Search

Background and purpose We investigated whether automated and reader-based ASPECTS in acute stroke patients are affected by different CT image reconstruction algorithms. Methods ASPECTS were assessed by commercial software and four independent blinded readers (two residents and two consultants) from different CT reconstructions (filtered back projection and two different iterative reconstruction [IR] levels) in 43 acute stroke patients with proximal middle cerebral artery occlusion. Ground truth was provided by an expert with unrestricted data access. Results The residents showed significant variations between IR levels and had a significantly lower internal consistency across different reconstructions compared to the software, which performed similarly to the consultants. The consultant as well as the software also showed different deviations from ground truth with different IR levels, which were least at IR strength level 2. Conclusions CT image postprocessing affects either automated or human ASPECTS in acute stroke patients. This effect was most pronounced in the less experienced readers, while the software had the most robust performance.

X-ray CT is increasingly being adopted in manufacturing as a non destructive inspection tool. Traditionally, industrial workflows follow a two step procedure of reconstruction followed by segmentation. Such workflows suffer from two main problems: (1) The reconstruction typically requires thousands of projections leading to increased data acquisition times. (2) The application of the segmentation process a posteriori is dependent on the quality of the original reconstruction and often does not preserve data fidelity. We present a fast iterative X-ray CT method which simultaneously reconstructs and segments an image from a limited number of projections called Fourier Null Space Regularization (FNSR). The novelty of the approach is in the explicit updating of the image null space with values derived from a regularized image from the previous iteration, thus compensating for any missing projections and effectively regularizing the reconstruction. The speed of the method is achieved by directly applying the Fourier Slice Theorem where the Non-Uniform Fast Fourier Transform (NUFFT) is used to compute the frequency spectrum of the projections at their positions in the image k-space. At each iteration a segmented image is computed which is used to populate the null values of the image k-space effectively steering the reconstruction towards a binary solution. The effectiveness of the method to generate accurate reconstructions is demonstrated and benchmarked against other iterative reconstruction techniques using a series of numerical examples. Finally, FNSR is validated using industrial X-ray CT data where accurate reconstructions were achieved with 18 or more projections, a significant reduction from the 5000 needed by filtered back projection.

Background Results of recent phantom studies show that variation in CT acquisition parameters and reconstruction techniques may make radiomic features largely nonreproduceable and of limited use for prognostic clinical studies. Purpose To investigate the effect of CT radiation dose and reconstruction settings on the reproducibility of radiomic features, as well as to identify correction factors for mitigating these sources of variability. Materials and Methods This was a secondary analysis of a prospective study of metastatic liver lesions in patients who underwent staging with single-energy dual-source contrast material-enhanced staging CT between September 2011 and April 2012. Technique parameters were altered, resulting in 28 CT data sets per patient that included different dose levels, section thicknesses, kernels, and reconstruction algorithm settings. By using a training data set (n = 76), reproducible intensity, shape, and texture radiomic features (reproducibility threshold, R2 ≥ 0.95) were selected and correction factors were calculated by using a linear model to convert each radiomic feature to its estimated value in a reference technique. By using a test data set (n = 75), the reproducibility of hierarchical clustering based on 106 radiomic features measured with different CT techniques was assessed. Results Data in 78 patients (mean age, 60 years ± 10; 33 women) with 151 liver lesions were included. The percentage of radiomic features deemed reproducible for any variation of the different technical parameters was 11% (12 of 106). Of all technical parameters, reconstructed section thickness had the largest impact on the reproducibility of radiomic features (12.3% [13 of 106]) if only one technical parameter was changed while all other technical parameters were kept constant. The results of the hierarchical cluster analysis showed improved clustering reproducibility when reproducible radiomic features with dedicated correction factors were used (ρ = 0.39-0.71 vs ρ = 0.14-0.47). Conclusion Most radiomic features are highly affected by CT acquisition and reconstruction settings, to the point of being nonreproducible. Selecting reproducible radiomic features along with study-specific correction factors offers improved clustering reproducibility. © RSNA, 2019 Online supplemental material is available for this article. See also the editorial by Sosna in this issue.

Limited angle CT reconstruction is an under-determined linear inverse problem that requires appropriate regularization techniques to be solved. In this work we study how pre-trained generative adversarial networks (GANs) can be used to clean noisy, highly artifact laden reconstructions from conventional techniques, by effectively projecting onto the inferred image manifold. In particular, we use a robust version of the popularly used GAN prior for inverse problems, based on a recent technique called corruption mimicking, that significantly improves the reconstruction quality. The proposed approach operates in the image space directly, as a result of which it does not need to be trained or require access to the measurement model, is scanner agnostic, and can work over a wide range of sensing scenarios.

Total variation based on the L0− and L1−norm regularizers is a popular and powerful method for computed tomography (CT) reconstruction from noisy data. However their solutions are difficult to compute because of non-differentiability and even discontinuity. In this paper, we propose smoothing L0- and L1-norm regularizers to approximate the original norms. Also, we provide two expectation maximization (EM) like iterative algorithms, which have properties similar to that of the EM algorithm, including the monotonous decrement of the cost function and the self-constraint within a non-negative space. Finally, we build the relationship between the non-local means (NLM) methods and the above methods. The simulated CT projections are used to evaluate the performance of the proposed algorithms. The results show the superiority of the new methods over the existing methods in terms of noise suppression and computational cost.

In this study, we combine dictionary learning and image gradient l_0-norm (l_0DL) for limited-angle CT reconstruction. The proposed l_0DL method can characterize image details and features by training an over-complete dictionary. It also employs the image gradient l_0-norm to protect image edges and reduce shadow artifacts in limited-angle CT. Real dataset experiments are performed to evaluate the outperformances of proposed l_0DL method than other state-of-the-art methods.

In iterative CT reconstruction, the regularization parameter is quite important because it balances the fidelity term and penalty term. Images reconstructed with the optimal regularization parameter will keep the detail preserved and the noise restrained at the same time. While in conventional CT reconstruction, the selection of the regularization parameter is very time-consuming. Besides, the fixed regularization parameter during the iterations is not suitable for every area. For example, the bone area contains more noise than soft tissue areas. With fixed regularization parameter may sacrifice the other resolution. In order to solve this question, in this paper, we proposed an automatic regularization parameter tuning strategy based on region variance. The proposed method based on the region variance tunes the regularization parameter automatically. Experiments show that the proposed method exhibits well in small detail preservation and noise reduction.

Limited-angle reconstruction is a very important but challenging problem in the field of computed tomography (CT) and it has been extensively studied for many years. However, some difficulties still remain. Based on the theory of visible and invisible boundary developed by Quinto et.al, we propose a reconstruction model for limited-angle CT, which encodes the visible edges as priors to recover the invisible ones. The new model introduces an anisotropic ep (0

Computed tomography (CT) image reconstruction and restoration are very important in medical image processing, and are associated together to be an inverse problem. Image iterative reconstruction is a key tool to increase the applicability of CT imaging and reduce radiation dose. Nevertheless, traditional image iterative reconstruction methods are limited by the sampling theorem and also the blurring of projection data will propagate unhampered artifact in the reconstructed image. To overcome these problems, image restoration techniques should be developed to accurately correct a wide variety of image degrading effects in order to effectively improve image reconstruction. In this paper, a blind image restoration technique is embedded in the compressive sensing CT image reconstruction, which can result in a high-quality reconstruction image using fewer projection data. Because a small amount of data can be obtained by radiation in a shorter time, high-quality image reconstruction with less data is equivalent to reducing radiation dose. Technically, both the blurring process and the sparse representation of the sharp CT image are first modeled as a serial of parameters. The sharp CT image will be obtained from the estimated sparse representation. Then, the model parameters are estimated by a hierarchical Bayesian maximum posteriori formulation. Finally, the estimated model parameters are optimized to obtain the final image reconstruction. We demonstrate the effectiveness of the proposed method with the simulation experiments in terms of the peak signal to noise ratio (PSNR), and structural similarity index (SSIM).

Objectives The accuracy of pedicle screw placement strongly affects the outcome of spinal surgery and has mainly relied on the surgeons’ experience. There is no simple, low-cost, and effective pedicle screw placement system to assist new spinal surgeons with less experience. Methods We designed a localization system with six parameters (starting point height [SP-H], starting point length [SP-L], transverse section angle, sagittal section angle [SSA], pedicle width [W] and height [H]) based on preoperative computed tomography reconstruction and combined it with the Roussouly classification to guide lumbar spine pedicle screw placement and analysed the change patterns of the six parameters in 50 participants. Results Based on the system, we confirmed that combining SP-H and SP-L can localize the entrance of the pedicle screw. Furthermore, we considered that SP-L and transverse section angle would be a new standard for determination of the transverse orientation of the pedicle screw. More importantly, the linear regression equations between H and W and SP-H and H were concealed. In addition, H and W can guide the appropriate selection of pedicle screw. Moreover, change patterns of SSA combined with the Roussouly classification indicate that SSA of L3 can be used as a benchmark to guide the establishment of sagittal alignment of the lumbar spine. Conclusions Understanding and applying the six-parameter localization system are essential for achieving accuracy in lumbar spine pedicle screw placement, and the system is a useful guide in the establishment of sagittal alignment. The translational potential of this article This study provides a new pedicle-screw placement system for accurate lumbar spine pedicle screw placement based on three-dimensional CT reconstruction, requiring six parameters to guide the system.

Model‐based iterative reconstruction (MBIR) reduces CT imaging dose while maintaining image quality. However, MBIR reduces noise while preserving edges which may impact intensity‐based tasks such as auto‐segmentation. This work evaluates the sensitivity of an auto‐contouring prostate atlas across multiple MBIR reconstruction protocols and benchmarks the results against filtered back projection (FBP). Images were created from raw projection data for 11 prostate cancer cases using FBP and nine different MBIR reconstructions (3 protocols/3 noise reduction levels) yielding 10 reconstructions/patient. Five bony structures, bladder, rectum, prostate, and seminal vesicles (SVs) were segmented using an auto‐segmentation pipeline that renders 3D binary masks for analysis. Performance was evaluated for volume percent difference (VPD) and Dice similarity coefficient (DSC), using FBP as the gold standard. Nonparametric Friedman tests plus post hoc all pairwise comparisons were employed to test for significant differences (P

Background Iterative computed tomography (CT) image reconstruction shows high potential for the preservation of image quality in diagnostic CT while reducing patients’ exposure; it has become available for low-dose CT (LD-CT) in high-end hybrid imaging systems (e.g. single-photon emission computed tomography [SPECT]-CT). Purpose To examine the effect of an iterative CT reconstruction algorithm on image quality, image noise, detectability, and the reader’s confidence for LD-CT data by a subjective assessment. Material and Methods The LD-CT data were validated for 40 patients examined by an abdominal hybrid SPECT-CT (U = 120 kV, I = 40 mA, pitch = 1.375). LD-CT was reconstructed using either filtered back projection (FBP) or an iterative image reconstruction algorithm (Adaptive Statistical Iterative Reconstruction [ASIR]®) with different parameters (ASIR levels 50% and 100%). The data were validated by two independent blinded readers using a scoring system for image quality, image noise, detectability, and reader confidence, for a predefined set of 16 anatomic substructures. Results The image quality was significantly improved by iterative reconstruction of the LD-CT data compared with FBP ( P ≤ 0.0001). While detectability increased in only 2/16 structures ( P ≤ 0.03), the reader’s confidence increased significantly due to iterative reconstruction ( P ≤ 0.002). Meanwhile, at the ASIR level of 100%, the detectability in bone structure was highly reduced ( P = 0.003). Conclusion An ASIR level of 50% represents a good compromise in abdominal LD-CT image reconstruction. The specific ASIR level improved image quality (reduced image noise) and reader confidence, while preserving detectability of bone structure.

Photon counting CT (PCCT) is an x-ray imaging technique that has undergone great development in the past decade. PCCT has the potential to improve dose efficiency and low-dose performance. In this paper, we propose a statistics-based iterative algorithm to perform a direct reconstruction of material-decomposed images. Compared with the conventional sinogram-based decomposition method which has degraded performance in low-dose scenarios, the multi-energy alternating minimization algorithm for photon counting CT (MEAM-PCCT) can generate accurate material-decomposed image with much smaller biases.

Tuning parameters in a reconstruction model is of central importance to iterative CT reconstruction, since it critically affects the resulting image quality. Manual parameter tuning is not only tedious, but becomes impractical when there exits a number of parameters. In this paper, we develop a novel deep reinforcement learning (DRL) framework to train a parameter-tuning policy network (PTPN) to automatically adjust parameters in a human-like manner. A quality assessment network (QAN) is trained together with PTPN to learn how to judge CT image quality, serving as a reward function to guide the reinforcement learning. We demonstrate our idea in an iterative CT reconstruction problem with pixel-wise total-variation regularization. Experimental results demonstrates the effectiveness of both PTPN and QAN, in terms of tuning parameter and evaluating image quality, respectively.

Background This study aimed to evaluate the safety and feasibility of subaxial cervical pedicle screw (CPS) insertion by comparing the morphological parameters between developmental canal stenosis (DCS) and non-developmental canal stenosis (NDCS) patients. Methods A total of 120 Chinese patients who had undergone cervical spinal multiplanar CT imaging from September 2010 to December 2014 were included in this study. According to the Pavlov ratio (PR), participants were classified into a DCS group (PR

There is an increasing need to reconstruct objects in four or more dimensions corresponding to space, time and other independent parameters. The best 4D reconstruction algorithms use regularized iterative reconstruction approaches such as model based iterative reconstruction (MBIR), which depends critically on the quality of the prior modeling. Recently, Plug-and-Play methods have been shown to be an effective way to incorporate advanced prior models using state-of-the-art denoising algorithms designed to remove additive white Gaussian noise (AWGN). However, state-of-the-art denoising algorithms such as BM4D and deep convolutional neural networks (CNNs) are primarily available for 2D and sometimes 3D images. In particular, CNNs are difficult and computationally expensive to implement in four or more dimensions, and training may be impossible if there is no associated high-dimensional training data. In this paper, we present Multi-Slice Fusion, a novel algorithm for 4D and higher-dimensional reconstruction, based on the fusion of multiple low-dimensional denoisers. Our approach uses multi-agent consensus equilibrium (MACE), an extension of Plug-and-Play, as a framework for integrating the multiple lower-dimensional prior models. We apply our method to the problem of 4D cone-beam X-ray CT reconstruction for Non Destructive Evaluation (NDE) of moving parts. This is done by solving the MACE equations using lower-dimensional CNN denoisers implemented in parallel on a heterogeneous cluster. Results on experimental CT data demonstrate that Multi-Slice Fusion can substantially improve the quality of reconstructions relative to traditional 4D priors, while also being practical to implement and train., Comment: 8 pages, 8 figures, IEEE International Conference on Computational Photography 2019, Tokyo

Sparse-view computed tomography (CT) has attracted much attention to reduce the potential radiation risk. However, image reconstruction from insufficient data suffers from distortion. As a discrete form of Radon transform, Mojette transform is exactly invertible even with few samples based on discrete projection and reconstruction lattice. The goal of this paper is to apply Mojette transform to reconstruct images from few views in practice. Due to the fact that its acquisition is incompatible with physical X-ray properties, this paper resorts to deep learning technique to map Radon projection into Mojette domain. Experimental results confirm the effectiveness of the proposed reconstruction scheme based on Mojette transform.

Treatments including selective neurectomy, muscle resections and botulinum toxin A (BTX) injections have been used to improve the stocky appearance of calves. BTX injection has the advantages of high efficiency and is almost noninvasive. However, criterion standards of injection are still missing.We aimed to establish a method to classify the hypertrophic calf for a personalized treatment and set up an injection protocol based on the findings.Three-dimensional CT reconstruction was used to measure the thickness and cross-sectional area of the triceps surae. B-mode ultrasound and palpation were used to evaluate the muscle thickness and determine the dosage. Patients were followed 3 and 6 months after the treatment.A total of 112 legs were classified into three degrees of thickness ( 15 mm, 15-25 mm and 25 mm). Twenty-seven subjects were treated with an individualized BTX (100-300 U). Maximal circumference decreased 0.33 ± 0.00 cm after 3 month (p 0.05) and 0.67 ± 0.11 cm after 6 months (p 0.01). The angulated calf contour was improved. No severe side effects were reported.Localizing and dosage are the key points when applying BTX. Dosage should be decided by muscle thickness instead of circumference. BTX treatment improves the prominent contour of the calf rather than reducing the volume.This journal requires that authors assign a level of evidence to each article. For a full description of these Evidence-Based Medicine ratings, please refer to the Table of Contents or the online Instructions to Authors www.springer.com/00266.

This paper analyzes the principle of general CT reconstruction. By comparing and combining the FBP and ART reconstruction algorithms, an improved algorithm of direct solution for CT reconstruction is studied. It is on the base of back projection method and the result satisfies the ART algebraic equations. The method should be used to obtain accurate CT reconstruction results without any iteration. It is a direct solution method.

Clevosaurus was an early sphenodontian, known from many specimens and species from the Late Triassic and Early Jurassic. Here we describe and compare the cranial morphology of Clevosaurus hudsoni, the type species, and Clevosaurus cambrica, both from the UK, based on digital reconstructions from segmented CT scan data of two skulls. C. hudsoni has been described in some detail, but the CT data reveal new information on cranial anatomy, and C. cambrica is a newly described species for which the CT scans enable us to identify many previously undescribed bones, re-identify bones whose identity had previously been uncertain, and refine certain cranial elements. The near complete preservation of the left side of the skull of C. cambrica, and the preservation of much of the left side of the skull of C. hudsoni, provided us with the opportunity to fully reconstruct both species’ skulls in three dimensions. For the first time, coronoid bones are identified in Clevosaurus species from the British Isles. These two species show the diversity of morphology in the genus Clevosaurus. We provide evidence for two morphotypes of C. hudsoni based on study of the dentary dentition of the syntypes and other attributed specimens.

In conventional X-ray-CT inspection of objects generated from a computer-aided design (CAD) model, a 3D CT reconstruction of the object is compared with the reference CAD model. This is a cost inefficient and tedious procedure, unsuitable for inline inspection. In this work, we propose an inspection scheme based on a limited set of radiographs, which are dynamically acquired during the scanning procedure. An efficient framework is described to determine the optimal view angle acquisition from a given CAD model and to automatically estimate the object pose in 3D with a fast, iterative algorithm that dynamically steers the acquisition geometry to acquire the optimal set of projections. We demonstrate the principle of our method on simulated data.

Model-Based Image Reconstruction (MBIR) methods significantly enhance the quality of computed tomographic (CT) reconstructions relative to analytical techniques, but are limited by high computational cost. In this article, we propose a multi-agent consensus equilibrium (MACE) algorithm for distributing both the computation and memory of MBIR reconstruction across a large number of parallel nodes. In MACE, each node stores only a sparse subset of views and a small portion of the system matrix, and each parallel node performs a local sparse-view reconstruction, which based on repeated feedback from other nodes, converges to the global optimum. Our distributed approach can also incorporate advanced denoisers as priors to enhance reconstruction quality. In this case, we obtain a parallel solution to the serial framework of Plug-n-play (PnP) priors, which we call MACE-PnP. In order to make MACE practical, we introduce a partial update method that eliminates nested iterations and prove that it converges to the same global solution. Finally, we validate our approach on a distributed memory system with real CT data. We also demonstrate an implementation of our approach on a massive supercomputer that can perform large-scale reconstruction in real-time.

In this paper, we propose a new iterative algorithm for Computed Tomography (CT) reconstruction when the problem has uncertainty in the view angles. The algorithm models this uncertainty by an additive model-discrepancy term leading to an estimate of the uncertainty in the likelihood function. This means we can combine state-of-the-art regularization priors such as total variation with this likelihood. To achieve a good reconstruction the algorithm alternates between updating the CT image and the uncertainty estimate in the likelihood. In simulated 2D numerical experiments, we show that our method is able to improve the relative reconstruction error and visual quality of the CT image for the uncertain-angle CT problem.

Computed tomography (CT) can provide a 3D view of the patient's internal organs, facilitating disease diagnosis, but it incurs more radiation dose to a patient and a CT scanner is much more cost prohibitive than an X-ray machine too. Traditional CT reconstruction methods require hundreds of X-ray projections through a full rotational scan of the body, which cannot be performed on a typical X-ray machine. In this work, we propose to reconstruct CT from two orthogonal X-rays using the generative adversarial network (GAN) framework. A specially designed generator network is exploited to increase data dimension from 2D (X-rays) to 3D (CT), which is not addressed in previous research of GAN. A novel feature fusion method is proposed to combine information from two X-rays. The mean squared error (MSE) loss and adversarial loss are combined to train the generator, resulting in a high-quality CT volume both visually and quantitatively. Extensive experiments on a publicly available chest CT dataset demonstrate the effectiveness of the proposed method. It could be a nice enhancement of a low-cost X-ray machine to provide physicians a CT-like 3D volume in several niche applications.

[EN] Algebraic factorization methods applied to the discipline of Computerized Tomography (CT) Medical Imaging Reconstruction involve a high computational cost. Since these techniques are significantly slower than the traditional analytical ones and time is critical in this field, we need to employ parallel implementations in order to exploit the machine resources and obtain efficient reconstructions. In this paper, we analyze the performance of the sparse QR decomposition implemented on SuiteSparseQR factorization package applied to the CT reconstruction problem. We explore both the parallelism provided by BLAS threads and the use of the Householder reflections to reconstruct multiple slices at once efficiently. Combining both strategies, we can boost the performance of the reconstructions and implement a reliable and competitive method that gets high-quality CT images., This research has been supported by Universitat Politècnica de València, Generalitat Valenciana under PROMETEO/2018/035 co-financed by FEDER funds, as well as ACIF/2017/075 predoctoral grant, and the Spanish Ministry of Economy and Competitiveness under Grant TIN2015-66972-C5-4-R and TIAMHA co-financed by FEDER funds.

We investigated whether the diagnostic performance of machine learning-based radiomics models for the discrimination of invasive pulmonary adenocarcinomas (IPAs) among subsolid nodules (SSNs) was affected by the proportion of images reconstructed with filtered back projection (FBP) and model-based iterative reconstruction (MBIR) in datasets used for feature extraction.This retrospective study included 60 patients (23 men and 37 women; mean age, 61.4 years) with 69 SSNs (54 part-solid and 15 pure ground-glass nodules). Preoperative CT scans were reconstructed with both FBP and MBIR. A total of 860 radiomics features were obtained from the entire nodule volume, and 70 resampled nodule datasets with an increasing proportion of nodules with MBIR-derived features (from 0/69 to 69/69) were prepared. After feature selection using neighborhood component analysis, support vector machines (SVMs) and an ensemble model were used as classifiers for the differentiation of IPAs. The diagnostic performances of all blending proportions of reconstruction algorithms were calculated and analyzed.The ROC AUC and the diagnostic accuracy of the radiomics models decreased significantly as the number of nodules with MBIR-derived features increased, and this relationship followed cubic functions (RInclusion of CT scans reconstructed with MBIR for radiomics modeling can significantly decrease diagnostic performance for the identification of IPAs.

Low dose and sparse view CT are effective approaches to reduce the radiation dose and accelerate scan speed. Images reconstructed from insufficient data acquired from low dose and sparse view CT are associated with severe streaking artifacts. Therefore, reducing the radiation dose will further degrade the imaging quality. Several attempts have been made to remove these artifacts using deep learning methods such as CNN. Although the deep learning methods for low dose and sparse view CT reconstruction have gained impressive successes, the reconstruction results are still over-smooth. In this work, we propose an artifacts reduction method for low dose and sparse-view CT via a single model trained by generative adversarial networks (GAN). Several numerical simulation experiments are implemented to test the performance of our network. The results show that our GAN can significantly reduce the streaking atrifacts campared with FBP mehtod, and carries more detailed information than CNN.

Objectives: In this study, we aimed to assess the impact of different CT reconstruction kernels on the stability of radiomic features and the transferability between different diseases and tissue types. Three lung diseases were evaluated, i.e. non-small cell lung cancer (NSCLC), malignant pleural mesothelioma (MPM) and interstitial lung disease related to systemic sclerosis (SSc-ILD) as well as four different tissue types, i.e. primary tumor, largest involved lymph node ipsilateral and contralateral lung. Methods: Pre-treatment non-contrast enhanced CT scans from 23 NSCLC, 10 MPM and 12 SSc-ILD patients were collected retrospectively. For each patient, CT scans were reconstructed using smooth and sharp kernel in filtered back projection. The regions of interest (ROIs) were contoured on the smooth kernel-based CT and transferred to the sharp kernel-based CT. The voxels were resized to the largest voxel dimension of each cohort. In total, 1386 features were analyzed. Feature stability was assessed using the intraclass correlation coefficient. Features above the stability threshold >0.9 were considered stable. Results: We observed a strong impact of the reconstruction method on stability of the features (at maximum 26% of the 1386 features were stable). Intensity features were the most stable followed by texture and wavelet features. The wavelet features showed a positive correlation between percentage of stable features and size of the ROI (R2 = 0.79, p = 0.005). Lymph node radiomics showed poorest stability (Conclusion: The robustness of radiomic features is strongly affected by different reconstruction kernels. The effect is largely influenced by the tissue type and less by the disease type. Advances in knowledge: The study presents to our knowledge the most complete analysis on the impact of convolution kernel on the robustness of CT-based radiomics for four relevant tissue types in three different lung diseases.

Limited-angle reconstruction is a very important but challenging problem in the field of computed tomography (CT) which has been extensively studied for many years. However, some difficulties still remain. Based on the theory of visible and invisible boundary developed by Quinto et.al, we propose a reconstruction model for limited-angle CT, which encodes the visible edges as priors to recover the invisible ones. The new model utilizes generalized shrinkage operators as regularizers to perform edge-preserving smoothing such that the visible edges are employed as anchors to recover piecewise-constant or piecewise-smooth reconstructions, while noises and artifacts are suppressed or removed. This work extends our previous research on limited-angle reconstruction which employs gradient \begin{document}$ \ell_0 $\end{document} and \begin{document}$ \ell_1 $\end{document} norm regularizers. The effectiveness of the proposed model and its corresponding solving algorithm shall be verified by numerical experiments with simulated data as well as real data.

The pore structure and seepage characteristics of coal are of great significance to the study of seepage effect of coal seam water injection. Because the current methods for comprehensively characterizing the pore structure and seepage effect of coal mass are imperfect, a multi-scale and visualization method for exploring coal structure and seepage effect is proposed. Through 3D CT reconstruction, Nuclear Magnetic Resonance (NMR) T2 spectrum and Magnetic Resonance Imaging (MRI), the pore size distribution, connectivity, and seepage characteristics of 4 coal samples are explored. The results show that the pore and fissure structures larger than 47.59 μm can be obtained through 3D reconstruction and the equivalent pore diameters (Deq) of the 4 coal samples are mainly concentrated between 100 and 500 μm. Because the pore size distribution in 0–100 μm is supplemented through NMR, it is found that the pore size of each coal sample is mainly distributed between 0.01 and 0.1 μm. The 3D reconstructed pore structure is basically consistent with the water signal spectrum obtained by NMR in the water saturated state, and both methods show that samples Q-1 and Q-2 have good pore connectivity. In addition, it is indicated from the NMR seepage experiments that Q-1 and Q-2 have more seepage space made of large-scale pores and fissures, and thus better connectivity. Therefore, the seepage effect of water injection for coal mass with different structures can be reflected by combining CT and NMR methods to comprehensively characterize the pore structure at different scales, which eliminates the limitations of a single method.

BACKGROUND: Decreasing the number of projections is an effective way to reduce the radiation dose exposed to patients in medical computed tomography (CT) imaging. However, incomplete projection data for CT reconstruction will result in artifacts and distortions. OBJECTIVE: In this paper, a novel dictionary learning algorithm operating in the gradient-domain (Grad-DL) is proposed for few-view CT reconstruction. Specifically, the dictionaries are trained from the horizontal and vertical gradient images, respectively and the desired image is reconstructed subsequently from the sparse representations of both gradients by solving the least-square method. METHODS: Since the gradient images are sparser than the image itself, the proposed approach could lead to sparser representations than conventional DL methods in the image-domain, and thus a better reconstruction quality is achieved. RESULTS: To evaluate the proposed Grad-DL algorithm, both qualitative and quantitative studies were employed through computer simulations as well as real data experiments on fan-beam and cone-beam geometry. CONCLUSIONS: The results show that the proposed algorithm can yield better images than the existing algorithms.

Introduction When one of the anterior cruciate ligament (ACL) bundles is torn, it seems appropriate to preserve the remaining bundle to improve the vascularization and proprioception of the graft, and to reproduce the attachment sites of the torn bundle. After ACL reconstruction, the functional result is worse when the tunnels are positioned further away from the native ACL's isometric attachment points. The goal of this study was to use CT 3D reconstructions to analyse the position of the femoral tunnel following partial ACL reconstruction and to compare it to complete ACL reconstruction cases. We hypothesized that the femoral tunnel positioning was optimal during partial ACL reconstruction. Methods In this prospective single-centre study, 16 patients who underwent isolated anteromedial bundle reconstruction were evaluated during the immediate postoperative period using 3D reconstruction of CT images. During this same period, 180 patients who underwent complete ACL reconstruction in the same surgery unit served as a control group. Results In the partial ACL reconstruction group, 6 tunnels (37.5%) were in the optimal position and 10 tunnels (62.5%) were not. In the complete ACL reconstruction group, 124 femoral (68.9%) were in the optimal position and 56 (31.1%) were not ( P Discussion Femoral tunnel positioning is not always optimal in patients who undergo partial ACL reconstruction. Three-dimensional CT reconstruction is a good tool to help surgeons detect and learn from their errors. Level of evidence III (case-control study).

Compressive Sensing (CS) theory has great potential for reconstructing Computed Tomography (CT) images from sparse-views projection data and Total Variation- (TV-) based CT reconstruction method is very popular. However, it does not directly incorporate prior images into the reconstruction. To improve the quality of reconstructed images, this paper proposed an improved TV minimization method using prior images and Split-Bregman method in CT reconstruction, which uses prior images to obtain valuable previous information and promote the subsequent imaging process. The images obtained asynchronously were registered via Locally Linear Embedding (LLE). To validate the method, two studies were performed. Numerical simulation using an abdomen phantom has been used to demonstrate that the proposed method enables accurate reconstruction of image objects under sparse projection data. A real dataset was used to further validate the method.

The $l_{p}$ ( $0 ) regularization has attracted a great attention in the compressive sensing field, because it can obtain sparser solutions than the well-known $l_{1}$ regularization. Recently, we developed an approximate general analytic thresholding representation for any $l_{p}$ regularization with $0 . The derived thresholding representations are exact for the well-known soft-threshold filtering for $l_{1}$ regularization and the hard-threshold filtering for $l_{0}$ regularization. Because the $l_{p}$ regularization is a nonconvex problem, an iterative algorithm can only converge to local optima instead of the global optimum. In this paper, we propose an alternating iteration algorithm for computed tomography reconstruction in a thresholding form based on our general analytic thresholding representation for better convergent properties. The alternating iteration algorithm alternatively minimizes one $l_{1}$ and one $l_{p}$ ( $0 ) regularized objective functions. While the $l_{p}$ regularization can help to find a sparser solution, the $l_{1}$ regularization can help to monitor the solution not away from the global optimum. Both numerical simulations and phantom experiments are performed to evaluate the proposed alternating iteration algorithm. Compared with the $l_{p}$ ( $0 ) regularization using a single $p$ , the proposed alternating iteration algorithm reduces more data measurements for accurate reconstruction and is more robust for projection noise.

To achieve high-quality low-dose computed tomography (CT) images, compressed sensing (CS)-based CT reconstructions recover the images using fewer projections; and wavelet inverse Radon algorithms recover wavelet subbands of CT images from locally scanned projections. Moreover, it has been shown that subband CS algorithms accelerate the convergence of the CS recovery methods. Here, we propose an innovative combination of a newly developed accelerated wavelet inverse Radon transform and non-convex CS formulation to recover the wavelet subbands of CT images from a reduced number of locally scanned X-ray projections. Fast pseudo-polar Fourier transform is used to decrease the computational complexity of CS recovery. Therefore, the proposed method, denoted by AWiR-SISTA, reduces the radiation dose by simultaneously decreasing the X-ray exposure area and the number of projections, decreases the CS computational complexity, and accelerates the CS recovery convergence rate. Phantom-based simulations show that high-quality ultra-low-dose local CT images can be reconstructed using the proposed method in few seconds, without numerical optimization. Clinical chest CT images are used to demonstrate the practical potential of the method.

Background With the advent of 3D volume rendered CT scans, more information is potentially available to aid the surgeon in complex calcaneal fractures. The primary aim was to determine if there is a difference in inter-observer and intra-observer reliability of 3D CT reconstructions compared to 2D CTs of calcaneus fractures based on classification, identification of specific fracture characteristics and proposed treatment. It is hypothesized that the correlation will be greater between experienced surgeons and trainees when using 3D CT. As a secondary aim, the authors wished to investigate the surgeons’ satisfaction with 3D CT. Patients and methods There were six raters, split into 2 groups: high and low surgical experience. Both 2D and 3D scans (10 patients) were reviewed by each rater by filling out a validated questionnaire. This sitting was over a period of six months and all scans were sent separately and randomly by our research coordinator. This process was repeated after a four-week break. Results The overall evaluation and satisfaction of CT scans of calcaneal fractures was improved by the additional use of volume rendered 3D images. Inexperienced surgeons benefited more than experienced surgeons in terms of reliabilities with fracture configuration. 3D CTs were more likely to change the surgeons’ operative strategy. The consistency of deciding on the fracture classification and satisfaction was more uniform between all the raters when 3D CT was used. Conclusion Inter-observer and intra-observer reliability of 3D CT reconstructions compared to 2D CTs of calcaneus fractures was greater based on classification, identification of specific fracture characteristics and proposed treatment. Inexperienced surgeons were more consistent when interpreting the scans using 3D CTs (improved inter-observer reliability) and were always more satisfied using the 3D CTs. As there is no extra cost or radiation exposure, we propose that 3D CTs may be valuable with preoperative teaching and planning.

Foreign bodies in the pediatric population are frequently seen in the nasal, aural and oral cavities, but are much less commonly reported in the rectum. While these cases are rarely reported in the literature, 8359 rectal foreign bodies were documented in the National Electronic Injury Surveillance System in patients 0–25 years old between 2008 and 2017, with a significant up trend in reported frequency annually [1]. Rectal foreign bodies may result in grave complications such as bleeding, perforation, abscess, sepsis or death [2]. Despite its growing incidence and potentially morbid outcomes, there are few documented case reports in the existing literature and there is no established management guideline for evaluation and management of rectal foreign bodies in the pediatric population. In this case report and review of the literature, we present an unusual case of a radiolucent foreign object in the rectum of a 10 year old child, only visualized with 3D reconstruction on CT scan, and the steps taken towards its removal. Keywords: Rectal foreign body, Sharp rectal objects, Child abuse, Abdominal pain, Proctalgia, Hematochezia

Traditional dictionary learning based CT reconstruction methods are patch-based and the features learned with these methods often contain shifted versions of the same features. To deal with these problems, the convolutional sparse coding (CSC) has been proposed and introduced into various applications. In this paper, inspired by the successful applications of CSC in the field of signal processing, we propose a novel sparse-view CT reconstruction method based on CSC with gradient regularization on feature maps. By directly working on the whole image, which need not to divide the image into overlapped patches like dictionary learning based methods, the proposed method can maintain more details and avoid the artifacts caused by patch aggregation. Experimental results demonstrate that the proposed method has better performance than several existing algorithms in both qualitative and quantitative aspects.

The measurement of temperature and species concentrations in combustion fields is very significant to develop the high-efficient combustion technologies for energy conservation and emission reduction. There are various measurement technologies including contact and non-contact measurement. Tunable diode laser absorption spectroscopy (TDLAS) technology is a proven non-contact method to detect the temperature and species concentrations by absorption measurement. To enable two-dimensional (2D) representation of temperature and species concentrations in combustion fields, the TDLAS technology is usually combined with computed tomography (CT). The latter is however considerably new in combustion research, especially in solid fuels reaction environment. In this paper, a 32-path, 2D CT-TDLAS system for temperature measurement in a pilot scale, coal-fired furnace was developed. The accuracy of CT algorithm to reconstruct 2D temperature distributions in different laser-paths arrangements was first analysed using SSD (sum of squared difference) and ZNCC (zero-mean normalized cross-correlation) by comparing to 2D temperature distribution of a full scale coal-fired furnace simulated using computational fluid dynamics (CFD). The accuracy was improved by 32-path reconstruction. The study was then progressed to investigate its accuracy for measurement in a simple CH4-air burner configuration with rounded and rectangular cells as well as sensitivity for flame shift detection whereby the reconstructed temperature distribution was compared to temperature measured using thermocouple. It is verified that this CT reconstruction was feasible for various measurement areas, even if the center of flame was shifted. Finally, a 32-path, 2D CT-TDLAS system with rectangular structure cell was developed and applied for a temperature measurement in a TNB Research’s pilot scale coal-fired furnace. 2D temperature distribution in coal-fired furnace was reconstructed accroding to the experimental results. It is demonstrated the potential of CT-TDLAS for online 2D temperature measurement for actual applications.

The objective of this study is to compare forward-projected model-based iterative reconstruction solution (FIRST), a newer fully iterative CT reconstruction method, with adaptive iterative dose reduction 3D (AIDR 3D) in low-dose screening CT for lung cancer. Differences in image noise, image quality, and pulmonary nodule detection, size, and characterization were specifically evaluated.Low-dose chest CT images obtained for 50 consecutive patients between December 2015 and January 2016 were retrospectively reviewed. Images were reconstructed using FIRST and AIDR 3D for both lung and soft-tissue reconstruction. Images were independently reviewed to assess image noise, subjective image quality (with use of a 5-point Likert scale, with 1 denoting far superior image quality; 2, superior quality; 3, equivalent quality; 4, inferior quality; and 5, far inferior quality), pulmonary nodule count, size of the largest pulmonary nodule, and characterization of the largest pulmonary nodule (i.e., solid, part solid, or ground glass).Across all 50 cases, measured image noise was lower with FIRST than with AIDR 3D (lung window, 44% reduction, 41 ± 7 vs 74 ± 8 HU, respectively; soft-tissue window, 32% reduction, 11 ± 2 vs 16 ± 2 HU, respectively). Readers subjectively rated images obtained with FIRST as comparable to images obtained with AIDR 3D (mean [± SD] Likert score for FIRST vs AIDR 3D, 3.2 ± 0.3 for soft-tissue reconstructions and 3.0 ± 0.3 for lung reconstructions). For each reader, very good agreement regarding nodule count was noted between FIRST and AIDR 3D (interclass correlation coefficient [ICC], 0.83 for reader 1 and 0.78 for reader 2). Excellent agreement regarding nodule size (ICC, 0.99 for reader 1 and 0.99 for reader 2) and characterization of the largest nodule (kappa value, 0.92 for reader 1 and 0.82 for reader 2) also existed.Images reconstructed with FIRST are superior to those reconstructed AIDR 3D with regard to image noise and are equivalent with regard to subjective image quality, pulmonary nodule count, and nodule characterization.

It has been proposed that the imaging workup of trauma patients be accelerated by omitting the initial chest radiography (CR) and directly performing a computed tomography (CT); however, the baseline CR is then lacking. The purpose of this study was to assess if coronal thick reconstructions generated from chest CT could present an adequate alternative for CR. Sixty trauma patients underwent bedside CR and multidetector row chest CT in the emergency room. The image quality of thoracic anatomical structures, the diagnostic accuracy for chest pathology, and the depiction of indwelling devices were assessed on both modalities. Main pulmonary arteries and perihilar bronchi were equally visualized with both modalities. Central bronchi, retrocardial lung parenchyma, diaphragm, descending aorta, and vertebral pedicles were better visualized on thick CT reconstructions, whereas peripheral lung vessels were better depicted on CR (p