Prior Image Research Articles

MRWM: A Multiple Residual Wasserstein Driven Model for Image Denoising

Residual histograms can provide valuable information for vision research. However, current image restoration methods have not fully exploited the potential of multiple residual histograms, especially their role as overall regularization constraints. In this paper, we propose a novel framework of multiple residual Wasserstein driven model (MRWM) that can organically combine multiple residual Wasserstein constraints and various natural image priors for image denoising. Specifically, by utilizing the Wasserstein distance derived from the optimal transmission theory, the multiple residual histograms of the observed images are forced to be as close as possible to the reference residual histogram, thereby improving the accuracy of residual estimation. Furthermore, the proposed concrete MRWM unifies the multiple residual Wasserstein distribution approximation and the image total variation prior knowledge to carry out image denoising. Alternating iterative algorithm of histogram matching and Chambolle dual projection has the characteristics of less parameters and easy implementation. Finally, our experiments confirm that compared with some representative image denoising algorithms, the MRWM can obtain better performance in objective evaluation, and can better preserve the details such as the image edges, making the image look more natural.

Deep Hyperspectral Image Fusion Network With Iterative Spatio-Spectral Regularization

Physical acquisition of high-resolution hyperspectral images (HR-HSI) has remained difficult, despite its potential of resolving material-related ambiguities in vision applications. Deep hyperspectral image fusion, aiming at reconstructing an HR-HSI from a pair of low-resolution hyperspectral image (LR-HSI) and high-resolution multispectral image (HR-MSI), has become an appealing computational alternative. Existing fusion methods either rely on hand-crafted image priors or treat fusion as a nonlinear mapping problem, ignoring important physical imaging models. In this paper, we propose a novel regularization strategy to fully exploit the spatio-spectral dependency by a spatially adaptive 3D filter. Moreover, the joint exploitation of spatio-spectral regularization and physical imaging models inspires us to formulate deep hyperspectral image fusion as a differentiable optimization problem. We show how to solve this optimization problem by an end-to-end training of a model-guided unfolding network named DHIF-Net. Unlike existing works of simply concatenating spatial with spectral regularization, our approach aims at an end-to-end optimization of <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">iterative</i> spatio-spectral regularization by multistage network implementations. Our extensive experimental results on both synthetic and real datasets have shown that our DHIF-Net outperforms other competing methods in terms of both objective and subjective visual quality.

Structure-Texture Parallel Embedding for Remote Sensing Image Super-Resolution

The structure and texture of images are crucial for remote sensing image super-resolution. Generative adversarial networks (GANs) recover image details through adversarial training. However, the recovered images always have structural distortions on the one hand, and GANs are difficult to train on the other hand. In addition, some methods assist reconstruction by introducing prior information of the image, but this brings additional computational cost. To address this issue, we propose a novel structure-texture parallel embedding (SPE) method for super-resolution (SR) of remote sensing images. Our method does not require additional image priors to reconstruct high-quality images. Specifically, we use the global structure information and local texture information of the image in the ascending space to guide the reconstruction result of the image. Firstly, we design a structure preserving block (SPB) to extract global structural features in the ascending space of the image, so as to obtain global structure information for a priori representation. Then, we design a local texture attention module (LTAM) to restore richer texture details. We have conducted lots of experiments on Draper public dataset. Experimental results show that our proposed method not only achieves a better trade-off between computational cost and performance, but also outperforms the existing several SR methods in terms of objective index evaluation and subjective visual effects.

Case Report of a Giant Tarlov Cyst and the Importance of Advanced Imaging Prior to Interventional Spine Procedures

Case Report of a Giant Tarlov Cyst and the Importance of Advanced Imaging Prior to Interventional Spine Procedures

Prior image-guided cone-beam computed tomography augmentation from under-sampled projections using a convolutional neural network.

Acquiring sparse-view cone-beam computed tomography (CBCT) is an effective way to reduce the imaging dose. However, images reconstructed by the conventional filtered back-projection method suffer from severe streak artifacts due to the projection under-sampling. Existing deep learning models have demonstrated feasibilities in restoring volumetric structures from the highly under-sampled images. However, because of the inter-patient variabilities, they failed to restore the patient-specific details with the common restoring pattern learned from the group data. Although the patient-specific models have been developed by training models using the intra-patient data and have shown effectiveness in restoring the patient-specific details, the models have to be retrained to be exclusive for each patient. It is highly desirable to develop a generalized model that can utilize the patient-specific information for the under-sampled image augmentation. In this study, we proposed a merging-encoder convolutional neural network (MeCNN) to realize the prior image-guided under-sampled CBCT augmentation. Instead of learning the patient-specific structures, the proposed model learns a generalized pattern of utilizing the patient-specific information in the prior images to facilitate the under-sampled image enhancement. Specifically, the MeCNN consists of a merging-encoder and a decoder. The merging-encoder extracts image features from both the prior CT images and the under-sampled CBCT images, and merges the features at multi-scale levels via deep convolutions. The merged features are then connected to the decoders via shortcuts to yield high-quality CBCT images. The proposed model was tested on both the simulated CBCTs and the clinical CBCTs. The predicted CBCT images were evaluated qualitatively and quantitatively in terms of image quality and tumor localization accuracy. Mann-Whitney U test was conducted for the statistical analysis. P<0.05 was considered statistically significant. The proposed model yields CT-like high-quality CBCT images from only 36 half-fan projections. Compared to other methods, CBCT images augmented by the proposed model have significantly lower intensity errors, significantly higher peak signal-to-noise ratio, and significantly higher structural similarity with respect to the ground truth images. Besides, the proposed method significantly reduced the 3D distance of the CBCT-based tumor localization errors. In addition, the CBCT augmentation is nearly real-time. With the prior-image guidance, the proposed method is effective in reconstructing high-quality CBCT images from the highly under-sampled projections, considerably reducing the imaging dose and improving the clinical utility of the CBCT.

Dual-wavelength in-line digital holography with untrained deep neural networks

Dual-wavelength in-line digital holography (DIDH) is one of the popular methods for quantitative phase imaging of objects with non-contact and high-accuracy features. Two technical challenges in the reconstruction of these objects include suppressing the amplified noise and the twin-image that respectively originate from the phase difference and the phase-conjugated wavefronts. In contrast to the conventional methods, the deep learning network has become a powerful tool for estimating phase information in DIDH with the assistance of noise suppressing or twin-image removing ability. However, most of the current deep learning-based methods rely on supervised learning and training instances, thereby resulting in weakness when it comes to applying this training to practical imaging settings. In this paper, a new DIDH network (DIDH-Net) is proposed, which encapsulates the prior image information and the physical imaging process in an untrained deep neural network. The DIDH-Net can effectively suppress the amplified noise and the twin-image of the DIDH simultaneously by automatically adjusting the weights of the network. The obtained results demonstrate that the proposed method with robust phase reconstruction is well suited to improve the imaging performance of DIDH.

Multispectral interlaced sparse sampling photoacoustic tomography based on directional total variation

Background and objectivePhotoacoustic tomography (PAT) is capable of obtaining cross-sectional images of small animals that represent the optical absorption of biological tissues. The multispectral Interlaced Sparse Sampling PAT, or ISS-PAT, is a previously proposed PAT imaging method that offered high quality images with much sparser transducer angular coverage. Although it provides superior imaging performance, the original ISS-PAT method suffered from a heavy computation burden, which hinders its practical application. MethodsHere, we propose a new regularization scheme based on the directional total variation (dTV) for ISS-PAT. This method efficiently imposes the structural information by considering both the edge position and direction information of the anatomical prior image in ISS-PAT. It does not require image segmentation, and can be conveniently solved by a modified alternating direction of multipliers (ADMM) algorithm. ResultsWe perform simulation, tissue mimicking phantom and in vivo small animal experiments to evaluate the proposed scheme. The reconstructed PAT images showed image quality and spectral un-mixing accuracy close to those obtained by non-local means based ISS-PAT, but with much shorter image reconstruction time. For a 1/6 sparse sampling rate, the average efficiency improvement is nearly 16-folds. ConclusionsThe experimental results demonstrate the feasibility of the dTV regularization scheme for ISS-PAT. Its efficient image reconstruction performance facilitates the potential of the hardware realization and practical applications of the ISS-PAT.

Automatic liver tumor localization using deep learning-based liver boundary motion estimation and biomechanical modeling (DL-Bio).

Recently, two-dimensional-to-three-dimensional (2D-3D) deformable registration has been applied to deform liver tumor contours from prior reference images onto estimated cone-beam computed tomography (CBCT) target images to automate on-board tumor localizations. Biomechanical modeling has also been introduced to fine-tune the intra-liver deformation-vector-fields (DVFs) solved by 2D-3D deformable registration, especially at low-contrast regions, using tissue elasticity information and liver boundary DVFs. However, the caudal liver boundary shows low contrast from surrounding tissues in the cone-beam projections, which degrades the accuracy of the intensity-based 2D-3D deformable registration there and results in less accurate boundary conditions for biomechanical modeling. We developed a deep-learning (DL)-based method to optimize the liver boundary DVFs after 2D-3D deformable registration to further improve the accuracy of subsequent biomechanical modeling and liver tumor localization. The DL-based network was built based on the U-Net architecture. The network was trained in a supervised fashion to learn motion correlation between cranial and caudal liver boundaries to optimize the liver boundary DVFs. Inputs of the network had three channels, and each channel featured the 3D DVFs estimated by the 2D-3D deformable registration along one Cartesian direction (x, y, z). To incorporate patient-specific liver boundary information into the DVFs, the DVFs were masked by a liver boundary ring structure generated from the liver contour of the prior reference image. The network outputs were the optimized DVFs along the liver boundary with higher accuracy. From these optimized DVFs, boundary conditions were extracted for biomechanical modeling to further optimize the solution of intra-liver tumor motion. We evaluated the method using 34 liver cancer patient cases, with 24 for training and 10 for testing. We evaluated and compared the performance of three methods: 2D-3D deformable registration, 2D-3D-Bio (2D-3D deformable registration with biomechanical modeling), and DL-Bio (DL model prediction with biomechanical modeling). The tumor localization errors were quantified through calculating the center-of-mass-errors (COMEs), DICE coefficients, and Hausdorff distance between deformed liver tumor contours and manually segmented "gold-standard" contours. The predicted DVFs by the DL model showed improved accuracy at the liver boundary, which translated into more accurate liver tumor localizations through biomechanical modeling. On a total of 90 evaluated images and tumor contours, the average (± sd) liver tumor COMEs of the 2D-3D, 2D-3D-Bio, and DL-Bio techniques were 4.7 ± 1.9mm, 2.9 ± 1.0mm, and 1.7 ± 0.4mm. The corresponding average (± sd) DICE coefficients were 0.60 ± 0.12, 0.71 ± 0.07, and 0.78 ± 0.03; and the average (± sd) Hausdorff distances were 7.0 ± 2.6mm, 5.4 ± 1.5mm, and 4.5 ± 1.3mm, respectively. DL-Bio solves a general correlation model to improve the accuracy of the DVFs at the liver boundary. With improved boundary conditions, the accuracy of biomechanical modeling can be further increased for accurate intra-liver low-contrast tumor localization.

On Architecture Selection for Linear Inverse Problems with Untrained Neural Networks.

In recent years, neural network based image priors have been shown to be highly effective for linear inverse problems, often significantly outperforming conventional methods that are based on sparsity and related notions. While pre-trained generative models are perhaps the most common, it has additionally been shown that even untrained neural networks can serve as excellent priors in various imaging applications. In this paper, we seek to broaden the applicability and understanding of untrained neural network priors by investigating the interaction between architecture selection, measurement models (e.g., inpainting vs. denoising vs. compressive sensing), and signal types (e.g., smooth vs. erratic). We motivate the problem via statistical learning theory, and provide two practical algorithms for tuning architectural hyperparameters. Using experimental evaluations, we demonstrate that the optimal hyperparameters may vary significantly between tasks and can exhibit large performance gaps when tuned for the wrong task. In addition, we investigate which hyperparameters tend to be more important, and which are robust to deviations from the optimum.

Association between diffuse idiopathic skeletal hyperostosis and long-term use of isotretinoin: case report

Diffuse Idiopathic Skeletal Hyperostosis (DISH) is a systemic condition that affects mainly the anterolateral spine, with abundant bone growth and formation of bridges between the vertebral bodies, but ossification in peripheral entheses may also occur. Its prevalence increases with age and is higher in males. Diagnosis at ages under 40 years are extremely rare. Hence, this report aims to describe a rare case of DISH in a young woman probably due to the long-term use of isotretinoin and discuss its presentation, bone involvement and differential diagnosis. A 37-year-old female patient was admitted to the Rheumatology service with progressive limitation of neck rotation and back flexion during the past 3 years, associated with great loss of function and impairment in daily activities. She had a previous diagnosis of congenital ichthyosiform erythroderma, managed with low doses of isotretinoin for 12 years. The current weekly isotretinoin dosage was 180mg. Prior image exams showed axial spondylopathy with diffuse peripheric enthesopathy since she was 33 years old and sacroiliitis with asymmetric syndesmophytes in the lumbar spine that resembled a psoriatic arthritis. Other diseases were denied. Family history included a case of rheumatoid arthritis but no history of psoriasis. Physical exam included erythema and diffuse skin peeling in the upper and lower limbs. Her hands were deformed with ulnar deviation but had no signs of arthritis. Cervical rotation was stiff and ranged 30o to the left and 35o to the right. Schober test distanced 1cm and Patrick’s Faber test was negative. Laboratory tests showed no alteration except for a serum 25-OH vitamin D of 7.25ng/mL (reference value: &gt;30ng/mL). Biochemical and bone turnover markers were normal, including serum total calcium, magnesium, phosphorus, parathyroid hormone and alkaline phosphatase. Inflammatory markers C-reactive protein and erythrocyte sedimentation rate were within normal ranges. Human leukocyte antigen B27 was negative. Conventional radiographies were performed and showed increased bone texture with calcification of anterior and lateral ligaments of the cervical, thoracic and lumbar spine. Bone mineral density, evaluated by dual-energy X-ray absorptiometry (DXA), showed a Z-score of +3.2 SD at lumbar spine. Trabecular bone score by DXA was normal and vertebral fracture assessment showed no fracture and confirmed the formation of bridges between vertebral bodies. Clinical presentation was not compatible with ankylosing spondylitis nor psoriatic arthritis as there was no typical inflammatory pain, no elevated inflammatory markers, no typical skin lesions and no family history of psoriasis. However, the hyperostosis affecting the spine and the diffuse enthesopathy were very exuberant for DISH, but the female sex, young age and bilateral spine involvement were uncommon finds. As this patient was on long-term isotretinoin therapy, DISH development is more likely an adverse effect of retinoids on bone metabolism. Thus, clinicians and dermatologists should be always alert to the adverse effects of prolonged use of isotretinoin and, in cases of movement restriction, possible bone involvement must be further evaluated, as DISH may be a differential diagnosis. Since there is no specific treatment, early diagnosis can be crucial to avoid extensive bone damage and irreversible morbidity.

Image Denoising Using Nonlocal Regularized Deep Image Prior

Deep neural networks have shown great potential in various low-level vision tasks, leading to several state-of-the-art image denoising techniques. Training a deep neural network in a supervised fashion usually requires the collection of a great number of examples and the consumption of a significant amount of time. However, the collection of training samples is very difficult for some application scenarios, such as the full-sampled data of magnetic resonance imaging and the data of satellite remote sensing imaging. In this paper, we overcome the problem of a lack of training data by using an unsupervised deep-learning-based method. Specifically, we propose a deep-learning-based method based on the deep image prior (DIP) method, which only requires a noisy image as training data, without any clean data. It infers the natural images with random inputs and the corrupted observation with the help of performing correction via a convolutional network. We improve the original DIP method as follows: Firstly, the original optimization objective function is modified by adding nonlocal regularizers, consisting of a spatial filter and a frequency domain filter, to promote the gradient sparsity of the solution. Secondly, we solve the optimization problem with the alternating direction method of multipliers (ADMM) framework, resulting in two separate optimization problems, including a symmetric U-Net training step and a plug-and-play proximal denoising step. As such, the proposed method exploits the powerful denoising ability of both deep neural networks and nonlocal regularizations. Experiments validate the effectiveness of leveraging a combination of DIP and nonlocal regularizers, and demonstrate the superior performance of the proposed method both quantitatively and visually compared with the original DIP method.

Change detection in sparse repeat CT scans with non-rigid deformations.

Detecting and interpreting changes in the images of follow-up CT scans by the clinicians is often time-consuming and error-prone due to changes in patient position and non-rigid anatomy deformations. Thus, reconstructed repeat scan images are required, precluding reduced dose sparse-view repeat scanning. A method to automatically detect changes in a region of interest of sparse-view repeat CT scans in the presence of non-rigid deformations of the patient's anatomy without reconstructing the original images. The proposed method uses the sparse sinogram data of two CT scans to distinguish between genuine changes in the repeat scan and differences due to non-rigid anatomic deformations. First, size and contrast level of the changed regions are estimated from the difference between the scans' sinogram data. The estimated types of changes in the repeat scan help optimize the method's parameter values. Two scans are then aligned using Radon space non-rigid registration. Rays which crossed changes in the ROI are detected and back-projected onto image space in a two-phase procedure. These rays form a likelihood map from which the binary changed region map is computed. Experimental studies on four pairs of clinical lung and liver CT scans with simulated changed regions yield a mean changed region recall rate > 86%and a mean precision rate > 83%when detecting large changes with low contrast, and high contrast changes, even when small. The new method outperforms image space methods using prior image constrained compressed sensing (PICCS) reconstruction, particularly for small, low contrast changes (recall = 15.8%, precision = 94.7%). Our method for automatic change detection in sparse-view repeat CT scans with non-rigid deformations may assist radiologists by highlighting the changed regions and may obviate the need for a high-quality repeat scan image when no changes are detected.

Prior ensemble learning : Theory and application to MR image priors.

Compressed sensing (CS) reduces the measurement time of magnetic resonance (MR) imaging, where the use of regularizers or image priors are key techniques to boost reconstruction precision. The optimal prior generally depends on the subject and the hand-building of priors is hard. A methodology of combining priors to create a better one would be useful for various forms of image processing that use image priors. We propose a theory, called prior ensemble learning (PEL), which combines many weak priors (not limited to images) efficiently and approximates the posterior mean (PM) estimate, which is Bayes optimal for minimizing the mean squared error (MSE). The way of combining priors is changed from that of an exponential family to a mixture family. We applied PEL to an undersampled (10%) multicoil MR image reconstruction task. We demonstrated that PEL could combine 136 image priors (norm-based priors such as total variation (TV) and wavelets with various regularization coefficient (RC) values) from only two training samples and that it was superior to the CS-SENSE-based method in terms of the MSE of the reconstructed image. The resulting combining weights were sparse (18% of the weak priors remained), as expected. By the theory, the PM estimator was decomposed into the sparse weighted sum of each weak prior's PM estimator, and the exponential computational complexity for RCs was reduced to polynomial order w.r.t. the number of weak priors. PEL is feasible and effective for a practical MR image reconstruction task.

Rapid noise removal based dual adversarial network for the Brillouin optical time domain analyzer

We propose a dual adversarial network (DANet) to improve the signal-to-noise ratio (SNR) of the Brillouin optical time domain analyzer. Rather than inferring the conditional posteriori distribution in the conventional maximum a posteriori (MAP) framework, DANet constructs a joint distribution from two different factorizations corresponding to the noise removal and generation tasks. This method utilizes all the information between the clean-noisy image pairs to preserve data completely without requiring traditional image priors and noise distribution assumptions. Additionally, the clean-noisy image pairs produced by the generator can expand the original dataset to retrain and enhance the denoising effect. The performance of DANet is verified using the simulated and experimental data. Without spatial resolution deterioration, an SNR improvement of 35.51 dB is observed in the simulation, and the Brillouin frequency shift (BFS) uncertainty along the fiber is reduced by 3.56 MHz. Experiments yield a maximum SNR improvement of 19.08 dB, with the BFS uncertainty along the fiber reduced by 0.93MHz. Significantly, DANet has a processing time of 1.26 s, which is considerably faster than conventional methods, demonstrating its potential for rapid noise removal tasks.

ACUTE MESENTERIC ISCHEMIA: A RARE BUT FATAL COMPLICATION IN COVID-19

TOPIC: Critical Care TYPE: Fellow Case Reports INTRODUCTION: Coronavirus disease 2019 (COVID-19) has initiated a global health-care crisis. It is known to cause multi-organ failure with respiratory failure at the helm of its clinical presentation. There is limited data known about patient's experiencing acute mesenteric thrombosis. We present a case of patient admitted with COVID-19 pneumonia who experienced pneumoperitoneum secondary to acute mesenteric ischemia. CASE PRESENTATION: 64-year-old male with past medical history of hypertension, chronic obstructive pulmonary disease, and former smoker presented with dyspnea, decreased appetite, nausea, and diarrhea for 10 days. On Admission he was hypoxic on room air. Lab work revealed polycythemia, lactic acidosis 3.3, D-dimer 0.87, ABG (respiratory alkalosis with underlying metabolic acidosis), elevated inflammatory markers, and positive COVID-19 PCR. Chest X ray showed diffuse pulmonary opacities. CT angiography of the chest ruled out pulmonary embolism but showed diffuse ground-glass opacities (GGOs). He was placed on High flow nasal cannula (HFNC) at 50% FiO2, and Dexamethasone 6 mg. On hospital day 8 he was tachycardic, more dyspneic and hypoxemic requiring escalation of HFNC. EKG revealed sinus tachycardia. He was empirically started on heparin drip for concern of pulmonary embolism and broad-spectrum antibiotics for suspected bacterial pneumonia. CT chest without contrast showed worsening GGOs and pneumoperitoneum not seen on prior images. He was taken to operating room for emergent exploratory laparotomy. Intraoperative findings revealed a thrombus at the branch of the mesenteric artery leading to perforated cecum with feculent peritonitis. Patient passed away several days post-op due to worsening clinical status. DISCUSSION: Pneumoperitoneum is usually caused by perforation of abdominal viscus. COVID19 tends to cause multi organ failure(6). Elevation of pro-inflammatory cytokines can damage the microvascular system leading to uncontrolled activation of coagulation cascades which can cause small-vessel vasculitis and extensive microthrombi. There are very few cases reported regarding perforation of abdominal viscus due to microvascular thrombosis from COVID19. It is perplexing if COVID19 infection can acutely trigger this mesenteric arterial thrombus leading to ischemia and eventually perforation. It has been proposed that it might be secondary to direct viral invasion of the enterocyte cells via ACE2 receptors or immune response leading to cytokine storms (7) and activation of pro-coagulant cascades or combination of both (1,2,3). Mesenteric Ischemia is a life threatening condition requiring resection of ischemic bowel, GI decompression, and fluid resuscitation (4). Sometimes, it can be managed conservatively (5). CONCLUSIONS: It is important to recognize that COVID19 can potentially increase the risk of bowel perforation and cause pneumoperitoneum at risk groups of patients. REFERENCE #1: Parry A.H., Wani A.H., Yaseen M. Acute Mesenteric Ischemia in Severe Coronavirus-19 (COVID-19): Possible Mechanisms and Diagnostic Pathway. Acad Radiol. 2020;27(8):1190 REFERENCE #2: Varga Z., Flammer A.J., Steiger P., Haberecker M., Andermatt R., Zinkernagel A.S. Endothelial cell infection and endotheliitis in COVID-19. Lancet. 2020;395(10234):1417–1418 REFERENCE #3: Panigada M., Bottino N., Tagliabue P., Grasselli G., Novembrino C., Chantarangkul V. Hypercoagulability of COVID-19 patients in intensive care unit: A report of thromboelastography findings and other parameters of hemostasis. J Thromb Haemost. 2020;18(7):1738–1742 DISCLOSURES: No relevant relationships by Khoi Paul Dang-Ho, source=Web Response No relevant relationships by Louis Gerolemou, source=Web Response No relevant relationships by Aneeta Kumari, source=Web Response No relevant relationships by Kyaw Zaw Linn, source=Web Response No relevant relationships by Nabil Mesiha, source=Web Response No relevant relationships by Kunal Nangrani, source=Web Response No relevant relationships by Viswanath Vasudeva , source=Web Response

Bayesian inference using hierarchical and spatial priors for intravoxel incoherent motion MR imaging in the brain: Analysis of cancer and acute stroke.

The intravoxel incoherent motion (IVIM) model allows to map diffusion (D) and perfusion-related parameters (F and D*). Parameter estimation is, however, error-prone due to the non-linearity of the signal model, the limited signal-to-noise ratio (SNR) and the small volume fraction of perfusion in the in-vivo brain. In the present work, the performance of Bayesian inference was examined in the presence of brain pathologies characterized by hypo- and hyperperfusion. In particular, a hierarchical and a spatial prior were combined. Performance was compared relative to conventional segmented least squares regression, hierarchical prior only (non-segmented and segmented data likelihoods) and a deep learning approach. Realistic numerical brain IVIM simulations were conducted to assess errors relative to ground truth. In-vivo, data of 11 central nervous system cancer patients and 9 patients with acute stroke were acquired. The proposed method yielded reduced error in simulations for both the cancer and acute stroke scenarios compared to other methods across the whole investigated SNR range. The contrast-to-noise ratio of the proposed method was better or on par compared to the other techniques in-vivo. The proposed Bayesian approach hence improves IVIM parameter estimation in brain cancer and acute stroke.

Accurate and robust sparse-view angle CT image reconstruction using deep learning and prior image constrained compressed sensing (DL-PICCS).

Sparse-view CT image reconstruction problems encountered in dynamic CT acquisitions are technically challenging. Recently, many deep learning strategies have been proposed to reconstruct CT images from sparse-view angle acquisitions showing promising results. However, two fundamental problems with these deep learning reconstruction methods remain to be addressed: (1) limited reconstruction accuracy for individual patients and (2) limited generalizability for patient statisticalcohorts. The purpose of this work is to address the previously mentioned challenges in current deep learningmethods. A method that combines a deep learning strategy with prior image constrained compressed sensing (PICCS) was developed to address these two problems. In this method, the sparse-view CT data were reconstructed by the conventional filtered backprojection (FBP) method first, and then processed by the trained deep neural network to eliminate streaking artifacts. The outputs of the deep learning architecture were then used as the needed prior image in PICCS to reconstruct the image. If the noise level from the PICCS reconstruction is not satisfactory, another light duty deep neural network can then be used to reduce noise level. Both extensive numerical simulation data and human subject data have been used to quantitatively and qualitatively assess the performance of the proposed DL-PICCS method in terms of reconstruction accuracy andgeneralizability. Extensive evaluation studies have demonstrated that: (1) quantitative reconstruction accuracy of DL-PICCS for individual patient is improved when it is compared with the deep learning methods and CS-based methods; (2) the false-positive lesion-like structures and false negative missing anatomical structures in the deep learning approaches can be effectively eliminated in the DL-PICCS reconstructed images; and (3) DL-PICCS enables a deep learning scheme to relax its working conditions to enhance itsgeneralizability. DL-PICCS offers a promising opportunity to achieve personalized reconstruction with improved reconstruction accuracy and enhanced generalizability.

Spatiotemporal structure-aware dictionary learning-based 4D CBCT reconstruction.

Four-dimensional cone-beam computed tomography (4D CBCT) is developed to reconstruct a sequence of phase-resolved images, which could assist in verifying the patient's position and offering information for cancer treatment planning. However, 4D CBCT images suffer from severe streaking artifacts and noise due to the extreme sparse-view CT reconstruction problem for each phase. As a result, it would cause inaccuracy of treatment estimation. The purpose of this paper was to develop a new 4D CBCT reconstruction method to generate a series of high spatiotemporal 4D CBCT images. Considering the advantage of (DL) on representing structural features and correlation between neighboring pixels effectively, we construct a novel DL-based method for the 4D CBCT reconstruction. In this study, both a motion-aware dictionary and a spatially structural 2D dictionary are trained for 4D CBCT by excavating the spatiotemporal correlation among ten phase-resolved images and the spatial information in each image, respectively. Specifically, two reconstruction models are produced in this study. The first one is the motion-aware dictionary learning-based 4D CBCT algorithm, called motion-aware DL based 4D CBCT (MaDL). The second one is the MaDL equipped with a prior knowledge constraint, called pMaDL. Qualitative and quantitative evaluations are performed using a 4D extended cardiac torso (XCAT) phantom, simulated patient data, and two sets of patient data sets. Several state-of-the-art 4D CBCT algorithms, such as the McKinnon-Bates (MKB) algorithm, prior image constrained compressed sensing (PICCS), and the high-quality initial image-guided 4D CBCT reconstruction method (HQI-4DCBCT) are applied for comparison to validate the performance of the proposed MaDL and prior constraint MaDL (pMaDL) pmadl reconstruction frameworks. Experimental results validate that the proposed MaDL can output the reconstructions with few streaking artifacts but some structural information such as tumors and blood vessels, may still be missed. Meanwhile, the results of the proposed pMaDL demonstrate an improved spatiotemporal resolution of the reconstructed 4D CBCT images. In these improved 4D CBCT reconstructions, streaking artifacts are suppressed primarily and detailed structures are also restored. Regarding the XCAT phantom, quantitative evaluations indicate that an average of 58.70%, 45.25%, and 40.10% decrease in terms of root-mean-square error (RMSE) and an average of 2.10, 1.37, and 1.37 times in terms of structural similarity index (SSIM) are achieved by the proposed pMaDL method when compared with piccs, PICCS, MaDL(2D), and MaDL(2D), respectively. Moreover the proposed pMaDL achieves a comparable performance with HQI-4DCBCT algorithm in terms of RMSE and SSIM metrics. However, pMaDL has a better ability to suppress streaking artifacts than HQI-4DCBCT. The proposed algorithm could reconstruct a set of 4D CBCT images with both high spatiotemporal resolution and detailed features preservation. Moreover the proposed pMaDL can effectively suppress the streaking artifacts in the resultant reconstructions, while achieving an overall improved spatiotemporal resolution by incorporating the motion-aware dictionary with a prior constraint into the proposed 4D CBCT iterative framework.

Limited-angle computed tomography with deep image and physics priors

Computed tomography is a well-established x-ray imaging technique to reconstruct the three-dimensional structure of objects. It has been used extensively in a variety of fields, from diagnostic imaging to materials and biological sciences. One major challenge in some applications, such as in electron or x-ray tomography systems, is that the projections cannot be gathered over all the angles due to the sample holder setup or shape of the sample. This results in an ill-posed problem called the limited angle reconstruction problem. Typical image reconstruction in this setup leads to distortion and artifacts, thereby hindering a quantitative evaluation of the results. To address this challenge, we use a generative model to effectively constrain the solution of a physics-based approach. Our approach is self-training that can iteratively learn the nonlinear mapping from partial projections to the scanned object. Because our approach combines the data likelihood and image prior terms into a single deep network, it is computationally tractable and improves performance through an end-to-end training. We also complement our approach with total-variation regularization to handle high-frequency noise in reconstructions and implement a solver based on alternating direction method of multipliers. We present numerical results for various degrees of missing angle range and noise levels, which demonstrate the effectiveness of the proposed approach.

Robotics, imaging, and artificial intelligence in the catheterisation laboratory.

The catheterisation laboratory today combines diagnosis and therapeutics, through various imaging modalities and a prolific list of interventional tools, led by balloons and stents. In this review, we focus primarily on advances in image-based coronary interventions. The X-ray images that are the primary modality for diagnosis and interventions are combined with novel tools for visualisation and display, including multi-imaging co-registration modalities with three- and four-dimensional presentations. Interpretation of the physiologic significance of coronary stenosis based on prior angiographic images is being explored and implemented. Major efforts to reduce X-ray exposure to the staff and the patients, using computer-based algorithms for image processing, and novel methods to limit the radiation spread are being explored. The use of artificial intelligence (AI) and machine learning for better patient care requires attention to universal methods for sharing and combining large data sets and for allowing interpretation and analysis of large cohorts of patients. Barriers to data sharing using integrated and universal protocols should be overcome to allow these methods to become widely applicable. Robotic catheterisation takes the physician away from the ionising radiation spot, enables coronary angioplasty and stenting without compromising safety, and may allow increased precision. Remote coronary procedures over the internet, that have been explored in virtual and animal studies and already applied to patients in a small pilot study, open possibilities for sharing experience across the world without travelling. Application of those technologies to neurovascular, and particularly stroke interventions, may be very timely in view of the need for expert neuro-interventionalists located mostly in central areas.