ct重建算法，投影图重建残差与FDK结果一起输入网络去伪影

Objective. This study aims to propose a dual-domain network that not only reduces scatter artifacts but also retains structure details in cone-beam computed tomography (CBCT). Approach. The proposed network comprises a projection-domain sub-network and an image-domain sub-network. The projection-domain sub-network utilizes a division residual network to amplify the difference between scatter signals and imaging signals, facilitating the learning of scatter signals. The image-domain sub-network contains dual encoders and a single decoder. The dual encoders extract features from two inputs parallelly, and the decoder fuses the extracted features from the two encoders and maps the fused features back to the final high-quality image. Of the two input images to the image-domain sub-network, one is the scatter-contaminated image analytically reconstructed from the scatter-contaminated projections, and the other is the pre-processed image reconstructed from the pre-processed projections produced by the projection-domain sub-network. Main results. Experimental results on both synthetic and real data demonstrate that our method can effectively reduce scatter artifacts and restore image details. Quantitative analysis using synthetic data shows the mean absolute error was reduced by 74% and peak signal-to-noise ratio increased by 57% compared to the scatter-contaminated ones. Testing on real data found a 38% increase in contrast-to-noise ratio with our method compared to the scatter-contaminated image. Additionally, our method consistently outperforms comparative methods such as U-Net, DSE-Net, deep residual convolution neural network (DRCNN) and the collimator-based method. Significance. A dual-domain network that leverages projection-domain division residual connection and image-domain feature fusion has been proposed for CBCT scatter correction. It has potential applications for reducing scatter artifacts and preserving image details in CBCT.

No abstract available

Cone-beam computed tomography (CBCT) is a widely used imaging technique. In practical applications, reducing projection views can decrease radiation exposure and accelerate scanning speed, with potential benefits for stationary CT systems. However, ultra-sparse-view acquisition (e.g., ≤ 30 views) introduces severe streak artifacts that degrade image quality. This poses a critical challenge because it is difficult to differentiate artifacts from real structures. We propose a sliding volume-based streak artifact reduction network (S-STAR Net) to remove artifacts while preserving structural details. Our method introduces three key technical innovations: (1) A sliding sampling sub-volume approach to process 3D sub-volumes, fully leveraging spatial context. (2) A difference enhancement (DE) loss to help separate artifacts from real structures. (3) Novel network includes volume-attention aided residual (VAR) blocks and Fourier transform convolution (FTC) blocks for multi-domain feature learning. Evaluated under 30 projection views, the method was tested on two datasets: a walnut dataset and the CQ500 head CT dataset. Quantitative metrics (PSNR/SSIM) and qualitative assessments (multi-planar visualization, residual error maps) demonstrate that S-STAR Net achieves superior performance in both artifacts suppression and detail preservation compared to existing approaches. The proposed method effectively addresses streak artifacts and recovers subtle structures in ultra-sparse-view CBCT reconstruction. Its robustness suggests broad applicability for medical image denoising, artifact reduction, and 3D image enhancement tasks.

Abstract This study aims to address and test a new residual learning algorithm in neural network applied to the projection data to generate high qualified imaging by reducing the streaking artifacts in cone-beam computed tomography (CBCT). Since the streaking artifacts have a large relationship with the noise on the projection, a residual objective upon Poisson noise corresponding to the image was proposed. As the prior, the convolution neural network (CNN) was constructed to residual learning based on the simulated label and exploited to eliminate the artifacts in the slice. To illustrate the robustness and applicability of CNN, the proposed method is evaluated using CBCT images. For the simulated projection, the PSNR and SSIM of the proposed method were dramatically increased by 15.4% and 85.9% of that with raw projection; for the true projection, the PSNR and SSIM were increased by 14.9% and 56.2%, respectively. Study results show effective results, and the proposed method is practical and attractive as a preferred solution to CT streaking artifacts suppression.

Cone-beam computed tomography (CBCT) is generally reconstructed with hundreds of two-dimensional X-Ray projections through the FDK algorithm, and its excessive ionizing radiation of X-Ray may impair patients' health. Two common dose-reduction strategies are to either lower the intensity of X-Ray, i.e., low-intensity CBCT, or reduce the number of projections, i.e., sparse-view CBCT. Existing efforts improve the low-dose CBCT images only under a single dose-reduction strategy. In this paper, we argue that applying the two strategies simultaneously can reduce dose in a gentle manner and avoid the extreme degradation of the projection data in a single dose-reduction strategy, especially under ultra-low-dose situations. Therefore, we develop a Joint Denoising and Interpolating Network (JDINet) in projection domain to improve the CBCT quality with the hybrid low-intensity and sparse-view projections. Specifically, JDINet mainly includes two important components, i.e., denoising module and interpolating module, to respectively suppress the noise caused by the low-intensity strategy and interpolate the missing projections caused by the sparse-view strategy. Because FDK actually utilizes the projection information after ramp-filtering, we develop a filtered structural similarity constraint to help JDINet focus on the reconstruction-required information. Afterward, we employ a Postprocessing Network (PostNet) in the reconstruction domain to refine the CBCT images that are reconstructed with denoised and interpolated projections. In general, a complete CBCT reconstruction framework is built with JDINet, FDK, and PostNet. Experiments demonstrate that our framework decreases RMSE by approximately 8 %, 15 %, and 17 %, respectively, on the 1/8, 1/16, and 1/32 dose data, compared to the latest methods. In conclusion, our learning-based framework can be deeply imbedded into the CBCT systems to promote the development of CBCT. Source code is available at https://github.com/LianyingChao/FusionLowDoseCBCT.

Extending cone-beam CT (CBCT) use toward dose accumulation and adaptive radiotherapy (ART) necessitates more accurate HU reproduction since cone-beam geometries are heavily degraded by photon scatter. This study proposes a novel method which aims to demonstrate how deep learning based on phantom data can be used effectively for CBCT intensity correction in patient images. Four anthropomorphic phantoms were scanned on a CBCT and conventional fan-beam CT system. Intensity correction is performed by estimating the cone-beam intensity deviations from prior information contained in the CT. Residual projections were extracted by subtraction of raw cone-beam projections from virtual CT projections. An improved version of U-net is utilized to train on a total of 2001 projection pairs. Once trained, the network could estimate intensity deviations from input patient head and neck raw projections. The results from our novel method showed that corrected CBCT images improved the (contrast-to-noise ratio) with respect to uncorrected reconstructions by a factor of 2.08. The mean absolute error and structural similarity index improved from 318 HU to 74 HU and 0.750 to 0.812 respectively. Visual assessment based on line-profile measurements and difference image analysis indicate the proposed method reduced noise and the presence of beam-hardening artefacts compared to uncorrected and manufacturer reconstructions. Projection domain intensity correction for cone-beam acquisitions of patients was shown to be feasible using a convolutional neural network trained on phantom data. The method shows promise for further improvements which may eventually facilitate dose monitoring and ART in the clinical radiotherapy workflow.

PURPOSE Sparse-data CT frequently occurs, such as breast tomosynthesis, C-arm CT, on-board 4D CBCT, and industrial CT. However, sparse-data image reconstruction remains challenging due to highly undersampled data. This work develops a data-driven image reconstruction method for sparse-data CT using deep neural networks (DNN). METHODS The new method so-called AirNet is designed to incorporate the benefits from analytical reconstruction method (AR), iterative reconstruction method (IR), and DNN. It is built upon fused analytical and iterative reconstruction (AIR) that synergizes AR and IR via the optimization framework of modified proximal forward-backward splitting (PFBS). By unrolling PFBS into IR updates of CT data fidelity and DNN regularization with residual learning, AirNet utilizes AR such as FBP during the data fidelity, introduces dense connectivity into DNN regularization, and learns PFBS coefficients and DNN parameters that minimize the loss function during the training stage. And then AirNet with trained parameters can be used for end-to-end image reconstruction. RESULTS A CT atlas of 100 prostate scans was used to validate the AirNet in comparison with state-of-art DNN-based post-processing and image reconstruction methods. The validation loss in AirNet had the fastest decreasing rate, owing to inherited fast convergence from AIR. AirNet was robust to noise in projection data and content differences between the training set and the images to be reconstructed. The impact of image quality on radiotherapy treatment planning was evaluated for both photon and proton therapy, and AirNet achieved the best treatment plan quality, especially for proton therapy. For example, with limited-angle data, the maximal target dose for AirNet was 109.5% in comparison with the ground truth 109.1%, while it was significantly elevated to 115.1% and 128.1% for FBPConvNet and LEARN respectively. CONCLUSIONS A new image reconstruction AirNet is developed for sparse-data CT image reconstruction. AirNet achieved the best image reconstruction quality both visually and quantitatively among all methods under comparison for all sparse-data scenarios (sparse-view and limited-angle), and provided the best photon and proton treatment plan quality based on sparse-data CT.

Objective. The quantitative and routine imaging capabilities of cone-beam CT (CBCT) are hindered from clinical applications due to the severe shading artifacts of scatter contamination. The scatter correction methods proposed in the literature only consider the anatomy of the scanned objects while disregarding the impact of incident x-ray energy spectra. The multiple-spectral model is in urgent need for CBCT scatter estimation. Approach. In this work, we incorporate the multiple spectral diagnostic multidetector CT labels into the pixel-to-pixel (Pix2pix) GAN to estimate accurate scatter distributions from CBCT projections acquired at various imaging volume sizes and x-ray energy spectra. The Pix2pix GAN combines the residual network as the generator and the PatchGAN as the discriminator to construct the correspondence between the scatter-contaminated projection and scatter distribution. The network architectures and loss function of Pix2pix GAN are optimized to achieve the best performance on projection-to-scatter transition. Results. The CBCT data of a head phantom and abdominal patients are applied to test the performance of the proposed method. The error of the corrected CBCT image using the proposed method is reduced from over 200 HU to be around 20 HU in both phantom and patient studies. The mean structural similarity index of the CT image is improved from 0.2 to around 0.9 after scatter correction using the proposed method compared with the MC-simulation method, which indicates a high similarity of the anatomy in the images before and after the proposed correction. The proposed method achieves higher accuracy of scatter estimation than using the Pix2pix GAN with the U-net generator. Significance. The proposed scheme is an effective solution to the multiple spectral CBCT scatter correction. The scatter-correction software using the proposed model will be available at: https://github.com/YangkangJiang/Cone-beam-CT-scatter-correction-tool.

PURPOSE Cone-beam computed tomography (CBCT) is widely utilized in modern radiotherapy; however, CBCT images exhibit increased scatter artifacts compared to planning CT (pCT), compromising image quality and limiting further applications. Scatter correction is thus crucial for improving CBCT image quality. METHODS In this study, we proposed an unsupervised contrastive learning method for CBCT scatter correction. Initially, we transformed low-quality CBCT into high-quality synthetic pCT (spCT) and generated forward projections of CBCT and spCT. By computing the difference between these projections, we obtained a residual image containing image details and scatter artifacts. Image details primarily comprise high-frequency signals, while scatter artifacts consist mainly of low-frequency signals. We extracted the scatter projection signal by applying a low-pass filter to remove image details. The corrected CBCT (cCBCT) projection signal was obtained by subtracting the scatter artifacts projection signal from the original CBCT projection. Finally, we employed the FDK reconstruction algorithm to generate the cCBCT image. RESULTS To evaluate cCBCT image quality, we aligned the CBCT and pCT of six patients. In comparison to CBCT, cCBCT maintains anatomical consistency and significantly enhances CT number, spatial homogeneity, and artifact suppression. The mean absolute error (MAE) of the test data decreased from 88.0623 ± 26.6700 HU to 17.5086 ± 3.1785 HU. The MAE of fat regions of interest (ROIs) declined from 370.2980 ± 64.9730 HU to 8.5149 ± 1.8265 HU, and the error between their maximum and minimum CT numbers decreased from 572.7528 HU to 132.4648 HU. The MAE of muscle ROIs reduced from 354.7689 ± 25.0139 HU to 16.4475 ± 3.6812 HU. We also compared our proposed method with several conventional unsupervised synthetic image generation techniques, demonstrating superior performance. CONCLUSIONS Our approach effectively enhances CBCT image quality and shows promising potential for future clinical adoption.

BACKGROUND Image-guided neurosurgery requires high localization and registration accuracy to enable effective treatment and avoid complications. However, accurate neuronavigation based on preoperative MR or CT images is challenged by brain deformation occurring during the surgical intervention. PURPOSE To facilitate intraoperative visualization of brain tissues and deformable registration with preoperative images, a 3D deep learning reconstruction framework (termed DL-Recon) was proposed for improved intraoperative cone-beam CT (CBCT) image quality. METHODS The DL-Recon framework combines physics-based models with deep learning CT synthesis and leverages uncertainty information to promote robustness to unseen features. A 3D generative adversarial network (GAN) with a conditional loss function modulated by aleatoric uncertainty was developed for CBCT-to-CT synthesis. Epistemic uncertainty of the synthesis model was estimated via Monte Carlo dropout. Using spatially varying weights derived from epistemic uncertainty, the DL-Recon image combines the synthetic CT with an artifact-corrected filtered back-projection (FBP) reconstruction. In regions of high epistemic uncertainty, DL-Recon includes greater contribution from the FBP image. Twenty paired real CT and simulated CBCT images of the head were used for network training and validation, and experiments evaluated the performance of DL-Recon on CBCT images containing simulated and real brain lesions not present in the training data. Performance among learning- and physics-based methods was quantified in terms of structural similarity (SSIM) of the resulting image to diagnostic CT and Dice similarity metric (DSC) in lesion segmentation compared to ground truth. A pilot study was conducted involving seven subjects with CBCT images acquired during neurosurgery to assess the feasibility of DL-Recon in clinical data. RESULTS CBCT images reconstructed via FBP with physics-based corrections exhibited the usual challenges to soft-tissue contrast resolution due to image non-uniformity, noise, and residual artifacts. GAN synthesis improved image uniformity and soft-tissue visibility but was subject to error in the shape and contrast of simulated lesions that were unseen in training. Incorporation of aleatoric uncertainty in synthesis loss improved estimation of epistemic uncertainty, with variable brain structures and unseen lesions exhibiting higher epistemic uncertainty. The DL-Recon approach mitigated synthesis errors while maintaining improvement in image quality, yielding 15-22% increase in SSIM (image appearance compared to diagnostic CT) and up to 25% increase in DSC in lesion segmentation compared to FBP. Clear gains in visual image quality were also observed in real brain lesions and in clinical CBCT images. CONCLUSIONS DL-Recon leveraged uncertainty estimation to combine the strengths of deep learning and physics-based reconstruction and demonstrated substantial improvements in the accuracy and quality of intraoperative CBCT. The improved soft-tissue contrast resolution could facilitate visualization of brain structures and support deformable registration with preoperative images, further extending the utility of intraoperative CBCT in image-guided neurosurgery. This article is protected by copyright. All rights reserved.

BACKGROUND Cone beam computed tomography (CBCT) plays an important role in many medical fields nowadays. Unfortunately, the potential of this imaging modality is hampered by lower image quality compared to the conventional CT, and producing accurate reconstructions remains challenging. A lot of recent research has been directed towards reconstruction methods relying on deep learning, which have shown great promise for various imaging modalities. However, practical application of deep learning to CBCT reconstruction is complicated by several issues, such as exceedingly high memory costs of deep learning methods when working with fully 3D data. Additionally, deep learning methods proposed in the literature are often trained and evaluated only on data from a specific region of interest, thus raising concerns about possible lack of generalization to other regions. PURPOSE In this work, we aim to address these limitations and propose LIRE: a learned invertible primal-dual iterative scheme for CBCT reconstruction. METHODS LIRE is a learned invertible primal-dual iterative scheme for CBCT reconstruction, wherein we employ a U-Net architecture in each primal block and a residual convolutional neural network (CNN) architecture in each dual block. Memory requirements of the network are substantially reduced while preserving its expressive power through a combination of invertible residual primal-dual blocks and patch-wise computations inside each of the blocks during both forward and backward pass. These techniques enable us to train on data with isotropic 2 mm voxel spacing, clinically-relevant projection count and detector panel resolution on current hardware with 24 GB video random access memory (VRAM). RESULTS Two LIRE models for small and for large field-of-view (FoV) setting were trained and validated on a set of 260 + 22 thorax CT scans and tested using a set of 142 thorax CT scans plus an out-of-distribution dataset of 79 head and neck CT scans. For both settings, our method surpasses the classical methods and the deep learning baselines on both test sets. On the thorax CT set, our method achieves peak signal-to-noise ratio (PSNR) of 33.84 ± 2.28 for the small FoV setting and 35.14 ± 2.69 for the large FoV setting; U-Net baseline achieves PSNR of 33.08 ± 1.75 and 34.29 ± 2.71 respectively. On the head and neck CT set, our method achieves PSNR of 39.35 ± 1.75 for the small FoV setting and 41.21 ± 1.41 for the large FoV setting; U-Net baseline achieves PSNR of 33.08 ± 1.75 and 34.29 ± 2.71 respectively. Additionally, we demonstrate that LIRE can be finetuned to reconstruct high-resolution CBCT data with the same geometry but 1 mm voxel spacing and higher detector panel resolution, where it outperforms the U-Net baseline as well. CONCLUSIONS Learned invertible primal-dual schemes with additional memory optimizations can be trained to reconstruct CBCT volumes directly from the projection data with clinically-relevant geometry and resolution. Such methods can offer better reconstruction quality and generalization compared to classical deep learning baselines.

Conebeam CT using a circular trajectory is quite often used for various applications due to its relative simple geometry. For conebeam geometry, Feldkamp, Davis and Kress algorithm is regarded as the standard reconstruction method, but this algorithm suffers from so-called conebeam artifacts as the cone angle increases. Various model-based iterative reconstruction methods have been developed to reduce the cone-beam artifacts, but these algorithms usually require multiple applications of computational expensive forward and backprojections. In this paper, we develop a novel deep learning approach for accurate conebeam artifact removal. In particular, our deep network, designed on the differentiated backprojection domain, performs a data-driven inversion of an ill-posed deconvolution problem associated with the Hilbert transform. The reconstruction results along the coronal and sagittal directions are then combined using a spectral blending technique to minimize the spectral leakage. Experimental results under various conditions confirmed that our method generalizes well and outperforms the existing iterative methods despite significantly reduced runtime complexity.

Photon-counting CT has gained significant attention in recent years; however, publicly available datasets for spectral reconstruction and deep learning training remain limited. Consequently, many image process algorithms and deep learning models are developed and validated using simulated rather than real spectral CT data. To address this gap, we present a cone-beam photon-counting CT (PCCT) dataset acquired using a custom-built micro-PCCT system and 15 walnut samples. Each walnut was scanned from four bed positions under dual energy thresholds (15 keV and 30 keV), resulting in a total of 172,800 raw projection images with a resolution of 2063 × 505 pixels. The dataset provides full access to raw multi-energy projections, system parameters, calibration tables, calibration phantom raw projection data and reconstruction code, enabling comprehensive spectral CT studies including spectral CT reconstruction, material decomposition, artifact correction, and deep learning-based methods. It addresses the scarcity of real PCCT datasets for developing and validating data-driven approaches and aims to foster fair and reproducible comparisons across spectral CT image process algorithms.

Sparse-view cone beam CT (CBCT) is crucial for lowering radiation dose in repeated scanning applications such as image-guided radiotherapy. However, image reconstructing with fewer projections is a challenge. In this work, a deep reconstruction framework is proposed for super sparse CBCT with ten projections based on projection and image domains, in which the CBCT geometric information is fully integrated in dual domains to enhance the details and reduce the reconstruction time. The projection network includes the feature extraction module of projections, feature fusion module across projections and image generation network. It is designed for extracting deep features from CBCT projections by introducing attention mechanisms from projection views and transformer. The image network includes the backprojection operator and image feature blocks based on 3D U-net, which fully integrates the projection geometry for transferring the projection domain to image domain and to maintain the spatial features. The experimental results with the LIDC-IDRI dataset show that this method is beneficial for improving the peak signal to noise ratio (PSNR) and structural similarity (SSIM) Index of reconstructed images, while also having low time cost.

No abstract available

X-ray scatter presents a significant challenge in computed tomography (CT) by introducing artifacts and diminishing contrast, ultimately affecting measurement accuracy in reconstructed images. While Monte Carlo (MC) simulation is the gold standard for scatter correction, its high computational cost limits clinical applications. Accelerated MC methods improve efficiency but may compromise accuracy. Scatter kernel-based approaches allow quick implementation but lack precision in measurement. Deep learning-based methods show promising results but require thorough evaluation for generalizability across diverse imaging protocols. This study presents a hybrid scatter correction method for cone-beam CT (CBCT) that integrates fast MC simulation with a deep neural network, aiming to enhance measurement precision and efficiency. Fast MC simulations are employed for low-dose, sparse-view scatter estimations, serving as references for missing-view scatter predictions. The network inputs and outputs are designed as differences relative to the reference signal, improving the method’s generalizability across varying imaging conditions. The proposed approach is validated through numerical simulations and physical experiments. Results demonstrate that the method achieves scatter predictions and CT reconstructions closer to ground truth, offering measurement accuracy comparable to standard MC simulations. Compared with scatter kernel-based methods and other deep learning-based approaches, the proposed method exhibits superior accuracy and robustness across diverse objects and system geometries, making it a reliable solution for accurate scatter measurement in clinical settings.

No abstract available

X-ray cone-beam computed tomography (CBCT) is a powerful tool for nondestructive testing and evaluation, yet the CT image quality can be compromised by artifact due to X-ray scattering within dense materials such as metals. This problem leads to the need for hardware- and software-based scatter artifact correction to enhance the image quality. Recently, deep learning techniques have merged as a promising approach to obtain scatter-free images efficiently. However, these deep learning techniques rely heavily on training data, often gathered through simulation. Simulated CT images, unfortunately, do not accurately reproduce the real properties of objects, and physically accurate X-ray simulation still requires significant computation time, hindering the collection of a large number of CT images. To address these problems, we propose a deep learning framework for scatter artifact correction using projections obtained solely by real CT scanning. To this end, we utilize a beam-hole array (BHA) to block the X-rays deviating from the primary beam path, thereby capturing scatter-free X-ray intensity at certain detector pixels. As the BHA shadows a large portion of detector pixels, we incorporate several regularization losses to enhance the training process. Furthermore, we introduce radiographic data augmentation to mitigate the need for long scanning time, which is a concern as CT devices equipped with BHA require two series of CT scans. Experimental validation showed that the proposed framework outperforms a baseline method that learns simulated projections where the entire image is visible and does not contain scattering artifacts.

Computed tomography (CT) provides non-invasive anatomical structures of the human body and is also widely used for clinical diagnosis, but excessive ionizing radiation in X-rays can cause harm to the human body. Therefore, the researchers obtained sparse sinograms reconstructed sparse view CT images (SVCT) by reducing the amount of X-ray projection, thereby reducing the radiological effects caused by radiation. This paper proposes a cascade-based dual-domain data correction network (CDDCN), which can effectively combine the complementary information contained in the sinogram domain and the image domain to reconstruct high-quality CT images from sparse view sinograms. Specifically, several encoder-decoder subnets are cascaded in the sinogram domain to reconstruct artifact-free and noise-free CT images. In the encoder-decoder subnets, spatial-channel domain learning is designed to achieve efficient feature fusion through a group merging structure, providing continuous and elaborate pixel-level features and improving feature extraction efficiency. At the same time, to ensure that the original sinogram data collected can be retained, a sinogram data consistency layer is proposed to ensure the fidelity of the sinogram data. To further maintain the consistency between the reconstructed image and the reference image, a multi-level composite loss function is designed for regularization to compensate for excessive smoothing and distortion of the image caused by pixel loss and preserve image details and texture. Quantitative and qualitative analysis shows that CDDCN achieves competitive results in artifact removal, edge preservation, detail restoration, and visual improvement for sparsely sampled data under different views.

Sparse-view computed tomography (SVCT) is regarded as a promising technique to accelerate data acquisition and reduce radiation dose. However, in the presence of metallic implants, SVCT inevitably makes the reconstructed CT images suffer from severe metal artifacts and streaking artifacts due to the lack of sufficient projection data. Previous stand-alone SVCT and metal artifact reduction (MAR) methods to solve the problem of simultaneously sparse-view and metal artifact reduction (SVMAR) are plagued by insufficient correction accuracy. To overcome this limitation, we propose a multi-domain deep unrolling network, called Mud-Net, for SVMAR. Specifically, we establish a joint sinogram, image, artifact, and coding domains deep unrolling reconstruction model to recover high-quality CT images from the under-sampled sinograms corrupted by metallic implants. To train this multi-domain network effectively, we embed multi-domain knowledge into the network training process. Comprehensive experiments demonstrate that our method is superior to both existing MAR methods in the full-view MAR task and previous SVCT methods in the SVMAR task.

Sparse-view computed tomography (CT) reduces radiation exposure by subsampling projection views, but conventional reconstruction methods produce severe streak artifacts with undersampled data. While deep-learning-based methods enable single-step artifact suppression, they often produce over-smoothed results under significant sparsity. Though diffusion models improve reconstruction via iterative refinement and generative priors, they require hundreds of sampling steps and struggle with stability in highly sparse regimes. To tackle these concerns, we present the Cross-view Generalized Diffusion Model (CvG-Diff), which reformulates sparse-view CT reconstruction as a generalized diffusion process. Unlike existing diffusion approaches that rely on stochastic Gaussian degradation, CvG-Diff explicitly models image-domain artifacts caused by angular subsampling as a deterministic degradation operator, leveraging correlations across sparse-view CT at different sample rates. To address the inherent artifact propagation and inefficiency of sequential sampling in generalized diffusion model, we introduce two innovations: Error-Propagating Composite Training (EPCT), which facilitates identifying error-prone regions and suppresses propagated artifacts, and Semantic-Prioritized Dual-Phase Sampling (SPDPS), an adaptive strategy that prioritizes semantic correctness before detail refinement. Together, these innovations enable CvG-Diff to achieve high-quality reconstructions with minimal iterations, achieving 38.34 dB PSNR and 0.9518 SSIM for 18-view CT using only \textbf{10} steps on AAPM-LDCT dataset. Extensive experiments demonstrate the superiority of CvG-Diff over state-of-the-art sparse-view CT reconstruction methods. The code is available at https://github.com/xmed-lab/CvG-Diff.

Sparse-view CT enables reduction of the dose but produces severe streaking artifacts and frequencybiased artifacts, which reduces this diagnosis. This document presents a proposal of a self-supervised framework approach to artifact suppression via frequency sensitivity refers to generative models, which requires no reference images of different views (only sparse-view training is given). The algorithm splits projections and reconstruction images into complementary frequency bands, which guides a generative network in selective attenuation of artifact-dominant bands at the cost of anatomically relevant structures. A consistency-based selfsupervising method applies forward and inverse CT physics to apply data fidelity by frequency band to stabilize training and avoid over-smoothing. The adversarial and perceptual loss is tuned to the frequency-awareness that facilitates the recovery and maintenance of natural textures and edges respectively. Broad experiments involving simulated and real-life sparseview CT data indicate that significant factors of artifact or structural regularity and clarity of diagnosis are observed when compared to the current supervised and unmated approaches. The proposed framework is a combination of over sampling pattern and level of noise, and this is an effective solution to low dose CT reconstruction as practiced in clinical practice.

No abstract available

Objective. X-ray computed tomography employing sparse projection views has emerged as a contemporary technique to mitigate radiation dose. However, due to the inadequate number of projection views, an analytic reconstruction method utilizing filtered backprojection results in severe streaking artifacts. Recently, deep learning (DL) strategies employing image-domain networks have demonstrated remarkable performance in eliminating the streaking artifact caused by analytic reconstruction methods with sparse projection views. Nevertheless, it is difficult to clarify the theoretical justification for applying DL to sparse view computed tomography (CT) reconstruction, and it has been understood as restoration by removing image artifacts, not reconstruction. Approach. By leveraging the theory of deep convolutional framelets (DCF) and the hierarchical decomposition of measurement, this research reveals the constraints of conventional image and projection-domain DL methodologies, subsequently, the research proposes a novel dual-domain DL framework utilizing hierarchical decomposed measurements. Specifically, the research elucidates how the performance of the projection-domain network can be enhanced through a low-rank property of DCF and a bowtie support of hierarchical decomposed measurement in the Fourier domain. Main results. This study demonstrated performance improvement of the proposed framework based on the low-rank property, resulting in superior reconstruction performance compared to conventional analytic and DL methods. Significance. By providing a theoretically justified DL approach for sparse-view CT reconstruction, this study not only offers a superior alternative to existing methods but also opens new avenues for research in medical imaging. It highlights the potential of dual-domain DL frameworks to achieve high-quality reconstructions with lower radiation doses, thereby advancing the field towards safer and more efficient diagnostic techniques. The code is available at https://github.com/hanyoseob/HDD-DL-for-SVCT.

Tomographic image reconstruction with deep learning is an emerging field of applied artificial intelligence. Reducing radiation dose with sparse views’ reconstruction is a significant task in cardiac imaging. Many efforts are contributing to sparse-view tomography imaging, but it is still a challenge for achieving good images from high sparse-view level, such as 60 views. In this study, we proposed a Deep Embedding-Attention-Refinement (DEAR) network to fundamentally address this challenge. DEAR consists of three modules including deep embedding, deep attention, and deep refinement. The measurement is extended by deep embedding network to generate artifact-reduction images. Then, the deep attention network is employed to remove sparse-view artifacts and correct wrong details introduced by deep embedding network. Finally, the deep refinement module is used to refine finer image features and structures. The results on clinical datasets demonstrate the efficiency of our proposed DEAR in edge preservation and feature recovery.

Sparse-view computed tomography (CT) is a promising solution for expediting the scanning process and mitigating radiation exposure to patients, the reconstructed images, however, contain severe streak artifacts, compromising subsequent screening and diagnosis. Recently, deep learning-based image post-processing methods along with their dual-domain counterparts have shown promising results. However, existing methods usually produce over-smoothed images with loss of details due to (1) the difficulty in accurately modeling the artifact patterns in the image domain, and (2) the equal treatment of each pixel in the loss function. To address these issues, we concentrate on the image post-processing and propose a simple yet effective FREquency-band-awarE and SElf-guidED network, termed FreeSeed, which can effectively remove artifact and recover missing detail from the contaminated sparse-view CT images. Specifically, we first propose a frequency-band-aware artifact modeling network (FreeNet), which learns artifact-related frequency-band attention in Fourier domain for better modeling the globally distributed streak artifact on the sparse-view CT images. We then introduce a self-guided artifact refinement network (SeedNet), which leverages the predicted artifact to assist FreeNet in continuing to refine the severely corrupted details. Extensive experiments demonstrate the superior performance of FreeSeed and its dual-domain counterpart over the state-of-the-art sparse-view CT reconstruction methods. Source code is made available at https://github.com/Masaaki-75/freeseed.

Sparse-view CT imaging can reduce radiation dose or shorten scanning time by reducing the number of projection views, thus methods for high-quality image reconstruction from under-sampled sinograms are actively sought after in the CT field. In this work, we present a BP-tensor domain fusion network (TensorFusion-Net) that can produce high-quality images from sparse-view sinograms using a cross-domain deep reconstruction framework. TensorFusion-Net holistically fuses the sparse-view BP-tensor derived from the sparse-view sinogram and the full-view BP-tensor calculated from the output image of a prior network (PriorNet). Furthermore, output from TensorFusion-Net can be used to update the full-view BP-tensor, which is fused again with the sparse-view BP-tensor iteratively. Our experimental results on AAPM Mayo datasets indicate that our algorithm performs better than several existing algorithms in artifact reduction and image detail preservation.

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 multi-view 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 can not be collected. This article is protected by copyright. All rights reserved.

Sparse-view CT reconstruction is a challenging ill-posed inverse problem, where insufficient projection data leads to degraded image quality with increased noise and artifacts. Recent deep learning approaches have shown promising results in CT reconstruction. However, existing methods often neglect projection data constraints and rely heavily on convolutional neural networks, resulting in limited feature extraction capabilities and inadequate adaptability. To address these limitations, we propose a Dual-domain deep Prior-guided Multi-scale fusion Attention (DPMA) model for sparse-view CT reconstruction, aiming to enhance reconstruction accuracy while ensuring data consistency and stability. First, we establish a residual regularization strategy that applies constraints on the difference between the prior image and target image, effectively integrating deep learning-based priors with model-based optimization. Second, we develop a multi-scale fusion attention mechanism that employs parallel pathways to simultaneously model global context, regional dependencies, and local details in a unified framework. Third, we incorporate a physics-informed consistency module based on range-null space decomposition to ensure adherence to projection data constraints. Experimental results demonstrate that DPMA achieves improved reconstruction quality compared to existing approaches, particularly in noise suppression, artifact reduction, and fine detail preservation.

To reduce the radiation dose, sparse-view computed tomography (CT) reconstruction has been proposed, aiming to recover high-quality CT images from sparsely sampled sinogram. To eliminate the artifacts present in sparse-view CT images, a new dual-domain diffusion model (DDDM) is proposed, which is composed of a sinogram upgrading module (SUM) and an image refining module (IRM) connected in series. In the sinogram domain, a novel degrading and upgrading framework is defined, in which SUM is trained to upgrade sparse-view sinograms step by step to reverse the degradation process of CT images caused by successive down-sampling of scanning views. In the image domain, IRM adopts an improved denoising diffusion framework to further reduce remaining artifacts and restore image details, where a skip connection from the original sparse-view sinogram is introduced to constrain the generation of details. Our DDDM shows significant improvement over deep-learning baseline models in both classical similarity metrics and perceptual loss, and has good generalization to untrained organs.

Sparse-view computed tomography aims to reduce radiation exposure but often suffers from degraded image quality due to insufficient projection data. Traditional methods struggle to balance data fidelity and detail preservation, particularly in high-frequency regions. In this paper, we propose a Cascaded Frequency-Encoded Multi-Scale Neural Fields (Ca-FMNF) framework. We reformulate the reconstruction task as refining high-frequency residuals upon a high-quality low-frequency foundation. It integrates a pre-trained iterative unfolding network for initial low-frequency estimation with a FMNF to represent high-frequency residuals. The FMNF parameters are optimized by minimizing the discrepancy between the measured projections and those estimated through the imaging forward model, thereby refining the residuals based on the initial estimation. This dual-stage strategy enhances data consistency and preserves fine structures. The extensive experiments on simulated and clinical datasets demonstrate that our method achieves the optimal results in both quantitative metrics and visual quality, effectively reducing artifacts and preserving structural details.

Sparse-view computed tomography (CT) becomes a major concern in the medical imaging field due to its reduced X-ray radiation dose. Recently, various convolutional neural network (CNN)-based approaches have been proposed, requiring the pairs of full and sparse-view CT images for network training. However, these paired data acquisition is impractical or difficult in clinical practice. To handle this problem, we propose the weakly-supervised learning for streak artifact reduction with unpaired sparse-view CT data. For CNN training dataset, we generate the pairs of input and target images from the given sparse-view CT data. Then, we iteratively apply the trained network to given sparse-view CT images and acquire the prior images. As the success factor of our novel framework, we estimate the original streak artifacts in the given sparse-view CT images from the prior images and subtract the estimated streak artifacts from the given sparse-view CT images. As a result, the proposed method has the best performance of lesion detection compared to the other methods.

Phase contrast computed tomography (PCCT) provides an effective non-destructive testing tool for weak absorption objects. Limited by the phase stepping principle and radiation dose requirement, sparse-view sampling is usually performed in PCCT, introducing severe artifacts in reconstruction. In this paper, we report a dual-domain (i.e., the projection sinogram domain and image domain) enhancement framework based on deep learning (DL) for PCCT with sparse-view projections. It consists of two convolutional neural networks (CNN) in dual domains and the phase contrast Radon inversion layer (PCRIL) to connect them. PCRIL can achieve PCCT reconstruction, and it allows the gradients to backpropagate from the image domain to the projection sinogram domain while training. Therefore, parameters of CNNs in dual domains are updated simultaneously. It could overcome the limitations that the enhancement in the image domain causes blurred images and the enhancement in the projection sinogram domain introduces unpredictable artifacts. Considering the grating-based PCCT as an example, the proposed framework is validated and demonstrated with experiments of the simulated datasets and experimental datasets. This work can generate high-quality PCCT images with given incomplete projections and has the potential to push the applications of PCCT techniques in the field of composite imaging and biomedical imaging.

BACKGROUND Artifact reduction or removal is a challenging task when the artifact creation physics are not well modeled mathematically. One of such situations is metal artifacts in x-ray CT when the metallic material is unknown, and the x-ray spectrum is wide. PURPOSE A neural network is used to act as the objective function for iterative artifact reduction when the artifact model is unknown. METHODS A hypothetical unpredictable projection data distortion model is used to illustrate the proposed approach. The model is unpredictable, because it is controlled by a random variable. A convolutional neural network is trained to recognize the artifacts. The trained network is then used to compute the objective function for an iterative algorithm, which tries to reduce the artifacts in a computed tomography (CT) task. The objective function is evaluated in the image domain. The iterative algorithm for artifact reduction is in the projection domain. A gradient descent algorithm is used for the objective function optimization. The associated gradient is calculated with the chain rule. RESULTS The learning curves illustrate the decreasing treads of the objective function as the number of iterations increases. The images after the iterative treatment show the reduction of artifacts. A quantitative metric, the Sum Square Difference (SSD), also indicates the effectiveness of the proposed method. CONCLUSION The methodology of using a neural network as an objective function has potential value for situations where a human developed model is difficult to describe the underlying physics. Real-world applications are expected to be benefit from this methodology.

OBJECTIVE The dual-domain deep learning-based reconstruction techniques have enjoyed many successful applications in the field of medical image reconstruction. Applying the analytical reconstruction based operator to transfer the data from the projection domain to the image domain, the dual-domain techniques may suffer from the insufficient suppression or removal of streak artifacts in areas with the missing view data, when addressing the sparse-view reconstruction problems. In this work, to overcome this problem, an intelligent sinogram synthesis based back-projection network (iSSBP-Net) was proposed for sparse-view computed tomography (CT) reconstruction. In the iSSBP-Net method, a convolutional neural network (CNN) was involved in the dual-domain method to inpaint the missing view data in the sinogram before CT reconstruction. METHODS The proposed iSSBP-Net method fused a sinogram synthesis sub-network (SS-Net), a sinogram filter sub-network (SF-Net), a back-projection layer, and a post-CNN into an end-to-end network. Firstly, to inpaint the missing view data, the SS-Net employed a CNN to synthesize the full-view sinogram in the projection domain. Secondly, to improve the visual quality of the sparse-view CT images, the synthesized sinogram was filtered by a CNN. Thirdly, the filtered sinogram was brought into the image domain through the back-projection layer. Finally, to yield images of high visual sensitivity, the post-CNN was applied to restore the desired images from the outputs of the back-projection layer. RESULTS The numerical experiments demonstrate that the proposed iSSBP-Net is superior to all competing algorithms under different scanning condintions for sparse-view CT reconstruction. Compared to the competing algorithms, the proposed iSSBP-Net method improved the peak signal-to-noise ratio of the reconstructed images about 1.21 dB, 0.26 dB, 0.01 dB, and 0.37 dB under the scanning conditions of 360, 180, 90, and 60 views, respectively. CONCLUSION The promising reconstruction results indicate that involving the SS-Net in the dual-domain method is could be an effective manner to suppress or remove the streak artifacts in sparse-view CT images. Due to the promising results reconstructed by the iSSBP-Net method, this study is intended to inspire the further development of sparse-view CT reconstruction by involving a SS-Net in the dual-domain method.

No abstract available

In previous deep learning based super-resolution techniques for CT images, only image domain data is used for training. However, image blurring can occur in image domain method which disrupts accurate diagnosis. In this work, we propose using both sinogram and image domain data to resolve the blurring issue. To predict upsampled sinogram more accurately, we use a convolutional neural network (CNN) as an encoder, which maps an input image to feature map for decoder. For decoder, we use dual multi-layer perceptron (MLP) structure. Our proposed dual-MLP structure consists of modulator and synthesizer MLP. Synthesizer MLP predicts the output pixel value by using coordinate-based information as an input, and modulator MLP helps synthesizer to estimate the output value accurately by using feature map information as an input. This network structure preserves high frequency components better than simple CNN structure. Through our proposed sinogram upsampling network (SUN) at sinogram domain, upsampled sinogram was generated, and image was reconstructed by filtered backprojection. The reconstructed image from upsampled sinogram preserves detailed textures compared to LR image. However, residual artifacts and blur still remain. Therefore, we train CNN using image domain data to reduce residual artifacts and blur. For the dataset, we acquire projection data from Mayo Clinic image using Siddon's algorithm in fan-beam CT geometry applying 4x1 detector binning. The binned sinogram is then used as an input for the SUN. The results show that our proposed hybrid domain method outperforms image domain and sinogram domain method with higher quantitative evaluation results.

Metal implants and other high-density objects cause significant artifacts in computed tomography (CT) images, hindering clinical diagnosis. Traditional metal artifact reduction methods often leave residual artifacts due to sinogram edges discontinuities. Supervised deep learning approaches struggle due to reliance on paired data, while unsupervised methods often lack multi-domain information. In this paper, we propose TDMAR-Net, a diffusion model-based three-domain neural network that leverages priors from projection, image, and Fourier domains for removing metal artifact and enhancing CT image quality. To enhance the model’s learning capability and gradient optimization while preventing reliance on a single data structure, we employ a two-stage training strategy that combines large-scale pretraining with masked data fine-tuning, improving both accuracy and adaptability in metal artifact removal. The specific process is to adjust the weight of the high frequency and low frequency components of the input image through the high-pass filter module in the Fourier domain, and process the image into blocks to extract the diffusion prior information. The prior information is then introduced iteratively into the sinogram and image domains to fill in the metal-induced artifacts. Our method overcomes the challenges of information sharing and complementarity across different domains, ensuring that each domain contributes effectively, thereby enhancing the precision and robustness of metal artifact elimination. Experiments show that our approach superior to existing unsupervised methods, which we have validated on both synthetic and clinical datasets.

Industrial computed tomography (CT) images reconstructed directly from projection data using the filtered back projection (FBP) method exhibit strong metal artifacts due to factors such as beam hardening, scatter, statistical noise, and deficiencies in the reconstruction algorithms. Traditional correction approaches, confined to either the projection domain or the image domain, fail to fully utilize the rich information embedded in the data. To leverage information from both domains, we propose a joint deep learning framework that integrates UNet and ResNet architectures for the correction of metal artifacts in CT images. Initially, the UNet network is employed to correct the imperfect projection data (sinograms), the output of which serves as the input for the CT image reconstruction unit. Subsequently, the reconstructed CT images are fed into the ResNet, with both networks undergoing a joint training process to optimize image quality. We take the projection data obtained by analytical simulation as the data set. The resulting optimized industrial CT images show a significant reduction in metal artifacts, with the average Peak Signal-to-Noise Ratio (PSNR) reaching 36.13 and the average Structural Similarity Index (SSIM) achieving 0.953. By conducting simultaneous correction in both the projection and image domains, our method effectively harnesses the complementary information from both, exhibiting a marked improvement in correction results over the deep learning-based single-domain corrections. The generalization capability of our proposed method is further verified in ablation experiments and multi-material phantom CT artifact correction.

The presence of metallic implants often introduces severe metal artifacts in the x-ray computed tomography (CT) images, which could adversely influence clinical diagnosis or dose calculation in radiation therapy. In this work, we present a novel deep-learning-based approach for metal artifact reduction (MAR). In order to alleviate the need for anatomically identical CT image pairs (i.e. metal artifact-corrupted CT image and metal artifact-free CT image) for network learning, we propose a self-supervised cross-domain learning framework. Specifically, we train a neural network to restore the metal trace region values in the given metal-free sinogram, where the metal trace is identified by the forward projection of metal masks. We then design a novel filtered backward projection (FBP) reconstruction loss to encourage the network to generate more perfect completion results and a residual-learning-based image refinement module to reduce the secondary artifacts in the reconstructed CT images. To preserve the fine structure details and fidelity of the final MAR image, instead of directly adopting convolutional neural network (CNN)-refined images as output, we incorporate the metal trace replacement into our framework and replace the metal-affected projections of the original sinogram with the prior sinogram generated by the forward projection of the CNN output. We then use the FBP algorithms for final MAR image reconstruction. We conduct an extensive evaluation on simulated and real artifact data to show the effectiveness of our design. Our method produces superior MAR results and outperforms other compelling methods. We also demonstrate the potential of our framework for other organ sites.

No abstract available

Sparse View Computed Tomography (SVCT) is a promising technology for reducing radiation dose and improving scan speed by reducing the number of required X-ray projection images. However, the limited projection data used in SVCT often results in reconstructed images with noise and artifacts. We propose a dual-domain SVCT image reconstruction network that leverages both residual attention mechanisms and convolutional networks. The network utilizes information from both the sinogram and image domains, enabling complementary information supplementation. To further enhance image restoration, we integrate a residual attention mechanism into the Swin-Transformer architecture to mitigate feature collapse. Additionally, a parallel convolutional branch is incorporated to enhance feature extraction completeness.By leveraging the remote correlation capability of attention mechanisms and the local information extraction capabilities of convolutional networks, our model effectively extracts valuable information for reconstruction. This integration of local and global information effectively guides the reconstruction process. Experiments on the NIH-AAPM dataset demonstrate that our model effectively reduces artifacts and improves the reconstruction of complex structures, outperforming other learning-based reconstruction models.

No abstract available

Interior tomography is a crucial technique in computed tomography (CT) that aims to minimize radiation exposure by limiting X-ray imaging to the region of interest (ROI) while maintaining diagnostic accuracy. However, traditional reconstruction algorithms often suffer from severe cupping artifacts caused by data truncation, which significantly degrades image quality. This study aims to develop a parallel network that effectively integrates information between the projection and image domains to improve interior tomography reconstruction. In this paper, we propose an end-to-end deep learning framework, the Two-Module Parallel Dual-Domain Network (TPDDN), which consists of two key modules. The Initial Restoration Module generates high-quality prior sinograms and images, providing a robust foundation for subsequent processing and effectively mitigating the impact of data truncation. The Interactive Fusion Module, the core of the network, employs two parallel and interactive branches that operate simultaneously on the projection and image domains. These branches enable bidirectional feature interaction and information fusion, significantly enhancing the accuracy and quality of the reconstructed images. Extensive experiments were conducted under both normal-dose and high-dose noise conditions to evaluate the performance of TPDDN. The results demonstrate that TPDDN achieves superior qualitative and quantitative performance compared to existing representative methods. The proposed TPDDN offers a robust and effective approach for interior tomography reconstruction by synergistically integrating information from both the projection and image domains. It effectively suppresses cupping artifacts and enhances reconstructed image quality under both normal-dose and high-noise conditions, demonstrating promising potential for safer and more accurate diagnostic imaging.

Computed tomography (CT) has been widely used for medical diagnosis, assessment, and therapy planning and guidance. In reality, CT images may be affected adversely in the presence of metallic objects, which could lead to severe metal artifacts and influence clinical diagnosis or dose calculation in radiation therapy. In this article, we propose a generalizable framework for metal artifact reduction (MAR) by simultaneously leveraging the advantages of image domain and sinogram domain-based MAR techniques. We formulate our framework as a sinogram completion problem and train a neural network (SinoNet) to restore the metal-affected projections. To improve the continuity of the completed projections at the boundary of metal trace and thus alleviate new artifacts in the reconstructed CT images, we train another neural network (PriorNet) to generate a good prior image to guide sinogram learning, and further design a novel residual sinogram learning strategy to effectively utilize the prior image information for better sinogram completion. The two networks are jointly trained in an end-to-end fashion with a differentiable forward projection (FP) operation so that the prior image generation and deep sinogram completion procedures can benefit from each other. Finally, the artifact-reduced CT images are reconstructed using the filtered backward projection (FBP) from the completed sinogram. Extensive experiments on simulated and real artifacts data demonstrate that our method produces superior artifact-reduced results while preserving the anatomical structures and outperforms other MAR methods.

The wide applications of X-ray computed tomography (CT) bring low-dose CT (LDCT) into a clinical prerequisite, but reducing the radiation exposure in CT often leads to significantly increased noise and artifacts, which might lower the judgment accuracy of radiologists. In this paper, we put forward a domain progressive 3D residual convolution network (DP-ResNet) for the LDCT imaging procedure that contains three stages: sinogram domain network (SD-net), filtered back projection (FBP), and image domain network (ID-net). Though both are based on the residual network structure, the SD-net and ID-net provide complementary effect on improving the final LDCT quality. The experimental results with both simulated and real projection data show that this domain progressive deep-learning network achieves significantly improved performance by combing the network processing in the two domains.

Low-dose computed tomography (LDCT) helps to reduce radiation risks in CT scanning while maintaining image quality, which involves a consistent pursuit of lower incident rays and higher reconstruction performance. Although deep learning approaches have achieved encouraging success in LDCT reconstruction, most of them treat the task as a general inverse problem in either the image domain or the dual (sinogram and image) domains. Such frameworks have not considered the original noise generation of the projection data and suffer from limited performance improvement for the LDCT task. In this paper, we propose a novel reconstruction model based on noise-generating and imaging mechanism in full-domain, which fully considers the statistical properties of intrinsic noises in LDCT and prior information in sinogram and image domains. To solve the model, we propose an optimization algorithm based on the proximal gradient technique. Specifically, we derive the approximate solutions of the integer programming problem on the projection data theoretically. Instead of hand-crafting the sinogram and image regularizers, we propose to unroll the optimization algorithm to be a deep network. The network implicitly learns the proximal operators of sinogram and image regularizers with two deep neural networks, providing a more interpretable and effective reconstruction procedure. Numerical results demonstrate our proposed method improvements of > 2.9 dB in peak signal to noise ratio, > 1.4% promotion in structural similarity metric, and > 9 HU decrements in root mean square error over current state-of-the-art LDCT methods.

The suppression of streak artifacts in computed tomography with a limited-angle configuration is challenging. Conventional analytical algorithms, such as filtered backprojection (FBP), are not successful due to incomplete projection data. Moreover, model-based iterative total variation algorithms effectively reduce small streaks but do not work well at eliminating large streaks. In contrast, FBP mapping networks and deep-learning-based postprocessing networks are outstanding at removing large streak artifacts; however, these methods perform processing in separate domains, and the advantages of multiple deep learning algorithms operating in different domains have not been simultaneously explored. In this paper, we present a hybrid-domain convolutional neural network (hdNet) for the reduction of streak artifacts in limited-angle computed tomography. The network consists of three components: the first component is a convolutional neural network operating in the sinogram domain, the second is a domain transformation operation, and the last is a convolutional neural network operating in the CT image domain. After training the network, we can obtain artifact-suppressed CT images directly from the sinogram domain. Verification results based on numerical, experimental and clinical data confirm that the proposed method can significantly reduce serious artifacts.

High pitch helical Computed Tomography (CT) scanning significantly reduces radiation dose while improving temporal resolution, offering substantial clinical benefits. However, the incomplete scanning data commonly leads to artifacts in the reconstructed images, degrading image quality and potentially affecting clinical diagnosis. Existing high pitch reconstruction methods primarily operate within the image domain or combine image-domain networks with traditional iterative algorithms, yet their performance remains limited. To address such limitations, we propose Delta-Net, a deep dual-domain alternating iterative optimization network for high pitch helical CT reconstruction. We introduce a novel optimization objective and develop an alternating iterative optimization framework, where each sub iteration consists of projection domain correction and image domain refinement. To enhance generalization and robustness, deep neural networks are employed to learn domain-specific priors, which are incorporated as regularization terms, with all hyper-parameters automatically optimized during training. Specifically, the image domain residual refinement network (IRN) and projection domain consistency enhanced network (PCN) regularize the intermediate results across both domains. Additionally, to improve the capability of artifact suppression and structure restoration, a structure-aware joint loss is tailored for the optimization of Delta-Net. Quantitative and qualitative evaluations on clinical datasets demonstrate that Delta-Net outperforms other competitive methods in artifact suppression, fine structure recovery, and generalization.

PURPOSE/OBJECTIVE(S) Limited-angle CBCT (LA-CBCT) is of great clinical interest, because the scanning time and the patient dose are proportional to the scanning range of gantry rotation angles of CBCT. However, the image reconstruction for LA-CBCT remains technically challenging, which suffers from severe wedge artifacts and image distortions. This work aims to improve LA-CBCT by developing deep learning (DL) methods for real clinical CBCT projection data, which is the first feasibility study of clinical-projection-data-based LA-CBCT, to the best of our knowledge. MATERIALS/METHODS Targeting at real clinical projection data, we have explored various DL methods such as image/data/hybrid-domain methods and finally developed a so-called Structure-Enhanced Attention Network (SEA-Net) method that has the best image quality from clinical projection data among the DL methods we have implemented. Specifically, the proposed SEA-Net employs a specialized structure enhancement sub-network to promote texture preservation. Based on the observation that the distribution of wedge artifacts in reconstruction images is non-uniform, the spatial attention module is utilized to emphasize the relevant regions while ignores the irrelevant ones, which leads to more accurate texture restoration. RESULTS SEA-Net was validated in comparison with analytic (FDK), iterative (TV), image-domain DL (DDNet and FED-INet, data-domain DL (DCAR), dual-domain DL (Sam'Net), and various unrolling DL (hdNet, CTNet, FSR-Net, CasRedSCAN) methods. Among all methods, the SEA-Net had the best image reconstruction quality as quantified by root-mean-square error (RMSE), peak signal-to-noise ratio (PSNR), and structural similarity index (SSIM), for various LA-CBCT problems of 90°-180° projection data. In addition, LA-CBCT via SEA-Net provided comparable accuracy for both patient setup (quantified by image registration accuracy from planning CT (pCT) to CBCT) and dose calculation (see the table), with full-view CBCT. CONCLUSION We explored various DL methods and developed an image-domain-based method termed SEA-Net that provided the best image quality for clinical projection data. To the best of our knowledge, this is the first feasibility study of the real clinical-projection-data-based LA-CBCT. Moreover, LA-CBCT via SEA-Net can potentially provide comparable accuracy for patient setup and dose calculation, with full-view CBCT.

Edges tend to be over-smoothed in total variation (TV) regularized under-sampled images. In this paper, symmetric residual convolutional neural network (SR-CNN), a deep learning based model, was proposed to enhance the sharpness of edges and detailed anatomical structures in under-sampled cone-beam computed tomography (CBCT). For training, CBCT images were reconstructed using TV-based method from limited projections simulated from the ground truth CT, and were fed into SR-CNN, which was trained to learn a restoring pattern from under-sampled images to the ground truth. For testing, under-sampled CBCT was reconstructed using TV regularization and was then augmented by SR-CNN. Performance of SR-CNN was evaluated using phantom and patient images of various disease sites acquired at different institutions both qualitatively and quantitatively using structure similarity (SSIM) and peak signal-to-noise ratio (PSNR). SR-CNN substantially enhanced image details in the TV-based CBCT across all experiments. In the patient study using real projections, SR-CNN augmented CBCT images reconstructed from as low as 120 half-fan projections to image quality comparable to the reference fully-sampled FDK reconstruction using 900 projections. In the tumor localization study, improvements in the tumor localization accuracy were made by the SR-CNN augmented images compared with the conventional FDK and TV-based images. SR-CNN demonstrated robustness against noise levels and projection number reductions and generalization for various disease sites and datasets from different institutions. Overall, the SR-CNN-based image augmentation technique was efficient and effective in considerably enhancing edges and anatomical structures in under-sampled 3D/4D-CBCT, which can be very valuable for image-guided radiotherapy.