Geometric Artifacts Research Articles

Deforming Patient-Specific Models of Vascular Anatomies to Represent Stent Implantation via Extended Position Based Dynamics.

Angioplasty with stent placement is a widely used treatment strategy for patients with stenotic blood vessels. However, it is often challenging to predict the outcomes of this procedure for individual patients. Image-based computational fluid dynamics (CFD) is a powerful technique for making these predictions. To perform CFD analysis of a stented vessel, a virtual model of the vessel must first be created. This model is typically made by manipulating two-dimensional contours of the vessel in its pre-stent state to reflect its post-stent shape. However, improper contour-editing can cause invalid geometric artifacts in the resulting mesh that then distort the subsequent CFD predictions. To address this limitation, we have developed a novel shape-editing method that deforms surface meshes of stenosed vessels to create stented models. Our method uses physics-based simulations via Extended Position Based Dynamics to guide these deformations. We embed an inflating stent inside a vessel and apply collision-generated forces to deform the vessel and expand its cross-section. We demonstrate that this technique is feasible and applicable for a wide range of vascular anatomies, while yielding clinically compatible results. We also illustrate the ability to parametrically vary the stented shape and create models allowing CFD analyses. Our stenting method will help clinicians predict the hemodynamic results of stenting interventions and adapt treatments to achieve target outcomes for patients. It will also enable generation of synthetic data for data-intensive applications, such as machine learning, to support cardiovascular research endeavors.

Self-calibration method for geometric parameters based on feature point projection trajectories in Cone-beam Computed Laminography

Cone-beam computed laminography (CBCL) is an X-ray three-dimensional imaging method that is suitable for large plate-like objects. Geometric parameter calibration is crucial during CL imaging to prevent geometric artifacts in the reconstructed image. This study introduced a self-calibration method employing feature-point projection trajectories for CL geometric parameters. The method constructed point projection trajectory models for the system’s geometric parameters and optimized the best fit of these models to the characteristic point projection trajectories to solve for the geometric parameters of the system. Furthermore, this study examined the impact of the feature point location in the object coordinate system on the parameter calibration accuracy and reduced the required feature point locations from 11°to 4°using the alternating optimization algorithm. The imaging results demonstrated the ability of the method to achieve high-accuracy calibration of CL system geometric parameters, reducing the calibration error of the rotational axis tilt angle α from 1.1% to 0.08% while exhibiting resilience to noise. The proposed method enabled the calibration of all CL system geometric parameters without the model calibration, laying the groundwork for rectifying image geometric artifacts and enhancing CL imaging quality. Moreover, this method presents a novel approach for CT system geometric parameter calibration.

Geometric Benchmarking of Metal Material Extrusion Technology: A Preliminary Study

Metal additive manufacturing technologies such as powder bed fusion (PBF) and direct energy deposition (DED) are experiencing fast development, due to the growing awareness of industries. However, high energy consumption, slow production processes, and high costs of both machines and feedstocks hamper their competitiveness, compared to conventional manufacturing techniques. Metal material extrusion (metal-MEX) can represent a cost- and energy-effective alternative for metal additive manufacturing. This article aims to assess the potential of such technology by addressing uncertainties related to product design and process stability through a preliminary geometric benchmarking study. The geometric tolerances and minimum achievable sizes of some simple geometries produced in 316L stainless steel were evaluated using geometric benchmark test artifacts (GBTAs). Process maps were also proposed to forecast the feasibility of achieving acceptable values of the investigated tolerances, based on the nominal dimensions of the features.

Geometric artifacts correction of equiangular fan-beam industrial CT system

To address the issue of geometric artifacts in equiangular fan-beam industrial computed tomography (CT) systems, this paper proposes a projection sinogram-based method for correcting these artifacts. Firstly, we discuss the formation mechanism of geometric artifacts in equiangular fan-beam industrial CT system when there is geometric deviation. Secondly, we establish a geometric model that incorporates these deviations and mathematically describe the projected position curve of a point within the plane. Subsequently, we use a needle gauge as a surrogate for the point, and fit the curve in the projection sinusoidal graph obtained from scanning the needle gauge to calibrate the geometric deviations. Finally, we introduce the calibrated geometric deviations into the filtered back projection (FBP) algorithm to achieve more accurate reconstructed images. The proposed method precisely calibrates geometrical deviations in the equiangular fan-beam industrial computed tomography (CT) system and effectively eliminates geometric artifacts in the reconstructed CT images, this is demonstrated by simulation and practical experiment. This method has also been successfully applied in engineering practice.

Improving Laser Powder Bed Fusion Printability of Tungsten Powders Using Simulation-Driven Process Optimization Algorithms.

This study applies numerical and experimental techniques to investigate the effect of process parameters on the density, structure and mechanical properties of pure tungsten specimens fabricated by laser powder bed fusion. A numerical model based on the simplified analysis of a thermal field generated in the powder bed by a moving laser source was used to calculate the melt pool dimensions, predict the density of printed parts and build a cost-effective plan of experiments. Specimens printed using a laser power of 188 W, a scanning speed of 188 mm/s, a hatching space of 80 µm and a layer thickness of 30 µm showed a maximum printed density of 93.2%, an ultimate compression strength of 867 MPa and a maximum strain to failure of ~7.0%, which are in keeping with the standard requirements for tungsten parts obtained using conventional powder metallurgy techniques. Using the optimized printing parameters, selected geometric artifacts were manufactured to characterize the printability limits. A complementary numerical study suggested that decreasing the layer thickness, increasing the laser power, applying hot isostatic pressing and alloying with rhenium are the most promising directions to further improve the physical and mechanical properties of printed tungsten parts.

A 3D dual-echo spiral sequence for simultaneous dynamic susceptibility contrast and dynamic contrast-enhanced MRI with single bolus injection.

Perfusion MRI reveals important tumor physiological and pathophysiologic information, making it a critical component in managing brain tumor patients. This study aimed to develop a dual-echo 3D spiral technique with a single-bolus scheme to simultaneously acquire both dynamic susceptibility contrast (DSC) and dynamic contrast-enhanced (DCE) data and overcome the limitations of current EPI-based techniques. A 3D spiral-based technique with dual-echo acquisition was implemented and optimized on a 3T MRI scanner with a spiral staircase trajectory and through-plane SENSE acceleration for improved speed and image quality, in-plane variable-density undersampling combined with a sliding-window acquisition and reconstruction approach for increased speed, and an advanced iterative deblurring algorithm. Four volunteers were scanned and compared with the standard of care (SOC) single-echo EPI and a dual-echo EPI technique. Two patients were scanned with the spiral technique during a preload bolus and compared with the SOC single-echo EPI collected during the second bolus injection. Volunteer data demonstrated that the spiral technique achieved high image quality, reduced geometric artifacts, and high temporal SNR compared with both single-echo and dual-echo EPI. Patient perfusion data showed that the spiral acquisition achieved accurate DSC quantification comparable to SOC single-echo dual-dose EPI, with the additional DCE information. A 3D dual-echo spiral technique was developed to simultaneously acquire both DSC and DCE data in a single-bolus injection with reduced contrast use. Preliminary volunteer and patient data demonstrated increased temporal SNR, reduced geometric artifacts, and accurate perfusion quantification, suggesting a competitive alternative to SOC-EPI techniques for brain perfusion MRI.

Convolutional network denoising for acceleration of multi-shot diffusion MRI

Multi-shot echo planar imaging is a promising technique to reduce geometric distortions and increase spatial resolution in diffusion-weighted MRI (DWI), at the expense of increased scan time. Moreover, performing DWI in the body requires multiple repetitions to obtain sufficient signal-to-noise ratio, which further increases the scan time. This work proposes to reduce the number of repetitions and perform denoising of high b-value images using a convolutional network denoising trained on single-shot DWI to accelerate the acquisition of multi-shot DWI. Convolutional network denoising is demonstrated to accelerate the acquisition of 2-shot DWI by a factor of 4 compared to the clinical standard on patients with rectal cancer. Image quality was evaluated using qualitative scores from expert body radiologists between accelerated and non-accelerated acquisition. Additionally, the effect of convolutional network denoising on each image quality score was analyzed using a Wilcoxon signed-rank test. Convolutional network denoising would enable to increase the number of shots without increasing scan time for significant geometric artifact reduction and spatial resolution increase.

A Complex of Stratifi ed Upper Paleolithic Sites in the Foothills of the Northern Tien Shan: General Data and Research Perspectives

This article examines the key cultural trends and events in the evolution of the Upper Paleolithic in the foothills of the Northern Tien Shan (Zailisky Alatau, Kazakhstan). It outlines the history of Paleolithic studies in southeastern Kazakhstan. We describe the geographic characteristics of the region, the geomorphological positions of sites, and features of sedimentation that infl uenced the preservation of cultural remains in situ. Archaeological materials from key Upper Paleolithic sites are reviewed, including those from stratifi ed sites—Maibulak, Rakhat, and Uzynagash-1, -2. Lithic industries and absolute dates suggest that Maibulak was permanently inhabited during the Early, Middle, and probably Late Upper Paleolithic, materials from the early stages being the most expressive. Early Upper Paleolithic industries display Aurignacian-like characteristics and are paralleled by certain Western Eurasian industries of the same age. The multicomponent site of Rakhat was peopled during the end of the Early Upper Paleolithic, in the Middle Upper Paleolithic, and at the beginning of the Late Upper Paleolithic, documenting the evolution of Upper Paleolithic cultures during the ~30–23 cal ka BP interval. The industries of Rakhat include an Aurignacianlike one, a Middle Upper Paleolithic complex with micro-Gravette-like points, and one with geometric artifacts shaped as scalene triangles. For the fi rst time, results of excavations and prospects of future studies at the new sites Uzynagash-1 and -2, dating to the late Early Upper Paleolithic, are outlined. We conclude that Upper Paleolithic cultures (or industries) of the foothills of the Northern Tien Shan are original, while following a single vector with the Upper Paleolithic of Western Eurasia.

Nonlinear current response and electric quantum oscillations in the Dirac semimetal Cd3As2

Chiral anomaly is a distinct quantum anomaly associated with chiral fermions in Dirac or Weyl semimetals. The use of negative magnetoresistance (negative MR) as a signature for this anomaly remains contentious, as trivial mechanisms such as current jetting and weak localization can also induce negative MR. In this study, we report a novel nonlinear behavior of the chiral anomaly in the longitudinal direction, which we observed by applying parallel current and magnetic field to the Dirac semimetal Cd3As2. This nonlinear characteristic peaks at an intermediate magnetic field of approximately 5 T, displaying a resistance-increasing property concomitant with strengthening of the current source. Through angle-dependence experiments, we were able to rule out trivial factors, such as thermal effects, geometric artifacts, and anisotropy. Furthermore, additional electric quantum oscillations were observed when the direct current (DC) was applied as high as 300 μA. Such an unusual phenomenon is ascribed to the formation of quantized levels due to Bloch oscillation in the high DC regime, suggesting that an oscillatory density distribution may arise as the electric field increases. The non-Ohmic electric quantum oscillations open a new avenue for exploring chiral anomaly and other nontrivial topological properties, which is also one of the salient features of nonequilibrium steady states in condensed matter physics.

Geometric parameters sensitivity evaluation based on projection trajectories for X-ray cone-beam computed laminography.

X-ray cone-beam computed laminography (CL) is widely used for large flat objects that computed tomography (CT) cannot investigate. The rotation angle of axis tilt makes geometric correction of CL system more complicated and has more uncertain factors. Therefore, it is necessary to evaluate sensitivity of the geometric parameters of CL system in advance. This study aims to objectively and comprehensively evaluate sensitivity of CL geometric parameters based on the projection trajectory. This study proposes the Minimum Deviation Unit (MDU) to evaluate sensitivity of CL geometric parameters. First, the projection trajectory formulas are derived according to the spatial relationship of CL system geometric parameters. Next, the MDU of the geometric parameters is obtained based on the projection trajectories and used as the evaluation index to measure the sensitivity of parameters. Then, the influence of the rotation angle of the axis tilt and magnification on the MDU of the parameters is analyzed. At low magnification, three susceptible parameters (η, u0, v0) with MDU less than 1 (° or mm) must be calibrated accurately to avoid geometric artifacts. The sensitivity of CL parameters increases as the magnification increases, and all parameters become highly sensitive when the magnification power is greater than 10. The results of this study have important guiding significance for the subsequent further parameter calibration.

Comparison of ZOOMit-DWI sequence and conventional DWI sequence in endometrial cancer.

Magnetic resonance diffusion-weighted imaging (DWI) has important clinical value in diagnosis and curative effect evaluation on endometrial carcinoma. How to improve the detection rate of endometrial small lesions by DWI is the research focus of MRI technology. This study aims to analyze the image quality of small field MRI ZOOMit-DWI sequence and conventional single-shot echo-planar imaging (SS-EPI) DWI sequence in the scanning of endometrial carcinoma, and to explore the clinical value of ZOOMit-DWI sequence. A total of 37 patients with endometrial carcinoma diagnosed by operation and pathology in the Second Xiangya Hospital of Central South University from July 2019 to May 2021 were collected. All patients were scanned with MRI ZOOMit-DWI sequence and SS-EPI DWI sequence before operation. Two radiologists subjectively evaluated the anatomical details, artifacts, geometric deformation and focus definition of the 2 groups of DWI images. At the same time, the signal intensity were measured and the signal-to-noise ratio (SNR), contrast to noise ratio (CNR), and apparent diffusion coefficient (ADC) of the 2 DWI sequences were calculated for objective evaluation. The differences of subjective score, objective score and ADC value of the 2 DWI sequences were analyzed. The SNR of the ZOOMit-DWI group was significantly higher than that of the SS-EPI DWI group (301.96±141.85 vs 94.66±41.26), and the CNR of the ZOOMit-DWI group was significantly higher than that of the SS-EPI DWI group (185.05±105.45 vs 57.91±31.54, P<0.05). There was no significant difference in noise standard deviation between the ZOOMit-DWI group and the SS-EPI DWI group (P>0.05). The subjective score of anatomical detail and focus definition in the ZOOMit-DWI group was significantly higher than that of the SS-EPI DWI group (both P<0.05). The subjective score of artifacts and geometric deformation of ZOOMit-DWI group was significantly lower than that of the SS-EPI DWI group (both P<0.05). ADC had no significant difference between the ZOOMit-DWI group and the SS-EPI DWI group (P>0.05). The image quality of ZOOMit-DWI is significantly higher than that of conventional SS-EPI DWI. In the MRI DWI examination of endometrial carcinoma, ZOOMit-DWI can effectively reduce the geometric deformation and artifacts of the image, which is more conducive to clinical diagnosis and treatment.

Geometric Parameter Self-Calibration Based on Projection Feature Matching for X-Ray Nanotomography

The mismatch of geometric parameters in a nanotomography system bears a significant impact on the reconstructed images. Moreover, projection image noise is increased due to limitations of the X-ray power source. The accuracy of the existing self-calibration method, which uses only the grayscale information of the projected image, is easily affected by noise and leads to reduced accuracy. This paper proposes a geometric parameter self-calibration method based on feature matching of mirror projection images. Firstly, the fast extraction and matching feature points in the mirror projection image are performed by speeded-up robust features (SURF). The feature triangle is then designed according to the stable position of the system’s rotation axis to further filter the feature points. In turn, the influence of the mismatched points on the calculation accuracy is reduced. Finally, the straight line where the rotation axis is located is fitted by the midpoint coordinates of the filtered feature points, thereby realizing geometric parameter calibration of the system. Simulation and actual data from the experimental results show that the proposed method effectively realizes the calibration of geometric parameters, and the blurring and ghosting caused by geometric artifacts are corrected. Compared with existing methods, the image clarity can be improved by up to 14.4%.

A C-arm photon counting CT prototype with volumetric coverage using multi-sweep step-and-shoot acquisitions

Objective. Existing clinical C-arm interventional systems use scintillator-based energy-integrating flat panel detectors (FPDs) to generate cone-beam CT (CBCT) images. Despite its volumetric coverage, FPD-CBCT does not provide sufficient low-contrast detectability desired for certain interventional procedures. The purpose of this work was to develop a C-arm photon counting detector (PCD) CT system with a step-and-shoot data acquisition method to further improve the tomographic imaging performance of interventional systems. Approach. As a proof-of-concept, a cadmium telluride-based 51 cm × 0.6 cm PCD was mounted in front of a FPD in an Artis Zee biplane system. A total of 10 C-arm sweeps (5 forward and 5 backward) were prescribed. A motorized patient table prototype was synchronized with the C-arm system such that it translates the object by a designated distance during the sub-second rest time in between gantry sweeps. To evaluate whether this multi-sweep step-and-shoot acquisition strategy can generate high-quality and volumetric PCD-CT images without geometric distortion artifacts, experiments were performed using physical phantoms, a human cadaver head, and an in vivo swine subject. Comparison with FPD-CT was made under matched narrow beam collimation and radiation dose conditions. Main results. Compared with FPD-CT images, PCD-CT images had lower noise and improved visualization of low-contrast lesion models, as well as improved visibility of small iodinated blood vessels. Fine structures were visualized more clearly by the PCD-CT than the highest-available resolution provided by FPD-CBCT and MDCT. No perceivable geometric distortion artifacts were observed in the multi-planar PCD-CT images. Significance. This work is the first demonstration of the feasibility of high-quality and multi-planar (volumetric) PCD-CT imaging with a rotating C-arm gantry.

Angle prediction model when the imaging plane is tilted about z-axis.

Computer Tomography (CT) is a complicated imaging system, requiring highly geometric positioning. We found a special artifact caused by detection plane tilted around z-axis. In short scan cone-beam reconstruction, this kind of geometric deviation result in half circle shaped fuzzy around highlighted particles in reconstructed slices. This artifact is distinct near the slice periphery, but deficient around the slice center. We generated mathematical models, and InceptionV3-R deep network to study the slice artifact features to estimate the detector z-axis tilt angle. The testing results are: mean absolute error of 0.08819 degree, the Root mean square error of 0.15221 degree and R-square of 0.99944. A geometric deviation recover formula was deduced, which can eliminate this artifact efficiently. This research enlarges the CT artifact knowledge hierarchy, and verifies the capability of machine learning in CT geometric deviation artifact recoveries.

Reproducibility and feasibility of optic nerve diffusion MRI techniques: single-shot echo-planar imaging (EPI), readout-segmented EPI, and reduced field-of-view diffusion-weighted imaging

BackgroundDiffusion-weighted imaging (DWI) is an essential technique for optic nerve diseases. However, the image quality of optic nerve DWI is decreased by the distortions and artifacts associated with conventional techniques. In order to establish this method as a critical tool in optic nerve diseases, reproducibility and feasibility of new technical and conventional approaches of DWI need to be systematically investigated.MethodsDWIs were acquired using ss-EPI, readout-segmented EPI (rs-EPI) DWI, and reduced field-of-view (rFOV) DWI. 26 volunteers (mean age 31.2 years) underwent repeated MRI examinations in order to assess scan–rescan reproducibility and accuracy. The apparent diffusion coefficient (ADC) values (three ROIs were measured on each side) were determined to evaluate the reproducibility of each sequence and the differences between the three techniques. To quantify the geometric distortion artifacts, the length of optic nerve and the maximum angle of optic nerve were defined and compared to T2-weighted imaging. In addition, two readers evaluated four different aspects of image quality on 5-point Likert scales.Resultsrs-EPI DWI (ICCs: 0.916, 0.797 and 0.781) and rFOV DWI (ICCs: 0.850, 0.595 and 0.750) showed higher reproducibility (ICCs: ROI1, ROI2 and ROI3) of mean ADC value in all three ROIs than ss-EPI DWI (ICCs: 0.810, 0.442 and 0.379). The quantitative analysis of geometric distortion yielded a higher agreement of both rs-EPI DWI and rFOV DWI with T2-weighted imaging than ss-EPI. rs-EPI DWI (2.38 ± 0.90) and rFOV DWI (2.46 ± 0.58) were superior to ss-EPI DWI (1.58 ± 0.64) with respect to overall image quality and other aspects of image quality, each with P < 0.05. The mean ADC values of rFOV DWI were significantly lower than those of rs-EPI DWI and ss-EPI DWI in all three ROIs (P < 0.001).ConclusionsBoth rs-EPI DWI and rFOV-EPI DWI are suitable techniques for the assessment of diffusion restriction and provide significantly improved image quality compared with ss-EPI DWI. For methods using the same acquisition time, rFOV DWI is superior to ss-EPI DWI, while rs-EPI showed an overall superiority, although this technique took 47% longer to perform.

Potential and Limitation of PlanetScope Images for 2-D and 3-D Earth Surface Monitoring With Example of Applications to Glaciers and Earthquakes

The Planet PlanetScope cubesat constellation acquires high-resolution optical images which cover the entire surface of the Earth daily, enabling an unprecedented capability to monitor the Earth surface changes. However, our analysis reveals artifacts of the geometry of PlanetScope images related to the imaging system and processing issues, limiting the usability of these data for various earth science applications, including the monitoring of glaciers, dune motion, or the measurement of ground deformation due to earthquakes and landslides. Here we analyze these artifacts and propose ways to remediate them. We use two examples to evaluate the data and assess the performance of our proposed approaches. The first is the ground deformation caused by the 2019 Ridgecrest earthquake sequence, California, and the second is the 2018-2019 surge of the Shisper glacier in the Karakorum. Using an image correlation technique, we show that PlanetScope images exhibit several geometric artifacts, such as scene-to-scene misregistration, inconsistence geolocation accuracy between spectral bands, and topographic artifacts. Altogether these artifacts make a quantitative analysis of ground displacement difficult and inaccurate. We present a method that remediates most of these geometric artifacts. In addition, we propose a framework for selecting the most appropriate images and a procedure for refining the Rational Function Model (RFM) of unrectified images to monitor surface displacements and topography changes in 3D. These tools should enhance the use of PlanetScope images for Earth Science applications.

Development of an Integrated C-Arm Interventional Imaging System With a Strip Photon Counting Detector and a Flat Panel Detector.

Modern interventional x-ray systems are often equipped with flat-panel detector-based cone-beam CT (FPD-CBCT) to provide tomographic, volumetric, and high spatial resolution imaging of interventional devices, iodinated vessels, and other objects. The purpose of this work was to bring an interchangeable strip photon-counting detector (PCD) to C-arm systems to supplement (instead of retiring) the existing FPD-CBCT with a high quality, spectral, and affordable PCD-CT imaging option. With minimal modification to the existing C-arm, a 51×0.6 cm2 PCD with a 0.75 mm CdTe layer, two energy thresholds, and 0.1 mm pixels was integrated with a Siemens Artis Zee interventional imaging system. The PCD can be translated in and out of the field-of-view to allow the system to switch between FPD and PCD-CT imaging modes. A dedicated phantom and a new algorithm were developed to calibrate the projection geometry of the narrow-beam PCD-CT system and correct the gantry wobbling-induced geometric distortion artifacts. In addition, a detector response calibration procedure was performed for each PCD pixel using materials with known radiological pathlengths to address concentric artifacts in PCD-CT images. Both phantom and human cadaver experiments were performed at a high gantry rotation speed and clinically relevant radiation dose level to evaluate the spectral and non-spectral imaging performance of the prototype system. Results show that the PCD-CT system has excellent image quality with negligible artifacts after the proposed corrections. Compared with FPD-CBCT images acquired at the same dose level, PCD-CT images demonstrated a 53% reduction in noise variance and additional quantitative imaging capability.

Proposal of an algorithm to evaluate geometric distortion and artifact spreading in digital breast tomosynthesis.

According to the European Reference Organization for Quality Assurance Breast Screening and European Diagnostic Services, the spatial accuracy of reconstructed images and reconstruction artifacts must be evaluated in digital breast tomosynthesis (DBT) quality control procedures. To propose a computational algorithm to evaluate the geometric distortion and artifact spreading (GDAS) in DBT images. The proposed algorithm analyzed tomosynthesis images of a phantom that contains aluminum spheres (1 mm in diameter) arranged in a rectangular matrix spaced 5 cm apart that was inserted in 5-mm-thick polymethylmethacrylate (PMMA). The obtained results were compared with the values provided by the algorithm developed by the National Coordinating Center for the Physics of Mammography (NCCPM). In the comparison, the results depended on the dimensions of the region of interest (ROI). This dependence proves the benefit of the proposed algorithm because it allows the user to select the ROI. The computational algorithm proved to be useful for the evaluation of GDAS in DBT images, in the same way as the reference algorithm (NCCPM), as well as allowing the selection of the ROI dimensions that best suit the spreading of the artifact in the analyzed images.

Geometric Artifacts Correction for Computed Tomography Exploiting A Generative Adversarial Network

Geometrical accuracy of an X-ray Computed Tomography (CT) system is crucial to achieve high quality tomographic reconstructions. Methods to correct the resulting geometric artifacts have been comprehensively described in the past few years. Deep convolution neural network is increasingly used in CT imaging, which has a great potential in image feature learning and processing tasks. In this work, a geometric artifact correction method exploiting generative adversarial networks (GAN) is developed. The U-Net structure is employed as the generator of the network to extract the CT image features with geometric artifacts. The convolutional neural network (CNN) based on image block perception acts as a discriminator for the network, further constraining the optimization of the generator. The proposed method has been shown experimentally feasible for geometric artifacts correction in circular cone beam CT by performing more accurate feature extraction. The peak signal to noise ratio (PSNR) of the corrected phantom images increased by 11.812 on average, and the root mean square error (RMSE) decreased by 9.982 on average.

Correction of geometric artifact in cone-beam computed tomography through a deep neural network

Cone-beam computed tomography is a noninvasive detection system that can obtain the three-dimensional structure of objects in a way that does not damage the object. It is widely applied in precision instruments, medical detection, and other fields. However, in the actual process, if a geometric artifact appears in the results, it will affect the quality of reconstructed images, including detail loss and decreased spatial resolution, which leads to inaccurate distinction of defects in detection. We propose a method for correcting a geometric artifact by means of data-driven projection and neural networks. The network designed is a deep neural network with six convolutional layers and six deconvolutional layers that can correct a geometric artifact by training a large number of labeled data and unlabeled data. Compared with other networks that require prior information for reconstructed images, the proposed method uses a projection data-driven approach that can avoid the requirement for prior information. The simulation data have been tested under varying degrees of noise, and satisfactory geometric artifact correction results have been obtained. Meanwhile, we use the actual data of line pairs and ball grid array solder joints to conduct experiments. The results obtained by our method are compared with two other phantom-based method and the U-net method, respectively. The results of similarity and spatial resolution show that the proposed method can achieve the comparable results as the two types of methods. At the same time, we apply a projection data-driven approach to avoid the requirement for prior information, which is more conducive to the correction of the geometric artifact in practical situations where prior information is lacking.