Image Reconstruction In Cone-beam CT Research Articles

SU‐FF‐I‐43: Iterative Methods of Cone‐Beam CT Image Reconstruction for Under‐Sampled and Truncated Projection Data

Purpose: In this work, iterative method with total variation (TV) minimization was proposed to reconstruct cone beam CT (CBCT) image from under sampled and truncated projection data. Method and Materials: Respiratory motion is a significant source of error in CBCT image reconstruction for the thorax and upper abdomen. To reduce motion artifacts in reconstructed image, four‐dimensional (4D) CT imaging has been studied. One category of 4D CT imaging is gated image reconstruction in which projection data acquired when free breathing were retrospective sorted into several subsets of projections that each corresponds to a certain breathing phase. The projection subsets are reconstructed into CBCT volume at different phases. Normally, CBCT scanning system acquires about 330 projections over half rotation. After sorting these 330 projections into 10 subsets, each subset is under sampled and only has 33 projections. Image reconstructed with this under sampled subset has severe streaking artifacts. In addition, the detector panel does not cover the extent of the body causing truncation of the projection data along axle and transaxle directions. The maximum likelihood expectation maximization was used with TV constraint and the results were compared with those of filtered backprojection (FBP) using computer simulations and patient data. Results: The MLEM‐TV method produced a larger reconstructed region‐of‐interest (ROI) than that of FBP in computer simulation. Also, the images of the proposed method appear to have better noise removal and streaking artifact suppression induced by under‐sampling in computer simulation and patient data. Signal to noise ratio (SNR) of three typical regions on a patient image of the FBP method are 5.4, 7.5 and 8, while those of MLEM‐TV are 9.7, 20.2 and 24.8 respectively. Conclusion: Iterative method is feasible in CBCT imaging and has an advantage over FBP method when the projection data are under sampled and truncated.

SU‐FF‐I‐44: Iterative Image Reconstruction for CBCT Using Edge‐Preserving Prior

Purpose: To improve the image quality of low‐dose cone‐beam CT (CBCT) using an iterative image reconstruction algorithm with an edge‐preserving prior. Method and Materials: We reconstructed CBCT image by minimizing the objective function based on the penalized weighted least‐squares (PWLS) criterion with an anisotropic quadratic penalty. The penalty or the prior constraint is applied on neighboring voxels in the image domain, where the gradient of image intensity controls the degree of smoothness in the reconstructed image. For neighbors associated with large gradient, such as the region around sharp edges, equivalence between them will be discouraged. CBCT projection data of a quality assurance phantom was acquired by a Varian Acuity simulator (Varian Medical Systems, Palo Alto, CA). Contrast‐to‐noise ratio (CNR) and modulation transfer function (MTF), which characterizes the spatial resolution of the reconstructed image, are used to quantitatively evaluate the proposed algorithm. Results: At matched spatial resolution, CNR of a low‐contrast region‐of‐interest in the image reconstructed by the proposed algorithm is 2 times higher than that in the corresponding image reconstructed by the analytical FDK algorithm. Compared with the projection domain smoothing algorithm, the iterative image reconstruction algorithm based on the image domain PWLS criterion produces higher resolution image at matched noise level. Comparison studies with conventional isotropic quadratic penalty and Huber penalty also show that the proposed anisotropic penalty generates images of better spatial resolution at matched noise level. Conclusion: CNR in the low‐dose CBCT image can be significantly improved without loss of spatial resolution by using the proposed iterative image reconstruction algorithm. Imaging dose of CBCT can be lowered using statistical image reconstruction algorithms.

GPU‐Based 3D Cone‐Beam CT Image Reconstruction for Large Data Volume

Currently, 3D cone-beam CT image reconstruction speed is still a severe limitation for clinical application. The computational power of modern graphics processing units (GPUs) has been harnessed to provide impressive acceleration of 3D volume image reconstruction. For extra large data volume exceeding the physical graphic memory of GPU, a straightforward compromise is to divide data volume into blocks. Different from the conventional Octree partition method, a new partition scheme is proposed in this paper. This method divides both projection data and reconstructed image volume into subsets according to geometric symmetries in circular cone-beam projection layout, and a fast reconstruction for large data volume can be implemented by packing the subsets of projection data into the RGBA channels of GPU, performing the reconstruction chunk by chunk and combining the individual results in the end. The method is evaluated by reconstructing 3D images from computer-simulation data and real micro-CT data. Our results indicate that the GPU implementation can maintain original precision and speed up the reconstruction process by 110–120 times for circular cone-beam scan, as compared to traditional CPU implementation.

Motion-compensated iterative cone-beam CT image reconstruction with adapted blobs as basis functions

This paper presents a three-dimensional method to reconstruct moving objects from cone-beam X-ray projections using an iterative reconstruction algorithm and a given motion vector field. For the image representation, adapted blobs are used, which can be implemented efficiently as basis functions. Iterative reconstruction requires the calculation of line integrals (forward projections) through the image volume, which are compared with the actual measurements to update the image volume. In the existence of a divergent motion vector field, a change in the volumes of the blobs has to be taken into account in the forward and backprojections. An efficient method to calculate the line integral through the adapted blobs is proposed. It solves the problem, how to compensate for the divergence in the motion vector field on a grid of basis functions. The method is evaluated on two phantoms, which are subject to three different known motions. Moreover, a motion-compensated filtered back-projection reconstruction method is used, and the reconstructed images are compared. Using the correct motion vector field with the iterative motion-compensated reconstruction, sharp images are obtained, with a quality that is significantly better than gated reconstructions.

GPU-based streaming architectures for fast cone-beam CT image reconstruction and demons deformable registration

This paper shows how to significantly accelerate cone-beam CT reconstruction and 3D deformable image registration using the stream-processing model. We describe data-parallel designs for the Feldkamp, Davis and Kress (FDK) reconstruction algorithm, and the demons deformable registration algorithm, suitable for use on a commodity graphics processing unit. The streaming versions of these algorithms are implemented using the Brook programming environment and executed on an NVidia 8800 GPU. Performance results using CT data of a preserved swine lung indicate that the GPU-based implementations of the FDK and demons algorithms achieve a substantial speedup—up to 80 times for FDK and 70 times for demons when compared to an optimized reference implementation on a 2.8 GHz Intel processor. In addition, the accuracy of the GPU-based implementations was found to be excellent. Compared with CPU-based implementations, the RMS differences were less than 0.1 Hounsfield unit for reconstruction and less than 0.1 mm for deformable registration.

A rebinned backprojection-filtration algorithm for image reconstruction in helical cone-beam CT

In the last few years, mathematically exact algorithms, including the backprojection-filtration (BPF) algorithm, have been developed for accurate image reconstruction in helical cone-beam CT. The BPF algorithm requires minimum data, and can reconstruct region-of-interest (ROI) images from data containing truncations. However, similar to other existing reconstruction algorithms for helical cone-beam CT, the BPF algorithm involves a backprojection with a spatially varying weighting factor, which is computationally demanding and, more importantly, can lead to undesirable numerical properties in reconstructed images. In this work, we develop a rebinned BPF algorithm in which the backprojection invokes no spatially varying weighting factor for accurate image reconstruction from helical cone-beam projections. This rebinned BPF algorithm is computationally more efficient and numerically more stable than the original BPF algorithm, while it also retains the nice properties of the original BPF algorithm such as minimum data requirement and ROI-image reconstruction from truncated data. We have also performed simulation studies to validate and evaluate the rebinned BPF algorithm.For more information on this article, see medicalphysicsweb.org

Mégavoltage cone-beam CT : récents développements et applications cliniques pour la radiothérapie conformationnelle avec modulation d'intensité

The Megavoltage cone-beam (MV CBCT) system consists of a new a-Si flat panel adapted for MV imaging and an integrated workflow application allowing the automatic acquisition of projection images, cone-beam CT image reconstruction, CT to CBCT image registration and couch position adjustment. This provides a 3D patient anatomy volume in the actual treatment position, relative to the treatment isocenter, moments before the dose delivery, that can be tightly aligned to the planning CT, allowing verification and correction of the patient position, detection of anatomical changes and dose calculation. In this paper, we present the main advantages and performance of this MV CBCT system and summarize the different clinical applications. Examples of the image-guided treatment process from the acquisition of the MV CBCT scan to the correction of the couch position and dose delivery will be presented for spinal and lung lesions and for head and neck, and prostate cancers.

SU‐FF‐J‐81: Clinical Integration of a MV Conebeam CT System for Image‐Guided Treatment

Purpose: To perform the integration of a newly developed image‐guidance system and to describe the main advantages and performance of the first Megavoltage Conebeam CT (MV CBCT) system. Method and Materials: The MV CBCT system, consisting of a new a‐Si flat panel adapted for MV imaging and an integrated workflow application allowing the automatic acquisition of projection images, conebeam CT image reconstruction, CT to CBCT image registration and couch position adjustment was recently introduced in clinic. Template protocols can be used for the acquisition of CBCT images at different dose ranging from 1 to 60 M.U. Geometrical calibration, gain image adjustment and defect pixels correction procedures are performed off‐line. Results: For a typical case, 200 projection portal images and a total exposure of 5 to 8 M.U. are acquired with the 6 MV beam in 45 seconds and the 256×256×256 MV CBCT image is reconstructed less than two minutes after the start of the acquisition. Examples of the image‐guided treatment process including the acquisition of projections images, the reconstruction of the MV CBCT image and its registration with the planning CT, followed by the couch position correction and dose delivery will be presented. Conclusion: MV CBCT provides a 3D patient anatomy volume in the actual treatment position, relative to the treatment isocenter, moments before the dose delivery, that can be tightly aligned to the planning CT, allowing verification and correction of the patient position. Research supported by Siemens Oncology Care Systems.

Parallel statistical image reconstruction for cone-beam x-ray CT on a shared memory computation platform

Statistical reconstruction methods offer possibilities of improving image quality as compared to analytical methods, but current reconstruction times prohibit routine clinical applications. To reduce reconstruction times we have parallelized a statistical reconstruction algorithm for cone-beam x-ray CT, the ordered subset convex algorithm (OSC), and evaluated it on a shared memory computer. Two different parallelization strategies were developed: one that employs parallelism by computing the work for all projections within a subset in parallel, and one that divides the total volume into parts and processes the work for each sub-volume in parallel. Both methods are used to reconstruct a three-dimensional mathematical phantom on two different grid densities. The reconstructed images are binary identical to the result of the serial (non-parallelized) algorithm. The speed-up factor equals approximately 30 when using 32 to 40 processors, and scales almost linearly with the number of cpus for both methods. The huge reduction in computation time allows us to apply statistical reconstruction to clinically relevant studies for the first time.

An extended data function and its generalized backprojection for image reconstruction in helical cone-beam CT

We have recently proposed a general formula (i.e., equations (9) to (11) in Zou and Pan (2004a Phys. Med. Biol. 49 941–59)) for image reconstruction from helical cone-beam data. On the basis of the formula, we have also developed two reconstruction algorithms, which are referred to as the backprojection filtration (BPF) algorithm (Zou and Pan 2004a) and the filtered backprojection (FBP) algorithm (Zou and Pan 2004b Phys. Med. Biol. 49 2717–31), respectively. The two algorithms have been implemented and evaluated in numerical studies. In this note, however, we point out that the data function previously used for proving the general formula in Zou and Pan (2004a) is incomplete and that, instead, an extended data function and its generalized backprojection, which are described in this note, should be used to complete the proof of the general formula. On the other hand, we also demonstrate in this note that the additional term in the extended data function has no effect on the previously developed BPF and FBP algorithms. The results can also be extended to general, smooth trajectories.

Image reconstruction with a shift‐variant filtration in circular cone‐beam CT

AbstractIt is well known that cone‐beam data acquired with a circular orbit are insufficient for exact image reconstruction. Despite this, because a cone‐beam scanning configuration with a circular orbit is easy to implement in practice, it has been widely employed for data acquisition in, e.g., micro‐CT and CT imaging in radiation therapy. The algorithm developed by Feldkamp, Davis, and Kress (FDK) and its modifications, such as the Tent–FDK (T‐FDK) algorithm, have been used for image reconstruction from circular cone‐beam data. In this work, we present an algorithm with spatially shift‐variant filtration for image reconstruction in circular cone‐beam CT. We performed computer‐simulation studies to compare the proposed and existing algorithms. Numerical results in these studies demonstrated that the proposed algorithm has resolution properties comparable to, and noise properties better than, the FDK algorithm. As compared to the T‐FDK algorithm, our proposed algorithm reconstructs images with an improved in‐plane spatial resolution. © 2005 Wiley Periodicals, Inc. Int J Imaging Syst Technol, 14, 213–221, 2004; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/ima.20026