Fast compressed sensing-based CBCT reconstruction using Barzilai-Borwein formulation for application to on-line IGRT

Iterative reconstruction for limited-angle CT using implicit neural representation

Objective.Limited-angle computed tomography (CT) presents a challenge due to its ill-posed nature. In such scenarios, analytical reconstruction methods often exhibit severe artifacts. To tackle this inverse problem, several supervised deep learning-based approaches have been proposed. However, they are constrained by limitations such as generalization issue and the difficulty of acquiring a large amount of paired CT images.Approach.In this work, we propose an iterative neural reconstruction framework designed for limited-angle CT. By leveraging a coordinate-based neural representation, we formulate tomographic reconstruction as a convex optimization problem involving a deep neural network. We then employ differentiable projection layer to optimize this network by minimizing the discrepancy between the predicted and measured projection data. In addition, we introduce a prior-based weight initialization method to ensure the network starts optimization with an informed initial guess. This strategic initialization significantly improves the quality of iterative reconstruction by stabilizing the divergent behavior in ill-posed neural fields. Our method operates in a self-supervised manner, thereby eliminating the need for extensive data.Main results.The proposed method outperforms other iterative and learning-based methods. Experimental results on XCAT and Mayo Clinic datasets demonstrate the effectiveness of our approach in restoring anatomical features as well as structures. This finding was substantiated by visual inspections and quantitative evaluations using NRMSE, PSNR, and SSIM. Moreover, we conduct a comprehensive investigation into the divergent behavior of iterative neural reconstruction, thus revealing its suboptimal convergence when starting from scratch. In contrast, our method consistently produced accurate images by incorporating an initial estimate as informed initialization.Significance.This work showcases the feasibility to reconstruct high-fidelity CT images from limited-angle x-ray projections. The proposed methodology introduces a novel data-free approach to enhance medical imaging, holding promise across various clinical applications.

Performance and usefulness evaluation of a software-based scatter correction technique for mammographic images

Although physical grids improve contrast in radiographic images by reducing scattered radiation, various artifacts such as grid shadow, moire, and cutoff result in increased patient doses. To overcome these problems, this study evaluates the applicability and usefulness of a material thickness-based scatter-correction technique for mammography. Specifically, this study aims to compare and evaluate the performance of mammography using the proposed software-based scatter correction framework and a physical grid. The proposed technique enables scatter correction based on pre-calculated parameters of a thickness-based scatter kernel at a water slab phantom and an empirical quantity of scatter components in a mammographic system. In the Monte Carlo simulation and experiment, the proposed framework displayed an intensity profile and full width at half maximum that closely approximated those seen in the physical grid. In addition, by applying the proposed framework to the ACR phantom, it was verified that all structures, including specks, were distinctly distinguished. The results demonstrate that the X-ray scatter-correction method with a software-based framework for mammography is applicable to the field of diagnostic imaging, as this approach yields image quality equivalent to that achieved with physical grids while also enabling a reduction in radiation doses for patients.

Sparsier2Sparse: Self-supervised convolutional neural network-based streak artifacts reduction in sparse-view CT images

BACKGROUND

Sparse-view computed tomography (CT) has attracted a lot of attention for reducing both scanning time and radiation dose. However, sparsely-sampled projection data generate severe streak artifacts in the reconstructed images. In recent decades, many sparse-view CT reconstruction techniques based on fully-supervised learning have been proposed and have shown promising results. However, it is not feasible to acquire pairs of full-view and sparse-view CT images in real clinical practice.

PURPOSE

In this study, we propose a novel self-supervised convolutional neural network (CNN) method to reduce streak artifacts in sparse-view CT images.

METHODS

We generate the training dataset using only sparse-view CT data and train CNN based on self-supervised learning. Since the streak artifacts can be estimated using prior images under the same CT geometry system, we acquire prior images by iteratively applying the trained network to given sparse-view CT images. We then subtract the estimated steak artifacts from given sparse-view CT images to produce the final results.

RESULTS

We validated the imaging performance of the proposed method using extended cardiac-torso (XCAT) and the 2016 AAPM Low-Dose CT Grand Challenge dataset from Mayo Clinic. From the results of visual inspection and modulation transfer function (MTF), the proposed method preserved the anatomical structures effectively and showed higher image resolution compared to the various streak artifacts reduction methods for all projection views.

CONCLUSIONS

We propose a new framework for streak artifacts reduction when only the sparse-view CT data are given. Although we do not use any information of full-view CT data for CNN training, the proposed method achieved the highest performance in preserving fine details. By overcoming the limitation of dataset requirements on fully-supervised-based methods, we expect that our framework can be utilized in the medical imaging field.

Chest tomosynthesis deblurring using CNN with deconvolution layer for vertebrae segmentation

BACKGROUND

Limited scan angles cause severe distortions and artifacts in reconstructed tomosynthesis images when the Feldkamp-Davis-Kress (FDK) algorithm is used for the purpose, which degrades clinical diagnostic performance. These blurring artifacts are fatal in chest tomosynthesis images because precise vertebrae segmentation is crucial for various diagnostic analyses, such as early diagnosis, surgical planning, and injury detection. Moreover, because most spinal pathologies are related to vertebral conditions, the development of methods for accurate and objective vertebrae segmentation in medical images is an important and challenging research area.

PURPOSE

The existing point-spread-function-(PSF)-based deblurring methods use the same PSF in all sub-volumes without considering the spatially varying property of tomosynthesis images. This increases the PSF estimation error, thus further degrading the deblurring performance. However, the proposed method estimates the PSF more accurately by using sub-CNNs that contain a deconvolution layer for each sub-system, which improves the deblurring performance.

METHODS

To minimize the effect of the spatially varying property, the proposed deblurring network architecture comprises four modules: (1) block division module, (2) partial PSF module, (3) deblurring block module, and (4) assembling block module. We compared the proposed DL-based method with the FDK algorithm, total-variation iterative reconstruction with GP-BB (TV-IR), 3D U-Net, FBPConvNet, and two-phase deblurring method. To investigate the deblurring performance of the proposed method, we evaluated its vertebrae segmentation performance by comparing the pixel accuracy (PA), intersection-over-union (IoU), and F-score values of reference images to those of the deblurred images. Also, pixel-based evaluations of the reference and deblurred images were performed by comparing their root mean squared error (RMSE) and visual information fidelity (VIF) values. In addition, 2D analysis of the deblurred images were performed by artifact spread function (ASF) and full width half maximum (FWHM) of the ASF curve.

RESULTS

The proposed method was able to recover the original structure significantly, thereby further improving the image quality. The proposed method yielded the best deblurring performance in terms of vertebrae segmentation and similarity. The IoU, F-score, and VIF values of the chest tomosynthesis images reconstructed using the proposed SV method were 53.5%, 28.7%, and 63.2% higher, respectively, than those of the images reconstructed using the FDK method, and the RMSE value was 80.3% lower. These quantitative results indicate that the proposed method can effectively restore both the vertebrae and the surrounding soft tissue.

CONCLUSIONS

We proposed a chest tomosynthesis deblurring technique for vertebrae segmentation by considering the spatially varying property of tomosynthesis systems. The results of quantitative evaluations indicated that the vertebrae segmentation performance of the proposed method was better than those of the existing deblurring methods.

A high-resolution cone beam computed tomography (HRCBCT) reconstruction framework for CBCT-guided online adaptive therapy

BACKGROUND

Kilo-voltage cone-beam computed tomography (CBCT) is a prevalent modality used for adaptive radiotherapy (ART) due to its compatibility with linear accelerators and ability to provide online imaging. However, the widely-used Feldkamp-Davis-Kress (FDK) reconstruction algorithm has several limitations, including potential streak aliasing artifacts and elevated noise levels. Iterative reconstruction (IR) techniques, such as total variation (TV) minimization, dictionary-based methods, and prior information-based methods, have emerged as viable solutions to address these limitations and improve the quality and applicability of CBCT in ART.

PURPOSE

One of the primary challenges in IR-based techniques is finding the right balance between minimizing image noise and preserving image resolution. To overcome this challenge, we have developed a new reconstruction technique called high-resolution CBCT (HRCBCT) that specifically focuses on improving image resolution while reducing noise levels.

METHODS

The HRCBCT reconstruction technique builds upon the conventional IR approach, incorporating three components: the data fidelity term, the resolution preservation term, and the regularization term. The data fidelity term ensures alignment between reconstructed values and measured projection data, while the resolution preservation term exploits the high resolution of the initial Feldkamp-Davis-Kress (FDK) algorithm. The regularization term mitigates noise during the IR process. To enhance convergence and resolution at each iterative stage, we applied Iterative Filtered Backprojection (IFBP) to the data fidelity minimization process.

RESULTS

We evaluated the performance of the proposed HRCBCT algorithm using data from two physical phantoms and one head and neck patient. The HRCBCT algorithm outperformed all four different algorithms; FDK, Iterative Filtered Back Projection (IFBP), Compressed Sensing based Iterative Reconstruction (CSIR), and Prior Image Constrained Compressed Sensing (PICCS) methods in terms of resolution and noise reduction for all data sets. Line profiles across three line pairs of resolution revealed that the HRCBCT algorithm delivered the highest distinguishable line pairs compared to the other algorithms. Similarly, the Modulation Transfer Function (MTF) measurements, obtained from the tungsten wire insert on the CatPhan 600 physical phantom, showed a significant improvement with HRCBCT over traditional algorithms.

CONCLUSION

The proposed HRCBCT algorithm offers a promising solution for enhancing CBCT image quality in adaptive radiotherapy settings. By addressing the challenges inherent in traditional IR methods, the algorithm delivers high-definition CBCT images with improved resolution and reduced noise throughout each iterative step. Implementing the HR CBCT algorithm could significantly impact the accuracy of treatment planning during online adaptive therapy.

Investigation of the usability of cone-beam computed tomography images using digital radiography equipment for boron neutron capture therapy treatment planning in the sitting position

In boron neutron capture therapy (BNCT), treatment planning images are acquired in the recumbent position. However, treatment is occasionally performed in the sitting position. For BNCT treatment planning, we investigated the usability of cone-beam computed tomography (CBCT) images using digital radiography equipment that allows imaging in the sitting position. The dose calculation results in both CBCT and fan beam CT were in good agreement. This method will eliminate the posture difference between planning and treatment.

A streak artifact reduction algorithm in sparse-view CT using a self-supervised neural representation

PURPOSE

Sparse-view computed tomography (CT) has been attracting attention for its reduced radiation dose and scanning time. However, analytical image reconstruction methods suffer from streak artifacts due to insufficient projection views. Recently, various deep learning-based methods have been developed to solve this ill-posed inverse problem. Despite their promising results, they are easily overfitted to the training data, showing limited generalizability to unseen systems and patients. In this work, we propose a novel streak artifact reduction algorithm that provides a system- and patient-specific solution.

METHODS

Motivated by the fact that streak artifacts are deterministic errors, we regenerate the same artifacts from a prior CT image under the same system geometry. This prior image need not be perfect but should contain patient-specific information and be consistent with full-view projection data for accurate regeneration of the artifacts. To this end, we use a coordinate-based neural representation that often causes image blur but can greatly suppress the streak artifacts while having multiview consistency. By employing techniques in neural radiance fields originally proposed for scene representations, the neural representation is optimized to the measured sparse-view projection data via self-supervised learning. Then, we subtract the regenerated artifacts from the analytically reconstructed original image to obtain the final corrected image.

RESULTS

To validate the proposed method, we used simulated data of extended cardiac-torso phantoms and the 2016 NIH-AAPM-Mayo Clinic Low-Dose CT Grand Challenge and experimental data of physical pediatric and head phantoms. The performance of the proposed method was compared with a total variation-based iterative reconstruction method, naive application of the neural representation, and a convolutional neural network-based method. In visual inspection, it was observed that the small anatomical features were best preserved by the proposed method. The proposed method also achieved the best scores in the visual information fidelity, modulation transfer function, and lung nodule segmentation.

CONCLUSIONS

The results on both simulated and experimental data suggest that the proposed method can effectively reduce the streak artifacts while preserving small anatomical structures that are easily blurred or replaced with misleading features by the existing methods. Since the proposed method does not require any additional training datasets, it would be useful in clinical practice where the large datasets cannot be collected.

Two-phase learning-based 3D deblurring method for digital breast tomosynthesis images

In digital breast tomosynthesis (DBT) systems, projection data are acquired from a limited number of angles. Consequently, the reconstructed images contain severe blurring artifacts that might heavily degrade the DBT image quality and cause difficulties in detecting lesions. In this study, we propose a two-phase learning approach for artifact compensation in a coarse-to-fine manner to mitigate blurring artifacts effectively along all viewing directions of the DBT image volume (i.e., along the axial, coronal, and sagittal planes) to improve the detection performance of lesions. The proposed method employs a convolutional neural network model comprising two submodels/phases, with Phase 1 performing three-dimensional (3D) deblurring and Phase 2 performing additional 2D deblurring. To investigate the effects of loss functions on the proposed model’s deblurring performance, we evaluated several loss functions, such as the pixel-based loss function, adversarial-based loss function, and perception-based loss function. Compared with the DBT image, the mean squared error of the image and the root mean squared errors of the gradient of the image decreased by 82.8% and 44.9%, respectively, and the contrast-to-noise ratio increased by 183.4% in the in-focus plane. We verified that the proposed method sequentially restored the missing frequency components as the DBT images were processed through the Phase 1 and Phase 2 steps. These results indicate that the proposed method performs effective 3D deblurring, significantly reducing the blurring artifacts in the in-focus plane and other planes of the DBT image, thus improving the detection performance of lesions.

Weakly-supervised progressive denoising with unpaired CT images

Although low-dose CT imaging has attracted a great interest due to its reduced radiation risk to the patients, it suffers from severe and complex noise. Recent fully-supervised methods have shown impressive performances on CT denoising task. However, they require a huge amount of paired normal-dose and low-dose CT images, which is generally unavailable in real clinical practice. To address this problem, we propose a weakly-supervised denoising framework that generates paired original and noisier CT images from unpaired CT images using a physics-based noise model. Our denoising framework also includes a progressive denoising module that bypasses the challenges of mapping from low-dose to normal-dose CT images directly via progressively compensating the small noise gap. To quantitatively evaluate diagnostic image quality, we present the noise power spectrum and signal detection accuracy, which are well correlated with the visual inspection. The experimental results demonstrate that our method achieves remarkable performances, even superior to fully-supervised CT denoising with respect to the signal detectability. Moreover, our framework increases the flexibility in data collection, allowing us to utilize any unpaired data at any dose levels.

Blind Deconvolution Based on Compressed Sensing with bi-l0-l2-norm Regularization in Light Microscopy Image

Blind deconvolution of light microscopy images could improve the ability of distinguishing cell-level substances. In this study, we investigated the blind deconvolution framework for a light microscope image, which combines the benefits of bi-l₀-l₂-norm regularization with compressed sensing and conjugated gradient algorithms. Several existing regularization approaches were limited by staircase artifacts (or cartooned artifacts) and noise amplification. Thus, we implemented our strategy to overcome these problems using the bi-l₀-l₂-norm regularization proposed. It was investigated through simulations and experiments using optical microscopy images including the background noise. The sharpness was improved through the successful image restoration while minimizing the noise amplification. In addition, quantitative factors of the restored images, including the intensity profile, root-mean-square error (RMSE), edge preservation index (EPI), structural similarity index measure (SSIM), and normalized noise power spectrum, were improved compared to those of existing or comparative images. In particular, the results of using the proposed method showed RMSE, EPI, and SSIM values of approximately 0.12, 0.81, and 0.88 when compared with the reference. In addition, RMSE, EPI, and SSIM values in the restored image were proven to be improved by about 5.97, 1.26, and 1.61 times compared with the degraded image. Consequently, the proposed method is expected to be effective for image restoration and to reduce the cost of a high-performance light microscope.

Incomplete angle reconstruction algorithm with the sparse optimization and the image optimal criterions

To solve the problem of artifact and image degradation caused by incomplete angle projection, this article presents an incomplete angle reconstruction algorithm based on sparse optimization and image optimization criterion (SO-IOC). Firstly, the joint objective function model is established based on the projection sparsity and the natural features of images. Secondly, by means of the idea of alternating direction method of multipliers, the augmented Lagrange method is used to decompose the reconstruction model into simple subproblems and the modified genetic algorithm is used for solving those subproblems. Finally, a multiobjective optimization operation is carried out to coordinate and select the candidate solutions to improve the quality of the reconstructed images. The algebraic reconstruction technique algorithm and the Split Bregman algorithm are compared with the SO-IOC algorithm. In the compared process, the mean relative error and the peak signal-to-noise ratio are used. The experimental results show the SO-IOC algorithm is best among the above three algorithms.

Cone-beam breast computed tomography using ultra-fast image reconstruction with constrained, total-variation minimization for suppression of artifacts

Compressed sensing based iterative reconstruction algorithms for computed tomography such as adaptive steepest descent-projection on convex sets (ASD-POCS) are attractive due to their applicability in incomplete datasets such as sparse-view data and can reduce radiation dose to the patients while preserving image quality. Although IR algorithms reduce image noise compared to analytical Feldkamp-Davis-Kress (FDK) algorithm, they may generate artifacts, particularly along the periphery of the object. One popular solution is to use finer image-grid followed by down-sampling. This approach is computationally intensive but may be compensated by reducing the field of view. Our proposed solution is to replace the algebraic reconstruction technique within the original ASD-POCS by ordered subsets-simultaneous algebraic reconstruction technique (OS-SART) and with initialization using FDK image. We refer to this method as Fast, Iterative, TV-Regularized, Statistical reconstruction Technique (FIRST). In this study, we investigate FIRST for cone-beam dedicated breast CT with large image matrix. The signal-difference to noise ratio (SDNR), the difference of the mean value and the variance of adipose and fibroglandular tissues for both FDK and FIRST reconstructions were determined. With FDK serving as the reference, the root-mean-square error (RMSE), bias, and the full-width at half-maximum (FWHM) of microcalcifications in two orthogonal directions were also computed. Our results suggest that FIRST is competitive to the finer image-grid method with shorter reconstruction time. Images reconstructed using the FIRST do not exhibit artifacts and outperformed FDK in terms of image noise. This suggests the potential of this approach for radiation dose reduction in cone-beam breast CT.

Analytical Low-Dose CBCT Reconstruction Using Non-local Total Variation Regularization for Image Guided Radiation Therapy

Purpose: Conventional iterative low-dose CBCT reconstruction techniques are slow and tend to over-smooth edges through uniform weighting of the image penalty gradient. In this study, we present a non-iterative analytical low-dose CBCT reconstruction technique by restoring the noisy low-dose CBCT projection with the non-local total variation (NLTV) method. Methods: We modeled the low-dose CBCT reconstruction as recovering high quality, high-dose CBCT x-ray projections (100 kVp, 1.6 mAs) from low-dose, noisy CBCT x-ray projections (100 kVp, 0.1 mAs). The restoration of CBCT projections was performed using the NLTV regularization method. In NLTV, the x-ray image is optimized by minimizing an energy function that penalizes gray-level difference between pair of pixels between noisy x-ray projection and denoising x-ray projection. After the noisy projection is restored by NLTV regularization, the standard FDK method was applied to generate the final reconstruction output. Results: Significant noise reduction was achieved comparing to original, noisy inputs while maintaining the image quality comparable to the high-dose CBCT projections. The experimental validations show the proposed NLTV algorithm can robustly restore the noise level of x-ray projection images while significantly improving the overall image quality. The improvement in normalized mean square error (NMSE) and peak signal-to-noise ratio (PSNR) measured from the non-local total variation-gradient projection (NLTV-GPSR) algorithm is noticeable compared to that of uncorrected low-dose CBCT images. Moreover, the difference of CNRs from the gains from the proposed algorithm is noticeable and comparable to high-dose CBCT. Conclusion: The proposed method successfully restores noise degraded, low-dose CBCT projections to high-dose projection quality. Such an outcome is a considerable improvement to the reconstruction result compared to the FDK-based method. In addition, a significant reduction in reconstruction time makes the proposed algorithm more attractive. This demonstrates the potential use of the proposed algorithm for clinical practice in radiotherapy.

GPU acceleration of a model-based iterative method for Digital Breast Tomosynthesis

Digital Breast Tomosynthesis (DBT) is a modern 3D Computed Tomography X-ray technique for the early detection of breast tumors, which is receiving growing interest in the medical and scientific community. Since DBT performs incomplete sampling of data, the image reconstruction approaches based on iterative methods are preferable to the classical analytic techniques, such as the Filtered Back Projection algorithm, providing fewer artifacts. In this work, we consider a Model-Based Iterative Reconstruction (MBIR) method well suited to describe the DBT data acquisition process and to include prior information on the reconstructed image. We propose a gradient-based solver named Scaled Gradient Projection (SGP) for the solution of the constrained optimization problem arising in the considered MBIR method. Even if the SGP algorithm exhibits fast convergence, the time required on a serial computer for the reconstruction of a real DBT data set is too long for the clinical needs. In this paper we propose a parallel SGP version designed to perform the most expensive computations of each iteration on Graphics Processing Unit (GPU). We apply the proposed parallel approach on three different GPU boards, with computational performance comparable with that of the boards usually installed in commercial DBT systems. The numerical results show that the proposed GPU-based MBIR method provides accurate reconstructions in a time suitable for clinical trials.

A performance comparison of convolutional neural network-based image denoising methods: The effect of loss functions on low-dose CT images

PURPOSE

Convolutional neural network (CNN)-based image denoising techniques have shown promising results in low-dose CT denoising. However, CNN often introduces blurring in denoised images when trained with a widely used pixel-level loss function. Perceptual loss and adversarial loss have been proposed recently to further improve the image denoising performance. In this paper, we investigate the effect of different loss functions on image denoising performance using task-based image quality assessment methods for various signals and dose levels.

METHODS

We used a modified version of U-net that was effective at reducing the correlated noise in CT images. The loss functions used for comparison were two pixel-level losses (i.e., the mean-squared error and the mean absolute error), Visual Geometry Group network-based perceptual loss (VGG loss), adversarial loss used to train the Wasserstein generative adversarial network with gradient penalty (WGAN-GP), and their weighted summation. Each image denoising method was applied to reconstructed images and sinogram images independently and validated using the extended cardiac-torso (XCAT) simulation and Mayo Clinic datasets. In the XCAT simulation, we generated fan-beam CT datasets with four different dose levels (25%, 50%, 75%, and 100% of a normal-dose level) using 10 XCAT phantoms and inserted signals in a test set. The signals had two different shapes (spherical and spiculated), sizes (4 and 12 mm), and contrast levels (60 and 160 HU). To evaluate signal detectability, we used a detection task SNR (tSNR) calculated from a non-prewhitening model observer with an eye filter. We also measured the noise power spectrum (NPS) and modulation transfer function (MTF) to compare the noise and signal transfer properties.

RESULTS

Compared to CNNs without VGG loss, VGG-loss-based CNNs achieved a more similar tSNR to that of the normal-dose CT for all signals at different dose levels except for a small signal at the 25% dose level. For a low-contrast signal at 25% or 50% dose, adding other losses to the VGG loss showed more improved performance than only using VGG loss. The NPS shapes from VGG-loss-based CNN closely matched that of normal-dose CT images while CNN without VGG loss overly reduced the mid-high-frequency noise power at all dose levels. MTF also showed VGG-loss-based CNN with better-preserved high resolution for all dose and contrast levels. It is also observed that additional WGAN-GP loss helps improve the noise and signal transfer properties of VGG-loss-based CNN.

CONCLUSIONS

The evaluation results using tSNR, NPS, and MTF indicate that VGG-loss-based CNNs are more effective than those without VGG loss for natural denoising of low-dose images and WGAN-GP loss improves the denoising performance of VGG-loss-based CNNs, which corresponds with the qualitative evaluation.

Low dose cone-beam computed tomography reconstruction via hybrid prior contour based total variation regularization (hybrid-PCTV)

Background

Previously, we developed a prior contour based total variation (PCTV) method to use edge information derived from prior images for edge enhancement in low-dose cone-beam computed tomography (CBCT) reconstruction. However, the accuracy of edge enhancement in PCTV is affected by the deformable registration errors and anatomical changes from prior to on-board images. In this study, we develop a hybrid-PCTV method to address this limitation to enhance the robustness and accuracy of the PCTV method.

Methods

Planning-CT is used as prior images and deformably registered with on-board CBCT reconstructed by the edge preserving TV (EPTV) method. Edges derived from planning CT are deformed based on the registered deformation vector fields to generate on-board edges for edge enhancement in PCTV reconstruction. Reference CBCT is reconstructed from the simulated projections of the deformed planning-CT. Image similarity map is then calculated between reference and on-board CBCT using structural similarity index (SSIM) method to estimate local registration accuracy. The hybrid-PCTV method enhances the edge information based on a weighted edge map that combines edges from both PCTV and EPTV methods. Higher weighting is given to PCTV edges at regions with high registration accuracy and to EPTV edges at regions with low registration accuracy. The hybrid-PCTV method was evaluated using both digital extended-cardiac-torso (XCAT) phantom and lung patient data. In XCAT study, breathing amplitude change, tumor shrinkage and new tumor were simulated from CT to CBCT. In the patient study, both simulated and real projections of lung patients were used for reconstruction. Results were compared with both EPTV and PCTV methods.

Results

EPTV led to blurring bony structures due to missing edge information, and PCTV led to blurring tumor edges due to inaccurate edge information caused by errors in the deformable registration. In contrast, hybrid-PCTV enhanced edges of both bone and tumor. In XCAT study using 30 half-fan CBCT projections, compared with ground truth, relative errors (REs) were 1.3%, 1.1% and 0.9% and edge cross-correlation were 0.66, 0.68 and 0.71 for EPTV, PCTV and hybrid-PCTV, respectively. Moreover, in the lung patient data, hybrid-PCTV avoided the wrong edge enhancement in the PCTV method while maintaining enhancements of the correct edges.

Conclusions

Hybrid-PCTV further improved the robustness and accuracy of PCTV by accounting for uncertainties in deformable registration and anatomical changes between prior and onboard images. The accurate edge enhancement in hybrid-PCTV will be valuable for target localization in radiation therapy.

4D digital tomosynthesis image reconstruction using brute force-based adaptive total variation (BF-ATV) in a prototype LINAC system

Respiratory-correlated cone-beam CT (CBCT) not only inhibits rapid scanning due to the slow speed of the LINAC head gantry rotation, but its implementation for routine patient imaging is impractical because of the high radiation dose delivered during the process. Digital tomosynthesis (DTS) is a potentially faster technique that delivers a much lower radiation dose by reducing the number of projections in a limited angular range. Unfortunately, 4D-DTS introduces strong aliasing artifacts in the reconstructed images due to the sparsely sampled projections in each respiratory phase bin. The authors hereby suggest a novel low-dose 4D-DTS image reconstruction method that achieves a compromise between the occurrence of aliasing artifacts and image smoothing using a brute force-based adaptive weighting parameter searching technique. We used a prototype LINAC system mounted with a flat-panel detector to acquire tomosynthesis projections of respiratory motion in a phantom in the anterior-posterior (AP) and lateral views. Three different 4D-DTS image reconstruction schemes that included conventional filtered back-projection (FBP), adaptive steepest descent projection onto convex sets (ASD-POCS), and the proposed brute force-based adaptive total variation (BF-ATV) were implemented in four different respiratory phase bins for both AP and lateral views. All reconstructions were accelerated using a single GPU card to reduce the computation time. To study the performance of the algorithm under various sparse conditions, we operated the prototype system in three different gantry sweep modes. The results indicate that the proposed BF-ATV method yields the largest structural similarities in the differenced image between the ground-truth dataset acquired using the slow gantry sweep mode and the sparse dataset from both moderate and fast sweep modes. In addition, the proposed method maintained the object sharpness with less streaking lines and small loss of sharpness compared to the conventional FBP and ASD-POCS methods. In conclusion, the proposed low-dose 4D-DTS reconstruction scheme may provide better performance due in part to its rapid scanning. Therefore, it is potentially applicable to practical 4D imaging for radiotherapy.

Multi-pass approach to reduce cone-beam artifacts in a circular orbit cone-beam CT system

We propose a multi-pass approach to reduce cone-beam artifacts in a circular orbit cone-beam computed tomography (CT) system. Employing a large 2D detector array reduces the scan time but produces cone-beam artifacts in the Feldkamp, Davis, and Kress (FDK) reconstruction because of insufficient sampling for exact reconstruction. While the two-pass algorithm proposed by Hsieh is effective at reducing cone-beam artifacts, the correction performance is degraded when the bone density is moderate and the cone angle is large. In this work, we treated the cone-beam artifacts generated from bone and soft tissue as if they were from less dense bone objects and corrected them iteratively. The proposed method was validated using a numerical Defrise phantom, XCAT phantom data, and experimental data from a pediatric phantom followed by image quality assessment for FDK, the two-pass algorithm, the proposed method, and the total variation minimization-based iterative reconstruction (TV-IR). The results show that the proposed method was superior to the two-pass algorithm in cone-beam artifact reduction and effectively reduced the overcorrection by the two-pass algorithm near bone regions. It can also be observed that the proposed method produced better correction performance with fewer iterations than the TV-IR algorithm. A qualitative evaluation with mean-squared error, structural similarity, and structural dissimilarity demonstrated the effectiveness of the proposed method.

A Compressed-Sensing Based Blind Deconvolution Method for Image Deblurring in Dental Cone-Beam Computed Tomography

In cone-beam computed tomography (CBCT), reconstructed images are inherently degraded, restricting its image performance, due mainly to imperfections in the imaging process resulting from detector resolution, noise, X-ray tube's focal spot, and reconstruction procedure as well. Thus, the recovery of CBCT images from their degraded version is essential for improving image quality. In this study, we investigated a compressed-sensing (CS)-based blind deconvolution method to solve the blurring problem in CBCT where both the image to be recovered and the blur kernel (or point-spread function) of the imaging system are simultaneously recursively identified. We implemented the proposed algorithm and performed a systematic simulation and experiment to demonstrate the feasibility of using the algorithm for image deblurring in dental CBCT. In the experiment, we used a commercially available dental CBCT system that consisted of an X-ray tube, which was operated at 90 kV_p and 5 mA, and a CMOS flat-panel detector with a 200-μm pixel size. The image characteristics were quantitatively investigated in terms of the image intensity, the root-mean-square error, the contrast-to-noise ratio, and the noise power spectrum. The results indicate that our proposed method effectively reduced the image blur in dental CBCT, excluding repetitious measurement of the system's blur kernel.

Early Changes in Serial CBCT-Measured Parotid Gland Biomarkers Predict Chronic Xerostomia After Head and Neck Radiation Therapy

PURPOSE

To determine whether serial cone beam computed tomography (CBCT) images taken during head and neck radiation therapy (HNR) can improve chronic xerostomia prediction.

METHODS AND MATERIALS

In a retrospective analysis, parotid glands (PGs) were delineated on daily kV CBCT images using deformable image registration for 119 HNR patients (60 or 70 Gy in 2 Gy fractions over 6 or 7 weeks). Deformable image registration accuracy for a subset of deformed contours was quantified using the Dice similarity coefficient and mean distance to agreement in comparison with manually drawn contours. Average weekly changes in CBCT-measured mean Hounsfield unit intensity and volume were calculated for each PG relative to week 1. Dose-volume histogram statistics were extracted from each plan, and interactions among dose, volume, and intensity were investigated. Univariable analysis and penalized logistic regression were used to analyze association with observer-rated xerostomia at 1 year after HNR. Models including CBCT delta imaging features were compared with clinical and dose-volume histogram-only models using area under the receiver operating characteristic curve (AUC) for grade ≥1 and grade ≥2 xerostomia prediction.

RESULTS

All patients experienced end-treatment PG volume reduction with mean (range) ipsilateral and contralateral PG shrinkage of 19.6% (0.9%-58.4%) and 17.7% (4.4%-56.3%), respectively. Midtreatment volume change was highly correlated with mean PG dose (r = -0.318, P < 1e^-6). Incidence of grade ≥1 and grade ≥2 xerostomia was 65% and 16%, respectively. For grade ≥1 xerostomia prediction, the delta-imaging model had an AUC of 0.719 (95% confidence interval [CI], 0.603-0.830), compared with 0.709 (95% CI, 0.603-0.815) for the dose/clinical model. For grade ≥2 xerostomia prediction, the dose/clinical model had an AUC of 0.692 (95% CI, 0.615-0.770), and the addition of contralateral PG changes modestly improved predictive performance, with an AUC of 0.776 (0.643-0.912).

CONCLUSIONS

The rate of CBCT-measured PG image feature changes improves prediction over dose alone for chronic xerostomia prediction. Analysis of CBCT images acquired for treatment positioning may provide an inexpensive monitoring system to support toxicity-reducing adaptive radiation therapy.

GPU based parallel acceleration for fast C-arm cone-beam CT reconstruction

BACKGROUND

With the introduction of Flat Panel Detector technology, cone-beam CT (CBCT) has become a novel image modality, and widely applied in clinical practices. C-arm mounted CBCT has shown extra suitability in image guided interventional surgeries. During practice, how to acquire high resolution and high quality 3D images with the real time requirement of clinical applications remain challenging.

METHODS

In this paper, we propose a GPU based accelerated method for fast C-arm CBCT 3D image reconstructions. A filtered back projection method is optimized and implemented with GPU parallel acceleration technique. A distributed system is designed to make full use of the image acquisition consumption to hide the reconstruction delay to further improve system performance.

RESULTS

With the acceleration both in algorithm and system design, we show that our method significantly increases system efficiency. The optimized GPU accelerated FDK algorithm improves the reconstruction efficiency. The system performance is further enhanced with the proposed system design by 26% and reconstruction delay is accelerated by 2.1 times when 90 frames of projections are used. When the number of frames used increases to 120, the numbers are 39% and 3.3 times. We also show that when the projection acquisition consumption increases, the reconstruction acceleration rate increases significantly.

Low dose CBCT reconstruction via prior contour based total variation (PCTV) regularization: a feasibility study

PURPOSE

compressed sensing reconstruction using total variation (TV) tends to over-smooth the edge information by uniformly penalizing the image gradient. The goal of this study is to develop a novel prior contour based TV (PCTV) method to enhance the edge information in compressed sensing reconstruction for CBCT.

METHODS

the edge information is extracted from prior planning-CT via edge detection. Prior CT is first registered with on-board CBCT reconstructed with TV method through rigid or deformable registration. The edge contours in prior-CT is then mapped to CBCT and used as the weight map for TV regularization to enhance edge information in CBCT reconstruction. The PCTV method was evaluated using extended-cardiac-torso (XCAT) phantom, physical CatPhan phantom and brain patient data. Results were compared with both TV and edge preserving TV (EPTV) methods which are commonly used for limited projection CBCT reconstruction. Relative error was used to calculate pixel value difference and edge cross correlation was defined as the similarity of edge information between reconstructed images and ground truth in the quantitative evaluation.

RESULTS

compared to TV and EPTV, PCTV enhanced the edge information of bone, lung vessels and tumor in XCAT reconstruction and complex bony structures in brain patient CBCT. In XCAT study using 45 half-fan CBCT projections, compared with ground truth, relative errors were 1.5%, 0.7% and 0.3% and edge cross correlations were 0.66, 0.72 and 0.78 for TV, EPTV and PCTV, respectively. PCTV is more robust to the projection number reduction. Edge enhancement was reduced slightly with noisy projections but PCTV was still superior to other methods. PCTV can maintain resolution while reducing the noise in the low mAs CatPhan reconstruction. Low contrast edges were preserved better with PCTV compared with TV and EPTV.

CONCLUSION

PCTV preserved edge information as well as reduced streak artifacts and noise in low dose CBCT reconstruction. PCTV is superior to TV and EPTV methods in edge enhancement, which can potentially improve the localization accuracy in radiation therapy.

Development of a chest digital tomosynthesis R/F system and implementation of low-dose GPU-accelerated compressed sensing (CS) image reconstruction

PURPOSE

This work describes the hardware and software developments of a prototype chest digital tomosynthesis (CDT) R/F system. The purpose of this study was to validate the developed system for its possible clinical application on low-dose chest tomosynthesis imaging.

METHODS

The prototype CDT R/F system was operated by carefully controlling the electromechanical subsystems through a synchronized interface. Once a command signal was delivered by the user, a tomosynthesis sweep started to acquire 81 projection views (PVs) in a limited angular range of ±20°. Among the full projection dataset of 81 images, several sets of 21 (quarter view) and 41 (half view) images with equally spaced angle steps were selected to represent a sparse view condition. GPU-accelerated and total-variation (TV) regularization strategy-based compressed sensing (CS) image reconstruction was implemented. The imaged objects were a flat-field using a copper filter to measure the noise power spectrum (NPS), a Catphan^® CTP682 quality assurance (QA) phantom to measure a task-based modulation transfer function (MTF_Task ) of three different cylinders' edge, and an anthropomorphic chest phantom with inserted lung nodules. The authors also verified the accelerated computing power over CPU programming by checking the elapsed time required for the CS method. The resultant absorbed and effective doses that were delivered to the chest phantom from two-view digital radiographic projections, helical computed tomography (CT), and the prototype CDT system were compared.

RESULTS

The prototype CDT system was successfully operated, showing little geometric error with fast rise and fall times of R/F x-ray pulse less than 2 and 10 ms, respectively. The in-plane NPS presented essential symmetric patterns as predicted by the central slice theorem. The NPS images from 21 PVs were provided quite different pattern against 41 and 81 PVs due to aliased noise. The voxel variance values which summed all NPS intensities were inversely proportional to the number of PVs, and the CS method gave much lower voxel variance by the factors of 3.97-6.43 and 2.28-3.36 compared to filtered backprojection (FBP) and 20 iterations of simultaneous algebraic reconstruction technique (SART). The spatial frequencies of the f₅₀ at which the MTF_Task reduced to 50% were 1.50, 1.55, and 1.67 cycles/mm for FBP, SART, and CS methods, respectively, in the case of Bone 20% cylinder using 41 views. A variety of ranges of TV reconstruction parameters were implemented during the CS method and we could observe that the NPS and MTF_Task preserved best when the regularization and TV smoothing parameters α and τ were in a range of 0.001-0.1. For the chest phantom data, the signal difference to noise ratios (SDNRs) were higher in the proposed CS scheme images than in the FBP and SART, showing the enhanced rate of 1.05-1.43 for half view imaging. The total averaged reconstruction time during 20 iterations of the CS scheme was 124.68 s, which could match-up a clinically feasible time (<3 min). This computing time represented an enhanced speed 386 times greater than CPU programming. The total amounts of estimated effective doses were 0.12, 0.53 (half view), and 2.56 mSv for two-view radiographs, the prototype CDT system, and helical CT, respectively, showing 4.49 times higher than conventional radiography and 4.83 times lower than a CT exam, respectively.

CONCLUSIONS

The current work describes the development and performance assessment of both hardware and software for tomosynthesis applications. The authors observed reasonable outcomes by showing a potential for low-dose application in CDT imaging using GPU acceleration.

Image reconstruction in region-of-interest (or interior) digital tomosynthesis (DTS) based on compressed-sensing (CS)

BACKGROUND AND OBJECTIVE

Digital tomosynthesis (DTS) based on filtered-backprojection (FBP) reconstruction requires a full field-of-view (FOV) scan and relatively dense projections, which results in high doses for medical imaging purposes. To overcome these difficulties, we investigated region-of-interest (ROI) or interior DTS reconstruction where the x-ray beam span covers only a small ROI containing a target area.

METHODS

An iterative method based on compressed-sensing (CS) scheme was compared with the FBP-based algorithm for ROI-DTS reconstruction. We implemented both algorithms and performed a systematic simulation and experiments on body and skull phantoms. The image characteristics were evaluated and compared.

RESULTS

The CS-based algorithm yielded much better reconstruction quality in ROI-DTS compared to the FBP-based algorithm, preserving superior image homogeneity, edge sharpening, and in-plane resolution. The image characteristics of the CS-reconstructed images in ROI-DTS were not significantly different from those in full-FOV DTS. The measured CNR value of the CS-reconstructed ROI-DTS image was about 12.3, about 1.9 times larger than that of the FBP-reconstructed ROI-DTS image.

CONCLUSIONS

ROI-DTS images of substantially high accuracy were obtained using the CS-based algorithm and at reduced imaging doses and less computational cost, compared to typical full-FOV DTS images. We expect that the proposed method will be useful for the development of new DTS systems.

Discriminative Prior - Prior Image Constrained Compressed Sensing Reconstruction for Low-Dose CT Imaging

X-ray computed tomography (CT) has been widely used to provide patient-specific anatomical information in the forms of tissue attenuation. However, the cumulative radiation induced in CT scan has raised extensive concerns in recently years. How to maintain reconstruction image quality is a major challenge for low-dose CT (LDCT) imaging. Generally, LDCT imaging can be greatly improved by incorporating prior knowledge in some specific forms. A joint estimation framework termed discriminative prior-prior image constrained compressed sensing (DP-PICCS) reconstruction is proposed in this paper. This DP-PICCS algorithm utilizes discriminative prior knowledge via two feature dictionary constraints which built on atoms from the samples of tissue attenuation feature patches and noise-artifacts residual feature patches, respectively. Also, the prior image construction relies on a discriminative feature representation (DFR) processing by two feature dictionary. Its comparison to other competing methods through experiments on low-dose projections acquired from torso phantom simulation study and clinical abdomen study demonstrated that the DP-PICCS method achieved promising improvement in terms of the effectively-suppressed noise and the well-retained structures.

A review of GPU-based medical image reconstruction

Tomographic image reconstruction is a computationally demanding task, even more so when advanced models are used to describe a more complete and accurate picture of the image formation process. Such advanced modeling and reconstruction algorithms can lead to better images, often with less dose, but at the price of long calculation times that are hardly compatible with clinical workflows. Fortunately, reconstruction tasks can often be executed advantageously on Graphics Processing Units (GPUs), which are exploited as massively parallel computational engines. This review paper focuses on recent developments made in GPU-based medical image reconstruction, from a CT, PET, SPECT, MRI and US perspective. Strategies and approaches to get the most out of GPUs in image reconstruction are presented as well as innovative applications arising from an increased computing capacity. The future of GPU-based image reconstruction is also envisioned, based on current trends in high-performance computing.

Combining Acceleration Techniques for Low-Dose X-Ray Cone Beam Computed Tomography Image Reconstruction

Background and Objective

Over the past decade, image quality in low-dose computed tomography has been greatly improved by various compressive sensing- (CS-) based reconstruction methods. However, these methods have some disadvantages including high computational cost and slow convergence rate. Many different speed-up techniques for CS-based reconstruction algorithms have been developed. The purpose of this paper is to propose a fast reconstruction framework that combines a CS-based reconstruction algorithm with several speed-up techniques.

Methods

First, total difference minimization (TDM) was implemented using the soft-threshold filtering (STF). Second, we combined TDM-STF with the ordered subsets transmission (OSTR) algorithm for accelerating the convergence. To further speed up the convergence of the proposed method, we applied the power factor and the fast iterative shrinkage thresholding algorithm to OSTR and TDM-STF, respectively.

Results

Results obtained from simulation and phantom studies showed that many speed-up techniques could be combined to greatly improve the convergence speed of a CS-based reconstruction algorithm. More importantly, the increased computation time (≤10%) was minor as compared to the acceleration provided by the proposed method.

Conclusions

In this paper, we have presented a CS-based reconstruction framework that combines several acceleration techniques. Both simulation and phantom studies provide evidence that the proposed method has the potential to satisfy the requirement of fast image reconstruction in practical CT.

Variable step size methods for solving simultaneous algebraic reconstruction technique (SART)-type cbct reconstructions

Compared to analytical reconstruction by Feldkamp-Davis-Kress (FDK), simultaneous algebraic reconstruction technique (SART) offers a higher degree of flexibility in input measurements and often produces superior quality images. Due to the iterative nature of the algorithm, however, SART requires intense computations which have prevented its use in clinical practice. In this paper, we developed a fast-converging SART-type algorithm and showed its clinical feasibility in CBCT reconstructions. Inspired by the quasi-orthogonal nature of the x-ray projections in CBCT, we implement a simple yet much faster algorithm by computing Barzilai and Borwein step size at each iteration. We applied this variable step-size (VS)-SART algorithm to numerical and physical phantoms as well as cancer patients for reconstruction. By connecting the SART algebraic problem to the statistical weighted least squares problem, we enhanced the reconstruction speed significantly (i.e., less number of iterations). We further accelerated the reconstruction speed of algorithms by using the parallel computing power of GPU.

Simultaneous deblurring and iterative reconstruction of CBCT for image guided brain radiosurgery

One of the limiting factors in cone-beam CT (CBCT) image quality is system blur, caused by detector response, x-ray source focal spot size, azimuthal blurring, and reconstruction algorithm. In this work, we develop a novel iterative reconstruction algorithm that improves spatial resolution by explicitly accounting for image unsharpness caused by different factors in the reconstruction formulation. While the model-based iterative reconstruction techniques use prior information about the detector response and x-ray source, our proposed technique uses a simple measurable blurring model. In our reconstruction algorithm, denoted as simultaneous deblurring and iterative reconstruction (SDIR), the blur kernel can be estimated using the modulation transfer function (MTF) slice of the CatPhan phantom or any other MTF phantom, such as wire phantoms. The proposed image reconstruction formulation includes two regularization terms: (1) total variation (TV) and (2) nonlocal regularization, solved with a split Bregman augmented Lagrangian iterative method. The SDIR formulation preserves edges, eases the parameter adjustments to achieve both high spatial resolution and low noise variances, and reduces the staircase effect caused by regular TV-penalized iterative algorithms. The proposed algorithm is optimized for a point-of-care head CBCT unit for image-guided radiosurgery and is tested with CatPhan phantom, an anthropomorphic head phantom, and 6 clinical brain stereotactic radiosurgery cases. Our experiments indicate that SDIR outperforms the conventional filtered back projection and TV penalized simultaneous algebraic reconstruction technique methods (represented by adaptive steepest-descent POCS algorithm, ASD-POCS) in terms of MTF and line pair resolution, and retains the favorable properties of the standard TV-based iterative reconstruction algorithms in improving the contrast and reducing the reconstruction artifacts. It improves the visibility of the high contrast details in bony areas and the brain soft-tissue. For example, the results show the ventricles and some brain folds become visible in SDIR reconstructed images and the contrast of the visible lesions is effectively improved. The line-pair resolution was improved from 12 line-pair/cm in FBP to 14 line-pair/cm in SDIR. Adjusting the parameters of the ASD-POCS to achieve 14 line-pair/cm caused the noise variance to be higher than the SDIR. Using these parameters for ASD-POCS, the MTF of FBP and ASD-POCS were very close and equal to 0.7 mm^-1 which was increased to 1.2 mm^-1 by SDIR, at half maximum.

Three-dimensional optoacoustic reconstruction using fast sparse representation

Optoacoustic tomography based on insufficient spatial sampling of ultrasound waves leads to loss of contrast and artifacts on the reconstructed images. Compared to reconstructions based on L2-norm regularization, sparsity-based reconstructions may improve contrast and reduce image artifacts but at a high computational cost, which has so far limited their use to 2D optoacoustic tomography. Here we propose a fast, sparsity-based reconstruction algorithm for 3D optoacoustic tomography, based on gradient descent with Barzilai-Borwein line search (L1-GDBB). Using simulations and experiments, we show that the L1-GDBB offers fourfold faster reconstruction than the previously reported L1-norm regularized reconstruction based on gradient descent with backtracking line search. Moreover, the new algorithm provides higher-quality images with fewer artifacts than the L2-norm regularized reconstruction and the back-projection reconstruction.

An efficient iterative CBCT reconstruction approach using gradient projection sparse reconstruction algorithm

The purpose of this study is to develop a fast and convergence proofed CBCT reconstruction framework based on the compressed sensing theory which not only lowers the imaging dose but also is computationally practicable in the busy clinic. We simplified the original mathematical formulation of gradient projection for sparse reconstruction (GPSR) to minimize the number of forward and backward projections for line search processes at each iteration. GPSR based algorithms generally showed improved image quality over the FDK algorithm especially when only a small number of projection data were available. When there were only 40 projections from 360 degree fan beam geometry, the quality of GPSR based algorithms surpassed FDK algorithm within 10 iterations in terms of the mean squared relative error. Our proposed GPSR algorithm converged as fast as the conventional GPSR with a reasonably low computational complexity. The outcomes demonstrate that the proposed GPSR algorithm is attractive for use in real time applications such as on-line IGRT.

Accelerated fast iterative shrinkage thresholding algorithms for sparsity-regularized cone-beam CT image reconstruction

PURPOSE

The development of iterative image reconstruction algorithms for cone-beam computed tomography (CBCT) remains an active and important research area. Even with hardware acceleration, the overwhelming majority of the available 3D iterative algorithms that implement nonsmooth regularizers remain computationally burdensome and have not been translated for routine use in time-sensitive applications such as image-guided radiation therapy (IGRT). In this work, two variants of the fast iterative shrinkage thresholding algorithm (FISTA) are proposed and investigated for accelerated iterative image reconstruction in CBCT.

METHODS

Algorithm acceleration was achieved by replacing the original gradient-descent step in the FISTAs by a subproblem that is solved by use of the ordered subset simultaneous algebraic reconstruction technique (OS-SART). Due to the preconditioning matrix adopted in the OS-SART method, two new weighted proximal problems were introduced and corresponding fast gradient projection-type algorithms were developed for solving them. We also provided efficient numerical implementations of the proposed algorithms that exploit the massive data parallelism of multiple graphics processing units.

RESULTS

The improved rates of convergence of the proposed algorithms were quantified in computer-simulation studies and by use of clinical projection data corresponding to an IGRT study. The accelerated FISTAs were shown to possess dramatically improved convergence properties as compared to the standard FISTAs. For example, the number of iterations to achieve a specified reconstruction error could be reduced by an order of magnitude. Volumetric images reconstructed from clinical data were produced in under 4 min.

CONCLUSIONS

The FISTA achieves a quadratic convergence rate and can therefore potentially reduce the number of iterations required to produce an image of a specified image quality as compared to first-order methods. We have proposed and investigated accelerated FISTAs for use with two nonsmooth penalty functions that will lead to further reductions in image reconstruction times while preserving image quality. Moreover, with the help of a mixed sparsity-regularization, better preservation of soft-tissue structures can be potentially obtained. The algorithms were systematically evaluated by use of computer-simulated and clinical data sets.

Structure-adaptive CBCT reconstruction using weighted total variation and Hessian penalties

The exposure of normal tissues to high radiation during cone-beam CT (CBCT) imaging increases the risk of cancer and genetic defects. Statistical iterative algorithms with the total variation (TV) penalty have been widely used for low dose CBCT reconstruction, with state-of-the-art performance in suppressing noise and preserving edges. However, TV is a first-order penalty and sometimes leads to the so-called staircase effect, particularly over regions with smooth intensity transition in the reconstruction images. A second-order penalty known as the Hessian penalty was recently used to replace TV to suppress the staircase effect in CBCT reconstruction at the cost of slightly blurring object edges. In this study, we proposed a new penalty, the TV-H, which combines TV and Hessian penalties for CBCT reconstruction in a structure-adaptive way. The TV-H penalty automatically differentiates the edges, gradual transition and uniform local regions within an image using the voxel gradient, and adaptively weights TV and Hessian according to the local image structures in the reconstruction process. Our proposed penalty retains the benefits of TV, including noise suppression and edge preservation. It also maintains the structures in regions with gradual intensity transition more successfully. A majorization-minimization (MM) approach was designed to optimize the objective energy function constructed with the TV-H penalty. The MM approach employed a quadratic upper bound of the original objective function, and the original optimization problem was changed to a series of quadratic optimization problems, which could be efficiently solved using the Gauss-Seidel update strategy. We tested the reconstruction algorithm on two simulated digital phantoms and two physical phantoms. Our experiments indicated that the TV-H penalty visually and quantitatively outperformed both TV and Hessian penalties.

Dual energy CT with one full scan and a second sparse-view scan using structure preserving iterative reconstruction (SPIR)

Conventional dual-energy CT (DECT) reconstruction requires two full-size projection datasets with two different energy spectra. In this study, we propose an iterative algorithm to enable a new data acquisition scheme which requires one full scan and a second sparse-view scan for potential reduction in imaging dose and engineering cost of DECT. A bilateral filter is calculated as a similarity matrix from the first full-scan CT image to quantify the similarity between any two pixels, which is assumed unchanged on a second CT image since DECT scans are performed on the same object. The second CT image from reduced projections is reconstructed by an iterative algorithm which updates the image by minimizing the total variation of the difference between the image and its filtered image by the similarity matrix under data fidelity constraint. As the redundant structural information of the two CT images is contained in the similarity matrix for CT reconstruction, we refer to the algorithm as structure preserving iterative reconstruction (SPIR). The proposed method is evaluated on both digital and physical phantoms, and is compared with the filtered-backprojection (FBP) method, the conventional total-variation-regularization-based algorithm (TVR) and prior-image-constrained-compressed-sensing (PICCS). SPIR with a second 10-view scan reduces the image noise STD by a factor of one order of magnitude with same spatial resolution as full-view FBP image. SPIR substantially improves over TVR on the reconstruction accuracy of a 10-view scan by decreasing the reconstruction error from 6.18% to 1.33%, and outperforms TVR at 50 and 20-view scans on spatial resolution with a higher frequency at the modulation transfer function value of 10% by an average factor of 4. Compared with the 20-view scan PICCS result, the SPIR image has 7 times lower noise STD with similar spatial resolution. The electron density map obtained from the SPIR-based DECT images with a second 10-view scan has an average error of less than 1%.

Low-dose CT reconstruction via L1 dictionary learning regularization using iteratively reweighted least-squares

Background

In order to reduce the radiation dose of CT (computed tomography), compressed sensing theory has been a hot topic since it provides the possibility of a high quality recovery from the sparse sampling data. Recently, the algorithm based on DL (dictionary learning) was developed to deal with the sparse CT reconstruction problem. However, the existing DL algorithm focuses on the minimization problem with the L₂-norm regularization term, which leads to reconstruction quality deteriorating while the sampling rate declines further. Therefore, it is essential to improve the DL method to meet the demand of more dose reduction.

Methods

In this paper, we replaced the L₂-norm regularization term with the L₁-norm one. It is expected that the proposed L₁-DL method could alleviate the over-smoothing effect of the L₂-minimization and reserve more image details. The proposed algorithm solves the L₁-minimization problem by a weighting strategy, solving the new weighted L₂-minimization problem based on IRLS (iteratively reweighted least squares).

Results

Through the numerical simulation, the proposed algorithm is compared with the existing DL method (adaptive dictionary based statistical iterative reconstruction, ADSIR) and other two typical compressed sensing algorithms. It is revealed that the proposed algorithm is more accurate than the other algorithms especially when further reducing the sampling rate or increasing the noise.

Conclusion

The proposed L₁-DL algorithm can utilize more prior information of image sparsity than ADSIR. By transforming the L₂-norm regularization term of ADSIR with the L₁-norm one and solving the L₁-minimization problem by IRLS strategy, L₁-DL could reconstruct the image more exactly.

Iterative CBCT reconstruction using Hessian penalty

Statistical iterative reconstruction algorithms have shown potential to improve cone-beam CT (CBCT) image quality. Most iterative reconstruction algorithms utilize prior knowledge as a penalty term in the objective function. The penalty term greatly affects the performance of a reconstruction algorithm. The total variation (TV) penalty has demonstrated great ability in suppressing noise and improving image quality. However, calculated from the first-order derivatives, the TV penalty leads to the well-known staircase effect, which sometimes makes the reconstructed images oversharpen and unnatural. In this study, we proposed to use a second-order derivative penalty that involves the Frobenius norm of the Hessian matrix of an image for CBCT reconstruction. The second-order penalty retains some of the most favorable properties of the TV penalty like convexity, homogeneity, and rotation and translation invariance, and has a better ability in preserving the structures of gradual transition in the reconstructed images. An effective algorithm was developed to minimize the objective function with the majorization-minimization (MM) approach. The experiments on a digital phantom and two physical phantoms demonstrated the priority of the proposed penalty, particularly in suppressing the staircase effect of the TV penalty.

A fast time-difference inverse solver for 3D EIT with application to lung imaging

A class of sparse optimization techniques that require solely matrix-vector products, rather than an explicit access to the forward matrix and its transpose, has been paid much attention in the recent decade for dealing with large-scale inverse problems. This study tailors application of the so-called Gradient Projection for Sparse Reconstruction (GPSR) to large-scale time-difference three-dimensional electrical impedance tomography (3D EIT). 3D EIT typically suffers from the need for a large number of voxels to cover the whole domain, so its application to real-time imaging, for example monitoring of lung function, remains scarce since the large number of degrees of freedom of the problem extremely increases storage space and reconstruction time. This study shows the great potential of the GPSR for large-size time-difference 3D EIT. Further studies are needed to improve its accuracy for imaging small-size anomalies.

Total Variation-Based Reduction of Streak Artifacts, Ring Artifacts and Noise in 3D Reconstruction from Optical Projection Tomography

Optical projection tomography (OPT) is a computed tomography technique at optical frequencies for samples of 0.5-15 mm in size, which fills an important "imaging gap" between confocal microscopy (for smaller samples) and large-sample methods such as fluorescence molecular tomography or micro magnetic resonance imaging. OPT operates in either fluorescence or transmission mode. Two-dimensional (2D) projections are taken over 360° with a fixed rotational increment around the vertical axis. Standard 3D reconstruction from 2D OPT uses the filtered backprojection (FBP) algorithm based on the Radon transform. FBP approximates the inverse Radon transform using a ramp filter that spreads reconstructed pixels to neighbor pixels thus producing streak and other types of artifacts, as well as noise. Artifacts increase the variation of grayscale values in the reconstructed images. We present an algorithm that improves the quality of reconstruction even for a low number of projections by simultaneously minimizing the sum of absolute brightness changes in the reconstructed volume (the total variation) and the error between measured and reconstructed data. We demonstrate the efficiency of the method on real biological data acquired on a dedicated OPT device.

Common-mask guided image reconstruction (c-MGIR) for enhanced 4D cone-beam computed tomography

Compared to 3D cone beam computed tomography (3D CBCT), the image quality of commercially available four-dimensional (4D) CBCT is severely impaired due to the insufficient amount of projection data available for each phase. Since the traditional Feldkamp-Davis-Kress (FDK)-based algorithm is infeasible for reconstructing high quality 4D CBCT images with limited projections, investigators had developed several compress-sensing (CS) based algorithms to improve image quality. The aim of this study is to develop a novel algorithm which can provide better image quality than the FDK and other CS based algorithms with limited projections. We named this algorithm 'the common mask guided image reconstruction' (c-MGIR).In c-MGIR, the unknown CBCT volume is mathematically modeled as a combination of phase-specific motion vectors and phase-independent static vectors. The common-mask matrix, which is the key concept behind the c-MGIR algorithm, separates the common static part across all phase images from the possible moving part in each phase image. The moving part and the static part of the volumes were then alternatively updated by solving two sub-minimization problems iteratively. As the novel mathematical transformation allows the static volume and moving volumes to be updated (during each iteration) with global projections and 'well' solved static volume respectively, the algorithm was able to reduce the noise and under-sampling artifact (an issue faced by other algorithms) to the maximum extent. To evaluate the performance of our proposed c-MGIR, we utilized imaging data from both numerical phantoms and a lung cancer patient. The qualities of the images reconstructed with c-MGIR were compared with (1) standard FDK algorithm, (2) conventional total variation (CTV) based algorithm, (3) prior image constrained compressed sensing (PICCS) algorithm, and (4) motion-map constrained image reconstruction (MCIR) algorithm, respectively. To improve the efficiency of the algorithm, the code was implemented with a graphic processing unit for parallel processing purposes.Root mean square error (RMSE) between the ground truth and reconstructed volumes of the numerical phantom were in the descending order of FDK, CTV, PICCS, MCIR, and c-MGIR for all phases. Specifically, the means and the standard deviations of the RMSE of FDK, CTV, PICCS, MCIR and c-MGIR for all phases were 42.64  ±  6.5%, 3.63  ±  0.83%, 1.31%  ±  0.09%, 0.86%  ±  0.11% and 0.52 %  ±  0.02%, respectively. The image quality of the patient case also indicated the superiority of c-MGIR compared to other algorithms.The results indicated that clinically viable 4D CBCT images can be reconstructed while requiring no more projection data than a typical clinical 3D CBCT scan. This makes c-MGIR a potential online reconstruction algorithm for 4D CBCT, which can provide much better image quality than other available algorithms, while requiring less dose and potentially less scanning time.

3D reconstruction based on compressed-sensing (CS)-based framework by using a dental panoramic detector

In this work, we propose a practical method that can combine the two functionalities of dental panoramic and cone-beam CT (CBCT) features in one by using a single panoramic detector. We implemented a CS-based reconstruction algorithm for the proposed method and performed a systematic simulation to demonstrate its viability for 3D dental X-ray imaging. We successfully reconstructed volumetric images of considerably high accuracy by using a panoramic detector having an active area of 198.4 mm × 6.4 mm and evaluated the reconstruction quality as a function of the pitch (p) and the angle step (Δθ). Our simulation results indicate that the CS-based reconstruction almost completely recovered the phantom structures, as in CBCT, for p≤2.0 and θ≤6°, indicating that it seems very promising for accurate image reconstruction even for large-pitch and few-view data. We expect the proposed method to be applicable to developing a cost-effective, volumetric dental X-ray imaging system.