Projection and backprojection are operations that arise frequently in tomographic imaging. Recently, we proposed a new method for projection backprojection, which call distance-driven, offers low arithmetic cost highly sequential memory access pattern. Furthermore, distance-driven avoid several artefact-inducing approximations characteristic of some other methods. We have previously demonstrated the application this to parallel fan beam geometries. In paper, extend framework three dimensions...

A new iterative maximum-likelihood reconstruction algorithm for X-ray computed tomography is presented. The prevents beam hardening artifacts by incorporating a polychromatic acquisition model. continuous spectrum of the tube modeled as number discrete energies. energy dependence attenuation taken into account decomposing linear coefficient photoelectric component and Compton scatter component. relative weight these components constrained based on prior material assumptions. Excellent...

Methods to overcome metal artifacts in computed tomography (CT) images have been researched and developed for nearly 40 years. When X-rays pass through a object, depending on its size density, different physical effects will negatively affect the measurements, most notably beam hardening, scatter, noise, non-linear partial volume effect. These phenomena severely degrade image quality hinder diagnostic power treatment outcomes many clinical applications. In this paper, we first review...

Metal streak artifacts are an important problem in X-ray computed tomography. A high-resolution 2D fan-beam tomography simulator is presented. Several potential causes of metal studied using phantom measurements and simulations. Beam hardening, scatter, noise exponential edge-gradient effects identified as artifacts. Furthermore, aliasing object motion can be also responsible for certain

Iterative reconstruction algorithms for helical CT are presented. The derived from two-dimensional algorithms, by adapting the projector/backprojector to orbit of source, and constraining axial frequencies with a Gaussian sieve. Simulations have been carried out performance iterative is compared that filtered backprojection synthetic (interpolated) sinograms. produce superior bias-noise curves. Axial resolution superior, but disturbing edge-artefacts introduced.

A maximum a posteriori algorithm for reduction of metal streak artifacts in X-ray computed tomography is presented. The uses Markov random field smoothness prior and applies increased sampling the reconstructed image. Good results are obtained simulations phantom measurements: reduced while small, line-shaped details preserved.

A significant and increasing number of patients receiving radiation therapy present with metal objects close to, or even within, the treatment area, resulting in artifacts computed tomography (CT) imaging, which is most commonly used imaging method for planning therapy. In presence implants, such as dental fillings head-and-neck tumors, spinal stabilization implants paraspinal hip replacements prostate cancer treatments, extreme photon absorption by object leads to prominent image artifacts....

X-ray detectors in clinical computed tomography (CT) usually operate current-integrating mode. Their complicated signal statistics often lead to intractable likelihood functions for practical use model-based image reconstruction (MBIR). It is therefore desirable design simplified statistical models without losing the essential factors. Depending on whether CT transmission data are logarithmically transformed, pre-log and post-log two major categories of choices MBIR. Both being...

We present a new simulation environment for X-ray computed tomography, called CatSim. CatSim provides research platform GE researchers and collaborators to explore reconstruction algorithms, CT architectures, source or detector technologies. The main requirements this simulator are accurate physics modeling, low computation times, geometrical flexibility. allows simulating complex analytic phantoms, such as the FORBILD including boxes, ellipsoids, elliptical cylinders, cones, cut planes....

A challenge for positron emission tomography/computed tomography (PET/CT) quantitation is patient respiratory motion, which can cause an underestimation of lesion activity uptake and overestimation volume. Several motion correction methods benefit from longer duration CT scans that are phase matched with PET scans. However, even the currently available, lowest dose techniques, extended cine impart a substantially high radiation dose. This study evaluates designed to reduce in PET/CT...

Artifacts resulting from metal objects have been a persistent problem in CT images over the last four decades. A common approach to overcome their effects is replace corrupt projection data with values synthesized an interpolation scheme or by reprojection of prior image. State-of-the-art correction methods, such as interpolation- and normalization-based algorithm NMAR, often do not produce clinically satisfactory results. Residual image artifacts remain challenging cases even new can be...

Machine learning and deep are rapidly finding applications in the medical imaging field. In this paper, we address long-standing problem of metal artifacts computed tomography (CT) images by training a dual-stream convolutional neural network for streak removal. While many artifact reduction methods exist, even state-of-the-art algorithms fall short some clinical applications. Specifically, proton therapy planning requires high image quality with accurate tumor volumes to ensure treatment...

. X-ray-based imaging modalities including mammography and computed tomography (CT) are widely used in cancer screening, diagnosis, staging, treatment planning, therapy response monitoring. Over the past few decades, improvements to these have resulted substantially improved efficacy efficiency, reduced radiation dose cost. However, such evolved more slowly than would be ideal because lengthy preclinical clinical evaluation is required. In many cases, new ideas cannot evaluated due high cost...

Cardiac CT plays an important role in diagnosing heart diseases but is conventionally limited by its complex workflow that requires dedicated phase and bolus tracking devices [e.g., electrocardiogram (ECG) gating]. This work reports first progress towards robust autonomous cardiac exams through joint deep learning (DL) analytical analysis of pulsed-mode projections (PMPs). To this end, uncertainty were simultaneously estimated using a novel projection domain estimation network (PhaseNet),...

Conventional single-spectrum computed tomography (CT) reconstructs a spectrally integrated attenuation image and reveals tissues morphology without any information about the elemental composition of tissues. Dual-energy CT (DECT) acquires two distinct datasets energy-selective (virtual monoenergetic [VM]) material-selective (material decomposition) images. However, DECT increases system complexity radiation dose compared with CT. In this paper, deep learning approach is presented to produce...

Projection and backprojection are important processes in computed tomography (CT). They used iterative reconstruction, simulation, artifact correction, as well routine (filtered-backprojection based) reconstruction. Existing methods either have poor performance or result artifacts. A new method for projecting backprojecting rays through pixels is presented that has good eliminates artifacts, could potentially enable reconstruction clinical CT systems. The method, which we call...

Distributed X-ray sources open the way to innovative system concepts in and computed tomography. They offer promising opportunities terms of performance, but they pose unique challenges source technologies. Several academic industrial teams have proposed a variety developed some prototypes. We present broad review multisource systems. also discuss components challenges. close with our perspective on future prospects imaging.

Machine Learning, especially deep learning, has been used in typical x‐ray computed tomography (CT) applications, including image reconstruction, enhancement, domain feature detection and characterization. To our knowledge, this is the first study on machine learning for analysis directly based CT projection data. Specifically, we present neural network methods blood vessel characterization sinogram avoiding any partial volume, beam hardening, or motion artifacts introduced during...