Model-based Iterative Reconstruction Research Articles

Reconstruction method suitable for fast CT imaging

Reconstructing computed tomography (CT) images from an extremely limited set of projections is crucial in practical applications. As the available projections significantly decrease, traditional reconstruction and model-based iterative reconstruction methods become constrained. This work aims to seek a reconstruction method applicable to fast CT imaging when available projections are highly sparse. To minimize the time and cost associated with projections acquisition, we propose a deep learning model, X-CTReNet, which parameterizes a nonlinear mapping function from orthogonal projections to CT volumes for 3D reconstruction. The proposed model demonstrates effective capability in inferring CT volumes from two-view projections compared to baseline methods, highlighting the significant potential for drastically reducing projection acquisition in fast CT imaging.

Evaluation of four computed tomography reconstruction algorithms using a coronary artery phantom.

Despite advancements in coronary computed tomography angiography (CTA), challenges in positive predictive value and specificity remain due to limited spatial resolution. The purpose of this experimental study was to investigate the effect of 2nd generation deep learning-based reconstruction (DLR) on the quantitative and qualitative image quality in coronary CTA. A vessel model with stepwise non-calcified plaque was scanned using 320-detector CT. Image reconstruction was performed using four techniques: hybrid iterative reconstruction (HIR), model-based iterative reconstruction (MBIR), DLR, and 2nd generation DLR. The luminal peak CT number, contrast-to-noise ratio (CNR), and edge rise slope (ERS) were quantitatively evaluated via profile curve analysis. Two observers qualitatively graded the graininess, lumen sharpness, and overall lumen visibility on the basis of the degree of confidence for the stenosis severity using a five-point scale. The image noise with HIR, MBIR, DLR, and 2nd generation DLR was 23.0, 21.0, 16.9, and 9.5 HU, respectively. The corresponding CNR (25% stenosis) was 15.5, 15.9, 22.1, and 38.3, respectively. The corresponding ERS (25% stenosis) was 203.2, 198.6, 228.9, and 262.4 HU/mm, respectively. Among the four reconstruction methods, the 2nd generation DLR achieved the significantly highest CNR and ERS values. The score of 2nd generation DLR in all evaluation points (graininess, sharpness, and overall lumen visibility) was higher than those of the other methods (overall vessel visibility score, 2.6±0.5, 3.8±0.6, 3.7±0.5, and 4.6±0.5 with HIR, MBIR, DLR, and 2nd generation DLR, respectively). 2nd generation DLR provided better CNR and ERS in coronary CTA than HIR, MBIR, and previous-generation DLR, leading to the highest subjective image quality in the assessment of vessel stenosis.

Super-resolution deep learning reconstruction to improve image quality of coronary CT angiography

Abstract Purpose To compare the objective and subjective image quality and diagnostic performance for coronary stenosis of normal-dose model-based iterative reconstruction and reduced-dose super-resolution deep learning reconstruction in coronary CT angiography. Materials and Methods This single-center retrospective study included 52 patients (mean age, 68 years ± 10 [SD]; 41 men) who underwent serial coronary CT angiography and subsequent invasive coronary angiography between January and November 2022. The first 25 patients were scanned with a standard dose using model-based iterative reconstruction. The last 27 patients were scanned with a reduced dose using super-resolution deep learning reconstruction. Per-patient objective and subjective image qualities were compared. Diagnostic performance of model-based iterative reconstruction and super-resolution deep learning reconstruction to diagnose significant stenosis on coronary angiography was compared per-vessel using receiver operating characteristics curve analysis. Results The median tube current of super-resolution deep learning reconstruction was lower than that of model-based iterative reconstruction (median [IQR], 890 mA [680, 900] vs. 900 mA [895, 900], P = 0.03). Image noise of super-resolution deep learning reconstruction was lower than that of model-based iterative reconstruction (14.6 Hounsfield units ± 1.3 vs. 22.7 Hounsfield units ± 4.4, P < .001). Super-resolution deep learning reconstruction improved the overall subjective image quality compared with model-based iterative reconstruction (median [IQR], 4 [3, 4] vs 3 [3, 3], P = .006). No difference in the area under the receiver operating characteristic curve in diagnosing coronary stenosis using super-resolution deep learning reconstruction (0.96; 95% CI, 0.92-0.99) and model-based iterative reconstruction (0.96; 95% CI, 0.92-0.98; P = .98) was observed. Conclusion Our exploratory analysis suggests that super-resolution deep learning reconstruction could improve image quality with lower tube current settings than model-based iterative reconstruction with similar diagnostic performance to diagnose coronary stenosis in coronary CT angiography.

Model-based iterative reconstruction for direct imaging with point spread function encoded echo planar MRI

BackgroundEcho planar imaging (EPI) is a fast measurement technique commonly used in magnetic resonance imaging (MRI), but is highly sensitive to measurement non-idealities in reconstruction. Point spread function (PSF)-encoded EPI is a multi-shot strategy which alleviates distortion, but acquisition of encodings suitable for direct distortion-free imaging prolongs scan time. In this work, a model-based iterative reconstruction (MBIR) framework is introduced for direct imaging with PSF-EPI to improve image quality and acceleration potential. MethodsAn MBIR platform was developed for accelerated PSF-EPI. The reconstruction utilizes a subspace representation, is regularized to promote local low-rankedness (LLR), and uses variable splitting for efficient iteration. Comparisons were made against standard reconstructions from prospectively accelerated PSF-EPI data and with retrospective subsampling. Exploring aggressive partial Fourier acceleration of the PSF-encoding dimension, additional comparisons were made against an extension of Homodyne to direct PSF-EPI in numerical experiments. A neuroradiologists' assessment was completed comparing images reconstructed with MBIR from retrospectively truncated data directly against images obtained with standard reconstructions from non-truncated datasets. ResultsImage quality results were consistently superior for MBIR relative to standard and Homodyne reconstructions. As the MBIR signal model and reconstruction allow for arbitrary sampling of the PSF space, random sampling of the PSF-encoding dimension was also demonstrated, with quantitative assessments indicating best performance achieved through nonuniform PSF sampling combined with partial Fourier. With retrospective subsampling, MBIR reconstructs high-quality images from sub-minute scan datasets. MBIR was shown to be superior in a neuroradiologists' assessment with respect to three of five performance criteria, with equivalence for the remaining two. ConclusionsA novel image reconstruction framework is introduced for direct imaging with PSF-EPI, enabling arbitrary PSF space sampling and reconstruction of diagnostic-quality images from highly accelerated PSF-encoded EPI data.

Submillimeter balanced SSFP BOLD-functional MRI accelerated with 3D stack-of-spirals at 9.4 T.

This work aims to improve the speed of balanced SSFP (bSSFP) acquisition with segmented 3D stack-of-spirals for functional brain studies at ultrahigh field. Functional experiments were performed with an accelerated 3D stack-of-spirals sequence with water excitation for fat suppression. The resulting data were reconstructed using an iterative algorithm with corrections for system imperfections such as trajectory deviations and B0 inhomogeneity. In the first set of experiments, we evaluated the signal change and stability with respect to echo and TR for a full-field checkerboard stimulus. To demonstrate the high spatio-temporal resolution of the developed method, the results of three optimized protocols at submillimeter resolution (0.6-mm isotropic and 0.8-mm isotropic) and at 1.2 mm isotropic resolution for whole-brain coverage were shown. Water excitation and the model-based iterative reconstruction improved image quality. The BOLD-related signal changes increased with longer TE and longer TR. We observed an increase in thermal noise performance at lower TE and higher TR. However, signal stability deteriorates at higher TE and TR. Therefore, optimized protocols used shorter TE and moderately long TR to maximize the sensitivity and speed. Reproducible activations were detected along the gray-matter gyri in the submillimeter protocols with a median signal change of approximately 4% across subjects. Three-dimensional stack-of-spirals enables passband balanced SSFP functional imaging at a much higher spatial and temporal scale, compared with conventional spoiled gradient-echo train sequences.

Image quality in CT thorax: effect of altering reconstruction algorithm and tube load.

Non-linear properties of iterative reconstruction (IR) algorithms can alter image texture. We evaluated the effect of a model-based IR algorithm (advanced modelled iterative reconstruction; ADMIRE) and dose on computed tomography thorax image quality. Dual-source scanner data were acquired at 20, 45 and 65 reference mAs in 20 patients. Images reconstructed with filtered back projection (FBP) and ADMIRE Strengths 3-5 were assessed independently by six radiologists and analysed using an ordinal logistic regression model. For all image criteria studied, the effects of tube load 20 mAs and all ADMIRE strengths were significant (p<0.001) when compared to reference categories 65 mAs and FBP. Increase in tube load from 45 to 65 mAs showed image quality improvement in three of six criteria. Replacing FBP with ADMIRE significantly improves perceived image quality for all criteria studied, potentially permitting a dose reduction of almost 70% without loss in image quality.

Feasibility of Low-Dose and Low-Contrast Media Volume Approach in Computed Tomography Cardiovascular Imaging Reconstructed with Model-Based Algorithm.

Aim: To evaluate the dose reduction and image quality of low-dose, low-contrast media volume in computed tomography (CT) examinations reconstructed with the model-based iterative reconstruction (MBIR) algorithm in comparison with the hybrid iterative (HIR) one. Methods: We prospectively enrolled a total of 401 patients referred for cardiovascular CT, evaluated with a 256-MDCT scan with a low kVp (80 kVp) reconstructed with an MBIR (study group) or a standard HIR protocol (100 kVp-control group) after injection of a fixed dose of contrast medium volume. Vessel contrast enhancement and image noise were measured by placing the region of interest (ROI) in the left ventricle, ascending aorta; left, right and circumflex coronary arteries; main, right and left pulmonary arteries; aortic arch; and abdominal aorta. The signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) were computed. Subjective image quality obtained by consensus was assessed by using a 4-point Likert scale. Radiation dose exposure was recorded. Results: HU values of the proximal tract of all coronary arteries; main, right and left pulmonary arteries; and of the aorta were significantly higher in the study group than in the control group (p < 0.05), while the noise was significantly lower (p < 0.05). SNR and CNR values in all anatomic districts were significantly higher in the study group (p < 0.05). MBIR subjective image quality was significantly higher than HIR in CCTA and CTPA protocols (p < 0.05). Radiation dose was significantly lower in the study group (p < 0.05). Conclusions: The MBIR algorithm combined with low-kVp can help reduce radiation dose exposure, reduce noise, and increase objective and subjective image quality.

Improving quantitative MRI using self-supervised deep learning with model reinforcement: Demonstration for rapid T1 mapping.

This paper proposes a novel self-supervised learning framework that uses model reinforcement, REference-free LAtent map eXtraction with MOdel REinforcement (RELAX-MORE), for accelerated quantitative MRI (qMRI) reconstruction. The proposed method uses an optimization algorithm to unroll an iterative model-based qMRI reconstruction into a deep learning framework, enabling accelerated MR parameter maps that are highly accurate and robust. Unlike conventional deep learning methods which require large amounts of training data, RELAX-MORE is a subject-specific method that can be trained on single-subject data through self-supervised learning, making it accessible and practically applicable to many qMRI studies. Using quantitative mapping as an example, the proposed method was applied to the brain, knee and phantom data. The proposed method generates high-quality MR parameter maps that correct for image artifacts, removes noise, and recovers image features in regions of imperfect image conditions. Compared with other state-of-the-art conventional and deep learning methods, RELAX-MORE significantly improves efficiency, accuracy, robustness, and generalizability for rapid MR parameter mapping. This work demonstrates the feasibility of a new self-supervised learning method for rapid MR parameter mapping, that is readily adaptable to the clinical translation of qMRI.

No-wait inversion—a novel model for T1 mapping from inversion recovery measurements without the waiting times

Introduction: Quantification of longitudinal relaxation time T1 gained interest as an important MR-inducible tissue property for tissue characterization. Standard inversion recovery (IR) measurements for T1 determination take a prohibitively long time, and signal models assume a perfect inversion. Acceleration is possible by using the Look–Locker (LL) technique or other accelerated, model-based algorithms. However, the calculation of real T1 values from LL acquisitions necessitates the knowledge of equilibrium magnetization M0. Thus, usually, a waiting time to allow for free relaxation between global inversion pulses must be implemented. This study aims to introduce a novel model-based fitting approach for T1 mapping without the need for such waiting times.Methods: Single-inversion spiral LL spoiled gradient echo acquisitions were performed in a phantom and eight healthy volunteers using a 1.5T magnetic resonance imaging (MRI) scanner. The measurements comprised two parts, one without magnetization preparation and a second featuring a global inversion pulse preparation before each of the 35 slices. Acquisition was performed without any waiting time in between slices, i.e., before the inversion pulses. T1 maps were calculated based on an iterative model-based reconstruction algorithm which combines the information from these two measurements, with and without inversion.Results: Accurate T1 maps were obtained in phantom and volunteer measurements. ROI-based mean T1 values differ by an average of 1.5% in the phantom and 5% in vivo between reference measurements and the proposed method. The combined fit benefits from both the information obtained in the inversion prepared and the unprepared measurements. The former provides a large dynamic range for accurate model-based fitting of the relaxation process, while the latter provides equilibrium magnetization M0, necessary to obtain accurate T1 values from a LL-like acquisition.Conclusion: The proposed model of a combined fit of an inversion-prepared and an unprepared measurement allows for robust fast T1 mapping, even in cases of imperfect inversion due to skipped waiting times for magnetization recovery. Thus, it can render long waiting times in between inversion pulses redundant.

TomoCAM: fast model-based iterative reconstruction via GPU acceleration and non-uniform fast Fourier transforms.

X-ray-based computed tomography is a well established technique for determining the three-dimensional structure of an object from its two-dimensional projections. In the past few decades, there have been significant advancements in the brightness and detector technology of tomography instruments at synchrotron sources. These advancements have led to the emergence of new observations and discoveries, with improved capabilities such as faster frame rates, larger fields of view, higher resolution and higher dimensionality. These advancements have enabled the material science community to expand the scope of tomographic measurements towards increasingly in situ and in operando measurements. In these new experiments, samples can be rapidly evolving, have complex geometries and restrictions on the field of view, limiting the number of projections that can be collected. In such cases, standard filtered back-projection often results in poor quality reconstructions. Iterative reconstruction algorithms, such as model-based iterative reconstructions (MBIR), have demonstrated considerable success in producing high-quality reconstructions under such restrictions, but typically require high-performance computing resources with hundreds of compute nodes to solve the problem in a reasonable time. Here, tomoCAM, is introduced, a new GPU-accelerated implementation of model-based iterative reconstruction that leverages non-uniform fast Fourier transforms to efficiently compute Radon and back-projection operators and asynchronous memory transfers to maximize the throughput to the GPU memory. The resulting code is significantly faster than traditional MBIR codes and delivers the reconstructive improvement offered by MBIR with affordable computing time and resources. tomoCAM has a Python front-end, allowing access from Jupyter-based frameworks, providing straightforward integration into existing workflows at synchrotron facilities.

SOUL-Net: A Sparse and Low-Rank Unrolling Network for Spectral CT Image Reconstruction.

Spectral computed tomography (CT) is an emerging technology, that generates a multienergy attenuation map for the interior of an object and extends the traditional image volume into a 4-D form. Compared with traditional CT based on energy-integrating detectors, spectral CT can make full use of spectral information, resulting in high resolution and providing accurate material quantification. Numerous model-based iterative reconstruction methods have been proposed for spectral CT reconstruction. However, these methods usually suffer from difficulties such as laborious parameter selection and expensive computational costs. In addition, due to the image similarity of different energy bins, spectral CT usually implies a strong low-rank prior, which has been widely adopted in current iterative reconstruction models. Singular value thresholding (SVT) is an effective algorithm to solve the low-rank constrained model. However, the SVT method requires a manual selection of thresholds, which may lead to suboptimal results. To relieve these problems, in this article, we propose a sparse and low-rank unrolling network (SOUL-Net) for spectral CT image reconstruction, that learns the parameters and thresholds in a data-driven manner. Furthermore, a Taylor expansion-based neural network backpropagation method is introduced to improve the numerical stability. The qualitative and quantitative results demonstrate that the proposed method outperforms several representative state-of-the-art algorithms in terms of detail preservation and artifact reduction.

Exploration of differentiability in a proton computed tomography simulation framework

Objective. Gradient-based optimization using algorithmic derivatives can be a useful technique to improve engineering designs with respect to a computer-implemented objective function. Likewise, uncertainty quantification through computer simulations can be carried out by means of derivatives of the computer simulation. However, the effectiveness of these techniques depends on how ‘well-linearizable’ the software is. In this study, we assess how promising derivative information of a typical proton computed tomography (pCT) scan computer simulation is for the aforementioned applications. Approach. This study is mainly based on numerical experiments, in which we repeatedly evaluate three representative computational steps with perturbed input values. We support our observations with a review of the algorithmic steps and arithmetic operations performed by the software, using debugging techniques. Main results. The model-based iterative reconstruction (MBIR) subprocedure (at the end of the software pipeline) and the Monte Carlo (MC) simulation (at the beginning) were piecewise differentiable. However, the observed high density and magnitude of jumps was likely to preclude most meaningful uses of the derivatives. Jumps in the MBIR function arose from the discrete computation of the set of voxels intersected by a proton path, and could be reduced in magnitude by a ‘fuzzy voxels’ approach. The investigated jumps in the MC function arose from local changes in the control flow that affected the amount of consumed random numbers. The tracking algorithm solves an inherently non-differentiable problem. Significance. Besides the technical challenges of merely applying AD to existing software projects, the MC and MBIR codes must be adapted to compute smoother functions. For the MBIR code, we presented one possible approach for this while for the MC code, this will be subject to further research. For the tracking subprocedure, further research on surrogate models is necessary.

Model-based deep CNN-regularized reconstruction for digital breast tomosynthesis with a task-based CNN image assessment approach

Objective. Digital breast tomosynthesis (DBT) is a quasi-three-dimensional breast imaging modality that improves breast cancer screening and diagnosis because it reduces fibroglandular tissue overlap compared with 2D mammography. However, DBT suffers from noise and blur problems that can lower the detectability of subtle signs of cancers such as microcalcifications (MCs). Our goal is to improve the image quality of DBT in terms of image noise and MC conspicuity. Approach. We proposed a model-based deep convolutional neural network (deep CNN or DCNN) regularized reconstruction (MDR) for DBT. It combined a model-based iterative reconstruction (MBIR) method that models the detector blur and correlated noise of the DBT system and the learning-based DCNN denoiser using the regularization-by-denoising framework. To facilitate the task-based image quality assessment, we also proposed two DCNN tools for image evaluation: a noise estimator (CNN-NE) trained to estimate the root-mean-square (RMS) noise of the images, and an MC classifier (CNN-MC) as a DCNN model observer to evaluate the detectability of clustered MCs in human subject DBTs. Main results. We demonstrated the efficacies of CNN-NE and CNN-MC on a set of physical phantom DBTs. The MDR method achieved low RMS noise and the highest detection area under the receiver operating characteristic curve (AUC) rankings evaluated by CNN-NE and CNN-MC among the reconstruction methods studied on an independent test set of human subject DBTs. Significance. The CNN-NE and CNN-MC may serve as a cost-effective surrogate for human observers to provide task-specific metrics for image quality comparisons. The proposed reconstruction method shows the promise of combining physics-based MBIR and learning-based DCNNs for DBT image reconstruction, which may potentially lead to lower dose and higher sensitivity and specificity for MC detection in breast cancer screening and diagnosis.

Accelerated model-based iterative reconstruction strategy for sparse-view photoacoustic tomography aided by multi-channel autoencoder priors.

Photoacoustic tomography (PAT) commonly works in sparse view due to data acquisition limitations. However, reconstruction suffers from serious deterioration (e.g., severe artifacts) using traditional algorithms under sparse view. Here, a novel accelerated model-based iterative reconstruction strategy for sparse-view PAT aided by multi-channel autoencoder priors was proposed. A multi-channel denoising autoencoder network was designed to learn prior information, which provides constraints for model-based iterative reconstruction. This integration accelerates the iteration process, leading to optimal reconstruction outcomes. The performance of the proposed method was evaluated using blood vessel simulation data and experimental data. The results show that the proposed method can achieve superior sparse-view reconstruction with a significant acceleration of iteration. Notably, the proposed method exhibits excellent performance under extremely sparse condition (e.g., 32 projections) compared with the U-Net method, with an improvement of 48% in PSNR and 12% in SSIM for in vivo experimental data.

Application of double low-dose mode in left atrial-pulmonary venous computed tomography angiography

This study adopted a 256-slice iCT scanner with the double low-dose mode in left atrial-pulmonary venous computed tomography angiography (CTA) and explored its effect on image quality. 120 patients were included and randomly classified into the Observation group and Control group. Patients in the Control group underwent routine left atrial CTA, while patients in the Observation group performed a double low-dose mode. Other scanning parameters were consistent in the two groups. The Full model-based iterative reconstruction (MBIR) technique was applied to fulfill image reconstruction in observation group. Continuous variables, ordered categorical variables were analyzed by statistical test. The CT values of left atrial in the Observation group were significantly higher than those in the Control group. The exposure doses (ED) and iodine intake were lower in the Observation group, as compared to the Control group. The left atrial-pulmonary venous CTA with the 256-slice iCT scanner in a double low-dose mode can reduce the ED of radiation and iodine contrast while providing high quality images. Comparatively, the ED in the Observation group was reduced by 13% compared with the control, and the iodine intake was reduced by approximately 33%.

Innovative advances in pediatric radiology: computed tomography reconstruction techniques, photon-counting detector computed tomography, and beyond.

In pediatric radiology, balancing diagnostic accuracy with reduced radiation exposure is paramount due to the heightened vulnerability of younger patients to radiation. Technological advancements in computed tomography (CT) reconstruction techniques, especially model-based iterative reconstruction and deep learning image reconstruction, have enabled significant reductions in radiation doses without compromising image quality. Deep learning image reconstruction, powered by deep learning algorithms, has demonstrated superiority over traditional techniques like filtered back projection, providing enhanced image quality, especially in pediatric head and cardiac CT scans. Photon-counting detector CT has emerged as another groundbreaking technology, allowing for high-resolution images while substantially reducing radiation doses, proving highly beneficial for pediatric patients requiring frequent imaging. Furthermore, cloud-based dose tracking software focuses on monitoring radiation exposure, ensuring adherence to safety standards. However, the deployment of these technologies presents challenges, including the need for large datasets, computational demands, and potential data privacy issues. This article provides a comprehensive exploration of these technological advancements, their clinical implications, and the ongoing efforts to enhance pediatric radiology's safety and effectiveness.

Synthetization of high-dose images using low-dose CT scans.

Radiation dose reduction has been the focus of many research activities in x-ray CT. Various approaches were taken to minimize the dose to patients, ranging from the optimization of clinical protocols, refinement of the scanner hardware design, and development of advanced reconstruction algorithms. Although significant progress has been made, more advancements in this area are needed to minimize the radiation risks to patients. Reconstruction algorithm-based dose reduction approaches focus mainly on the suppression of noise in the reconstructed images while preserving detailed anatomical structures. Such an approach effectively produces synthesized high-dose images (SHD) from the data acquired with low-dose scans. A representative example is the model-based iterative reconstruction (MBIR). Despite its widespread deployment, its full adoption in a clinical environment is often limited by an undesirable image texture. Recent studies have shown that deep learning image reconstruction (DLIR) can overcome this shortcoming. However, the limited availability of high-quality clinical images for training and validation is often the bottleneck for its development. In this paper, we propose a novel approach to generate SHD with existing low-dose clinical datasets that overcomes both the noise texture issue and the data availability issue. Our approach is based on the observation that noise in the image can be effectively reduced by performing image processing orthogonal to the imaging plane. This process essentially creates an equivalent thick-slice image (TSI), and the characteristics of TSI depend on the nature of the image processing. An advantage of this approach is its potential to reduce impact on the noise texture. The resulting image, however, is likely corrupted by the anatomical structural degradation due to partial volume effects. Careful examination has shown that the differential signal between the original and the processed image contains sufficient information to identify regions where anatomical structures are modified. The differential signal, unfortunately, contains significant noise and has to be removed. The noise removal can be accomplished by performing iterative noise reduction to preserve structural information. The processed differential signal is subsequently subtracted from TSI to arrive at SHD. The algorithm was evaluated extensively with phantom and clinical datasets. For better visual inspection, difference images between the original and SHD were generated and carefully examined. Negligible residual structure could be observed. In addition to the qualitative inspection, quantitative analyses were performed on clinical images in terms of the CT number consistency and the noise reduction characteristics. Results indicate that no CT number bias is introduced by the proposed algorithm. In addition, noise reduction capability is consistent across different patient anatomical regions. Further, simulated water phantom scans were utilized in the generation of the noise power spectrum (NPS) to demonstrate the preservation of the noise-texture. We present a method to generate SHD datasets from regularly acquired low-dose CT scans. Images produced with the proposed approach exhibit excellent noise-reduction with the desired noise-texture. Extensive clinical and phantom studies have demonstrated the efficacy and robustness of our approach. Potential limitations of the current implementation are discussed and further research topics are outlined.

Super-resolution deep learning reconstruction at coronary computed tomography angiography to evaluate the coronary arteries and in-stent lumen: an initial experience

A super-resolution deep learning reconstruction (SR-DLR) algorithm trained using data acquired on the ultrahigh spatial resolution computed tomography (UHRCT) has the potential to provide better image quality of coronary arteries on the whole-heart, single-rotation cardiac coverage on a 320-detector row CT scanner. However, the advantages of SR-DLR at coronary computed tomography angiography (CCTA) have not been fully investigated. The present study aimed to compare the image quality of the coronary arteries and in-stent lumen between SR-DLR and model-based iterative reconstruction (MBIR). We prospectively enrolled 70 patients (median age, 69 years; interquartile range [IQR], 59–75 years; 50 men) who underwent CCTA using a 320-detector row CT scanner between January and August 2022. The image noise in the ascending aorta, left atrium, and septal wall of the ventricle was measured, and the signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) in the proximal coronary arteries were calculated. Of the twenty stents, stent strut thickness and luminal diameter were quantitatively evaluated. The image noise on SR-DLR was significantly lower than that on MBIR (median 22.1 HU; IQR, 19.3–24.9 HU vs. 27.4 HU; IQR, 24.2–31.2 HU, p < 0.01), whereas the SNR (median 16.3; IQR, 11.8–21.8 vs. 13.7; IQR, 9.9–18.4, p = 0.01) and CNR (median 24.4; IQR, 15.5–30.2 vs. 19.2; IQR, 14.1–23.2, p < 0.01) on SR-DLR were significantly higher than that on MBIR. Stent struts were significantly thinner (median, 0.68 mm; IQR, 0.61–0.78 mm vs. 0.81 mm; IQR, 0.72–0.96 mm, p < 0.01) and in-stent lumens were significantly larger (median, 1.84 mm; IQR, 1.65–2.26 mm vs. 1.52 mm; IQR, 1.28–2.25 mm, p < 0.01) on SR-DLR than on MBIR. Although further large-scale studies using invasive coronary angiography as the reference standard, comparative studies with UHRCT, and studies in more challenging population for CCTA are needed, this study’s initial experience with SR-DLR would improve the utility of CCTA in daily clinical practice due to the better image quality of the coronary arteries and in-stent lumen at CCTA compared with conventional MBIR.

Physical characteristics of deep learning-based image processing software in computed tomography: a phantom study.

This study aimed to assess the image characteristics of deep-learning-based image processing software (DLIP; FCT PixelShine, FUJIFILM, Tokyo, Japan) and compare it with filtered back projection (FBP), model-based iterative reconstruction (MBIR), and deep-learning-based reconstruction (DLR). This phantom study assessed the object-specific spatial resolution (task-based transfer function [TTF]), noise characteristics (noise power spectrum [NPS]), and low-contrast detectability (low-contrast object-specific contrast-to-noise ratio [CNRLO]) at three different output doses (standard: 10mGy; low: 3.9mGy; ultralow: 2.0mGy). The processing strength of DLIPFBP with A1, A4, and A9 was compared with those of FBP, MBIR, and DLR. The standard dose with high-contrast TTFs of DLIPFBP exceeded that of FBP. Low-contrast TTFs were comparable to or lower than that of FBP. The NPS peak frequency (fP) of DLIPFBP shifts to low spatial frequencies of up to 8.6% at ultralow doses compared to the standard FBP dose. MBIR shifted the most fP compared to FBP-a marked shift of up to 49%. DLIPFBP showed a CNRLO equal to or greater than that of DLR in standard or low doses. In contrast, the CNRLO of the DLIPFBP was equal to or lower than that of the DLR in ultralow doses. DLIPFBP reduced image noise while maintaining a resolution similar to commercially available MBIR and DLR. The slight spatial frequency shift of fP in DLIPFBP contributed to the noise texture degradation suppression. The NPS suppression in the low spatial frequency range effectively improved the low-contrast detectability.

Sparse-view reconstruction for photoacoustic tomography combining diffusion model with model-based iteration

As a non-invasive hybrid biomedical imaging technology, photoacoustic tomography combines high contrast of optical imaging and high penetration of acoustic imaging. However, the conventional standard reconstruction under sparse view could result in low-quality image in photoacoustic tomography. Here, a novel model-based sparse reconstruction method for photoacoustic tomography via diffusion model was proposed. A score-based diffusion model is designed for learning the prior information of the data distribution. The learned prior information is utilized as a constraint for the data consistency term of an optimization problem based on the least-square method in the model-based iterative reconstruction, aiming to achieve the optimal solution. Blood vessels simulation data and the animal in vivo experimental data were used to evaluate the performance of the proposed method. The results demonstrate that the proposed method achieves higher-quality sparse reconstruction compared with conventional reconstruction methods and U-Net. In particular, under the extreme sparse projection (e.g., 32 projections), the proposed method achieves an improvement of ∼ 260 % in structural similarity and ∼ 30 % in peak signal-to-noise ratio for in vivo data, compared with the conventional delay-and-sum method. This method has the potential to reduce the acquisition time and cost of photoacoustic tomography, which will further expand the application range.