Iterative CT Reconstruction Research Articles

RBP-DIP: Residual back projection with deep image prior for ill-posed CT reconstruction

The success of deep image prior (DIP) in a number of image processing tasks has motivated their application in image reconstruction problems in computed tomography (CT). In this paper, we introduce a residual back projection technique (RBP) that improves the performance of deep image prior framework in iterative CT reconstruction, especially when the reconstruction problem is highly ill-posed. The RBP-DIP framework uses an untrained U-net in conjunction with a novel residual back projection connection to minimize the objective function while improving reconstruction accuracy. In each iteration, the weights of the untrained U-net are optimized, and the output of the U-net in the current iteration is used to update the input of the U-net in the next iteration through the proposed RBP connection. The introduction of the RBP connection strengthens the regularization effects of the DIP framework in the context of iterative CT reconstruction leading to improvements in accuracy. Our experiments demonstrate that the RBP-DIP framework offers improvements over other state-of-the-art conventional IR methods, as well as pre-trained and untrained models with similar network structures under multiple conditions. These improvements are particularly significant in the few-view and limited-angle CT reconstructions, where the corresponding inverse problems are highly ill-posed and the training data is limited. Furthermore, RBP-DIP has the potential for further improvement. Most existing IR algorithms, pre-trained models, and enhancements applicable to the original DIP algorithm can also be integrated into the RBP-DIP framework.

Artificial intelligence iterative reconstruction in abdominal CT of patients with irregular arm positioning: a case-by-case evaluation.

Streak artifacts induced by irregular arm positioning have been an issue in diagnosing the abdomen. To illustrate the risk of misdiagnosis in abdominal computed tomography (CT) of patients with irregular arm positioning through a case-by-case evaluation and to test if it can be solved by the artificial intelligence iterative reconstruction (AIIR) algorithm. By reviewing 5220 cases of chest and thoracoabdominal CT, 64 patients with irregular arm positioning were enrolled, whose image data were reconstructed using AIIR in addition to routine hybrid iterative reconstruction (HIR). Lesion detection for livers, spleens, kidneys, gallbladders, and pancreas on AIIR images, performed by two radiologists, was compared with those on HIR images. Discrepancies arising from AIIR images included both cases with additional abnormalities and those with corrections made on previous detections. For cases with discrepancies, artifact scores for organs where discrepancies were found, and contrast-to-noise ratios (CNRs) of cysts with discrepancies were compared between two image sets. Additional abnormalities were detected for 15 cases: additional liver cirrhosis (n=2); additional gallbladder stone (n=1); additional cholecystitis (n=1), additional spleen nodule (n=1); additional kidney cysts (n=8); additional liver cysts (3); and additional spleen cyst (n=1). A spleen contusion was corrected for one case. All involved artifact scores were improved on AIIR images. CNRs of involved liver, kidney, and spleen cysts were improved by up to 539.7%, 538.5%, and 245.5%, respectively. Irregular arm positioning may induce a variety of misdiagnoses in abdominal CT, which is almost totally avoidable by the AIIR algorithm.

2255: Iterative CT Reconstruction with Generative AI

2255: Iterative CT Reconstruction with Generative AI

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.

Robust residual-guided iterative reconstruction for sparse-view CT in small animal imaging

Objective. We introduce a robust image reconstruction algorithm named residual-guided Golub–Kahan iterative reconstruction technique (RGIRT) designed for sparse-view computed tomography (CT), which aims at high-fidelity image reconstruction from a limited number of projection views. Approach. RGIRT utilizes an inner-outer dual iteration framework, with a flexible least square QR (FLSQR) algorithm implemented in the inner iteration and a restarted iterative scheme applied in the outer iteration. The inner FLSQR employs a flexible Golub–Kahan bidiagonalization method to reduce the size of the inverse problem, and a weighted generalized cross-validation method to adaptively estimate the regularization hyper-parameter. The inner iteration efficiently yields the intermediate reconstruction result, while the outer iteration minimizes the residual and refines the solution by using the result obtained from the inner iteration. Main results. The reconstruction performance of RGIRT is evaluated and compared to other reference methods (FBPConvNet, SART-TV, and FLSQR) using projection data from both numerical phantoms and real experimental Micro-CT data. The experimental findings, from testing various numbers of projection views and different noise levels, underscore the robustness of RGIRT. Meanwhile, theoretical analysis confirms the convergence of residual for our approach. Significance. We propose a robust iterative reconstruction algorithm for x-ray CT scans with sparse views, thereby shortening scanning time and mitigating excessive ionizing radiation exposure to small animals.

Sparse-view cone-beam computed tomography iterative reconstruction based on new multi-gradient direction total variation.

The accurate reconstruction of cone-beam computed tomography (CBCT) from sparse projections is one of the most important areas for study. The compressed sensing theory has been widely employed in the sparse reconstruction of CBCT. However, the total variation (TV) approach solely uses information from the i-coordinate, j-coordinate, and k-coordinate gradients to reconstruct the CBCT image. It is well recognized that the CBCT image can be reconstructed more accurately with more gradient information from different directions. Thus, this study introduces a novel approach, named the new multi-gradient direction total variation minimization method. The method uses gradient information from the ij-coordinate, ik-coordinate, and jk-coordinate directions to reconstruct CBCT images, which incorporates nine different types of gradient information from nine directions. This study assessed the efficacy of the proposed methodology using under-sampled projections from four different experiments, including two digital phantoms, one patient's head dataset, and one physical phantom dataset. The results indicated that the proposed method achieved the lowest RMSE index and the highest SSIM index. Meanwhile, we compared the voxel intensity curves of the reconstructed images to assess the edge structure preservation. Among the various methods compared, the curves generated by the proposed method exhibited the highest level of consistency with the gold standard image curves. In summary, the proposed method showed significant potential in enhancing the quality and accuracy of CBCT image reconstruction.

CT target scanning in the diagnosis of solid pseudopapillary tumor of the pancreas

ObjectiveTo discuss the value of computed tomography (CT) iterative reconstruction technique combined with target scanning in the diagnosis of solid pseudopapillary tumor of the pancreas (SPTP).MethodsThe clinical information and CT examination data of 27 patients with SPTP were retrospectively analyzed, and the general condition and CT performance of the patients were observed. The CT image reconstruction algorithm of all patients used iterative reconstruction technique combined with the application of target scanning technique.ResultsA total of 27 patients were included in this study, including 6 males and 21 females, aged 14–72 years with a mean age of 39.6 ± 13.6 years. SPTP was more common in young and middle-aged females, with a low level of tumor markers, dominated by cystic-solid tumors. The combination of CT iterative reconstruction technology and targeted scanning revealed the following: the capsule of SPTP was clear and complete, where calcifications were visible, solid components were progressively enhanced, and rare pancreatic and bile duct dilation was seen. Tumors were cystic-solid in 18 of 27 patients with SPTP, of which the solid components showed isodensity or slightly low-density, with calcifications. The solid components and cyst walls were mildly enhanced during the arterial phase, and were progressively enhanced during the parenchymal phase, portal vein phase, and delayed phase, with their enhancement degree lower than that of normal pancreatic parenchyma, and pancreatic and bile duct dilation was rare. There were no statistical differences in tumor location, morphology, growth pattern, integrity of capsule, cystic or solid, calcifications, and enhancement features between the male group and the female group (P > 0.05).ConclusionThe iterative reconstruction combined with target scanning clearly displayed the CT features of tumors, helping the diagnosis and clinical treatment of the disease.

Image Quality Improvement in Deep Learning Image Reconstruction of Head Computed Tomography Examination.

In this study, we assess image quality in computed tomography scans reconstructed via DLIR (Deep Learning Image Reconstruction) and compare it with iterative reconstruction ASIR-V (Adaptive Statistical Iterative Reconstruction) in CT (computed tomography) scans of the head. The CT scans of 109 patients were subjected to both objective and subjective evaluation of image quality. The objective evaluation was based on the SNR (signal-to-noise ratio) and CNR (contrast-to-noise ratio) of the brain's gray and white matter. The regions of interest for our study were set in the BGA (basal ganglia area) and PCF (posterior cranial fossa). Simultaneously, a subjective assessment of image quality, based on brain structure visibility, was conducted by experienced radiologists. In the assessed scans, we obtained up to a 54% increase in SNR for gray matter and a 60% increase for white matter using DLIR in comparison to ASIR-V. Moreover, we achieved a CNR increment of 58% in the BGA structures and 50% in the PCF. In the subjective assessment of the obtained images, DLIR had a mean rating score of 2.8, compared to the mean score of 2.6 for ASIR-V images. In conclusion, DLIR shows improved image quality compared to the standard iterative reconstruction of CT images of the head.

Motion-compensated scheme for sequential scanned statistical iterative dual-energy CT reconstruction

Objective. Dual-energy computed tomography (DECT) has been widely used to reconstruct numerous types of images due its ability to better discriminate tissue properties. Sequential scanning is a popular dual-energy data acquisition method as it requires no specialized hardware. However, patient motion between two sequential scans may lead to severe motion artifacts in DECT statistical iterative reconstructions (SIR) images. The objective is to reduce the motion artifacts in such reconstructions. Approach. We propose a motion-compensation scheme that incorporates a deformation vector field into any DECT SIR. The deformation vector field is estimated via the multi-modality symmetric deformable registration method. The precalculated registration mapping and its inverse or adjoint are then embedded into each iteration of the iterative DECT algorithm. Main results. Results from a simulated and clinical case show that the proposed framework is capable of reducing motion artifacts in DECT SIRs. Percentage mean square errors in regions of interest in the simulated and clinical cases were reduced from 4.6% to 0.5% and 6.8% to 0.8%, respectively. A perturbation analysis was then performed to determine errors in approximating the continuous deformation by using the deformation field and interpolation. Our findings show that errors in our method are mostly propagated through the target image and amplified by the inverse matrix of the combination of the Fisher information and Hessian of the penalty term. Significance. We have proposed a novel motion-compensation scheme to incorporate a 3D registration method into the joint statistical iterative DECT algorithm in order to reduce motion artifacts caused by inter-scan motion, and successfully demonstrate that interscan motion corrections can be integrated into the DECT SIR process, enabling accurate imaging of radiological quantities on conventional SECT scanners, without significant loss of either computational efficiency or accuracy.

CT Image Reconstruction by Boltzmann Machine for Effective Cancer Classification

Depending on technology is a surprisingly easy task for a person since the course of parameter change can be calculated intuitively by the consistency of the solution. However, manual parameter modification in many situations is varied. It becomes unworkable when specific parameters occur in a crisis. The model’s performance was evaluated using generalized data throughout the testing step. According to cross-validation studies, a 5-fold method might successfully hamper the overfitting problem. This paper aims to overcome this issue and create a system that changes its parameters automatically in the way humans do. This concept can be illustrated as an optimization-based iterative CT reconstruction model using a pixel-savvy regularisation term. A network of parameter-tuned policies maps an Image data patch to an output defining the position and amplitude of the patch center’s parameter is also setup. The PTPN is designed for a complete strengthening phase. It can be proved that replicated ct images achieve comparable quality or good performance to those reconstructed with electronic parameters under the guidance of the professional PTPN.

Visual grading assessments of model-based iterative reconstruction in abdominal CT: Can radiologists learn to adapt to the unusual image quality?

Visual grading assessments of model-based iterative reconstruction in abdominal CT: Can radiologists learn to adapt to the unusual image quality?

Iterative computed tomography reconstruction of void fraction in a fuel bundle on unstructured mesh

The experimental investigation of internal two-phase flow structures in high pressure/high temperature systems is technologically challenging and limited to a few instrumentation techniques. For complex fuel bundle geometries, the use of Computed Tomography (CT) allows for relatively high spatial resolution (0.2 to 1 mm) and non-intrusive measurement of time-averaged void fraction. One of the main issues using standard structured reconstruction mesh is due to the loss of accuracy near the test section boundaries, for instance within the pixels overlapping both the two-phase coolant and the heated rods. In this work, a void fraction reconstruction methodology based on unstructured mesh is developed to accurately bound relevant attenuation regions within the considered test section. This method eliminates any overlapping reconstruction pixels and allows straightforward applications of knowledge-based, low noise, algebraic iterative reconstruction techniques (ART, SIRT, etc.). Within the same consistent framework, CT scan forward projections can be simulated for various selected cases (single-phase, arbitrary-two-phase void fraction distributions, etc.) and reconstructed to assess the performance of both the CT scan setup (e.g., projection data sampling and number of projections) and the reconstruction technique. The method is applied to the void fraction reconstruction of gamma CT data in a 5 × 5 fuel bundle geometry measured at the Westinghouse FRIGG thermal–hydraulic test facility operating under prototypical conditions of Boiling Water Reactor. The unstructured mesh is designed to match the features of the test section (isolation, pressure vessel, water gap, fuel channel, heated rods and coolant) with high-resolution in the domain of interest (coolant) and very low-resolution in all other domains with known attenuation coefficients. As compared to previous void fraction reconstruction attempts for this data set, the new reconstruction methodology allows significant improvements in void fraction resolution and noise reduction.

Effects of Model-Based Iterative Reconstruction in Low-Dose Paranasal Computed Tomography: A Comparison with Filtered Back Projection and Hybrid Iterative Reconstruction.

Iterative reconstruction (IR) improves image quality compared with filtered back projection (FBP). This study investigated the usefulness of model-based IR (forward-projected model-based iterative reconstruction solution [FIRST]) in comparison with FBP and hybrid IR (adaptive iterative dose reduction three-dimensional processing [AIDR 3D]) in low-dose paranasal CT. Twenty-four patients with paranasal sinusitis who underwent standard-dose CT (120 kV) and low-dose CT (100 kV) scanning before and after medical treatment were enrolled. Standard-dose CT scans were reconstructed with FBP (FBP120), and low-dose CT scans with FBP (FBP100), AIDR 3D, and FIRST. The signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) in three anatomical structures and effective doses were compared using Mann-Whitney U test. Two radiologists independently evaluated the visibility of 16 anatomical structures, overall image quality, and artifacts. Effective doses in lowdose CT were significantly reduced compared with those in standard-dose CT (0.24 vs 0.43 mSv, p<0.001). FIRST achieved significantly higher SNR (p<0.01, respectively) and CNR (p<0.001, respectively) of evaluated structures and significant improvement in overall image quality (p<0.001), artifacts (p<0.001), and visibility related to muscles (p<0.05) compared to FBP120, FBP100, and AIDR 3D. FIRST allowed radiation-dose reduction, while maintaining objective and subjective image quality in low-dose paranasal CT.

Truncation effect reduction for fast iterative reconstruction in cone-beam CT

BackgroundIterative reconstruction for cone-beam computed tomography (CBCT) has been applied to improve image quality and reduce radiation dose. In a case where an object’s actual projection is larger than a flat panel detector, CBCT images contain truncated data or incomplete projections, which degrade image quality inside the field of view (FOV). In this work, we propose truncation effect reduction for fast iterative reconstruction in CBCT imaging.MethodsThe volume matrix size of the FOV and the height of projection images were extrapolated to a suitable size. These extended projections were reconstructed by fast iterative reconstruction. Moreover, a smoothing parameter for noise regularization in iterative reconstruction was modified to reduce the accumulated error while processing. The proposed work was evaluated by image quality measurements and compared with conventional filtered backprojection (FBP). To validate the proposed method, we used a head phantom for evaluation and preliminarily tested on a human dataset.ResultsIn the experimental results, the reconstructed images from the head phantom showed enhanced image quality. In addition, fast iterative reconstruction can be run continuously while maintaining a consistent mean-percentage-error value for many iterations. The contrast-to-noise ratio of the soft-tissue images was improved. Visualization of low contrast in the ventricle and soft-tissue images was much improved compared to those from FBP using the same dose index of 5 mGy.ConclusionsOur proposed method showed satisfactory performance to reduce the truncation effect, especially inside the FOV with better image quality for soft-tissue imaging. The convergence of fast iterative reconstruction tends to be stable for many iterations.

Iterative phase contrast CT reconstruction with novel tomographic operator and data-driven prior.

Breast cancer remains the most prevalent malignancy in women in many countries around the world, thus calling for better imaging technologies to improve screening and diagnosis. Grating interferometry (GI)-based phase contrast X-ray CT is a promising technique which could make the transition to clinical practice and improve breast cancer diagnosis by combining the high three-dimensional resolution of conventional CT with higher soft-tissue contrast. Unfortunately though, obtaining high-quality images is challenging. Grating fabrication defects and photon starvation lead to high noise amplitudes in the measured data. Moreover, the highly ill-conditioned differential nature of the GI-CT forward operator renders the inversion from corrupted data even more cumbersome. In this paper, we propose a novel regularized iterative reconstruction algorithm with an improved tomographic operator and a powerful data-driven regularizer to tackle this challenging inverse problem. Our algorithm combines the L-BFGS optimization scheme with a data-driven prior parameterized by a deep neural network. Importantly, we propose a novel regularization strategy to ensure that the trained network is non-expansive, which is critical for the convergence and stability analysis we provide. We empirically show that the proposed method achieves high quality images, both on simulated data as well as on real measurements.

Fast algorithm for Joseph’s forward projection in iterative computed tomography reconstruction

Fast algorithm for Joseph’s forward projection in iterative computed tomography reconstruction

Low-dose CT iterative reconstruction based on image block classification and dictionary learning

Low-dose CT iterative reconstruction based on image block classification and dictionary learning

Deep learning reconstruction allows low-dose imaging while maintaining image quality: comparison of deep learning reconstruction and hybrid iterative reconstruction in contrast-enhanced abdominal CT.

We aimed to compare the radiation dose and image quality of a low-dose abdominal computed tomography (CT) protocol reconstructed with deep learning reconstruction (DLR) with those of a routine-dose protocol reconstructed with hybrid-iterative reconstruction. This retrospective study enrolled 71 patients [61 men; average age, 71.9 years; mean body mass index (BMI), 24.3 kg/m2] who underwent both low-dose abdominal CT with DLR [advanced intelligent clear-IQ engine (AiCE)] and routine-dose abdominal CT with hybrid-iterative reconstruction [adaptive iterative dose reduction 3D (AIDR 3D)]. Radiation dose parameters included volume CT dose index (CTDIvol), effective dose (ED), and size-specific dose estimate (SSDE). Mean image noise and contrast-to-noise ratio (CNR) were calculated. Image noise was measured in the hepatic parenchyma and bilateral erector spinae muscles. Moreover, subjective assessment of perceived image quality and diagnostic acceptability was performed. The low-dose protocol helped reduce the CTDIvol by 44.3%, ED by 43.7%, and SSDE by 44.9%. Moreover, the noise was significantly lower and CNR significantly higher with the low-dose protocol than with the normal-dose protocol (P<0.001). In the subjective assessment of image quality, there was no significant difference between the protocols with regard to image noise. Overall, AiCE was superior to AIDR 3D in terms of diagnostic acceptability (P=0.001). The use of AiCE can reduce overall radiation dose by more than 40% without loss of image quality compared to routine-dose abdominal CT with AIDR 3D.

Ultralow-dose CT of the petrous bone using iterative reconstruction technique, tin filter and high resolution detectors allows an adequate assessment of the petrous bone structures

ObjectivesTo assess image quality and radiation dose in computed tomography (CT) studies of the petrous bone done with a scanner using a tin filter, high-resolution detectors, and iterative reconstruction, and to compare versus in studies done with another scanner without a tin filter using filtered back projection reconstruction. Material and methodsThirty two patients (group 1) were acquired with an ultra-low dose CT (32-MDCT, 130kV, tin filter and iterative reconstruction). Images and radiation doses were compared to 36 patients (group 2) acquired in a 16-MDCT (120kV and filtered back-projection). Muscle density, bone density, and background noise were measured. Signal-to-noise ratio (SNR) was calculated. To assess image quality, two independent radiologists subjectively evaluated the visualization of the different structures of the middle and inner ear (0=not visualized, 3=perfectly identified and delimited). Interobserver agreement was calculated. Effective dose at different anatomical levels with the dose-length product was recorded. ResultsIn the quantitative analysis, there were no significant differences in image noise between the two groups. In the qualitative analysis, a similar or slightly lower subjective score was obtained in the delimitation of different structures of the ossicular chain and cochlea in the 32-MDCT, compared to 16-MDCT, with statistically significant differences. Mean effective dose (±standard deviation) was 0.16±0.04mSv for the 32-MDCT and 1.25±0.30mSv for the 16-MDCT. ConclusionsThe use of scanners with tin filters, high-resolution detectors, and iterative reconstruction allows to obtain images with adequate quality for the evaluation of the petrous bone structures with ultralow doses of radiation (0.16±0.04mSv).

Analytical covariance estimation for iterative CT reconstruction methods

Covariance of reconstruction images are useful to analyze the magnitude and correlation of noise in the evaluation of systems and reconstruction algorithms. The covariance estimation requires a big number of image samples that are hard to acquire in reality. A covariance propagation method from projection by a few noisy realizations is studied in this work. Based on the property of convergent points of cost funtions, the proposed method is composed of three steps, (1) construct a relationship between the covariance of projection and corresponding reconstruction from cost functions at its convergent point, (2) simplify the covariance relationship constructed in (1) by introducing an approximate gradient of penalties, and (3) obtain an analytical covariance estimation according to the simplified relationship in (2). Three approximation methods for step (2) are studied: the linear approximation of the gradient of penalties (LAM), the Taylor apprximation (TAM), and the mixture of LAM and TAM (MAM). TV and qGGMRF penalized weighted least square methods are experimented on. Results from statistical methods are used as reference. Under the condition of unstable 2nd derivative of penalties such as TV, the covariance image estimated by LAM accords to reference well but of smaller values, while the covarianc estimation by TAM is quite off. Under the conditon of relatively stable 2nd derivative of penalties such as qGGMRF, TAM performs well and LAM is again with a negative bias in magnitude. MAM gives a best performance under both conditions by combining LAM and TAM. Results also show that only one noise realization is enough to obtain reasonable covariance estimation analytically, which is important for practical usage. This work suggests the necessity and a new way to estimate the covariance for non-quadratically penalized reconstructions. Currently, the proposed method is computationally expensive for large size reconstructions.Computational efficiency is our future work to focus.