Evaluation of State-of-the-Art Hardware Architectures for Fast Cone-Beam CT Reconstruction

Scherl, Holger; Kowarschik, Markus; Hofmann, Hannes; Keck, Benjamin; Hornegger, Joachim

doi:10.1007/978-3-8348-8259-2

Cited by 12 publications

(10 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Implementation of cosine weighting follows from equations (15)- (17). Following the convolution theorem, the one-dimensional filtration of each radiograph (equation (18)) is performed in the frequency domain [17]. First, the one-dimensional fast Fourier transform (FFT) of each row in the radiograph is calculated.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Software-based compensation of instrument misalignments for X-ray computed tomography dimensional metrology

Ametova

Ferrucci

Chilingaryan

et al. 2018

Precision Engineering

View full text Add to dashboard Cite

X-ray computed tomography (CT) is an imaging technique that allows the reconstruction of an imaged part in the form of a three-dimensional attenuation map. The CT data acquisition process consists of acquiring X-ray transmission images from multiple perspectives. Analysis of the reconstructed attenuation map can provide dimensional and material information about the measured part(s). Therefore, CT is recognized as a solution for quality control tasks, for example dimensional inspection of complex objects with intricate inner geometries. CT measurements can suffer from various sources of error in the measurement procedure. One such influence is the geometrical alignment of the CT instrument components. Typical tomographic reconstruction algorithms impose strict requirements on the relative position and orientation of the three main components: X-ray source, rotation axis of the sample stage, and X-ray detector. Any discrepancy in the actual CT geometry from the geometry assumed by the reconstruction algorithm will contribute to errors in measurements performed on the reconstructed data. There is currently no standardized or easily implementable method for users to compensate geometrical misalignments of the CT instrument. In many cases, the procedure of mechanical adjustment of CT instrument is time consuming and impractical. In this paper, we show that software-based compensation of deviations in CT instrument geometry is an effective alternative to mechanical adjustment of CT instrument. Through computer simulations, we compare qualitatively and quantitatively two methods to compensate CT instrument misalignment: radiographic re-binning (interpolation) and a modified conventional reconstruction algorithm with embedded misalignment compensation.

show abstract

Section: Methodsmentioning

confidence: 99%

“…combination of weighted contributions of filtered projections from all rotation positions. Hence, implementation of equation 3 consists of two consecutive steps [17,18]: 1. Preprocessing: cosine weighting and convolution to obtain filtered projection ; 2.…”

Section: Prerequisitesmentioning

confidence: 99%

Software-based compensation of instrument misalignments for X-ray computed tomography dimensional metrology

Ametova

Ferrucci

Chilingaryan

et al. 2018

Precision Engineering

View full text Add to dashboard Cite

show abstract

“…Since these early works, the field of GPU-accelerated CT reconstruction has exploded (Noel et al, 2010;Okitsu et al, 2010;Xu et al, 2010a;Vintache et al, 2010). Recent examples include several papers on using multiple GPUs to further accelerate reconstruction (Jang et al, 2009;Liria et al, 2012;Zhang et al, 2012;Zhu et al, 2012), including the incorporation of scatter correction into the reconstruction process (Sisniega et al, 2011), SART based reconstruction combined with motion compensation (Pang et al, 2011), CT reconstruction with OpenCL ) and an up-to-date and thorough comparison of different hardware implementations (CPU, GPU, FPGA and the Cell Broadband Architecture) of FBP (Scherl et al, 2012).…”

Section: Ctmentioning

confidence: 99%

Medical image processing on the GPU – Past, present and future

Eklund

Dufort

Forsberg

et al. 2013

Medical Image Analysis

348

198

View full text Add to dashboard Cite

Graphics processing units (GPUs) are used today in a wide range of applications, mainly because they can dramatically accelerate parallel computing, are affordable and energy efficient. In the field of medical imaging, GPUs are in some cases crucial for enabling practical use of computationally demanding algorithms. This review presents the past and present work on GPU accelerated medical image processing, and is meant to serve as an overview and introduction to existing GPU implementations. The review covers GPU acceleration of basic image processing operations (filtering, interpolation, histogram estimation and distance transforms), the most commonly used algorithms in medical imaging (image registration, image segmentation and image denoising) and algorithms that are specific to individual modalities (CT, PET, SPECT, MRI, fMRI, DTI, ultrasound, optical imaging and microscopy). The review ends by highlighting some future possibilities and challenges.

show abstract

“…The equation 16 using the Fourier transform, F , for finding the MTF given the point spread function (PSF) data is shown in Eq. (11). The solution for the MTF of the system is shown in Eqs.…”

Section: Frequency Responsementioning

confidence: 99%

“…The backprojection operation does not perform any filtering or weighting of the projection data. This work implemented the p-matrix voxel driven backprojection algorithm as described in Scherl et al, 11 Wiesent et al, 12 and Galigekere et al 13 The Hadamard product "°" of the weighting volume, i W , and the constraining image, i C , generates raw 4-D time frame, i 4D , data, which can be further processed. The postprocessing includes corrections for vessel overlap in a projection, which causes a reconstruction artifact, higher than actual values during overlap periods, in the TACs of 4-D DSA.…”

Section: Introductionmentioning

confidence: 99%

Volumetric limiting spatial resolution analysis of four-dimensional digital subtraction angiography

et al. 2016

View full text Add to dashboard Cite

Abstract. C-Arm CT three-dimensional (3-D) digital subtraction angiography (DSA) reconstructions cannot provide temporal information to radiologists. Four-dimensional (4-D) DSA provides a time series of 3-D volumes utilizing temporal dynamics in the two-dimensional (2-D) projections using a constraining image reconstruction approach. Volumetric limiting spatial resolution (VLSR) of 4-D DSA is quantified and compared to a 3-D DSA. The effects of varying 4-D DSA parameters of 2-D projection blurring kernel size and threshold of the 3-D DSA (constraining image) of an in silico phantom (ISPH) and physical phantom (PPH) were investigated. The PPH consisted of a 76-micron tungsten wire. An 8-s∕248-frame∕198-deg scan protocol acquired the projection data. VLSR was determined from MTF curves generated from each 2-D transverse slice of every (248) Published by SPIE under a Creative Commons Attribution 3.0 Unported License. Distribution or reproduction of this work in whole or in part requires full attribution of the original publication, including its DOI.

show abstract

Evaluation of State-of-the-Art Hardware Architectures for Fast Cone-Beam CT Reconstruction

Cited by 12 publications

References 10 publications

Software-based compensation of instrument misalignments for X-ray computed tomography dimensional metrology

Software-based compensation of instrument misalignments for X-ray computed tomography dimensional metrology

Medical image processing on the GPU – Past, present and future

Volumetric limiting spatial resolution analysis of four-dimensional digital subtraction angiography

Contact Info

Product

Resources

About