrected image sets. Dose distribution differences were evaluated by comparing dose-volume histogram (DVH) metrics, including planning target volume D95 and clinical target volume D99 coverages, V100, D0.03cc, and heterogeneity indexes, along with a qualitative and water equivalent thickness (WET) analysis.

RESULTS

Uncorrected CT metal artifacts and commercial MAR algorithms negatively impacted the proton dose distributions of all five target shapes and locations in an inconsistent manner, sometimes overdosing by as much as 11.1% (D0.03) or underdosing by as much as 11.7% (V100) the planning target volumes. The AMPP-corrected images, however, provided dose distributions that consistently agreed with the baseline dose distribution. The dosimetry results also suggest that the commercial MAR algorithms' performances varied more with target location and less with target shape. Once relocated further from the metal, the target showed dose distributions that agreed more with the baseline for all commercial solutions, improving the overdosing by as much as 6%, implying inadequate HU correction from commercial MAR algorithms. In comparison to the baseline, HU profile shapes were considerably altered by commercial algorithms and reference values showed differences that represent stopping power percentage differences of 2.7-10%. The AMPP algorithm plans showed the smallest WET differences with the baseline (0.06 cm on average), while the commercial image sets created differences that ranged from 0.11 to 0.54 cm.

CONCLUSIONS

Computed tomography metal artifacts negatively impacted proton dose distributions on all five targets analyzed. The commercial MAR solutions performed inconsistently throughout all targets compared to the metal-free baseline. A lack of CTV coverage and an increased number of hotspots were observed throughout all commercial solutions. Dose distribution errors were related to the proximity to the artifacts, demonstrating the inability of commercial techniques to adequately correct severe artifacts. In contrast, AMPP consistently showed dose distributions that best matched the baseline, likely because it makes use of accurate HU information, as opposed to interpolated data like commercial algorithms.

Superiorized method for metal artifact reduction

PURPOSE

Metal artifact reduction (MAR) is a challenging problem in computed tomography (CT) imaging. A popular class of MAR methods replace sinogram measurements that are corrupted by metal with artificial data, typically generated from some combination of interpolation along with other heuristics. While these "projection completion" approaches are successful in eliminating severe artifacts, secondary artifacts may be introduced by the artificial data. In this paper, we propose an approach which uses projection completion to generate a prior image, which is then incorporated into an iterative reconstruction algorithm based on the superiorization framework. The rationale is that the image produced by the iterative algorithm can inherit the desirable properties of the prior image, while also reducing secondary artifacts.

METHODS

The prior image is reconstructed using normalized metal artifact reduction (NMAR), a popular projection completion approach. The iterative algorithm is a modified version of the simultaneous algebraic reconstruction technique (SART), which reduces artifacts by incorporating a polyenergetic forward model, least-squares weighting, and superiorization. The penalty function used for superiorization is a weighted average between a total variation (TV) term and a term promoting similarity with the prior image, similar to penalty functions used in prior image constrained compressive sensing (PICCS). Because the prior is largely free of severe metal artifacts, these artifacts are discouraged from arising during iterative reconstruction; additionally, because the iterative approach uses the original projection data, it is able to recover information that is lost during the NMAR process.

RESULTS

We perform numerical experiments modeling a simple geometric object, as well as several more realistic scenarios such as metal pins, bilateral hip implants, and dental fillings placed within an anatomical phantom. The proposed iterative algorithm is largely successful at eliminating severe metal artifacts as well as secondary artifacts introduced by the NMAR process, especially lost edges of bone structures in the neighborhood of the metal regions. In one case modeling severe photon starvation, the NMAR algorithm is found to provide better results.

CONCLUSION

The proposed algorithm is effective in applying the superiorization methodology to the problem of MAR, providing better results than both NMAR and a purely total variation-based superiorization approach in nearly all cases.

The dosimetric impact of stabilizing spinal implants in radiotherapy treatment planning with protons and photons: standard titanium alloy vs. radiolucent carbon-fiber-reinforced PEEK systems

Background

Throughout the last years, carbon‐fibre‐reinforced PEEK (CFP) pedicle screw systems were introduced to replace standard titanium alloy (Ti) implants for spinal instrumentation, promising improved radiotherapy (RT) treatment planning accuracy. We compared the dosimetric impact of both implants for intensity modulated proton (IMPT) and volumetric arc photon therapy (VMAT), with the focus on uncertainties in Hounsfield unit assignment of titanium alloy.

Methods

Retrospective planning was performed on CT data of five patients with Ti and five with CFP implants. Carbon‐fibre‐reinforced PEEK systems comprised radiolucent pedicle screws with thin titanium‐coated regions and titanium tulips. For each patient, one IMPT and one VMAT plan were generated with a nominal relative stopping power (SP) (IMPT) and electron density (ρ) (VMAT) and recalculated onto the identical CT with increased and decreased SP or ρ by ±6% for the titanium components.

Results

Recalculated VMAT dose distributions hardly deviated from the nominal plans for both screw types. IMPT plans resulted in more heterogeneous target coverage, measured by the standard deviation σ inside the target, which increased on average by 7.6 ± 2.3% (Ti) vs 3.4 ± 1.2% (CFP). Larger SPs lead to lower target minimum doses, lower SPs to higher dose maxima, with a more pronounced effect for Ti screws.

Conclusions

While VMAT plans showed no relevant difference in dosimetric quality between both screw types, IMPT plans demonstrated the benefit of CFP screws through a smaller dosimetric impact of CT‐value uncertainties compared to Ti. Reducing metal components in implants will therefore improve dose calculation accuracy and lower the risk for tumor underdosage.

Impact of different iterative metal artifact reduction (iMAR) algorithms on PET/CT attenuation correction after port implementation

PURPOSE

To evaluate the effect of various interactive metal artifact reduction (iMAR) algorithms on attenuation correction in the vicinity of port chambers in PET/CT.

MATERIAL AND METHODS

In this prospective study, 30 oncological patients (12 female, 18 male, mean age 59.6 ± 10.5y) with implanted port chambers undergoing 18F-FDG PET/CT were included. CT images were reconstructed with standard weighted filtered back projection (WFBP) and three different iMAR algorithms (hip, dental filling (DF) and pacemaker (PM)). PET attenuation correction was performed with all four CT datasets. SUVmean, SUVmax and HU measurements were performed in fat and muscle tissue in the vicinity of the port chamber at the location of the strongest bright and dark band artifacts. Differences between HU and SUV values across all CT- and PET-images were investigated using a paired t-test. Bonferroni correction was used to prevent alpha-error accumulation (p < 0.008).

RESULTS

In comparison to WFBP (fat: 94.2 ± 53.9 HU, muscle: 197.6 ± 49.2 HU) all three iMAR algorithms led to a decrease of HU in bright band artifacts. iMAR-DF led to a decrease of 159.2 % (fat: -51.9 ± 58.5 HU, muscle: 94.5 ± 55.3 HU), iMAR-hip of 138.3 % (fat: -30.3 ± 58.5, muscle: 70.4 ± 28.8) and iMAR-PM of 122.3 % (fat: -21.2 ± 47.2 HU, muscle: 72.5 ± 25.1 HU; for all p < 0.008). There was no significant effect of iMAR on SUV measurements in comparison to WFBP.

CONCLUSION

iMAR leads to a significant change of HU values in artifacts caused by port catheter chambers in comparison to WFBP. However, no significant differences in attenuation correction and consecutive changes in SUV measurements can be observed.

Clinical application of multi-material artifact reduction (MMAR) technique in Revolution CT to reduce metallic dental artifacts

Background

This study aimed to explore the performance of Revolution CT virtual monoenergetic images (VMI) combined with the multi-material artifact reduction (MMAR) technique in reducing metal artifacts in oral and maxillofacial imaging.

Results

There were significant differences in image quality scores between VMI + MMAR images and VMI+MARS (multiple artifact reduction system) images at each monochromatic energy level (p = 0.000). Compared with the MARS technology, the MMAR technology further reduced metal artifacts and improved the image quality. At VMI_{90 keV} and VMI_{110 keV}, the SD, CNR, and AI in the Revolution CT group were significantly lower than in the Discovery CT, but no significant differences in these parameters were found between two groups at VMI_{50 keV}, VMI_{70 keV}, and VMI_{130 keV} (p > 0.05). The attenuation was comparable between two groups at any energy level (p > 0.05).

Conclusions

Compared with the MARS reconstruction technique of Discovery CT, the MMAR technique of Revolution CT is better to reduce the artifacts of dental implants in oral and maxillofacial imaging, which improves the image quality and the diagnostic value of surrounding soft tissues.

Impact of different metal artifact reduction techniques on attenuation correction in 18F-FDG PET/CT examinations

OBJECTIVE

To evaluate the impact of different metal artifact reduction (MAR) algorithms on Hounsfield unit (HU) and standardized uptake values (SUV) in a phantom setting and verify these results in patients with metallic implants undergoing oncological PET/CT examinations.

METHODS AND MATERIALS

In this prospective study, PET-CT examinations of 28 oncological patients (14 female, 14 male, mean age 69.5 ± 15.2y) with 38 different metal implants were included. CT datasets were reconstructed using standard weighted filtered back projection (WFBP) without MAR, MAR in image space (MARIS) and iterative MAR (iMAR, hip algorithm). The three datasets were used for PET attenuation correction. SUV and HU measurements were performed at the site of the most prominent bright and dark band artifacts. Differences between HU and SUV values across the different reconstructions were compared using paired t-tests. Bonferroni correction was used to prevent alpha-error accumulation (p < 0.017).

RESULTS

For bright band artifacts, MARIS led to a non-significant mean decrease of 12.0% (345 ± 315 HU) in comparison with WFBP (391 ± 293 HU), whereas iMAR led to a significant decrease of 68.3% (125 ± 185 HU, p < 0.017). For SUVmean, MARIS showed no significant effect in comparison with WFBP (WFBP: 0.99 ± 0.40, MARIS: 0.96 ± 0.39), while iMAR led to a significant decrease of 11.1% (0.88 ± 0.35, p < 0.017). Similar results were observed for dark band artifacts.

CONCLUSION

iMAR significantly reduces artifacts caused by metal implants in CT and thus leads to a significant change of SUV measurements in bright and dark band artifacts compared with WFBP and MARIS, thus probably improving PET quantification.

ADVANCES IN KNOWLEDGE

The present work indicates that MAR algorithms such as iMAR algorithm in integrated PET/CT scanners are useful to improve CT image quality as well as PET quantification in the evaluation of tracer uptake adjacent to large metal implants. A detailed analysis of oncological patients with various large metal implants using different MAR algorithms in PET/CT has not been conducted yet.

Known-component 3D image reconstruction for improved intraoperative imaging in spine surgery: A clinical pilot study

PURPOSE

Intraoperative imaging plays an increased role in support of surgical guidance and quality assurance for interventional approaches. However, image quality sufficient to detect complications and provide quantitative assessment of the surgical product is often confounded by image noise and artifacts. In this work, we translated a three-dimensional model-based image reconstruction (referred to as "Known-Component Reconstruction," KC-Recon) for the first time to clinical studies with the aim of resolving both limitations.

METHODS

KC-Recon builds upon a penalized weighted least-squares (PWLS) method by incorporating models of surgical instrumentation ("known components") within a joint image registration-reconstruction process to improve image quality. Under IRB approval, a clinical pilot study was conducted with 17 spine surgery patients imaged under informed consent using the O-arm cone-beam CT system (Medtronic, Littleton MA) before and after spinal instrumentation. Volumetric images were generated for each patient using KC-Recon in comparison to conventional filtered backprojection (FBP). Imaging performance prior to instrumentation ("preinstrumentation") was evaluated in terms of soft-tissue contrast-to-noise ratio (CNR) and spatial resolution. The quality of images obtained after the instrumentation ("postinstrumentation") was assessed by quantifying the magnitude of metal artifacts (blooming and streaks) arising from pedicle screws. The potential low-dose advantages of the algorithm were tested by simulating low-dose data (down to one-tenth of the dose of standard protocols) from images acquired at normal dose.

RESULTS

Preinstrumentation images (at normal clinical dose and matched resolution) exhibited an average 24.0% increase in soft-tissue CNR with KC-Recon compared to FBP (N = 16, P = 0.02), improving visualization of paraspinal muscles, major vessels, and other soft-tissues about the spine and abdomen. For a total of 72 screws in postinstrumentation images, KC-Recon yielded a significant reduction in metal artifacts: 66.3% reduction in overestimation of screw shaft width due to blooming (P < 0.0001) and reduction in streaks at the screw tip (65.8% increase in attenuation accuracy, P < 0.0001), enabling clearer depiction of the screw within the pedicle and vertebral body for an assessment of breach. Depending on the imaging task, dose reduction up to an order of magnitude appeared feasible while maintaining soft-tissue visibility and metal artifact reduction.

CONCLUSIONS

KC-Recon offers a promising means to improve visualization in the presence of surgical instrumentation and reduce patient dose in image-guided procedures. The improved soft-tissue visibility could facilitate the use of cone-beam CT to soft-tissue surgeries, and the ability to precisely quantify and visualize instrument placement could provide a valuable check against complications in the operating room (cf., postoperative CT).

The effects of the orthopedic metal artifact reduction (O-MAR) algorithm on contouring and dosimetry of head and neck radiotherapy patients

Metallic objects, such as dental fillings, cause artifacts in computed tomography (CT) scans. We quantify the contouring and dosimetric effects of Orthopedic Metal Artifact Reduction (O-MAR), in head and neck radiotherapy. The ease of organ contouring was assessed by having a radiation oncologist identify the CT data set with or without O-MAR for each of 28 patients that was easier to contour. The effect on contouring was quantified further by having the physician recontour parotid glands, previously drawn by him on the O-MAR scans, on uncorrected scans, and calculating the Dice coefficent (a measure of overlap) for the contours. Radiotherapy plans originally generated on scans reconstructed with O-MAR were recalculated on scans without metal artifact correction. The study was done using the Analytical Anisotropic Algorithm (AAA) dose calculation algorithm. The 15 patients with a planning target volume (PTV) extending to the same slice as the artifacts were used for this part of the study. The normal tissue doses were not significantly affected. The PTV mean dose and V95 were not affected, but the cold spots became less severe in the O-MAR corrected plans, with the minimum point dose on average being 4.1% higher. In 79% of the cases, the radiation oncologist identified the O-MAR scan as easier to contour; in 11% he chose the uncorrected scan and in 11% the scans were judged to have equal quality. A total of nine parotid glands (on both scans-18 contours in total) in 5 patients were recontoured. The average Dice coefficient for parotids drawn with and without O-MAR was found to be 0.775 +/- 0.045. The O-MAR algorithm does not produce a significant dosimetric effect in head and neck plans when using the AAA dose calculation algorithm. It can therefore be used for improved contouring accuracy without updating the critical structure tolerance doses and target coverage expectations.

An additional tilted-scan-based CT metal-artifact-reduction method for radiation therapy planning

Purpose

As computed tomography (CT) imaging is the most commonly used modality for treatment planning in radiation therapy, metal artifacts in the planning CT images may complicate the target delineation and reduce the dose calculation accuracy. Although current CT scanners do provide certain correction steps, it is a common understanding that there is not a universal solution yet to the metal artifact reduction (MAR) in general. Particularly noting the importance of MAR for radiation treatment planning, we propose a novel MAR method in this work that recruits an additional tilted CT scan and synthesizes nearly metal‐artifact‐free CT images.

Methods

The proposed method is based on the facts that the most pronounced metal artifacts in CT images show up along the x‐ray beam direction traversing multiple metallic objects and that a tilted CT scan can provide complementary information free of such metal artifacts in the earlier scan. Although the tilted CT scan would contain its own metal artifacts in the images, the artifacts may manifest in a different fashion leaving a chance to concatenate the two CT images with the metal artifacts much suppressed. We developed an image processing technique that uses the structural similarity (SSIM) for suppressing the metal artifacts. On top of the additional scan, we proposed to use an existing MAR method for each scan if necessary to further suppress the metal artifacts.

Results

The proposed method was validated by a simulation study using the pelvic region of an XCAT numerical phantom and also by an experimental study using the head part of the Rando phantom. The proposed method was found to effectively reduce the metal artifacts. Quantitative analyses revealed that the proposed method reduced the mean absolute percentages of the error by up to 86% and 89% in the simulation and experimental studies, respectively.

Conclusions

It was confirmed that the proposed method, using complementary information acquired from an additional tilted CT scan, can provide nearly metal‐artifact‐free images for the treatment planning.

Sequential 3-dimensional computed tomography analysis of implant position following total shoulder arthroplasty

BACKGROUND

Detection of postoperative component position and implant shift following total shoulder arthroplasty (TSA) can be challenging using routine imaging. The purpose of this study was to evaluate glenoid component position over time using 3-dimensional computed tomography (CT) analysis with minimum 2-year follow-up.

METHODS

Twenty patients underwent primary TSA with sequential CT scanning of the shoulder: a preoperative study, an immediate postoperative study within 2 weeks of surgery, and a postoperative study performed at minimum 2-year follow-up (CT3). Postoperative glenoid component position and central peg osteolysis were assessed across the immediate postoperative CT scan and CT3. Glenoids with evidence of component shift and/or grade 1 central peg osteolysis on CT3 were considered at risk of loosening.

RESULTS

Of the patients, 7 (35%) showed evidence of glenoid components at risk of loosening on CT3, 6 with component shift (3 with increased inclination alone, 1 with increased retroversion alone, and 2 with both increased inclination and retroversion). Significantly more patients with glenoid component shift had grade 1 central peg osteolysis on CT3 compared with those without shift (83% vs 7%, P = .002). One clinical failure occurred, with the patient undergoing revision to reverse TSA for rotator cuff deficiency.

CONCLUSIONS

Three-dimensional CT imaging analysis following TSA identified changes in glenoid component position over time, with inclination being the most common direction of shift and grade 1 central peg osteolysis commonly associated with shift. These findings raise concern for glenoids at risk of loosening, but further follow-up is needed to determine the long-term clinical impact of these findings.

Convolutional Neural Network Based Metal Artifact Reduction in X-Ray Computed Tomography

In the presence of metal implants, metal artifacts are introduced to x-ray computed tomography CT images. Although a large number of metal artifact reduction (MAR) methods have been proposed in the past decades, MAR is still one of the major problems in clinical x-ray CT. In this paper, we develop a convolutional neural network (CNN)-based open MAR framework, which fuses the information from the original and corrected images to suppress artifacts. The proposed approach consists of two phases. In the CNN training phase, we build a database consisting of metal-free, metal-inserted and pre-corrected CT images, and image patches are extracted and used for CNN training. In the MAR phase, the uncorrected and pre-corrected images are used as the input of the trained CNN to generate a CNN image with reduced artifacts. To further reduce the remaining artifacts, water equivalent tissues in a CNN image are set to a uniform value to yield a CNN prior, whose forward projections are used to replace the metal-affected projections, followed by the FBP reconstruction. The effectiveness of the proposed method is validated on both simulated and real data. Experimental results demonstrate the superior MAR capability of the proposed method to its competitors in terms of artifact suppression and preservation of anatomical structures in the vicinity of metal implants.

Markov random field segmentation for industrial computed tomography with metal artefacts

X-ray Computed Tomography (XCT) has become an important tool for industrial measurement and quality control through its ability to measure internal structures and volumetric defects. Segmentation of constituent materials in the volume acquired through XCT is one of the most critical factors that influence its robustness and repeatability. Highly attenuating materials such as steel can introduce artefacts in CT images that adversely affect the segmentation process, and results in large errors during quantification. This paper presents a Markov Random Field (MRF) segmentation method as a suitable approach for industrial samples with metal artefacts. The advantages of employing the MRF segmentation method are shown in comparison with Otsu thresholding on CT data from two industrial objects.

Dual-energy CT with virtual monochromatic images and metal artifact reduction software for reducing metallic dental artifacts

Background Metallic dental prostheses may degrade image quality on head and neck computed tomography (CT). However, there is little information available on the use of dual-energy CT (DECT) and metal artifact reduction software (MARS) in the head and neck regions to reduce metallic dental artifacts. Purpose To assess the usefulness of DECT with virtual monochromatic imaging and MARS to reduce metallic dental artifacts. Material and Methods DECT was performed using fast kilovoltage (kV)-switching between 80-kV and 140-kV in 20 patients with metallic dental prostheses. CT data were reconstructed with and without MARS, and with synthesized monochromatic energy in the range of 40-140-kiloelectron volt (keV). For quantitative analysis, the artifact index of the tongue, buccal, and parotid areas was calculated for each scan. For qualitative analysis, two radiologists evaluated 70-keV and 100-keV images with and without MARS for tongue, buccal, parotid areas, and metallic denture. The locations and characteristics of the MARS-related artifacts, if any, were also recorded. Results DECT with MARS markedly reduced metallic dental artifacts and improved image quality in the buccal area ( P < 0.001) and the tongue ( P < 0.001), but not in the parotid area. The margin and internal architecture of the metallic dentures were more clearly delineated with MARS ( P < 0.001) and in the higher-energy images than in the lower-energy images ( P = 0.042). MARS-related artifacts most commonly occurred in the deep center of the neck. Conclusion DECT with MARS can reduce metallic dental artifacts and improve delineation of the metallic prosthesis and periprosthetic region.

Improving image quality for digital breast tomosynthesis: an automated detection and diffusion-based method for metal artifact reduction

In digital breast tomosynthesis (DBT), the high-attenuation metallic clips marking a previous biopsy site in the breast cause errors in the estimation of attenuation along the ray paths intersecting the markers during reconstruction, which result in interplane and inplane artifacts obscuring the visibility of subtle lesions. We proposed a new metal artifact reduction (MAR) method to improve image quality. Our method uses automatic detection and segmentation to generate a marker location map for each projection (PV). A voting technique based on the geometric correlation among different PVs is designed to reduce false positives (FPs) and to label the pixels on the PVs and the voxels in the imaged volume that represent the location and shape of the markers. An iterative diffusion method replaces the labeled pixels on the PVs with estimated tissue intensity from the neighboring regions while preserving the original pixel values in the neighboring regions. The inpainted PVs are then used for DBT reconstruction. The markers are repainted on the reconstructed DBT slices for radiologists' information. The MAR method is independent of reconstruction techniques or acquisition geometry. For the training set, the method achieved 100% success rate with one FP in 19 views. For the test set, the success rate by view was 97.2% for core biopsy microclips and 66.7% for clusters of large post-lumpectomy markers with a total of 10 FPs in 58 views. All FPs were large dense benign calcifications that also generated artifacts if they were not corrected by MAR. For the views with successful detection, the metal artifacts were reduced to a level that was not visually apparent in the reconstructed slices. The visibility of breast lesions obscured by the reconstruction artifacts from the metallic markers was restored.

A metal artifact reduction algorithm in CT using multiple prior images by recursive active contour segmentation

We propose a novel metal artifact reduction (MAR) algorithm for CT images that completes a corrupted sinogram along the metal trace region. When metal implants are located inside a field of view, they create a barrier to the transmitted X-ray beam due to the high attenuation of metals, which significantly degrades the image quality. To fill in the metal trace region efficiently, the proposed algorithm uses multiple prior images with residual error compensation in sinogram space. Multiple prior images are generated by applying a recursive active contour (RAC) segmentation algorithm to the pre-corrected image acquired by MAR with linear interpolation, where the number of prior image is controlled by RAC depending on the object complexity. A sinogram basis is then acquired by forward projection of the prior images. The metal trace region of the original sinogram is replaced by the linearly combined sinogram of the prior images. Then, the additional correction in the metal trace region is performed to compensate the residual errors occurred by non-ideal data acquisition condition. The performance of the proposed MAR algorithm is compared with MAR with linear interpolation and the normalized MAR algorithm using simulated and experimental data. The results show that the proposed algorithm outperforms other MAR algorithms, especially when the object is complex with multiple bone objects.

Polyenergetic known-component CT reconstruction with unknown material compositions and unknown x-ray spectra

Metal artifacts can cause substantial image quality issues in computed tomography. This is particularly true in interventional imaging where surgical tools or metal implants are in the field-of-view. Moreover, the region-of-interest is often near such devices which is exactly where image quality degradations are largest. Previous work on known-component reconstruction (KCR) has shown the incorporation of a physical model (e.g. shape, material composition, etc) of the metal component into the reconstruction algorithm can significantly reduce artifacts even near the edge of a metal component. However, for such approaches to be effective, they must have an accurate model of the component that include energy-dependent properties of both the metal device and the CT scanner, placing a burden on system characterization and component material knowledge. In this work, we propose a modified KCR approach that adopts a mixed forward model with a polyenergetic model for the component and a monoenergetic model for the background anatomy. This new approach called Poly-KCR jointly estimates a spectral transfer function associated with known components in addition to the background attenuation values. Thus, this approach eliminates both the need to know component material composition a prior as well as the requirement for an energy-dependent characterization of the CT scanner. We demonstrate the efficacy of this novel approach and illustrate its improved performance over traditional and model-based iterative reconstruction methods in both simulation studies and in physical data including an implanted cadaver sample.

Image-based Metal Artifact Reduction in X-ray Computed Tomography utilizing Local Anatomical Similarity

X-ray computed tomography (CT) is widely used in radiation therapy treatment planning in recent years. However, metal implants such as dental fillings and hip prostheses can cause severe bright and dark streaking artifacts in reconstructed CT images. These artifacts decrease image contrast and degrade HU accuracy, leading to inaccuracies in target delineation and dose calculation. In this work, a metal artifact reduction method is proposed based on the intrinsic anatomical similarity between neighboring CT slices. Neighboring CT slices from the same patient exhibit similar anatomical features. Exploiting this anatomical similarity, a gamma map is calculated as a weighted summation of relative HU error and distance error for each pixel in an artifact-corrupted CT image relative to a neighboring, artifact-free image. The minimum value in the gamma map for each pixel is used to identify an appropriate pixel from the artifact-free CT slice to replace the corresponding artifact-corrupted pixel. With the proposed method, the mean CT HU error was reduced from 360 HU and 460 HU to 24 HU and 34 HU on head and pelvis CT images, respectively. Dose calculation accuracy also improved, as the dose difference was reduced from greater than 20% to less than 4%. Using 3%/3mm criteria, the gamma analysis failure rate was reduced from 23.25% to 0.02%. An image-based metal artifact reduction method is proposed that replaces corrupted image pixels with pixels from neighboring CT slices free of metal artifacts. This method is shown to be capable of suppressing streaking artifacts, thereby improving HU and dose calculation accuracy.

Iterative metal artefact reduction in CT: can dedicated algorithms improve image quality after spinal instrumentation?

AIM

To investigate the value of dedicated computed tomography (CT) iterative metal artefact reduction (iMAR) algorithms in patients after spinal instrumentation.

MATERIALS AND METHODS

Post-surgical spinal CT images of 24 patients performed between March 2015 and July 2016 were retrospectively included. Images were reconstructed with standard weighted filtered back projection (WFBP) and with two dedicated iMAR algorithms (iMAR-Algo1, adjusted to spinal instrumentations and iMAR-Algo2, adjusted to large metallic hip implants) using a medium smooth kernel (B30f) and a sharp kernel (B70f). Frequencies of density changes were quantified to assess objective image quality. Image quality was rated subjectively by evaluating the visibility of critical anatomical structures including the central canal, the spinal cord, neural foramina, and vertebral bone.

RESULTS

Both iMAR algorithms significantly reduced artefacts from metal compared with WFBP (p<0.0001). Results of subjective image analysis showed that both iMAR algorithms led to an improvement in visualisation of soft-tissue structures (median iMAR-Algo1=3; interquartile range [IQR]:1.5-3; iMAR-Algo2=4; IQR: 3.5-4) and bone structures (iMAR-Algo1=3; IQR:3-4; iMAR-Algo2=4; IQR:4-5) compared to WFBP (soft tissue: median 2; IQR: 0.5-2 and bone structures: median 2; IQR: 1-3; p<0.0001). Compared with iMAR-Algo1, objective artefact reduction and subjective visualisation of soft-tissue and bone structures were improved with iMAR-Algo2 (p<0.0001).

CONCLUSION

Both iMAR algorithms reduced artefacts compared with WFBP, however, the iMAR algorithm with dedicated settings for large metallic implants was superior to the algorithm specifically adjusted to spinal implants.

Metal artifact reduction through MVCBCT and kVCT in radiotherapy

This study proposes a new method for removal of metal artifacts from megavoltage cone beam computed tomography (MVCBCT) and kilovoltage CT (kVCT) images. Both images were combined to obtain prior image, which was forward projected to obtain surrogate data and replace metal trace in the uncorrected kVCT image. The corrected image was then reconstructed through filtered back projection. A similar radiotherapy plan was designed using the theoretical CT image, the uncorrected kVCT image, and the corrected image. The corrected images removed most metal artifacts, and the CT values were accurate. The corrected image also distinguished the hollow circular hole at the center of the metal. The uncorrected kVCT image did not display the internal structure of the metal, and the hole was misclassified as metal portion. Dose distribution calculated based on the corrected image was similar to that based on the theoretical CT image. The calculated dose distribution also evidently differed between the uncorrected kVCT image and the theoretical CT image. The use of the combined kVCT and MVCBCT to obtain the prior image can distinctly improve the quality of CT images containing large metal implants.

CT of facial fracture fixation: an experimental study of artefact reducing methods

OBJECTIVES

This study aimed to determine the optimal post-operative CT imaging method that enables best visualization of facial bony structures in the vicinity of osteosynthesis material.

METHODS

Conducted at Töölö Hospital (Helsinki, Finland), this study relied on scanning a phantom with CBCT, 64-slice CT and high-definition multislice CT with dual-energy scan (providing monochromatic images of 70-, 100-, 120- and 140-keV energy levels) and iterative reconstruction (IR) methods. Two radiologists assessed the image quality, and the assessments were analyzed. In addition, a physicist performed a semi-quantitative analysis of the metal-induced artefacts.

RESULTS

The three subjects most easily assessed were the loose screw and both the bone structure and the fracture further away from the screw and the plate. Soft tissues adjacent to the screw and the plate remained more difficult for assessment. Both image interpreters agreed that the artefacts disturbed their assessments under dual energy. Metal artefacts disturbed the least under multislice CT with IR [adaptive statistical iterative reconstruction (ASiR) and VEO]. Neither interpreter found metal suppression helpful in CBCT.

CONCLUSIONS

CBCT with or without a metal artefact reduction algorithm was not optimal for post-operative facial imaging compared with multislice CT with IR. Multislice CT with ASiR filtering offered good image quality performance with fast image volume reconstruction, representing the current sweet spot in post-operative maxillofacial imaging.

Noninvasive microwave ablation zone radii estimation using x-ray CT image analysis

PURPOSE

The aims of this study were to noninvasively and automatically estimate both the radius of the ablated liver tissue and the radius encircling the treated zone, which also defines where the tissue is definitely untreated during a microwave (MW) thermal ablation procedure.

METHODS

Fourteen ex vivo bovine fresh liver specimens were ablated at 40 W using a 14 G microwave antenna, for durations of 3, 6, 8, and 10 min. The tissues were scanned every 5 s during the ablation using an x-ray CT scanner. In order to estimate the radius of the ablation zone, the acquired images were transformed into a polar presentation by displaying the Hounsfield units (HU) as a function of angle and radius. From this polar presentation, the average HU radial profile was analyzed at each time point and the ablation zone radius was estimated. In addition, textural analysis was applied to the original CT images. The proposed algorithm identified high entropy regions and estimated the treated zone radius per time. The estimated ablated zone radii as a function of treatment durations were compared, by means of correlation coefficient and root mean square error (RMSE) to gross pathology measurements taken immediately post-treatment from similarly ablated tissue.

RESULTS

Both the estimated ablation radii and the treated zone radii demonstrated strong correlation with the measured gross pathology values (R(2) ≥ 0.89 and R(2) ≥ 0.86, respectively). The automated ablation radii estimation had an average discrepancy of less than 1 mm (RMSE = 0.65 mm) from the gross pathology measured values, while the treated zone radii showed a slight overestimation of approximately 1.5 mm (RMSE = 1.6 mm).

CONCLUSIONS

Noninvasive monitoring of MW ablation using x-ray CT and image analysis is feasible. Automatic estimations of the ablation zone radius and the radius encompassing the treated zone that highly correlate with actual ablation measured values can be obtained. This technique can therefore potentially be used to obtain real time monitoring and improve the clinical outcome.