Prior Image Research Articles

PET Image Reconstruction with Kernel and Kernel Space Composite Regularizer.

Led by the kernelized expectation maximization (KEM) method, the kernelized maximum-likelihood (ML) expectation maximization (EM) methods have recently gained prominence in PET image reconstruction, outperforming many previous state-of-the-art methods. But they are not immune to the problems of non-kernelized MLEM methods in potentially large reconstruction variance and high sensitivity to iteration numbers, and the difficulty in preserving image details and suppressing image variance simultaneously. To solve these problems, this paper derives, using the ideas of data manifold and graph regularization, a novel regularized KEM (RKEM) method with a kernel space composite regularizer for PET image reconstruction. The composite regularizer consists of a convex kernel space graph regularizer that smooths the kernel coefficients, a concave kernel space energy regularizer that enhances the coefficients' energy, and a composition constant that is analytically set to guarantee the convexity of composite regularizer. The composite regularizer renders easy use of PET-only image priors to overcome KEM's difficulty caused by the mismatch of MR prior and underlying PET images. Using this kernel space composite regularizer and the technique of optimization transfer, a globally convergent iterative algorithm is derived for RKEM reconstruction. Tests and comparisons on the simulated and in vivo data are presented to validate and evaluate the proposed algorithm, and demonstrate its better performance and advantages over KEM and other conventional methods.

Comparison of artificial intelligence versus real-time physician assessment of pulmonary edema with lung ultrasound

BackgroundLung ultrasound can evaluate for pulmonary edema, but data suggest moderate inter-rater reliability among users. Artificial intelligence (AI) has been proposed as a model to increase the accuracy of B line interpretation. Early data suggest a benefit among more novice users, but data are limited among average residency-trained physicians. The objective of this study was to compare the accuracy of AI versus real-time physician assessment for B lines. MethodsThis was a prospective, observational study of adult Emergency Department patients presenting with suspected pulmonary edema. We excluded patients with active COVID-19 or interstitial lung disease. A physician performed thoracic ultrasound using the 12-zone technique. The physician recorded a video clip in each zone and provided an interpretation of positive (≥3 B lines or a wide, dense B line) or negative (<3 B lines and the absence of a wide, dense B line) for pulmonary edema based upon the real-time assessment. A research assistant then utilized the AI program to analyze the same saved clip to determine if it was positive versus negative for pulmonary edema. The physician sonographer was blinded to this assessment. The video clips were then reviewed independently by two expert physician sonographers (ultrasound leaders with >10,000 prior ultrasound image reviews) who were blinded to the AI and initial determinations. The experts reviewed all discordant values and reached consensus on whether the field (i.e., the area of lung between two adjacent ribs) was positive or negative using the same criteria as defined above, which served as the gold standard. Results71 patients were included in the study (56.3% female; mean BMI: 33.4 [95% CI 30.6–36.2]), with 88.3% (752/852) of lung fields being of adequate quality for assessment. Overall, 36.1% of lung fields were positive for pulmonary edema. The physician was 96.7% (95% CI 93.8%–98.5%) sensitive and 79.1% (95% CI 75.1%–82.6%) specific. The AI software was 95.6% (95% CI 92.4%–97.7%) sensitive and 64.1% (95% CI 59.8%–68.5%) specific. ConclusionBoth the physician and AI software were highly sensitive, though the physician was more specific. Future research should identify which factors are associated with increased diagnostic accuracy.

Self-Supervised Remote Sensing Image Dehazing Network Based on Zero-Shot Learning

Traditional dehazing approaches that rely on prior knowledge exhibit limited efficacy when confronted with the intricacies of real-world hazy environments. While learning-based dehazing techniques necessitate large-scale datasets for effective model training, the acquisition of these datasets is time-consuming and laborious, and the resulting models may encounter a domain shift when processing real-world hazy images. To overcome the limitations of prior-based and learning-based dehazing methods, we propose a self-supervised remote sensing (RS) image-dehazing network based on zero-shot learning, where the self-supervised process avoids dense dataset requirements and the learning-based structures refine the artifacts in extracted image priors caused by complex real-world environments. The proposed method has three stages. The first stage involves pre-processing the input hazy image by utilizing a prior-based dehazing module; in this study, we employed the widely recognized dark channel prior (DCP) to obtain atmospheric light, a transmission map, and the preliminary dehazed image. In the second stage, we devised two convolutional neural networks, known as RefineNets, dedicated to enhancing the transmission map and the initial dehazed image. In the final stage, we generated a hazy image using the atmospheric light, the refined transmission map, and the refined dehazed image by following the haze imaging model. The meticulously crafted loss function encourages cycle-consistency between the regenerated hazy image and the input hazy image, thereby facilitating a self-supervised dehazing model. During the inference phase, the model undergoes training in a zero-shot manner to yield the haze-free image. These thorough experiments validate the substantial improvement of our method over the prior-based dehazing module and the zero-shot training efficiency. Furthermore, assessments conducted on both uniform and non-uniform RS hazy images demonstrate the superiority of our proposed dehazing technique.

Overlapping group prior for image deconvolution using patch-wise gradient statistics

In image deconvolution problems, global gradient statistics serve as powerful image priors that separate high-quality images from degraded observations. However, it is unclear whether the important detailed information is sufficiently regularized in these approaches. Owing to the limitations, the restored images exhibit sharply reconstructed edges around structures, but commonly suffer from the loss of significant textures and details. In this study, we introduced patch-wise gradient statistics as an image prior strategy to restore disregarded regions in consideration of regional characteristics. The analytic relation between the prior and the restored local statistics was investigated through the Bayesian interpretation of overlapping group sparsity. The regularization function that incorporates the contextual information was derived in the maximum a posteriori framework. Subsequently, the unknown prior information and latent image were alternately estimated from the observations based on the patch-wise statistical analysis of the non-blind deconvolution process. In the experimental results, the superiority of the proposed algorithm was verified by more original textures in the restored images compared to the conventional approaches.

3D directional gradient L0 norm minimization guided limited-view reconstruction in a dual-panel positron emission mammography

BackgroundDual-panel PET is often used for local organ imaging, especially breast imaging, due to its simple structure, high sensitivity, good in-plane resolution, and straightforward fusion with other imaging modalities. Nevertheless, because of data loss caused by the dual-panel structure, using conventional image reconstruction methods results in limited-view artifacts and low image quality in dual-panel positron emission mammography (PEM), which may seriously affect the diagnosis. To mitigate the limited-view artifacts in the dual-panel PEM, we propose a 3D directional gradient L0 norm minimization (3D-DL0) guided reconstruction method. MethodsThe detailed derivation and reasonable simplification of the 3D-DL0 algorithm are given first. Using this algorithm, we then obtain a prior image with edge recovery but contrast loss. To limit the solution space, the 3D-DL0 prior is introduced into the Maximum a Posteriori reconstruction. Meanwhile, a space-invariant point spread function is also implemented to restore image contrast and boundaries. Finally, the reconstructed images with limited-view artifact suppression are obtained. The proposed method was evaluated using the data acquired from physical phantoms and patients with breast tumors on a commercial dual-panel PET system. ResultsThe qualitative and quantitative studies for phantom data and the blind reader study for clinical data show that the proposed method is more effective in reaching a balance between artifact elimination and image contrast improvement compared with various limited-view reconstruction methods. In addition, the iteration process of the method is proved convergent numerically. ConclusionsThe image quality improvement confirms the potential value of the proposed reconstruction algorithm to address the limited-view problem, and thus improve diagnostic accuracy in dual-panel PEM imaging.

What does degeneration at the cervicothoracic junction tell us? A kinematic MRI study of 93 individuals.

Current decision-making in multilevel cervical fusion weighs the potential to protect adjacent levels and reduce reoperation risk by crossing the cervicothoracic junction (C7/T1) against increased operative time and risk of complication. Careful planning is required, and the planned distal and adjacent levels should be assessed for degenerative disc disease (DDD). This study assessed whether DDD at the cervicothoracic junction was associated with DDD, disc height, translational motion, or angular variation in the adjacent superior (C6/C7) or inferior (T1/T2) levels. This study retrospectively analyzed 93 cases with kinematic MRI. Cases were randomly selected from a database with inclusion criteria being no prior spine surgery and images having sufficient quality for analysis. DDD was assessed using Pfirrmann classification. Vertebral body bone marrow lesions were assessed using Modic changes. Disc height was measured at the mid-disc in neutral and extension. Translational motion and angular variation were calculated by assessing translational or angular motion segment integrity respectively in flexion and extension. Statistical associations were assessed with scatterplots and Kendall's tau. DDD at C7/T1 was positively associated with DDD at C6/C7 (tau = 0.53, p < 0.01) and T1/T2 (tau = 0.58, p < 0.01), with greater disc height in neutral position at T1/T2 (tau = 0.22, p < 0.01), and with greater disc height in extended position at C7/T1 (tau = 0.17, p = 0.04) and at T1/T2 (tau = 0.21, p < 0.01). DDD at C7/T1 was negatively associated with angular variation at C6/C7 (tau = - 0.23, p < 0.01). No association was appreciated between DDD at C7/T1 and translational motion. The association of DDD at the cervicothoracic junction with DDD at the adjacent levels emphasizes the necessity for careful selection of the distal level in multilevel fusion in the distal cervical spine.

Triple modality image reconstruction of PET data using SPECT, PET, CT information increases lesion uptake in images of patients treated with radioembolization with ^{90}Y micro-spheres

PurposeNuclear medicine imaging modalities like computed tomography (CT), single photon emission CT (SPECT) and positron emission tomography (PET) are employed in the field of theranostics to estimate and plan the dose delivered to tumors and the surrounding tissues and to monitor the effect of the therapy. However, therapeutic radionuclides often provide poor images, which translate to inaccurate treatment planning and inadequate monitoring images. Multimodality information can be exploited in the reconstruction to enhance image quality. Triple modality PET/SPECT/CT scanners are particularly useful in this context due to the easier registration process between images. In this study, we propose to include PET, SPECT and CT information in the reconstruction of PET data. The method is applied to Yttrium-90 (^{90}Y) data.MethodsData from a NEMA phantom filled with ^{90}Y were used for validation. PET, SPECT and CT data from 10 patients treated with Selective Internal Radiation Therapy (SIRT) were used. Different combinations of prior images using the Hybrid kernelized expectation maximization were investigated in terms of VOI activity and noise suppression.ResultsOur results show that triple modality PET reconstruction provides significantly higher uptake when compared to the method used as standard in the hospital and OSEM. In particular, using CT-guided SPECT images, as guiding information in the PET reconstruction significantly increases uptake quantification on tumoral lesions.ConclusionThis work proposes the first triple modality reconstruction method and demonstrates up to 69% lesion uptake increase over standard methods with SIRT ^{90}Y patient data. Promising results are expected for other radionuclide combination used in theranostic applications using PET and SPECT.

A Triple Deep Image Prior Model for Image Denoising Based on Mixed Priors and Noise Learning

Image denoising poses a significant challenge in computer vision due to the high-level visual task’s dependency on image quality. Several advanced denoising models have been proposed in recent decades. Recently, deep image prior (DIP), using a particular network structure and a noisy image to achieve denoising, has provided a novel image denoising method. However, the denoising performance of the DIP model still lags behind that of mainstream denoising models. To improve the performance of the DIP denoising model, we propose a TripleDIP model with internal and external mixed images priors for image denoising. The TripleDIP comprises of three branches: one for content learning and two for independent noise learning. We firstly use a Transformer-based supervised model (i.e., Restormer) to obtain a pre-denoised image (used as external prior) from a given noisy image, and then take the noisy image and the pre-denoised image as the first and second target image, respectively, to perform the denoising process under the designed loss function. We add constraints between two-branch noise learning and content learning, allowing the TripleDIP to employ external prior while enhancing independent noise learning stability. Moreover, the automatic stop criterion we proposed prevents the model from overfitting the noisy image and improves the execution efficiency. The experimental results demonstrate that TripleDIP outperforms the original DIP by an average of 2.79 dB and outperforms classical unsupervised methods such as N2V by an average of 2.68 dB and the latest supervised models such as SwinIR and Restormer by an average of 0.63 dB and 0.59 dB on the Set12 dataset. This can mainly be attributed to the fact that two-branch noise learning can obtain more stable noise while constraining the content learning branch’s optimization process. Our proposed TripleDIP significantly enhances DIP denoising performance and has broad application potential in scenarios with insufficient training datasets.

Accelerated submillimeter wave-encoded magnetic resonance imaging via deep untrained neural network.

Wave gradient encoding can adequately utilize coil sensitivity profiles to facilitate higher accelerations in parallel magnetic resonance imaging (pMRI). However, there are limitations in mainstream pMRI and a few deep learning (DL) methods for recovering missing data under wave encoding framework: the former is prone to introduce errors from the auto-calibration signals (ACS) signal acquisition and is time-consuming, while the latter requires a large amount of trainingdata. To tackle the above issues, an untrained neural network (UNN) model incorporating wave-encoded physical properties and deep generative model, named WDGM, was proposed with additional ACS- and trainingdata-free. Generally, the proposed method can provide powerful missing data interpolation capability using the wave physical encoding framework and designed UNN to characterize the MR image (k-space data) priors. Specifically, the MRI reconstruction combining physical wave encoding and elaborate UNN is modeled as a generalized minimization problem. The designation of UNN is driven by the coil sensitivity maps (CSM) smoothness and k-space linear predictability. And then, the iterative paradigm to recover the full k-space signal is determined by the projected gradient descent, and the complex computation is unrolled to the network with optimized parameters by the optimizer. Simulated wave encoding and in vivo experiments are exploited to demonstrate the feasibility of the proposed method. The best quantitative metrics RMSE/SSIM/PSNR of 0.0413, 0.9514, and 37.4862 gave competitive results in all experiments with at least six-fold acceleration, respectively. In vivo experiments of human brains and knees showed that the proposed method can achieve comparable reconstruction quality and even has superiority relative to the comparison, especially at a high resolution of 0.67 mm and fewer ACS. In addition, the proposed method has a higher computational efficiency achieving a computation time of 9.6 s/perslice. The model proposed in this work addresses two limitations of MRI reconstruction in the wave encoding framework. The first is to eliminate the need for ACS signal acquisition to perform the time-consuming calibration process and to avoid errors such as motion during the acquisition procedure. Furthermore, the proposed method has clinical application friendly without the need to prepare large training datasets, which is difficult in the clinical. All results of the proposed method demonstrate more confidence in both quantitative and qualitative metrics. In addition, the proposed method can achieve higher computationalefficiency.

Pie-Net: Prior-information-enabled deep learning noise reduction for coronary CT angiography acquired with a photon counting detector CT.

Photon-counting-detector CT (PCD-CT) enables the production of virtual monoenergetic images (VMIs) at a high spatial resolution (HR) via simultaneous acquisition of multi-energy data. However, noise levels in these HR VMIs are markedly increased. To develop a deep learning technique that utilizes a lower noise VMI as prior information to reduce image noise in HR, PCD-CT coronary CT angiography (CTA). Coronary CTA exams of 10 patients were acquired using PCD-CT (NAEOTOM Alpha, Siemens Healthineers). A prior-information-enabled neural network (Pie-Net) was developed, treating one lower-noise VMI (e.g., 70keV) as a prior input and one noisy VMI (e.g., 50keV or 100keV) as another. For data preprocessing, noisy VMIs were reconstructed by filtered back-projection (FBP) and iterative reconstruction (IR), which were then subtracted to generate "noise-only" images. Spatial decoupling was applied to the noise-only images to mitigate overfitting and improve randomization. Thicker slice averaging was used for the IR and prior images. The final training inputs for the convolutional neural network (CNN) inside the Pie-Net consisted of thicker-slice signal images with the reinsertion of spatially decoupled noise-only images and the thicker-slice prior images. The CNN training labels consisted of the corresponding thicker-slice label images without noise insertion. Pie-Net's performance was evaluated in terms of image noise, spatial detail preservation, and quantitative accuracy, and compared to a U-net-based method that did not include prior information. Pie-Net provided strong noise reduction, by 95±1% relative to FBP and by 60±8% relative to IR. For HR VMIs at different keV (e.g., 50keV or 100keV), Pie-Net maintained spatial and spectral fidelity. The inclusion of prior information from the PCD-CT data in the spectral domain was able to improve a robust deep learning-based denoising performance compared to the U-net-based method, which caused some loss of spatial detail and introduced some artifacts. The proposed Pie-Net achieved substantial noise reduction while preserving HR VMI's spatial and spectral properties.

GeoGlue: feature matching with self-supervised geometric priors for high-resolution UAV images

ABSTRACT We present GeoGlue, a novel method using high-resolution UAV imagery for accurate feature matching, which is normally challenging due to the complicated scenes. Current feature detection methods are performed without guidance of geometric priors (e.g., geometric lines), lacking enough attention given to salient geometric features which are indispensable for accurate matching due to their stable existence across views. In this work, geometric lines are firstly detected by a CNN-based geometry detector (GD) which is pre-trained in a self-supervised manner through automatically generated images. Then, geometric lines are naturally vectorized based on GD and thus non-significant features can be disregarded as judged by their disordered geometric morphology. A graph attention network (GAT) is utilized for final feature matching, spanning across the image pair with geometric priors informed by GD. Comprehensive experiments show that GeoGlue outperforms other state-of-the-art methods in feature-matching accuracy and performance stability, achieving pose estimation with maximum rotation and translation errors under 1% in challenging scenes from benchmark datasets, Tanks & Temples and ETH3D. This study also proposes the first self-supervised deep-learning model for curved line detection, generating geometric priors for matching so that more attention is put on prominent features and improving the visual effect of 3D reconstruction.

MP41-01 EVALUATION OF GROWTH RATES FOR SMALL RENAL MASSES IN ELDERLY PATIENTS UNDERGOING ACTIVE SURVEILLANCE

You have accessJournal of UrologyCME1 Apr 2023MP41-01 EVALUATION OF GROWTH RATES FOR SMALL RENAL MASSES IN ELDERLY PATIENTS UNDERGOING ACTIVE SURVEILLANCE Ridwan Alam, Sunil Patel, Peter Chang, Andrew Wagner, James McKiernan, Phillip Pierorazio, Mohamad Allaf, and Nirmish Singla Ridwan AlamRidwan Alam More articles by this author , Sunil PatelSunil Patel More articles by this author , Peter ChangPeter Chang More articles by this author , Andrew WagnerAndrew Wagner More articles by this author , James McKiernanJames McKiernan More articles by this author , Phillip PierorazioPhillip Pierorazio More articles by this author , Mohamad AllafMohamad Allaf More articles by this author , and Nirmish SinglaNirmish Singla More articles by this author View All Author Informationhttps://doi.org/10.1097/JU.0000000000003279.01AboutPDF ToolsAdd to favoritesDownload CitationsTrack CitationsPermissionsReprints ShareFacebookLinked InTwitterEmail Abstract INTRODUCTION AND OBJECTIVE: Emerging data suggests that tumor microenvironments may differ among patients based on age, which can be of concern to patients who are enrolled in active surveillance (AS) for a prolonged period. We sought to examine if particular age cutoffs were associated with an increased growth rate (GR) for patients enrolled in AS for small renal masses (SRMs). METHODS: The Delayed Intervention and Surveillance for Small Renal Masses registry is a prospective, multi-institutional study examining outcomes in patients undergoing AS for SRMs. Two definitions of GR were examined: growth rate from the initial image (GRi) and growth rate from the prior image (GRp). Image measurements were dichotomized based on patient age at the time of imaging. Multiple age cutoffs were examined: 65, 70, 75, and 80 years. Mixed effects linear regression examined associations between age and GR, with controlling to account for multiple measurements from the same individual. RESULTS: We examined 2542 measurements from 571 patients [Figure]. The median age of patients at enrollment was 70.9 years (IQR 63.2-77.4) with a median tumor diameter of 1.8 cm (IQR 1.4-2.5). When examined as a continuous variable, age was not associated with GRi (-0.0001 cm/year, 95% CI -0.007 – 0.007, P=0.97) or GRp (0.008 cm/year, 95% CI -0.004 – 0.020, P=0.17) after adjustment. The only age thresholds associated with an increased GR were 65 years for GRi and 70 years for GRp[Table]. Patients who underwent delayed intervention had increased GRi (0.282 cm/year, 95% CI 0.078 – 0.487, P=0.007) but no difference in GRp (0.319 cm/year, 95% CI -0.032 – 0.670, P=0.07) compared to those who remained on AS. CONCLUSIONS: Patients with SRMs on AS do not exhibit a clear association between tumor GR and patient age, suggesting that AS is a safe and durable management option for aging patients with SRMs. Source of Funding: None. © 2023 by American Urological Association Education and Research, Inc.FiguresReferencesRelatedDetails Volume 209Issue Supplement 4April 2023Page: e554 Advertisement Copyright & Permissions© 2023 by American Urological Association Education and Research, Inc.MetricsAuthor Information Ridwan Alam More articles by this author Sunil Patel More articles by this author Peter Chang More articles by this author Andrew Wagner More articles by this author James McKiernan More articles by this author Phillip Pierorazio More articles by this author Mohamad Allaf More articles by this author Nirmish Singla More articles by this author Expand All Advertisement PDF downloadLoading ...

Unsupervised MRI Super Resolution Using Deep External Learning and Guided Residual Dense Network With Multimodal Image Priors

Deep learning techniques have led to state-of-the-art image super resolution with natural images. Normally, pairs of high-resolution and low-resolution images are used to train the deep learning models. These techniques have also been applied to medical image super-resolution. The characteristics of medical images differ significantly from natural images in several ways. First, it is difficult to obtain high-resolution images for training in real clinical applications due to the limitations of imaging systems and clinical requirements. Second, other modal high-resolution images are available (e.g., high-resolution T1-weighted images are available for enhancing low-resolution T2-weighted images). In this paper, we propose an unsupervised image super-resolution technique based on simple prior knowledge of the human anatomy. This technique does not require target T2WI high-resolution images for training. Furthermore, we present a guided residual dense network, which incorporates a residual dense network with a guided deep convolutional neural network for enhancing the resolution of low-resolution images by referring to different modal high-resolution images of the same subject. Experiments on a publicly available brain MRI database showed that our proposed method achieves better performance than the state-of-the-art methods.

A semiautomatic segmentation method framework for pelvic bone tumors based on CT-MR multimodal images.

The pelvic structure is complex and the tumor is poorly defined from the surrounding tissues. Finding the exact tumor resection margin based on the surgeon's clinical experience alone is a time-consuming and difficult task, which is a major factor leading to surgical failure. An accurate method for segmenting pelvic bone tumors is needed. In this paper, a semiautomatic segmentation method for pelvic bone tumors based on CT-MR multimodal images is presented. The method combines multiple medical prior knowledge and image segmentation algorithms. Finally, the segmentation results are visualized in three dimensions. We tested the proposed method on a collection of 10 cases (97 tumor MR images in total). The segmentation results were compared with the manual annotation of the physicians. On average, our method has an accuracy of 0.9358, a recall of 0.9278, an IOU value of 0.8697, a Dice value of 0.9280, and an AUC value of 0.9632. The average error of the 3D model was within the allowable range of the surgery. The proposed algorithm can accurately segment bone tumors in pelvic MR images regardless of tumor location, size, and other factors. It provides the possibility to assist pelvic bone tumor preservation surgery.

Super-resolution image reconstruction from sparsity regularization and deep residual-learned priors.

Computed tomography (CT) plays an important role in the field of non-destructive testing. However, conventional CT images often have blurred edge and unclear texture, which is not conducive to the follow-up medical diagnosis and industrial testing work. This study aims to generate high-resolution CT images using a new CT super-resolution reconstruction method combining with the sparsity regularization and deep learning prior. The new method reconstructs CT images through a reconstruction model incorporating image gradient L0-norm minimization and deep image priors using a plug-and-play super-resolution framework. The deep learning priors are learned from a deep residual network and then plugged into the proposed new framework, and alternating direction method of multipliers is utilized to optimize the iterative solution of the model. The simulation data analysis results show that the new method improves the signal-to-noise ratio (PSNR) by 7% and the modulation transfer function (MTF) curves show that the value of MTF50 increases by 0.02 factors compared with the result of deep plug-and-play super-resolution. Additionally, the real CT image data analysis results show that the new method improves the PSNR by 5.1% and MTF50 by 0.11 factors. Both simulation and real data experiments prove that the proposed new CT super-resolution method using deep learning priors can reconstruct CT images with lower noise and better detail recovery. This method is flexible, effective and extensive for low-resolution CT image super-resolution.

3-D Computed Laminography Based on Prior Images and Total Variation

The 3-D sparse-view computed laminography (CL) is an important research issue due to its considerable potential in lowering costs. To solve the artifacts and noise caused by sparse-view scanning, the total variation (TV) method is widely used. However, TV only utilizes the prior information of the reconstructed image itself. Therefore, the prior-image-based reconstruction method (PIBR) is proposed to reduce the artifacts by using a prior image. In this study, we proposed a 3-D CL method based on the final prior image and TV for sparse-view scanning, which is divided into two subproblems corresponding to two solution steps: the novel PIBR step and the denoising step. In the first step, we devise the final prior image and its corresponding prior mask to ensure the effectiveness of the PIBR. The final prior image is obtained based on the prior images, which are reconstructed from two different oblique angle projections. The prior mask is a binary mask, which is used to record the position of the prior image information. Then, the denoising step is based on the TV minimization method using the split-Bregman (SB) frame. Soft shrinkage operations and fast Fourier transform (FFT) are used to efficiently implement the SB frame. Compared with several iterative reconstruction methods, the experimental results demonstrate the effectiveness of the proposed method in terms of preserving edges, suppressing interslices aliasing and denoising.

Evaluation of Growth Rates for Small Renal Masses in Elderly Patients Undergoing Active Surveillance

BackgroundAs the adoption of active surveillance (AS) for small renal masses (SRMs) grows, the number of elderly patients enrolled for a prolonged period of time will increase. However, our understanding of comparative growth rates (GRs) in aging patients with SRMs remains poor. ObjectiveTo examine whether particular age cutoffs are associated with an increased GR for patients undergoing AS for SRMs. Design, setting, and participantsWe identified all patients with SRMs enrolled in the multi-institutional, prospective Delayed Intervention and Surveillance for Small Renal Masses (DISSRM) registry since 2009 who elected for AS. Outcome measurements and statistical analysisTwo definitions of GR were examined: GR from the initial image (GRi) and GR from the prior image (GRp). Image measurements were dichotomized based on patient age at the time of imaging. Multiple age cutoffs were examined: 65, 70, 75, and 80 yr. Mixed-effect linear regression examined the associations between age and GR, with controlling to account for multiple measurements from the same individual. Results and limitationsWe examined 2542 measurements from 571 patients. The median age at enrollment was 70.9 yr (interquartile range [IQR] 63.2–77.4) with a median tumor diameter of 1.8 cm (IQR 1.4–2.5). As a continuous variable, age was not associated with GRi (–0.0001 cm/yr, 95% confidence interval [CI] –0.007 to 0.007, p = 0.97) or GRp (0.008 cm/yr, 95% CI –0.004 to 0.020, p = 0.17) after adjustment. The only age thresholds associated with an increased GR were 65 yr for GRi and 70 yr for GRp. Limitations include the one-dimensional nature of the measurements used. ConclusionsIncreased age for patients on AS for SRMs is not associated with increased GRs. Patient summaryWe examined whether patients undergoing active surveillance (AS) exhibited accelerated growth of their small renal masses (SRMs) after a certain age. No demonstrable change was seen, suggesting that AS is a safe and durable management option for aging patients with SRMs.

Source-detector trajectory optimization for CBCT metal artifact reduction based on PICCS reconstruction.

Precise instrument placement plays a critical role in all interventional procedures, especially percutaneous procedures such as needle biopsies, to achieve successful tumor targeting and increased diagnostic accuracy. C-arm cone beam computed tomography (CBCT) has the potential to precisely visualize the anatomy in direct vicinity of the needle and evaluate the adequacy of needle placement during the intervention, allowing for instantaneous adjustment in case of misplacement. However, even with the most advanced C-arm CBCT devices, it can be difficult to identify the exact needle position on CBCT images due to the strong metal artifacts around the needle. In this study, we proposed a framework for customized trajectory design in CBCT imaging based on Prior Image Constrained Compressed Sensing (PICCS) reconstruction with the goal of reducing metal artifacts in needle-based procedures. We proposed to optimize out-of-plane rotations in three-dimensional (3D) space and minimize projection views while reducing metal artifacts at specific volume of interests (VOIs). An anthropomorphic thorax phantom with a needle inserted inside and two tumor models as the imaging targets were used to validate the proposed approach. The performance of the proposed approach was also evaluated for CBCT imaging under kinematic constraints by simulating some collision areas on the geometry of the C-arm. We compared the result of optimized 3D trajectories using the PICCS algorithm and 20 projections with the result of a circular trajectory with sparse view using PICCS and Feldkamp, Davis, and Kress (FDK), both using 20 projections, and the circular FDK method with 313 projections. For imaging targets 1 and 2, the highest values of structural similarity index measure (SSIM) and universal quality index (UQI) between the reconstructed image from the optimized trajectories and the initial CBCT image at the VOI was calculated 0.7521, 0.7308 and 0.7308, 0.7248 respectively. These results significantly outperformed the FDK method (with 20 and 313 projections) and the PICCS method (20 projections) both using the circular trajectory. Our results showed that the proposed optimized trajectories not only significantly reduce metal artifacts but also suggest a dose reduction for needle-based CBCT interventions, considering the small number of projections used. Furthermore, our results showed that the optimized trajectories are compatible with spatially constrained situations and enable CBCT imaging under kinematic constraints when the standard circular trajectory is not feasible.

Fabric defect detection based on low-rank decomposition with factor group-sparse regularizer

Recently, many low-rank-based methods in terms of detecting defects in fabric images have been proposed. However, there are two disadvantages of these methods. First, current low-rank based methods use the nuclear norm as the surrogate of rank, which causes inefficient optimization process and sub-optimal performance. Second, low-rank defective regions cannot be detected by low-rank based models. Thus, we propose a factor group-sparse regularized low-rank decomposition model (FGSRLRD) to solve these problems. This method takes the factor group-sparse regularizer as the surrogate of rank, which is more efficient as singular value decomposition (SVD) is not applied in the optimization process. Better performance is achieved as the factor group-sparse regularizer is a more accurate approximation of the rank. In addition, the weight matrix generated by the lightweight autoencoder is incorporated into the object function of FGSRLRD to guide locating defective regions. Besides, as low-rank defective regions cannot be segmented by low-rank models, this method constructs a fusion image of the prior image and the sparse image to highlight the defective regions. The performance of the proposed method is evaluated on two standard datasets, and the results indicate that the suggested method outperforms the existing state-of-the-art methods in locating the defective regions on fabric images.

Pill Box Text Identification Using DBNet-CRNN.

The recognition process of natural scenes is complicated at present, and images themselves may be complex owing to the special features of natural scenes. In this study, we use the detection and recognition of pill box text as an application scenario and design a deep-learning-based text detection algorithm for such natural scenes. We propose an end-to-end graphical text detection and recognition model and implement a detection system based on the B/S research application for pill box recognition, which uses DBNet as the text detection framework and a convolutional recurrent neural network (CRNN) as the text recognition framework. No prior image preprocessing is required in the detection and recognition processes. The recognition result from the back-end is returned to the front-end display. Compared with traditional methods, this recognition process reduces the complexity of preprocessing prior to image detection and improves the simplicity of the model application. Experiments on the detection and recognition of 100 pill boxes demonstrate that the proposed method achieves better accuracy in text localization and recognition results than the previous CTPN + CRNN method. The proposed method is significantly more accurate and easier to use than the traditional approach in terms of both training and recognition processes.