X-ray Computed Tomography Reconstruction Research Articles

Projection-Angle-Sensor-Assisted X-ray Computed Tomography for Cylindrical Lithium-Ion Batteries.

X-ray computed tomography (XCT) has become a powerful technique for studying lithium-ion batteries, allowing non-destructive 3D imaging across multiple spatial scales. Image quality is particularly important for observing the internal structure of lithium-ion batteries. During multiple rotations, the existence of cumulative errors and random errors in the rotary table leads to errors in the projection angle, affecting the imaging quality of XCT. The accuracy of the projection angle is an important factor that directly affects imaging. However, the impact of the projection angle on XCT reconstruction imaging is difficult to quantify. Therefore, the required precision of the projection angle sensor cannot be determined explicitly. In this research, we selected a common 18650 cylindrical lithium-ion battery for experiments. By setting up an XCT scanning platform and installing an angle sensor to calibrate the projection angle, we proceeded with image reconstruction after introducing various angle errors. When comparing the results, we found that projection angle errors lead to the appearance of noise and many stripe artifacts in the image. This is particularly noticeable in the form of many irregular artifacts in the image background. The overall variation and residual projection error in detection indicators can effectively reflect the trend in image quality. This research analyzed the impact of projection angle errors on imaging and improved the quality of XCT imaging by installing angle sensors on a rotary table.

Orchestration of materials science workflows for heterogeneous resources at large scale

In the era of big data, materials science workflows need to handle large-scale data distribution, storage, and computation. Any of these areas can become a performance bottleneck. We present a framework for analyzing internal material structures (e.g., cracks) to mitigate these bottlenecks. We demonstrate the effectiveness of our framework for a workflow performing synchrotron X-ray computed tomography reconstruction and segmentation of a silica-based structure. Our framework provides a cloud-based, cutting-edge solution to challenges such as growing intermediate and output data and heavy resource demands during image reconstruction and segmentation. Specifically, our framework efficiently manages data storage, scaling up compute resources on the cloud. The multi-layer software structure of our framework includes three layers. A top layer uses Jupyter notebooks and serves as the user interface. A middle layer uses Ansible for resource deployment and managing the execution environment. A low layer is dedicated to resource management and provides resource management and job scheduling on heterogeneous nodes (i.e., GPU and CPU). At the core of this layer, Kubernetes supports resource management, and Dask enables large-scale job scheduling for heterogeneous resources. The broader impact of our work is four-fold: through our framework, we hide the complexity of the cloud’s software stack to the user who otherwise is required to have expertise in cloud technologies; we manage job scheduling efficiently and in a scalable manner; we enable resource elasticity and workflow orchestration at a large scale; and we facilitate moving the study of nonporous structures, which has wide applications in engineering and scientific fields, to the cloud. While we demonstrate the capability of our framework for a specific materials science application, it can be adapted for other applications and domains because of its modular, multi-layer architecture.

Synthetic Data Generation for Automatic Segmentation of X-ray Computed Tomography Reconstructions of Complex Microstructures.

The greatest challenge when using deep convolutional neural networks (DCNNs) for automatic segmentation of microstructural X-ray computed tomography (XCT) data is the acquisition of sufficient and relevant data to train the working network. Traditionally, these have been attained by manually annotating a few slices for 2D DCNNs. However, complex multiphase microstructures would presumably be better segmented with 3D networks. However, manual segmentation labeling for 3D problems is prohibitive. In this work, we introduce a method for generating synthetic XCT data for a challenging six-phase Al-Si alloy composite reinforced with ceramic fibers and particles. Moreover, we propose certain data augmentations (brightness, contrast, noise, and blur), a special in-house designed deep convolutional neural network (Triple UNet), and a multi-view forwarding strategy to promote generalized learning from synthetic data and therefore achieve successful segmentations. We obtain an overall Dice score of 0.77. Lastly, we prove the detrimental effects of artifacts in the XCT data on achieving accurate segmentations when synthetic data are employed for training the DCNNs. The methods presented in this work are applicable to other materials and imaging techniques as well. Successful segmentation coupled with neural networks trained with synthetic data will accelerate scientific output.

Fused x-ray and fast neutron CT reconstruction for imaging large and dense objects

Megavolt x-ray computed tomography (CT) is a powerful tool for three-dimensional characterization. However, its utility is limited for large objects composed of high-atomic number (Z) materials, where x rays fail to penetrate. Information from fast neutron CT (FNCT) can complement x-ray CT reconstructions since fast neutrons can more readily penetrate high-Z objects. Here, we demonstrate a method for combining FNCT and x-ray CT data to create a single reconstruction, more accurate than could be achieved with either x rays or fast neutrons alone. The algorithm was tested on an exemplar comprising multiple concentric, nested cylinders of different materials. Simulated and empirical x-ray CT data were acquired for the exemplar using a 9 MV bremsstrahlung spectrum. Additional simulated and empirical FNCT data were acquired using an accelerator based fast neutron source. The FNCT data were used to synthesize x-ray CT data and augment the x-ray CT data missing due to lack of penetration. This approach mitigates artifacts that would otherwise negatively affect the accuracy and resolution of a single-modality reconstructed volume.

Critique of ”MemXCT: Memory-Centric X-Ray CT Reconstruction With Massive Parallelization” by SCC Team From Nanyang Technological University

In this technical report, we focus on reproducing the results reported in the paper MemXCT: Memory-Centric X-ray CT Reconstruction with Massive Parallelization. MemXCT is a scalable approach to X-ray Computed Tomography reconstruction which removes redundant computation. We reproduced the single CPU/GPU performance as well as strong scaling experiments. We set up our congurations on Microsoft Azure CycleCloud and have two clusters. One cluster has 4 nodes with 60 CPUs on each node and the other cluster has 4 nodes with 4 NVIDIA V100 GPUs on each node. Both clusters come with InniBand. The original author conducted his experiments on Theta and Blue Waters supercomputers. We were able to reproduce part of the results in the original paper, however, failed to produce similar performance on other experiments. This report was submitted as part of the reproducibility challenge in SC20 Student Cluster Competition.

Exact gram filtering and efficient backprojection for iterative CT reconstruction.

Forward and backprojections are the basis of all model-based iterative reconstruction (MBIR) methods. However, computing these accurately is time-consuming. In this paper, we present a method for MBIR in parallel X-ray beam geometry that utilizes a Gram filter to efficiently implement forward andbackprojection. We propose using voxel-basis and modeling its footprint in a box spline framework to calculate the Gram filter exactly and improve the performance of backprojection. In the special case of parallel X-ray beam geometry, the forward and backprojection can be implemented by an estimated Gram filter efficiently if the sinogram signal is bandlimited. In this paper, a specialized sinogram interpolation method is proposed to eliminate the bandlimited prerequisite and thus improve the reconstruction accuracy. We build on this idea by utilizing the continuity of the voxel-basis' footprint, which provides a more accurate sinogram interpolation and further improves the efficiency and quality of backprojection. In addition, the detector blur effect can be efficiently accounted for in our method to better handle realisticscenarios. The proposed method is tested on both phantom and real computed tomography (CT) images under different resolutions, sinogram sampling steps, and noise levels. The proposed method consistently outperforms other state-of-the-art projection models in terms of speed and accuracy for both backprojection andreconstruction. We proposed a iterative reconstruction methodology for 3D parallel-beam X-ray CT reconstruction. Our experimental results demonstrate that the proposed methodology is accurate, fast, and reproducible, and outperforms alternative state-of-the-art projection models on both backprojection and reconstruction resultssignificantly.

Unraveling pore evolution in post-processing of binder jetting materials: X-ray computed tomography, computer vision, and machine learning

Quality control in metal additive manufacturing prioritizes the development of advanced inspection schemes to characterize the defect evolution during processing and post-processing. This involves grand challenges in detecting internal defects and analyzing large and complex defect datasets in macroscopic samples. Here, we present an inspection pipeline that integrates (i) fast, micro X-ray computed tomography reconstruction, (ii) automated 3D morphology analysis, and (iii) machine learning-based big data analysis. X-ray computed tomography and automated computer vision result in a holistic defect morphology database for the inspected macroscopic volume, based on which machine learning analysis is employed to reveal quantitative insights into the global evolution of defect characteristics beyond qualitative human observations. We demonstrate this pipeline by examining the global-scale pore evolution in post-processing of binder jetting additive manufacturing, from the green state, to the sintered state, and to the hot isostatic pressed state of copper. The pipeline is shown to be effective at detecting and processing the information associated with a large number (∼105) of pores in macroscopic volumes. The subsequent principal component analysis and clustering analysis extract the key morphological descriptors and categorize the detected pores into four morphological groups. By quantifying the evolution of (i) the weight of pore morphology parameters and (ii) the pore number and volume fraction of each categorized group, new understandings are developed regarding the effects of sintering and hot isostatic pressing on pore decomposition, shrinkage, and smoothing during post-processing of binder jetting.

Deformation mechanisms and evolution of mechanical properties in damaged advanced ceramics

We investigated the uniaxial compressive behavior of damaged and intact alumina using quantitative X-ray computed tomography (XCT) analysis coupled with digital image correlation (DIC) for mechanical characterization. Internal three-dimensional crack characteristics such as crack surface area and orientation were quantified using XCT to assess the level of damage. From the quasi-static and dynamic stress–strain results, the primary effects of crack damage are to reduce the initial stiffness and rate of lateral expansion in damaged alumina. With increasing axial strain, crack closure was found to lead to a recovery of elastic properties, in some cases to intact levels, in the damaged specimens. Localized deformation mechanisms related to the crack structure, including lateral crack closure, axial crack opening and closing, and inclined crack sliding, were visualized in-situ and connected to XCT reconstructions. High-speed imaging also revealed a mixed fracture mode for damaged alumina that included axial splitting and failure along pre-existing cracks.

A Splitting-Based Iterative Algorithm for GPU-Accelerated Statistical Dual-Energy X-Ray CT Reconstruction

Non-smooth regularization is widely used in image reconstruction to eliminate the noise while preserving subtle image structures. In this work, we investigate the use of proximal Newton (PN) method to solve an optimization problem with a smooth data-fidelity term and total variation (TV) regularization arising from image reconstruction applications. Specifically, we consider a nonlinear Poisson-modeled single-energy X-ray computed tomography reconstruction problem with the data-fidelity term given by the I-divergence. The PN algorithm is compared to state-of-the-art first-order proximal algorithms, such as the wellestablished fast iterative shrinkage and thresholding algorithm (FISTA), both in terms of number of iterations and time to solutions. We discuss the key factors that influence the performance of PN, including the strength of regularization, the stopping criterion for both sub-problem and main-problem, and the use of exact or approximated Hessian operators.

Insights into immiscible supercritical CO2 EOR: An XCT scanner assisted flow behaviour in layered sandstone porous media

Reservoir heterogeneity and possible associated crossflow between low and high permeability zones may trigger intrinsic uncertainties to manage and predict reservoir performance. To understand the contribution of crossflow to oil recovery during immiscible CO2 flooding, we conducted core flooding experiments using layered core samples with the aid of a high-resolution medical X-Ray computed tomography (XCT) scanner. Our results showed that core scale heterogeneity plays a significant role in dictating the spatial distribution of injected CO2 during flooding thus the oil recovery factor. The experimental results reveal fluid distributions during flooding to be correlated reasonably well with the permeability of each layer. In other words, the higher permeability region of the rock sample contained a higher CO2 saturation at all times compared with the lower permeability one. X-ray images have facilitated and visually demonstrated the channelling of CO2 into the high permeability layer leaving a considerable amount of bypassed oil in the low permeability one. We also observed that crossflow indeed arises between layers of the heterogeneous cores suggesting that capillarity plays a noticeable role in increasing oil recovery (i.e. 4.8%). Three dimensional XCT reconstruction of cores together with 2D cross sectional images validated the occurrence of crossflow that would divert CO2 flow from the high permeability to low permeability zone enhancing oil recovery. The images show that the diverted CO2 into the low permeability layer would be trapped by associated trapping mechanisms (e.g. snapped-off) and would not form a continuous phase along the length of the sample.

Use of X‐Ray Computed Tomography for Validation of Random Packs of Composite Solid Propellants

AbstractThis paper presents a comparison of actual solid propellants with a computationally generated propellant pack, which is intended for use in combustion modelling studies. X‐ray computed tomography (XCT) was used to obtain a 3D reconstruction of an actual propellant sample. Random distribution of spherical particles of ammonium perchlorate (AP) has been used to model a composite solid propellant (random pack). The simulated random pack and 3D reconstructed propellant are compared based on three parameters capturing volume, area and length metrics, viz., density, exposed AP area and AP/binder intercept lengths. The density match is within 1.5 % between the actual propellant and the simulated propellant pack. The exposed AP surface area lies in the range 53–63 % for the 80–86 % solid loaded non‐aluminized propellants considered in the study. The histogram of the intercept lengths matched reasonably as well capturing the variation in particle size and solid loading of the propellant. However, the comparison between simulated and XCT reconstructed packs was found to be limited by the resolution of the XCT reconstruction of actual propellant samples.

Structural analysis of three-dimensional mesh fabric by Micro X-ray computed tomography

Three-dimensional mesh fabric is a special type of knitted spacer fabric with sandwich structure consisting of two separate meshed multifilament outer layers linked together with a layer of spacer monofilaments. It has found wide applications in home textiles and automotive industry due to its good energy-absorbing performance and ventilation property. This article presents a comprehensive structural analysis on a mesh spacer fabric of thickness 10.3 mm in order to place a solid foundation to enhance the understanding of the structural features and mechanical behavior. The real geometric structure was reconstructed by scanning the fabric via Micro X-ray computed tomography. The spacer monofilaments and meshed outer layers were analyzed quantitatively based on the Micro X-ray computed tomography reconstruction. It was found that the spacer monofilaments are different in length, curvature, and torsion. The spacer monofilament lengths can be ranging from 9.43 to 10.77 mm. Their curvatures varied from around 0.1 to 0.3 mm−1. The torsions of spacer monofilaments can be positive and negative with different directions of twisting and had a certain degree of symmetry between two adjacent wales. The monofilament loops in the outer layers also varied in height, width, and enclosed area. Three-dimensional mesh fabric is a periodic but highly inhomogeneous structure.

Improved X-ray computed tomography reconstruction of the largest fragment of the Antikythera Mechanism, an ancient Greek astronomical calculator.

The Antikythera Mechanism is an extraordinarily complex ancient Greek astronomical calculating device whose mode of operation is now relatively well understood particularly since imaging studies in 2005 revealed gears and inscriptions which were previously illegible. Unfortunately, the highest resolution X-ray computed tomography image of the largest fragment had some errors which meant that the reconstructed images were not as clear as had been expected. Here, the original X-ray data have been reanalysed and reconstructed. The new X-ray computed tomography images have improved contrast and resolution, leading to better clarity and legibility. The improvement in image quality is characterised and some examples of writing on the Mechanism which can now be read with increased confidence are given.

Low-Dose X-ray Computed Tomography Reconstruction Using Curvelet Sparse Regularization

Low-Dose X-ray Computed Tomography Reconstruction Using Curvelet Sparse Regularization

Using high resolution X-ray computed tomography to create an image based model of a lymph node

Lymph nodes are an important part of the immune system. They filter the lymphatic fluid as it is transported from the tissues before being returned to the blood stream. The fluid flow through the nodes influences the behaviour of the immune cells that gather within the nodes and the structure of the node itself. Measuring the fluid flow in lymph nodes experimentally is challenging due to their small size and fragility. In this paper, we present high resolution X-ray computed tomography images of a murine lymph node. The impact of the resulting visualized structures on fluid transport are investigated using an image based model. The high contrast between different structures within the lymph node provided by phase contrast X-ray computed tomography reconstruction results in images that, when related to the permeability of the lymph node tissue, suggest an increased fluid velocity through the interstitial channels in the lymph node tissue. Fluid taking a direct path from the afferent to the efferent lymphatic vessel, through the centre of the node, moved faster than the fluid that flowed around the periphery of the lymph node. This is a possible mechanism for particles being moved into the cortex.

Using Laser Ultrasound to Detect Subsurface Defects in Metal Laser Powder Bed Fusion Components

Laser powder bed fusion offers many advantages over conventional manufacturing methods, such as the integration of multiple parts that can result in significant weight-savings. The increased design freedom that layer-wise manufacture allows has also been seen to enhance component performance at little or no added cost. For such benefits to be realized, however, the material quality must first be assured. Laser ultrasonic testing is a noncontact inspection technique that has been proposed as suitable for in situ monitoring of metal additive manufacturing processes. This article explores the current capability of this technique to detect manufactured, subsurface defects in Ti-6Al-4V samples, ex situ. The results are compared with x-ray computed tomography reconstructions and focus variation microscopy. Although laser ultrasound has been used to identify material discontinuities, further work is required before this technique could be implemented in situ.

Characterising and linking X-ray CT derived macroporosity parameters to infiltration in soils with contrasting structures

Soils deliver the regulating ecosystem services of water infiltration and distribution, which can be controlled by macropores. Parameterizing macropore hydraulic properties is challenging due to the lack of direct measurement methods. With tension-disc infiltrometry hydraulic properties near saturation can be measured. Differentiating between hydrologically active and non-active pores, at a given water potential, indirectly assesses macropore continuity. Water flow through macropores is controlled by macropore size distribution, tortuosity, and connectivity, which can be directly derived by X-ray computed tomography (CT). Our objective was to parameterize macropore hydraulic properties based on the imaged macropore network of three horizons of an Andosol and a Gleysol. Hydraulic conductivity Kunsat was derived from infiltration measurements. Soil cores from the infiltration areas were scanned with X-ray CT. Kunsat was significantly higher in the Andosol than in the Gleysol at all water potentials, and decreased significantly with depth in both soils. The in situ measurements guided the definition of new macroporosity parameters from the X-ray CT reconstructions. For the Andosol, Kunsat was best predicted using the imaged-limited macroporosity. A low total macroporosity, coupled with a high macropore density, indicated the abundance of smaller macropores, leading to homogeneous matrix flux. Imaged macropores were not well connected. In contrast, the Gleysol had a bi-modal macropore system with few very large, but well-connected macropores. Kunsat was best predicted using the imaged macroporosity consisting only of macropores with diameters between 0.75 and 3mm. Our research demonstrates that linking traditional soil physical measurements with soil-visualization techniques has a huge potential to improve parameterizing macropore hydraulic properties. The relevance of the relationships found in this study for larger scales and other soil types still needs to be tested, for example by a multi-scale investigation including a much wider range of different soils.

Iterative reconstruction for sparse-view x-ray CT using alpha-divergence constrained total generalized variation minimization.

Accurate statistical model of the measured projection data is essential for computed tomography (CT) image reconstruction. The transmission data can be described by a compound Poisson distribution upon an electronic noise background. However, such a statistical distribution is numerically intractable for image reconstruction. Although the sinogram data is easily manipulated, it lacks a statistical description for image reconstruction. To address this problem, we present an alpha-divergence constrained total generalized variation (AD-TGV) method for sparse-view x-ray CT image reconstruction. The AD-TGV method is formulated as an optimization problem, which balances the alpha-divergence (AD) fidelity and total generalized variation (TGV) regularization in one framework. The alpha-divergence is used to measure the discrepancy between the measured and estimated projection data. The TGV regularization can effectively eliminate the staircase and patchy artifacts which is often observed in total variation (TV) regularization. A modified proximal forward-backward splitting algorithm was proposed to minimize the associated objective function. Qualitative and quantitative evaluations were carried out on both phantom and patient data. Compared with the original TV-based method, the evaluations clearly demonstrate that the AD-TGV method achieves higher accuracy and lower noise, while preserving structural details. The experimental results show that the presented AD-TGV method can achieve more gains over the AD-TV method in preserving structural details and suppressing image noise and undesired patchy artifacts. The authors can draw the conclusion that the presented AD-TGV method is potential for radiation dose reduction by lowering the milliampere-seconds (mAs) and/or reducing the number of projection views.

X-ray Computed Tomography using a Sparsity Enforcing Prior Model Based on Haar Transformation in a Bayesian Framework

X-ray Computed Tomography (CT) has become a hot topic in both medical and industrial applications in recent decades. Reconstruction by using a limited number of projections is a significant research domain. In this paper, we propose to solve the X-ray CT reconstruction problem by using the Bayesian approach with a hierarchical structured prior model basing on the multilevel Haar transformation. In the proposed model, the multilevel Haar transformation is used as the sparse representation of a piecewise continuous image, and a generalized Student-t distribution is used to enforce its sparsity. The simulation results compare the performance of the proposed method with some state-of-the-art methods.

Areal surface texture data extraction from X-ray computed tomography reconstructions of metal additively manufactured parts

Many applications that exploit the manufacturing flexibility of additive manufacturing (AM) produce surfaces, primarily internal features, which cannot be measured using conventional contact or line-of-sight optical methods. This paper evaluates the capability of a novel technique to extract areal surface data from micro-focus X-ray computed tomography (XCT) from AM components and then generate surface parameter data per ISO 25178-2. This non-destructive evaluation of internal features has potential advantages during AM product research and commercial production. The data extracted from XCT is compared with data extracted using a focus variation instrument. A reference dimensional artefact is included in all XCT measurements to evaluate XCT surface determination performance and dimensional scaling accuracy. Selected areal parameters generated using the extraction technique are compared, including Sa, for which the nominal difference between the value obtained using XCT and used the focus variation method was less than 2.5%.