Streak Artifacts Research Articles

Performance of free-breathing dynamic T1-weighted sequences in patients at risk of developing motion artifacts undergoing gadoxetic acid-enhanced liver MRI.

To evaluate the recall rate and performance of free-breathing T1W dynamic imaging in patients who underwent gadoxetic acid-enhanced liver MRI. We retrospectively reviewed patients who underwent free-breathing dynamic T1WI liver MRI using Cartesian (XD-VIBE) or self-gated radial (SG-GRASP) sequences at two institutions. Four radiologists independently reviewed the overall image quality, streak, and motion artifacts for precontrast, arterial, and portal venous phases on a 4-point scale. Hepatic observations were annotated and assessed according to LI-RADS v2018. In total, 360 patients were included (XD-VIBE [n = 253], SG-GRASP [n = 107]). The overall image quality of free-breathing T1WI was 3.4 ± 0.4, 3.2 ± 0.4, and 3.5 ± 0.4 for precontrast, arterial, and portal venous phases, respectively. The actual recall rate was 0.6% (2/360). The SG-GRASP group showed fewer motion artifacts and more streak artifacts than the XD-VIBE group in all phases (p < 0.001 for all). The overall image quality was not significantly different between the two sequences in arterial (3.2 ± 0.4 in both, p = 0.607) and portal venous phases (3.5 ± 0.4 in XD-VIBE, 3.4 ± 0.4 in SG-GRASP, p = 0.214). Two sequences did not show significant differences in the lesion detection rate (figure of merit, FOM: 0.67 vs. 0.68, p = 0.876) or diagnostic performance for hepatocellular carcinoma (FOM: 0.55 vs. 0.62, p = 0.105). Both XD-VIBE and SG-GRASP provided sufficient image quality for patients at risk of developing motion artifacts, without significant differences in image quality or the lesion detection rate between sequences. • The overall image quality of free-breathing T1-weighted images using Cartesian or radial sequences was 3.4 ± 0.4, 3.2 ± 0.4, and 3.5 ± 0.4 for precontrast, arterial, and portal venous phases, respectively. • Only 0.3% (1/360) had undiagnostic exams and the actual recall rate was 0.6% (2/360) in patients who underwent free-breathing dynamic T1WI. • The overall lesion detection rate was 0.67 without a significant difference between Cartesian and radial sequences (figure of merit: 0.67 vs. 0.68, respectively, p = 0.876).

EPRI sparse reconstruction method based on deep learning

Electron paramagnetic resonance imaging (EPRI) is an advanced tumor oxygen concentration imaging method. Now, the bottleneck problem of EPRI is that the scanning time is too long. Sparse reconstruction is an effective and fast imaging method, which means reconstructing images from sparse-view projections. However, the EPRI images sparsely reconstructed by the classic filtered back projection (FBP) algorithm often contain severe streak artifacts, which affect subsequent image processing. In this work, we propose a feature pyramid attention-based, residual, dense, deep convolutional network (FRD-Net) to suppress the streak artifacts in the FBP-reconstructed images. This network combines residual connection, attention mechanism, dense connections and introduces perceptual loss. The EPRI image with streak artifacts is used as the input of the network and the output-label is the corresponding high-quality image densely reconstructed by the FBP algorithm. After training, the FRD-Net gets the capability of suppressing streak artifacts. The real data reconstruction experiments show that the FRD-Net can better improve the sparse reconstruction accuracy, compared with three existing representative deep networks.

Deep learning-based motion compensation for four-dimensional cone-beam computed tomography (4D-CBCT) reconstruction.

Motion-compensated (MoCo) reconstruction shows great promise in improving four-dimensional cone-beam computed tomography (4D-CBCT) image quality. MoCo reconstruction for a 4D-CBCT could be more accurate using motion information at the CBCT imaging time than that obtained from previous 4D-CT scans. However, such data-driven approaches are hampered by the quality of initial 4D-CBCT images used for motion modeling. This study aims to develop a deep-learning method to generate high-quality motion models for MoCo reconstruction to improve the quality of final 4D-CBCT images. A 3D artifact-reduction convolutional neural network (CNN) was proposed to improve conventional phase-correlated Feldkamp-Davis-Kress (PCF) reconstructions by reducing undersampling-induced streaking artifacts while maintaining motion information. The CNN-generated artifact-mitigated 4D-CBCT images (CNN enhanced) were then used to build a motion model which was used by MoCo reconstruction (CNN+MoCo). The proposed procedure was evaluated using in-vivo patient datasets, an extended cardiac-torso (XCAT) phantom, and the public SPARE challenge datasets. The quality of reconstructed images for XCAT phantom and SPARE datasets was quantitatively assessed using root-mean-square-error (RMSE) and normalized cross-correlation (NCC). The trained CNN effectively reduced the streaking artifacts of PCF CBCT images for all datasets. More detailed structures can be recovered using the proposed CNN+MoCo reconstruction procedure. XCAT phantom experiments showed that the accuracy of estimated motion model using CNN enhanced images was greatly improved over PCF. CNN+MoCo showed lower RMSE and higher NCC compared to PCF, CNN enhanced and conventional MoCo. For the SPARE datasets, the average(±standard deviation) RMSE in mm-1 for body region of PCF, CNN enhanced, conventional MoCo and CNN+MoCo were 0.0040±0.0009, 0.0029±0.0002, 0.0024±0.0003 and 0.0021±0.0003. Corresponding NCC were 0.84±0.05, 0.91±0.05, 0.91±0.05 and 0.93±0.04. CNN-based artifact reduction can substantially reduce the artifacts in the initial 4D-CBCT images. The improved images could be used to enhance the motion modeling and ultimately improve the quality of the final 4D-CBCT images reconstructed using MoCo.

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.

Omni-Directional Semi-Global Stereo Matching with Reliable Information Propagation

High efficiency and accuracy of semi-global matching (SGM) make it widely used in many stereo vision applications. However, SGM not only struggles in dealing with pixels in homogeneous area, but also suffers from streak artifacts. In this paper, we propose a novel omni-directional SGM (OmniSGM) with a cost volume update scheme to aggregate costs from paths along all directions and to encourage reliable information to propagate across entire image. Specifically, we perform SGM along four tree structures, namely trees in the left, right, top and bottom of root node, and then fuse the outputs to obtain final result. The contributions of pixels on each tree can be recursively computed from leaf nodes to root node, ensuring our method has linear time computational complexity. Moreover, An iterative cost volume update scheme is proposed using aggregated cost in the last pass to enhance the robustness of initial matching cost. Thus, useful information is more likely to propagate in a long distance to handle the ambiguities in low textural area. Finally, we present an efficient strategy to propagate disparities of stable pixels along the minimum spanning tree (MST) for disparity refinement. Extensive experiments in stereo matching on Middlebury and KITTI datasets demonstrate that our method outperforms typical traditional SGM-based cost aggregation methods.

Microlocal Analysis of $d$-Plane Transform on the Euclidean Space

We study the basic properties of the $d$-plane transform on the Euclidean space as a Fourier integral operator and its application to the microlocal analysis of streaking artifacts in its filtered back-projection. The $d$-plane transform is defined by integrals of functions on the $n$-dimensional Euclidean space over all the $d$-dimensional planes, where $0<d<n$. This maps functions on the Euclidean space to those on the affine Grassmannian $G(d,n)$. This is said to be X-ray transform if $d=1$ and Radon transform if $d=n-1$. When $n=2$ the X-ray transform is thought to be measurements of computed tomography (CT) scanners. In this paper we obtain a concrete expression of the canonical relation between the $d$-plane transform and the quantitative properties of the filtered back-projection of the product of the images of the $d$-plane transform. The FBP of the product is related to the metal streaking artifacts of CT images. Our result is a generalization of recent results of Park, Choi, and Seo [Comm. Pure Appl. Math., 70 (2017), pp. 2191--2217] and Palacios, Uhlmann, and Wang [SIAM J. Math. Anal., 50 (2018), pp. 4914--4936] for the X-ray transform on the plane.

Balanced Steady-State Free Precession Cine MR Imaging in the Presence of Cardiac Devices: Value of Interleaved Radial Linear Combination Acquisition With Partial Dephasing.

Balanced steady-state free precession (bSSFP) is important in cardiac MRI but suffers from off-resonance artifacts. The interpretation-limiting artifacts in patients with cardiac implants remain an unsolved issue. To develop an interleaved radial linear combination bSSFP (lcSSFP) method with partial dephasing (PD) for improved cardiac cine imaging when implanted cardiovascular devices are present. Prospective. Flow phantom adjacent to a pacemaker and 10 healthy volunteers (mean age ± standard deviation: 31.9 ± 2.9 years, 4 females) with a cardioverter-defibrillator (ICD) positioned extracorporeally at the left chest in the prepectoral region. A 3-T, 1) Cartesian bSSFP, 2) Cartesian gradient echo (GRE), 3) Cartesian lcSSFP, and 4) radial lcSSFP cine sequences. Flow artifacts mitigation using PD was validated with phantom experiments. Undersampled radial lcSSFP with interleaving across phase-cyclings and cardiac phases (RLC-SSFP), combined with PD, was then employed for achieving improved quality of cine images from left ventricular short-axis view. The image quality in the presence of cardiac devices was qualitatively assessed by three independent raters (1=worst, 5=best), regarding five criteria (banding artifacts, streak artifacts, flow artifacts, cavity visibility, and overall image quality). Wilcoxon rank-sum test for the five criteria between Cartesian bSSFP cine and RLC-SSFP with PD. Fleiss kappa test for inter-reader agreement. A P value < 0.05 was considered statistically significant. Based on simulations and phantom experiments, 60 projections per phase cycling and 1/6 PD were chosen. The in vivo experiments demonstrated significantly reduced banding artifacts (4.8 ± 0.4 vs. 2.7 ± 0.7), fewer streak artifacts (3.7 ± 0.6 vs. 2.6 ± 0.7) and flow artifacts (4.4 ± 0.4 vs. 3.7 ± 0.6), therefore improved cavity visibility (4.1 ± 0.4 vs. 2.9 ± 0.9) and overall quality (4.0 ± 0.4 vs. 2.7 ± 0.7). RLC-SSFP method with PD may improve cine image quality in subjects with cardiac devices. 2. Stage 1.

PRIOR: Prior-Regularized Iterative Optimization Reconstruction For 4D CBCT.

4D cone-beam computed tomography (CBCT) is an important imaging modality in image-guided radiation therapy to address the motion-induced artifacts caused by organ movements during the respiratory process. However, due to the extremely sparse projection data for each temporal phase, 4D CBCT reconstructions will suffer from severe streaking artifacts. Therefore, to tackle the streak artifacts and provide high-quality images, we proposed a framework termed Prior-Regularized Iterative Optimization Reconstruction (PRIOR) for 4D CBCT. The PRIOR framework combines the physics-based model and data-driven method simultaneously, with powerful feature extracting capacity, significantly promoting the image quality compared to single model-based or deep learning-based methods. Besides, we designed a specialized deep learning model named PRIOR-Net, which can effectively excavate the static information in the prior image reconstructed from the fully-sampled projections at the encoding stage to improve the reconstruction performance for individual phase-resolved images. Both the simulated and clinical 4D CBCT datasets were performed to evaluate the performance of the PRIOR-Net and the PRIOR framework. Compared with the advanced 4D CBCT reconstruction methods, the proposed methods achieve promising results quantitatively and qualitatively in streak artifact suppression, soft tissue restoration, and tiny detail preservation.

Artifact Reduction for Sparse-View CT Using Deep Learning With Band Patch

Sparse-view computed tomography (CT), an imaging technique that reduces the number of projections, can reduce the total scan duration and radiation dose. However, sparse data sampling causes streak artifacts on images reconstructed with analytical algorithms. In this article, we propose an artifact reduction method for sparse-view CT using deep learning. We developed a lightweight fully convolutional network to estimate a fully sampled sinogram from a sparse-view sinogram by enlargement in the vertical direction. Furthermore, we introduced the band patch, a rectangular region cropped in the vertical direction, as an input image for the network based on the sinogram’s characteristics. Comparison experiments using a swine rib dataset of micro-CT scans and a chest dataset of clinical CT scans were conducted to compare the proposed method, improved U-net from a previous study, and the U-net with band patches. The experimental results showed that the proposed method achieved the best performance and the U-net with band patches had the second-best result in terms of accuracy and prediction time. In addition, the reconstructed images of the proposed method suppressed streak artifacts while preserving the object’s structural information. We confirmed that the proposed method and band patch are useful for artifact reduction for sparse-view CT.

PGNet: Projection generative network for sparse-view reconstruction of projection-based magnetic particle imaging.

Magnetic particle imaging (MPI) is a novel tomographic imaging modality that scans the distribution of superparamagnetic iron oxide nanoparticles. However, it is time-consuming to scan multiview two-dimensional (2D) projections for three-dimensional (3D) reconstruction in projection MPI, such as computed tomography (CT). An intuitive idea is to use the sparse-view projections for reconstruction to improve the temporal resolution. Tremendous progress has been made toward addressing the sparse-view problem in CT, because of the availability of large data sets. For the novel tomography of MPI, to the best of our knowledge, studies on the sparse-view problem have not yet been reported. The acquisition of multiview projections for 3D MPI imaging is time-consuming. Our goal is to only acquire sparse-view projections for reconstruction to improve the 3D imaging temporal resolution of projection MPI. We propose to address the sparse-view problem in projection MPI by generating novel projections. The data set we constructed consists of three parts: simulation data set (including 3000 3D data), four phantoms data, and an in vivo mouse data. The simulation data set is used to train and validate the network, and the phantoms and in vivo mouse data are used to test the network. When the number of novel generated projections meets the requirements of filtered back projection, the streaking artifacts will be absent from MPI tomographic imaging. Specifically, we propose a projection generativenetwork (PGNet), that combines an attention mechanism, adversarial training strategy, and a fusion loss function and can generate novel projections based on sparse-view real projections. To the best of our knowledge, we are the first to propose a deep learning method to attempt to overcome the sparse-view problem in projection MPI. We compare our method with several sparse-view methods on phantoms and in vivo mouse data and validate the advantages and effectiveness of our proposed PGNet. Our proposed PGNet enables the 3D imaging temporal resolution of projection MPI to be improved by 6.6 times, while significantly suppressing the streaking artifacts. We proposed a deep learning method operated in projection domain to address the sparse-view reconstruction of MPI, and the data scarcity problem in projection MPI reconstruction is alleviated by constructing a sparse-dense simulated projection data set. By our proposed method, the number of acquisitions of real projections can be reduced. The advantage of our method is that it prevents the generation of streaking artifacts at the source. Our proposed sparse-view reconstruction method has great potential for application to time-sensitive in vivo 3D MPI imaging.

Metal artifact correction in photon-counting detector computed tomography: metal trace replacement using high-energy data.

Metal artifacts have been an outstanding issue in computed tomography (CT) since its first uses in the clinic and continue to interfere. Metal artifact reduction (MAR) methods continue to be proposed and photon-counting detectors (PCDs) have recently been the subject of research toward this purpose. PCDs offer the ability to distinguish the energy of incident x-rays and sort them in a set number of energy bins. High-energy data captured using PCDs have been shown to reduce metal artifacts in reconstructions due to reduced beam hardening. High-energy reconstructions using PCD-CT have their drawbacks, such as reduced image contrast and increased noise. Here, we demonstrate a MAR algorithm, trace replacement MAR (TRMAR), in which the data corrupted by metal artifacts in full energy spectrum projections are corrected using the high-energy data captured during the same scan. The resulting reconstructions offer similar MAR to that seen in high-energy reconstructions, but with improved image quality. Experimental data were collected using a bench-top PCD-CT system with a cadmium zinc telluride PCD. Simulations were performed to determine the optimal high-energy threshold and to test TRMAR in simulations using the XCAT phantom and a biological sample. For experiments a 100-mm diameter cylindrical phantom containing vials of water, two screws, various densities of Ca(ClO4 )2 , and a spatial resolution phantom was imaged with and without the screws. The screws were segmented in the initial reconstruction and forward projected to identify them in the sinogram space in order to perform TRMAR. The resulting reconstructions were compared to the control and to reconstructions corrected using normalized metal artifact reduction (NMAR). Additionally, a beef short rib was imaged with and without metal to provide a more realistic phantom. XCAT simulations showed a reduction in the streak artifact from -978HU in uncorrected images to -10HU with TRMAR. The magnitude of the metal artifact in uncorrected images of the 100-mm phantom was -442 HU, compared to the desired -81HU with no metal. TRMAR reduced the magnitude of the artifact to -142 HU, with NMAR reducing the magnitude to -96 HU. Relative image noise was reduced from 176% in the high-energy image to 56% using TRMAR. Density quantification was better with NMAR, with the Ca(ClO4 )2 vial affected most by metal artifacts showing 0.8% error compared to 2.1% with TRMAR. Small features were preserved to a greater extent with TRMAR, with the limiting spatial frequency at 20% of the MTF fully maintained at 1.31 lp/mm, while with NMAR it was reduced to 1.22 lp/mm. Images of the beef short rib showed better delineation of the shape of the metal using TRMAR. NMAR offers slightly better performance compared to TRMAR in streak reduction and image quality metrics. However, TRMAR is less susceptible to metal segmentation errors and can closely approximate the reduction in the streak metal artifact seen in NMAR at 1/3 the computation time. With the recent introduction of PCD-CT into the clinic, TRMAR offers notable potential for fast, effective MAR.

Streaking artifact reduction for CBCT-based synthetic CT generation in adaptive radiotherapy.

Cone-beam computed tomography (CBCT) is widely used for daily image guidance in radiation therapy, enhancing the reproducibility of patient setup. However, its application in adaptive radiotherapy (ART) is limited by many imaging artifacts and inaccurate Hounsfield units (HUs). The correction of CBCT image is necessary and of great value for CBCT-based ART. To explore the synthetic CT (sCT) generation from CBCT images of thorax and abdomen patients, which usually surfer from serious artifacts duo to organ state changes. In this study, a streaking artifact reduction network (SARN) is proposed to reduce artifacts and combine with cycleGAN to generate high-quality sCT images from CBCT and achieve an accurate dose calculation. The proposed SARN was trained in a self-supervised manner. Artifact-CT images were generated from planning CT by random deformation and projection replacement, and SARN was trained based on paired artifact-CT and CT images. The planning CT and CBCT images of 260 patients with cancer, including 120 thoracic and 140 abdominal CT scans, were used to train and evaluate neural networks. The CBCT images of another 12 patients in late treatment fractions, which contained large anatomy changes, were also tested by trained models. The trained models include commonly used U-Net, cycleGAN, attention-gated cycleGAN (cycAT), and cascade models combined SARN with cycleGAN or cycAT. The generated sCT images were compared in terms of image quality and dose calculation accuracy. The sCT images generated by SARN combined with cycleGAN and cycAT showed the best image quality, removed the most artifacts, and retained the normal anatomical structure. The SARN+cycleGAN performed best in streaking artifacts removal with the maximum percent integrity uniformity (PIUm ) of 91.0% and minimum standard deviation (SD) of 35.4 HU for delineated artifact regions among all models. The mean absolute error (MAE) of CBCT images in the thorax and abdomen were 71.6 and 55.2 HU, respectively, using planning CT images after deformable registration as ground truth. Compared with CBCT, the thoracic and abdominal sCT images generated by each model had significantly improved image quality with smaller MAE (p < 0.05). The SARN+cycAT obtained the minimum MAEs of 42.5 HU in the thorax while SARN+cycleGAN got the minimum MAEs of 32.0 HU in the abdomen. The sCT generated by U-Net had a remarkably lower anatomical structure accuracy compared with the other models. The thoracic and abdominal sCT images generated by SARN+cycleGAN showed optimal dose calculation accuracy with gamma passing rates (2 mm/2%) of 98.2% and 96.9%, respectively. The proposed SARN can reduce serious streaking artifacts in CBCT images. The SARN combined with cycleGAN can generate high-quality sCT images with fewer artifacts, high-accuracy HU values, and accurate anatomical structures, thus providing reliable dose calculation in ART.

Undersampling artifact reduction for free-breathing 3D stack-of-radial MRI based on a deep adversarial learning network

PurposeStack-of-radial MRI allows free-breathing abdominal scans, however, it requires relatively long acquisition time. Undersampling reduces scan time but can cause streaking artifacts and degrade image quality. This study developed deep learning networks with adversarial loss and evaluated the performance of reducing streaking artifacts and preserving perceptual image sharpness. MethodsA 3D generative adversarial network (GAN) was developed for reducing streaking artifacts in stack-of-radial abdominal scans. Training and validation datasets were self-gated to 5 respiratory states to reduce motion artifacts and to effectively augment the data. The network used a combination of three loss functions to constrain the anatomy and preserve image quality: adversarial loss, mean-squared-error loss and structural similarity index loss. The performance of the network was investigated for 3–5 times undersampled data from 2 institutions. The performance of the GAN for 5 times accelerated images was compared with a 3D U-Net and evaluated using quantitative NMSE, SSIM and region of interest (ROI) measurements as well as qualitative scores of radiologists. ResultsThe 3D GAN showed similar NMSE (0.0657 vs. 0.0559, p = 0.5217) and significantly higher SSIM (0.841 vs. 0.798, p < 0.0001) compared to U-Net. ROI analysis showed GAN removed streaks in both the background air and the tissue and was not significantly different from the reference mean and variations. Radiologists' scores showed GAN had a significant improvement of 1.6 point (p = 0.004) on a 4-point scale in streaking score while no significant difference in sharpness score compared to the input. Conclusion3D GAN removes streaking artifacts and preserves perceptual image details.

Joint application of hybrid iterative reconstruction and adaptive filters on neck-and-shoulder CT imaging: A clinical evaluation.

To assess whether the joint application of hybrid iterative reconstruction (HIR) and an adaptive filter (AF) could reduce streak artifacts and improve image quality of neck-and-shoulder computed tomography (CT). This study included 96 patients with suspicious neck lesions who underwent a routine nonenhanced scan on a 64-slice CT scanner. The raw data were reconstructed using four different settings: filtered back projection (FBP), HIR, FBP+AF, and HIR+AF. Regions of interest were manually drawn in erector spine, axillary fat, latissimus dorsi, and dorsal cervical fat. Mean and standard deviation (SD) of the CT number, signal-to-noise ratio (SNR), and contrast-to-noise ratio (CNR) were obtained and compared using Wilcoxon signed-rank tests. The qualitative assessments of five factors were compared by two independent investigators. Compared to the other three settings, HIR+AF reduced noise in the area where the streak artifact of the lower neck were most serious (SD; all p≤0.001). The SNR and CNR were improved significantly (all p≤0.001). Compared to the other three settings, HIR+AF showed a significant improvement in CT image quality regarding the visibility of suspicious lesions, the extent of streaking artifacts, noise, soft-tissue contrast, and visualization of small structures (all p≤0.02). The combination of HIR and AF can significantly reduce streaking artifacts and improve image quality in neck-and-shoulder CT imaging.

A dual-domain neural network based on sinogram synthesis for sparse-view CT reconstruction

ObjectiveThe dual-domain deep learning-based reconstruction techniques have enjoyed many successful applications in the field of medical image reconstruction. Applying the analytical reconstruction based operator to transfer the data from the projection domain to the image domain, the dual-domain techniques may suffer from the insufficient suppression or removal of streak artifacts in areas with the missing view data, when addressing the sparse-view reconstruction problems. In this work, to overcome this problem, an intelligent sinogram synthesis based back-projection network (iSSBP-Net) was proposed for sparse-view computed tomography (CT) reconstruction. In the iSSBP-Net method, a convolutional neural network (CNN) was involved in the dual-domain method to inpaint the missing view data in the sinogram before CT reconstruction. MethodsThe proposed iSSBP-Net method fused a sinogram synthesis sub-network (SS-Net), a sinogram filter sub-network (SF-Net), a back-projection layer, and a post-CNN into an end-to-end network. Firstly, to inpaint the missing view data, the SS-Net employed a CNN to synthesize the full-view sinogram in the projection domain. Secondly, to improve the visual quality of the sparse-view CT images, the synthesized sinogram was filtered by a CNN. Thirdly, the filtered sinogram was brought into the image domain through the back-projection layer. Finally, to yield images of high visual sensitivity, the post-CNN was applied to restore the desired images from the outputs of the back-projection layer. ResultsThe numerical experiments demonstrate that the proposed iSSBP-Net is superior to all competing algorithms under different scanning condintions for sparse-view CT reconstruction. Compared to the competing algorithms, the proposed iSSBP-Net method improved the peak signal-to-noise ratio of the reconstructed images about 1.21 dB, 0.26 dB, 0.01 dB, and 0.37 dB under the scanning conditions of 360, 180, 90, and 60 views, respectively. ConclusionThe promising reconstruction results indicate that involving the SS-Net in the dual-domain method is could be an effective manner to suppress or remove the streak artifacts in sparse-view CT images. Due to the promising results reconstructed by the iSSBP-Net method, this study is intended to inspire the further development of sparse-view CT reconstruction by involving a SS-Net in the dual-domain method.

Deep learning based correction of low performing pixel in computed tomography

Low Performing Pixel (LPP)/bad pixel in CT detectors cause ring and streaks artifacts, structured non-uniformities and deterioration of the image quality. These artifacts make the image unusable for diagnostic purposes. A missing/defective detector pixel translates to a channel missing across all views in sinogram domain and its effect gets spill over entire image in reconstruction domain as artifacts. Most of the existing ring and streak removal algorithms perform correction only in the reconstructed image domain. In this work, we propose a supervised deep learning algorithm that operates in sinogram domain to remove distortions cause by the LPP. This method leverages CT scan geometry, including conjugate ray information to learn the interpolation in sinogram domain. While the experiments are designed to cover the entire detector space, we emphasize on LPPs near detector iso-center as these have most adverse impact on image quality specially if the LPPs fall on the high frequency region (bone-tissue interface). We demonstrated efficacy of the proposed method using data acquired on GE RevACT multi-slice CT system with flat-panel detector. Experimental results on head scans show significant reduction in ring artifacts regardless of LPP location in the detector geometry. We have simulated isolated LPPs accounting for 5% and 10% of total channels. Detailed statistical analysis illustrates approximately 5dB improvement in SNR in both sinogram and reconstruction domain as compared to classical bicubic and Lagrange interpolation methods. Also, with reduction in ring and streak artifacts, the perceptual image quality is improved across all the test images.

Description and Use of Three-Dimensional Numerical Phantoms of Cardiac Computed Tomography Images

The World Health Organization indicates the top cause of death is heart disease. These diseases can be detected using several imaging modalities, especially cardiac computed tomography (CT), whose images have imperfections associated with noise and certain artifacts. To minimize the impact of these imperfections on the quality of the CT images, several researchers have developed digital image processing techniques (DPIT) by which the quality is evaluated considering several metrics and databases (DB), both real and simulated. This article describes the processes that made it possible to generate and utilize six three-dimensional synthetic cardiac DBs or voxels-based numerical phantoms. An exhaustive analysis of the most relevant features of images of the left ventricle, belonging to a real CT DB of the human heart, was performed. These features are recreated in the synthetic DBs, generating a reference phantom or ground truth free of imperfections (DB1) and five phantoms, in which Poisson noise (DB2), stair-step artifact (DB3), streak artifact (DB4), both artifacts (DB5) and all imperfections (DB6) are incorporated. These DBs can be used to determine the performance of DPIT, aimed at decreasing the effect of these imperfections on the quality of cardiac images.

Nonlinearly scaled prior image-controlled frequency split for high-frequency metal artifact reduction in computed tomography.

This paper introduces a new approach for the dedicated reduction of high-frequency metal artifacts, which applies a nonlinear scaling (NLS) transfer function on the high-frequency projection domain to reduce artifacts, while preserving edge information and anatomic detail by incorporating prior image information. An NLS function is applied to suppress high-frequency streak artifacts, but to restrict the correction to metal projections only, scaling is performed in the sinogram domain. Anatomic information should be preserved and is excluded from scaling by incorporating a prior image from tissue classification. The corrected high-frequency sinogram is reconstructed and combined with the low-frequency component of a normalized metal artifact reduction (NMAR) image. Scans of different anthropomorphic phantoms were acquired (unilateral hip, bilateral hip, dental implants, and embolization coil). Multiple regions of interest (ROIs) were drawn around the metal implants and hounsfield unit (HU) deviations were analyzed. Clinical data sets including single image slices of dental fillings, a bilateral hip implant, spinal fixation screws, and an aneurysm coil were reconstructed and assessed. The prior image-controlled NLS can remove streak artifacts while preserving anatomic detail within the bone and soft tissue. The qualitative analysis of clinical cases showed a tremendous enhancement within dental fillings and neuro coils, and a significant enhancement within spinal screws or hip implants. The phantom scan measurements support this observation. In all phantom setups, the NLS-corrected result showed lowest HU derivation and the best visualization of thedata. The prior image-controlled NLS provides a method to reduce high-frequency streaks in metal-corrupted computed tomography (CT)data.

Convolutional neural network-based metal and streak artifacts reduction in dental CT images with sparse-view sampling scheme.

Sparse-view sampling has attracted attention for reducing the scan time and radiation dose of dental cone-beam computed tomography (CBCT). Recently, various deep learning-based image reconstruction techniques for sparse-view CT have been employed to produce high-quality image while effectively reducing streak artifacts caused by the lack of projection views. However, most of these methods do not fully consider the effects of metal implants. As sparse-view sampling strengthens the artifacts caused by metal objects, simultaneously reducing both metal and streak artifacts in sparse-view CT images has been challenging. To solve this problem, in this study, we propose a novelframework. The proposed method was based on the normalized metal artifact reduction (NMAR) method, and its performance was enhanced using two convolutional neural networks (CNNs). The first network reduced the initial artifacts while preserving the fine details to generate high-quality priors for NMAR processing. Subsequently, the second network was employed to reduce the streak artifacts after NMAR processing of sparse-view CT data. To validate the proposed method, we generated training and test data by computer simulations using both extended cardiac-torso (XCAT) and clinical data sets. Visual inspection and quantitative evaluations demonstrated that the proposed method effectively reduced both metal and streak artifacts while preserving the details of anatomical structures compared with the conventional metal artifact reduction methods. We propose a framework for reconstructing accurate CT images in metal-inserted sparse-view CT. The proposed method reduces streak artifacts from both metal objects and sparse-view sampling while recovering the anatomical details, indicating the feasibility of fast-scan dental CBCTimaging.

A streak artifact reduction algorithm in sparse-view CT using a self-supervised neural representation.

Sparse-view computed tomography (CT) has been attracting attention for its reduced radiation dose and scanning time. However, analytical image reconstruction methods suffer from streak artifacts due to insufficient projection views. Recently, various deep learning-based methods have been developed to solve this ill-posed inverse problem. Despite their promising results, they are easily overfitted to the training data, showing limited generalizability to unseen systems and patients. In this work, we propose a novel streak artifact reduction algorithm that provides a system- and patient-specificsolution. Motivated by the fact that streak artifacts are deterministic errors, we regenerate the same artifacts from a prior CT image under the same system geometry. This prior image need not be perfect but should contain patient-specific information and be consistent with full-view projection data for accurate regeneration of the artifacts. To this end, we use a coordinate-based neural representation that often causes image blur but can greatly suppress the streak artifacts while having multiview consistency. By employing techniques in neural radiance fields originally proposed for scene representations, the neural representation is optimized to the measured sparse-view projection data via self-supervised learning. Then, we subtract the regenerated artifacts from the analytically reconstructed original image to obtain the final correctedimage. To validate the proposed method, we used simulated data of extended cardiac-torso phantoms and the 2016 NIH-AAPM-Mayo Clinic Low-Dose CT Grand Challenge and experimental data of physical pediatric and head phantoms. The performance of the proposed method was compared with a total variation-based iterative reconstruction method, naive application of the neural representation, and a convolutional neural network-based method. In visual inspection, it was observed that the small anatomical features were best preserved by the proposed method. The proposed method also achieved the best scores in the visual information fidelity, modulation transfer function, and lung nodulesegmentation. The results on both simulated and experimental data suggest that the proposed method can effectively reduce the streak artifacts while preserving small anatomical structures that are easily blurred or replaced with misleading features by the existing methods. Since the proposed method does not require any additional training datasets, it would be useful in clinical practice where the large datasets cannot becollected.