Streak Artifacts Research Articles

Low-Dose CT Image Denoising Based on Improved DD-Net and Local Filtered Mechanism.

Low-dose CT (LDCT) images can reduce the radiation damage to the patients; however, the unavoidable information loss will influence the clinical diagnosis under low-dose conditions, such as noise, streak artifacts, and smooth details. LDCT image denoising is a significant topic in medical image processing to overcome the above deficits. This work proposes an improved DD-Net (DenseNet and deconvolution-based network) joint local filtered mechanism, the DD-Net is enhanced by introducing improved residual dense block to strengthen the feature representation ability, and the local filtered mechanism and gradient loss are also employed to effectively restore the subtle structures. First, the LDCT image is inputted into the network to obtain the denoised image. The original loss between the denoised image and normal-dose CT (NDCT) image is calculated, and the difference image between the NDCT image and the denoised image is obtained. Second, a mask image is generated by taking a threshold operation to the difference image, and the filtered LDCT and NDCT images are obtained by conducting an elementwise multiplication operation with LDCT and NDCT images using the mask image. Third, the filtered image is inputted into the network to obtain the filtered denoised image, and the correction loss is calculated. At last, the sum of original loss and correction loss of the improved DD-Net is used to optimize the network. Considering that it is insufficient to generate the edge information using the combination of mean square error (MSE) and multiscale structural similarity (MS-SSIM), we introduce the gradient loss that can calculate the loss of the high-frequency portion. The experimental results show that the proposed method can achieve better performance than conventional schemes and most neural networks. Our source code is made available at https://github.com/LHE-IT/Low-dose-CT-Image-Denoising/tree/main/Local Filtered Mechanism.

Combining natural and artificial intelligence for robust automatic anatomy segmentation: Application in neck and thorax auto-contouring.

Automatic segmentation of 3D objects in computed tomography (CT) is challenging. Current methods, based mainly on artificial intelligence (AI) and end-to-end deep learning (DL) networks, are weak in garnering high-level anatomic information, which leads to compromised efficiency and robustness. This can be overcome by incorporating natural intelligence (NI) into AI methods via computational models of human anatomic knowledge. We formulate a hybrid intelligence (HI) approach that integrates the complementary strengths of NI and AI for organ segmentation in CT images and illustrate performance in the application of radiation therapy (RT) planning via multisite clinical evaluation. The system employs five modules: (i) body region recognition, which automatically trims a given image to a precisely defined target body region; (ii) NI-based automatic anatomy recognition object recognition (AAR-R), which performs object recognition in the trimmed image without DL and outputs a localized fuzzy model for each object; (iii) DL-based recognition (DL-R), which refines the coarse recognition results of AAR-R and outputs a stack of 2D bounding boxes (BBs) for each object; (iv) model morphing (MM), which deforms the AAR-R fuzzy model of each object guided by the BBs output by DL-R; and (v) DL-based delineation (DL-D), which employs the object containment information provided by MM to delineate each object. NI from (ii), AI from (i), (iii), and (v), and their combination from (iv) facilitate the HI system. The HI system was tested on 26 organs in neck and thorax body regions on CT images obtained prospectively from 464 patients in a study involving four RT centers. Data sets from one separate independent institution involving 125 patients were employed in training/model building for each of the two body regions, whereas 104 and 110 data sets from the 4 RT centers were utilized for testing on neck and thorax, respectively. In the testing data sets, 83% of the images had limitations such as streak artifacts, poor contrast, shape distortion, pathology, or implants. The contours output by the HI system were compared to contours drawn in clinical practice at the four RT centers by utilizing an independently established ground-truth set of contours as reference. Three sets of measures were employed: accuracy via Dice coefficient (DC) and Hausdorff boundary distance (HD), subjective clinical acceptability via a blinded reader study, and efficiency by measuring human time saved in contouring by the HI system. Overall, the HI system achieved a mean DC of 0.78 and 0.87 and a mean HD of 2.22 and 4.53mm for neck and thorax, respectively. It significantly outperformed clinical contouring in accuracy and saved overall 70% of human time over clinical contouring time, whereas acceptability scores varied significantly from site to site for both auto-contours and clinically drawn contours. The HI system is observed to behave like an expert human in robustness in the contouring task but vastly more efficiently. It seems to use NI help where image information alone will not suffice to decide, first for the correct localization of the object and then for the precise delineation of the boundary.

RAD-UNet: a Residual, Attention-Based, Dense UNet for CT Sparse Reconstruction.

To suppress the streak artifacts in images reconstructed from sparse-view projections in computed tomography (CT), a residual, attention-based, dense UNet (RAD-UNet) deep network is proposed to achieve accurate sparse reconstruction. The filtered back projection (FBP) algorithm is used to reconstruct the CT image with streak artifacts from sparse-view projections. Then, the image is processed by the RAD-UNet to suppress streak artifacts and obtain high-quality CT image. Those images with streak artifacts are used as the input of the RAD-UNet, and the output-label images are the corresponding high-quality images. Through training via the large-scale training data, the RAD-UNet can obtain the capability of suppressing streak artifacts. This network combines residual connection, attention mechanism, dense connection and perceptual loss. This network can improve the nonlinear fitting capability and the performance of suppressing streak artifacts. The experimental results show that the RAD-UNet can improve the reconstruction accuracy compared with three existing representative deep networks. It may not only suppress streak artifacts but also better preserve image details. The proposed networks may be readily applied to other image processing tasks including image denoising, image deblurring, and image super-resolution.

The evaluation of the reduction of radiation dose via deep learning-based reconstruction for cadaveric human lung CT images

To compare the quality of CT images of the lung reconstructed using deep learning-based reconstruction (True Fidelity Image: TFI ™; GE Healthcare) to filtered back projection (FBP), and to determine the minimum tube current–time product in TFI without compromising image quality. Four cadaveric human lungs were scanned on CT at 120 kVp and different tube current–time products (10, 25, 50, 75, 100, and 175 mAs) and reconstructed with TFI and FBP. Two image evaluations were performed by three independent radiologists. In the first experiment, using the same tube current–time product, a side-by-side TFI and FBP comparison was performed. Images were evaluated with regard to noise, streak artifacts, and overall image quality. Overall image quality was evaluated in view of whole image quality. In the second experiment, CT images reconstructed using TFI and FBP with five different tube current–time products were displayed in random order, which were evaluated with reference to the 175 mAs-FBP image. Images were scored with regard to normal structure, abnormal findings, noise, streak artifacts, and overall image quality. Median scores from three radiologists were statistically analyzed. Quantitative evaluation of noise was performed by setting regions of interest (ROIs) in air. In first experiment, overall image quality was improved, and noise was decreased in images of TFI compared to that of FBP for all tube current–time products. In second experiment, scores of all evaluation items except for small vessels in images of 25 mAs-TFI were almost the same as that of 175 mAs-FBP (all p > 0.31). Using TFI instead of FBP, at least 85% radiation dose reduction could be possible without any degradation in the image quality.

Imaging feature of cosmetic fillers in cone-beam computed tomography and its dental consideration

BackgroundAs the application of cone-beam computed tomography (CBCT) in head and neck area increases for dental treatment purposes, cosmetic filler materials are incidentally observed. Since the materials are very diverse, unnecessary referrals or additional examination may be performed when clinicians are unfamiliar with the imaging findings. Thus, this study aimed to introduce the imaging characteristics of cosmetic fillers and grafts shown in dental CBCT with dental considerations that the clinician should be aware of.MethodsCBCT obtained for dental purpose presenting cosmetic material were selected. The location of the material was identified as buccal, retroantral, parotid space, nose, zygoma, and symphysis. The material was classified as single or multiple, and grouped according to morphology: speckle, round, eggshell, linear, and amorphous. The radiopacity was classified as similar to soft tissue, between soft and hard tissue, similar to hard tissue, and metal.ResultsTwenty-one patients were reviewed, and all patients were female with mean age of 50.5 years. The buccal space was the most frequent location for multiple filler materials. The symphysis was the next frequent location and only single material were shown in this location. Cases having multiple filler showed diverse shapes while all single materials showed round shape. Fillers showing radiopacity of hard tissue were similar to diseases producing soft tissue calcifications. Metal-density material distributed in spaces induced white and dark streak artifacts in the CBCT image. All single materials presented radiopacity between soft and hard tissue and attached to the bone surface causing mandibular bone resorption.ConclusionsCosmetic materials displayed various imaging features in CBCT acquired during dental procedure. Clinicians should consider that cosmetic material may cause mandibular bone resorption and imaging artifacts on CBCT. Knowledge of the imaging characteristics of cosmetic fillers may help correct diagnosis.

Coronary computed tomography and cardiac devices : Diagnostic results or nothing but artifacts?

Coronary computed tomography (CT) angiography has become amajor cornerstone in the diagnostic workup of cardiologic patients, particularly for evaluation of the coronary arteries and preprocedural planning of interventions for structural heart disease. Despite the possible problems that intensive electromagnetic radiation (including X‑rays) might cause when directly impacting on implanted cardiac devices, cardiac CT is asafe diagnostic test and should not be withheld from patients with devices if properly indicated. Sufficient image quality is paramount for the evaluation; hence, special attention should be paid to alow heart rate (< 60 bpm) and sufficient compliance with breathing instructions. Furthermore, pacemaker or implantable cardioverter-defibrillator (ICD) leads may cause metal artifacts, especially around the lead tip. Their dense material causes beam hardening and streak artifacts which may result in reduced image quality and limited diagnostic assessability. The prevalence of such artifacts depends not only on lead material but also on lead positioning relative to the gantry plane. Metal artifacts are more frequent in patients with unipolar leads and shock coils, which can impair the assessment of coronary arteries, mainly of the right coronary artery (RCA). Artifacts caused by left ventricular (LV) leads of cardiac resynchronization therapy (CRT) systems tend to affect assessment of the left circumflex artery (LCX). By using dual energy CT and postprocessing algorithms, the impact of artifacts can be reduced and diagnostic image quality can be achieved in most cases. Unfortunately, the actual occurrence of such artifacts or the degree of impairment of image quality cannot be reliably predicted.

Multisite reproducibility of quantitative susceptibility mapping and effective transverse relaxation rate in deep gray matter at 3 T using locally optimized sequences in 24 traveling heads

Iron concentration in the human brain plays a crucial role in several neurodegenerative diseases and can be monitored noninvasively using quantitative susceptibility mapping (QSM) and effective transverse relaxation rate (R2 *) mapping from multiecho T2 *-weighted images. Large population studies enable better understanding of pathologies and can benefit from pooling multisite data. However, reproducibility may be compromised between sites and studies using different hardware and sequence protocols. This work investigates QSM and R2 * reproducibility at 3T using locally optimized sequences from three centers and two vendors, and investigates possible reduction of cross-site variability through postprocessing approaches. Twenty-four healthy subjects traveled between three sites and were scanned twice at each site. Scan-rescan measurements from seven deep gray matter regions were used for assessing within-site and cross-site reproducibility using intraclass correlation coefficient (ICC) and within-subject standard deviation (SDw) measures. In addition, multiple QSM and R2 * postprocessing options were investigated with the aim to minimize cross-site sequence-related variations, including: mask generation approach, echo-timing selection, harmonizing spatial resolution, field map estimation, susceptibility inversion method, and linear field correction for magnitude images. The same-subject cross-site region of interest measurements for QSM and R2 * were highly correlated (R2 ≥ 0.94) and reproducible (mean ICC of 0.89 and 0.82 for QSM and R2 *, respectively). The mean cross-site SDw was 4.16 parts per billion (ppb) for QSM and 1.27 s-1 for R2 *. For within-site measurements of QSM and R2 *, the mean ICC was 0.97 and 0.87 and mean SDw was 2.36 ppb and 0.97 s-1 , respectively. The precision level is regionally dependent and is reduced in the frontal lobe, near brain edges, and in white matter regions. Cross-site QSM variability (mean SDw) was reduced up to 46% through postprocessing approaches, such as masking out less reliable regions, matching available echo timings and spatial resolution, avoiding the use of the nonconsistent magnitude contrast between scans in field estimation, and minimizing streaking artifacts.

Image quality improvement and motion degradation reduction in shoulder MR imaging: comparison of BLADE and rectilinear techniques at 3-Tesla scanning.

MR imaging of joints, particularly shoulder, requires a high degree of spatial resolution to ascertain anatomy and pathology. Unfortunately, motion artifacts can reduce the clinical quality of the examinations. BLADE sequence reduces motion degradation improving overall diagnostic imaging quality. The objective was to compare standard, rectilinear k-space coverage turbo spin echo (TSE), and BLADE sequences. Over a 4-month period, fifty-seven consecutive patients (22 males, 35 females; mean age: 48.5years, range: 23-64years) were scanned using traditional intermediate-weighted spin echo and BLADE sequences. Qualitative evaluation was performed by three musculoskeletal fellowship trained radiologists, each with more than 5years of experience. Image sequences were evaluated using a Likert scale for each of the following five categories: motion degradation, ghosting/phase misregistration artifacts, star/radial encoding artifacts, fat suppression quality, and overall diagnostic quality. Additionally, image sequences were evaluated for signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) using manually drawn regions of interest (ROI) analysis. Ghosting and phase artifacts were lower within BLADE sequence while streak artifacts were higher (p < 0.001). Image fat suppression, tendon and labral appearances, and the overall SNR and CNR were comparable on both sequences (p > 0.05). Addition of BLADE reduces motion degradation and improves overall diagnostic imaging quality. Application of BLADE in patient scans suspected of motion artifacts can reduce the frequency of repeat imaging in patients with claustrophobia or those where motion is a concern. By reducing overall imaging time and call backs, it could reduce the cost burden to patients and healthcare system.

Can deep learning improve image quality of low-dose CT: a prospective study in interstitial lung disease.

To investigate whether deep learning reconstruction (DLR) could keep image quality and reduce radiation dose in interstitial lung disease (ILD) patients compared with HRCT reconstructed with hybrid iterative reconstruction (hybrid-IR). Seventy ILD patients were prospectively enrolled and underwent HRCT (120 kVp, automatic tube current) and LDCT (120 kVp, 30 mAs) scans. HRCT images were reconstructed with hybrid-IR (Adaptive Iterative Dose Reduction 3-Dimensional [AIDR3D], standard-setting); LDCT images were reconstructed with DLR (Advanced Intelligence Clear-IQ Engine [AiCE], lung/bone, mild/standard/strong setting). Image noise, streak artifact, overall image quality, and visualization of normal and abnormal features of ILD were evaluated. The mean radiation dose of LDCT was 38% of HRCT. Objective image noise of reconstructed LDCT images was 33.6 to 111.3% of HRCT, and signal-to-noise ratio (SNR) was 0.9 to 3.1 times of the latter (p < 0.001). LDCT-AiCE was not significantly different from or even better than HRCT in overall image quality and visualization of normal lung structures. LDCT-AiCE (lung, mild/standard/strong) showed progressively better recognition of ground glass opacity than HRCT-AIDR3D (p < 0.05, p < 0.01, p < 0.001), and LDCT-AiCE (lung, mild/standard/strong; bone, mild) was superior to HRCT-AIDR3D in visualization of architectural distortion (p < 0.01, p < 0.01, p < 0.01; p < 0.05). LDCT-AiCE (bone, strong) was better than HRCT-AIDR3D in the assessment of bronchiectasis and/or bronchiolectasis (p < 0.05). LDCT-AiCE (bone, mild/standard/strong) was significantly better at the visualization of honeycombing than HRCT-AIDR3D (p < 0.05, p < 0.05, p < 0.01). Deep learning reconstruction could effectively reduce radiation dose and keep image quality in ILD patients compared to HRCT with hybrid-IR. • Deep learning reconstruction was a novel image reconstruction algorithm based on deep convolutional neural networks. It was applied in chest CT studies and received auspicious results. • HRCT plays an essential role in the whole process of diagnosis, treatment efficacy evaluation, and follow-ups for interstitial lung disease patients. However, cumulative radiation exposure could increase the risks of cancer. • Deep learning reconstruction method could effectively reduce the radiation dose and keep the image quality compared with HRCT reconstructed with hybrid iterative reconstruction in patients with interstitial lung disease.

Metal artefact reduction in low-dose computed tomography: Benefits of tin prefiltration versus postprocessing of dual-energy datasets over conventional CT imaging

IntroductionThe purpose of this study was to determine the potential for metal artefact reduction in low-dose multidetector CT as these pose a frequent challenge in clinical routine. Investigations focused on whether spectral shaping via tin prefiltration, virtual monoenergetic imaging or virtual blend imaging (VBI) offers superior image quality in comparison with conventional CT imaging. MethodsUsing a third-generation dual-source CT scanner, two cadaveric specimens with different metal implants (dental, cervical spine, hip, knee) were examined with acquisition protocols matched for radiation dose with regards to tube voltage and current. In order to allow for precise comparison, and due to the relatively short scan lengths, automatic tube current modulation was disabled. Specifically, the following scan protocals were examined: conventional CT protocols (100/120 kVp), tin prefiltration (Sn 100/Sn 150 kVp), VBI and virtual monoenergetic imaging (VME 100/120/150 keV). Mean attenuation and image noise were measured in hyperdense and hypodense artefacts, in artefact-impaired and artefact-free soft tissue. Subjective image quality was rated independently by three radiologists. ResultsObjectively, Sn 150 kVp allowed for the best reduction of hyperdense streak artefacts (p < 0.001), while VME 150 keV and Sn 150 kVp protocols facilitated equally good reduction of hypodense artefacts (p = 0.173). Artefact-impaired soft tissue attenuation was lowest in Sn 150 kVp protocols (p ≤ 0.011), whereas all VME showed significantly less image noise compared to conventional or tin-filtered protocols (p ≤ 0.001). Subjective assessment favoured Sn 150 kVp regarding hyperdense streak artefacts and delineation of cortical bone (p ≤ 0.005). The intraclass correlation coefficient was 0.776 (95% confidence interval: 0.712–0.831; p < 0.001) indicating good interrater reliability. ConclusionIn the presence of metal implants in our cadaveric study, tin prefiltration with 150 kVp offers superior artefact reduction for low-dose CT imaging of osseous tissue compared with virtual monoenergetic images of dual-energy datasets. The delineation of cortical boundaries seems to benefit particularly from spectral shaping. Implications for practiceLow-dose CT imaging of osseous tissue in combination with tin prefiltration allows for superior metal artefact reduction when compared to virtual monoenergetic images of dual-energy datasets. Employing this technique ought to be considered in daily routine when metal implants are present within the scan volume as findings suggest it allows for radiation dose reduction and facilitates diagnosis relevant to further treatment.

Patch-based artifact reduction for three-dimensional volume projection data of sparse-view micro-computed tomography.

Micro-computed tomography (micro-CT) enables the non-destructive acquisition of three-dimensional (3D) morphological structures at the micrometer scale. Although it is expected to be used in pathology and histology to analyze the 3D microstructure of tissues, micro-CT imaging of tissue specimens requires a long scan time. A high-speed imaging method, sparse-view CT, can reduce the total scan time and radiation dose; however, it causes severe streak artifacts on tomographic images reconstructed with analytical algorithms due to insufficient sampling. In this paper, we propose an artifact reduction method for 3D volume projection data from sparse-view micro-CT. Specifically, we developed a patch-based lightweight fully convolutional network to estimate full-view 3D volume projection data from sparse-view 3D volume projection data. We evaluated the effectiveness of the proposed method using physically acquired datasets. The qualitative and quantitative results showed that the proposed method achieved high estimation accuracy and suppressed streak artifacts in the reconstructed images. In addition, we confirmed that the proposed method requires both short training and prediction times. Our study demonstrates that the proposed method has great potential for artifact reduction for 3D volume projection data under sparse-view conditions.

Cancellation of streak artifacts in radial abdominal imaging using interference null space projection.

In radial abdominal imaging, it has been commonly observed that signal from the arms cause streaks due to system imperfections. We previously introduced a streak removal technique (B-STAR), which is inherently spatially variant and limited to work in image space. In this work, we propose a spatially invariant streak cancellation technique (CACTUS), which can be applied in either image space or k-space and is compatible with iterative reconstructions. Streak sources are typically spatially localized and can be represented using a low-dimensional subspace. CACTUS identifies the streak subspace by leveraging the spatial redundancy of receiver coils and projects the data onto the streak null space to eliminate the streaks. When applied in k-space, CACTUS can be combined with iterative reconstructions. CACTUS was tested in phantoms and in vivo abdominal imaging using a radial turbo spin-echo pulse sequence. In phantoms, CACTUS improved T2 estimation in comparison to previous de-streaking methods. In vivo experiments showed that CACTUS reduced streaks and yielded T2 estimation, in regions affected by streaks, closer to a streak-free reference. Evaluation using a clinical abdominal dataset (n=20) showed that CACTUS is comparable to B-STAR and yields significantly better signal preservation and streak cancellation than coil removal and suppression methods. CACTUS provides superior signal preservation and streak reduction performance compared to coil removal and suppression methods. As a clear advantage over B-STAR, CACTUS can be integrated with iterative reconstruction methods. In abdominal T2 mapping, CACTUS improves the accuracy of parameter estimation in areas affected by streaks.

Suppressing multi-material and streak artifacts with an accelerated 3D iterative image reconstruction algorithm for in-line X-ray phase-contrast computed tomography.

In-line X-ray phase-contrast computed tomography typically contains two independent procedures: phase retrieval and computed tomography reconstruction, in which multi-material and streak artifacts are two important problems. To address these problems simultaneously, an accelerated 3D iterative image reconstruction algorithm is proposed. It merges the above-mentioned two procedures into one step, and establishes the data fidelity term in raw projection domain while introducing 3D total variation regularization term in image domain. Specifically, a transport-of-intensity equation (TIE)-based phase retrieval method is updated alternately for different areas of the multi-material sample. Simulation and experimental results validate the effectiveness and efficiency of the proposed algorithm.

Radiation Dose Reduction for 80-kVp Pediatric CT Using Deep Learning-Based Reconstruction: A Clinical and Phantom Study.

BACKGROUND. Deep learning-based reconstruction (DLR) may facilitate CT radiation dose reduction, but a paucity of literature has compared lower-dose DLR images with standard-dose iterative reconstruction (IR) images or explored application of DLR to low-tube-voltage scanning in children. OBJECTIVE. The purpose of this study was to assess whether DLR can be used to reduce radiation dose while maintaining diagnostic image quality in comparison with hybrid IR (HIR) and model-based IR (MBIR) for low-tube-voltage pediatric CT. METHODS. This retrospective study included children 6 years old or younger who underwent contrast-enhanced 80-kVp CT with a standard-dose or lower-dose protocol. Standard images were reconstructed with HIR, and lower-dose images were reconstructed with HIR, MBIR, and DLR. Size-specific dose estimate (SSDE) was calculated for both protocols. Image noise, signal-to-noise ratio (SNR), and contrast-to-noise ratio (CNR) were quantified. Two radiologists independently evaluated noise magnitude, noise texture, streak artifact, edge sharpness, and overall quality. Interreader agreement was assessed, and mean values were calculated. To evaluate task-based object detection performance, a phantom was imaged with 80-kVp CT at six doses (SSDE, 0.6-5.3 mGy). Detectability index (d') was calculated from the noise power spectrum and task-based transfer function. Reconstruction methods were compared. RESULTS. Sixty-five children (mean age, 25.0 ± 25.2 months) who underwent CT with standard- (n = 31) or lower-dose (n = 34) protocol were included. SSDE was 54% lower for the lower-dose than for the standard-dose group (1.9 ± 0.4 vs 4.1 ± 0.8 mGy). Lower-dose DLR and MBIR yielded lower image noise and higher SNR and CNR than standard-dose HIR (p < .05). Interobserver agreement on subjective features ranged from a kappa coefficient of 0.68 to 0.78. The readers subjectively scored noise texture, edge sharpness, and overall quality lower for lower-dose MBIR than for standard-dose HIR (p < .001), though higher for lower-dose DLR than for standard-dose HIR (p < .001). In the phantom, DLR provided higher d' than HIR and MBIR at each dose. Object detectability was greater for 2.0-mGy DLR than for 4.0-mGy HIR for low-contrast (3.67 vs 3.57) and high-contrast (1.20 vs 1.04) objects. CONCLUSION. Compared with IR algorithms, DLR results in substantial dose reduction with preserved or even improved image quality for low-tube-voltage pediatric CT. CLINICAL IMPACT. Use of DLR at 80 kVp allows greater dose reduction for pediatric CT than do current IR techniques.

MRCON-Net: Multiscale reweighted convolutional coding neural network for low-dose CT imaging

Background and objectiveLow-dose computed tomography (LDCT) has become increasingly important for alleviating X-ray radiation damage. However, reducing the administered radiation dose may lead to degraded CT images with amplified mottle noise and nonstationary streak artifacts. Previous studies have confirmed that deep learning (DL) is promising for improving LDCT imaging. However, most DL-based frameworks are built intuitively, lack interpretability, and suffer from image detail information loss, which has become a general challenging issue. MethodsA multiscale reweighted convolutional coding neural network (MRCON-Net) is developed to address the above problems. MRCON-Net is compact and more explainable than other networks. First, inspired by the learning-based reweighted iterative soft thresholding algorithm (ISTA), we extend traditional convolutional sparse coding (CSC) to its reweighted convolutional learning form. Second, we use dilated convolution to extract multiscale image features, allowing our single model to capture the correlations between features of different scales. Finally, to automatically adjust the elements in the feature code to correct the obtained solution, a channel attention (CA) mechanism is utilized to learn appropriate weights. ResultsThe visual results obtained based on the American Association of Physicians in Medicine (AAPM) Challenge and United Image Healthcare (UIH) clinical datasets confirm that the proposed model significantly reduces serious artifact noise while retaining the desired structures. Quantitative results show that the average structural similarity index measurement (SSIM) and peak signal-to-noise ratio (PSNR) achieved on the AAPM Challenge dataset are 0.9491 and 40.66, respectively, and the SSIM and PSNR achieved on the UIH clinical dataset are 0.915 and 42.44, respectively; these are promising quantitative results. ConclusionCompared with recent state-of-the-art methods, the proposed model achieves subtle structure-enhanced LDCT imaging. In addition, through ablation studies, the components of the proposed model are validated to achieve performance improvements.

Free-breathing dynamic T1WI using compressed sensing-golden angle radial sparse parallel imaging for liver MRI in patients with limited breath-holding capability

PurposeEvaluating the performance of free-breathing dynamic T1-weighted imaging (T1WI) using compressed sensing golden-angle radial sparse parallel imaging (CS-GRASP) in non-cooperative patients compared with a general group. MethodThis retrospective study included patients who underwent gadoxetic acid-enhanced liver MRI using CS-GRASP at 3 T between March 2018 and October 2019. Patients were divided chronologically, into one of two groups: Group 1, who underwent MRI during the sequence implementation period regardless of breath-hold capability; and Group 2, who underwent MRI from June 2018 due to limited breath-hold capability. Three radiologists evaluated motion and streak artifacts as well as overall image quality on a four-point scale at the precontrast phase, late arterial phase (LAP) and portal venous phase (PVP). Intra-individual comparisons were made between sequences in each group. ResultsWe identified 102 patients, who were divided into either Group 1 (n = 41) or 2 (n = 61). For the LAP, the former group had higher image quality (3.22 ± 0.65 vs. 2.95 ± 0.61, P < 0.001) and less streak artifact (2.96 ± 0.56 vs. 2.74 ± 0.57, P = 0.001) than the latter. However, there was no significant difference between the two groups regarding either the proportion of patients with acceptable motion artifact, at 92.7% (38/41) for Group 1 vs. 96.7% (59/61) for Group 2, or that of patients with acceptable image quality at 80.5% (33/41) for Group 1 vs. 65.6% (40/61) for Group 2 (P > 0.05). In intra-individual comparisons, portal phase showed the highest image quality than the others in both groups (P < 0.001). ConclusionsAcceptable image quality for the LAP in non-cooperative patients was provided with a success rate of over 50% via free-breathing T1WI using CS-GRASP.

Streak Metal Artifact Reduction Based on Sinogram Fusion and Tissue-Class Model in CT Images

The presence of streak metal artifacts seriously degrades the diagnostic value and deteriorates the qualities of CT images. Analyzing the causes and classical streak metal artifact reduction (MAR) methods, the paper proposes the streak metal artifact reduction method based on sinogram fusion and tissue-class mode for CT images (F-MAR). Firstly, the original CT images are corrected using a linear interpolation streak metal artifact reduction (L-MAR) scheme in the raw data domain. Subsequently, to preserve the edge information, the metal artifact-reduced images are then smoothed into smoothed images (tissue-class model) by using the mean filter. Segment the original CT image that contained the streak artifacts. The original CT image and the CT image that contained high-density material are projected into the original sinogram and the high density material sinogram, respectively. Secondly, the simple linear interpolation is used to correct the CT original CT image into the corrected CT image. The mean filter is applied in the corrected CT image. The corrected CT image is projected into the corrected sinogram. Thirdly, according to the position of the high density material sinogram located in the original sinogram and the corrected sinogram, the original position sinogram included in the original sinogram and the corrected position sinogram included in the corrected sinogram are, respectively, obtained. The two sinograms are fused into the fused sinogram. The fused sinogram, the original sinogram, and the high-density material sinogram are fused into the final sinogram. Finally, the filtered back projection reconstruction algorithm is used to reconstruct the final sinogram into the reconstructed CT image. The reconstructed CT image and high density material image are fused into the final image. The experimentation results show that the method proposed in the paper can obtain better correction effect than the classical correction methods in vision.

Empirical scatter correction: CBCT scatter artifact reduction without prior information.

The image quality of cone beam CT (CBCT) scans severely suffers from scattered radiation if no countermeasures are taken. Scatter artifacts may induce cupping and streak artifacts and lead to a reduced image contrast and wrong CT values of the reconstructed volumes. Established software-based approaches for a correction of scattered radiation typically rely on prior knowledge of the CT system, scan parameters, the scanned object, or all of theaforementioned. This study proposes a simple and effective postprocessing software-based correction method of scatter artifacts in CBCT scans without specific priorknowledge. We propose the empirical scatter correction (ESC), which generates scatter-like basis images from each projection image by convolution operations. A linear combination of these basis images is subtracted from the original projection image. The logarithm is taken and an FDK reconstruction is performed. The coefficients needed for the linear combination are determined automatically by a downhill simplex algorithm such that the resulting reconstructed images show no scatter artifacts. We demonstrate the potential of ESC by correcting simulated volumes with Monte Carlo scatter artifacts, a head phantom scan performed on our table-top CBCT, and a pelvis scan from a Varian Edge CBCTscanner. ESC is able to improve the image quality of CBCT scans, which is shown on the basis of our simulations and on measured data. For a simulated head CT, the CT value difference to the scatter-free reference image was as low as -6 HU after using ESC, whereas the uncorrected data deviated by more than -200 HU from the reference data. Simulations of thorax and abdomen CT scans show that although scatter artifacts are not fully removed, anatomical features which were hard to discover prior to the correction become clearly visible and better segmentable with ESC. Similar results are obtained in the phantom measurement, where a comparison to a slit scan of our head phantom shows only small differences. The CT values in soft tissue are improved in this measurement, as well. In soft tissue areas with severe scatter artifacts, the CT values agree well with those of the slit scan (difference to slit scan: 35HU corrected and -289 HU uncorrected). Scatter artifacts in measured patient data can also be reduced using the proposed ESC. The results are comparable to those achieved with designated correction algorithms installed on the Varian Edge CBCT system. ESC allows to reduce artifacts caused by patient scatter solely based on the projectiondata.

Multi-domain integrative Swin transformer network for sparse-view tomographic reconstruction

SummaryDecreasing projection views to a lower X-ray radiation dose usually leads to severe streak artifacts. To improve image quality from sparse-view data, a multi-domain integrative Swin transformer network (MIST-net) was developed and is reported in this article. First, MIST-net incorporated lavish domain features from data, residual data, image, and residual image using flexible network architectures, where a residual data and residual image sub-network was considered as a data consistency module to eliminate interpolation and reconstruction errors. Second, a trainable edge enhancement filter was incorporated to detect and protect image edges. Third, a high-quality reconstruction Swin transformer (i.e., Recformer) was designed to capture image global features. The experimental results on numerical and real cardiac clinical datasets with 48 views demonstrated that our proposed MIST-net provided better image quality with more small features and sharp edges than other competitors.

Sparse-View CT Reconstruction Based on a Hybrid Domain Model with Multi-Level Wavelet Transform.

The reconstruction of sparsely sampled projection data will generate obvious streaking artifacts, resulting in image quality degradation and affecting medical diagnosis results. Wavelet transform can effectively decompose directional components of image, so the artifact features and edge details with high directionality can be better detected in the wavelet domain. Therefore, a hybrid domain method based on wavelet transform is proposed in this paper for the sparse-view CT reconstruction. The reconstruction model combines wavelet, spatial, and radon domains to restore the projection consistency and enhance image details. In addition, the global distribution of artifacts requires the network to have a large receptive field, so that a multi-level wavelet transform network (MWCNN) is applied to the hybrid domain model. Wavelet transform is used in the encoding part of the network to reduce the size of feature maps instead of pooling operation and inverse wavelet transform is deployed in the decoding part to recover image details. The proposed method can achieve PSNR of 41.049 dB and SSIM of 0.958 with 120 projections of three angular intervals, and obtain the highest values in this paper. Through the results of numerical analysis and reconstructed images, it shows that the hybrid domain method is superior to the single-domain methods. At the same time, the multi-level wavelet transform model is more suitable for CT reconstruction than the single-level wavelet transform.