Search

Purpose: Computer‐aided diagnosis (CAD) systems on breast ultrasound (BUS) aim to increase the efficiency and effectiveness of breast screening, helping specialists to detect and classify breast lesions. CAD system development requires a set of annotated images, including lesion segmentation, biopsy results to specify benign and malignant cases, and BI‐RADS categories to indicate the likelihood of malignancy. Besides, standardized partitions of training, validation, and test sets promote reproducibility and fair comparisons between different approaches. Thus, we present a publicly available BUS dataset whose novelty is the substantial increment of cases with the above‐mentioned annotations and the inclusion of standardized partitions to objectively assess and compare CAD systems. Acquisition and Validation Methods: The BUS dataset comprises 1875 anonymized images from 1064 female patients acquired via four ultrasound scanners during systematic studies at the National Institute of Cancer (Rio de Janeiro, Brazil). The dataset includes biopsy‐proven tumors divided into 722 benign and 342 malignant cases. Besides, a senior ultrasonographer performed a BI‐RADS assessment in categories 2 to 5. Additionally, the ultrasonographer manually outlined the breast lesions to obtain ground truth segmentations. Furthermore, 5‐ and 10‐fold cross‐validation partitions are provided to standardize the training and test sets to evaluate and reproduce CAD systems. Finally, to validate the utility of the BUS dataset, an evaluation framework is implemented to assess the performance of deep neural networks for segmenting and classifying breast lesions. Data Format and Usage Notes: The BUS dataset is publicly available for academic and research purposes through an open‐access repository under the name BUS‐BRA: A Breast Ultrasound Dataset for Assessing CAD Systems. BUS images and reference segmentations are saved in Portable Network Graphic (PNG) format files, and the dataset information is stored in separate Comma‐Separated Value (CSV) files. Potential Applications: The BUS‐BRA dataset can be used to develop and assess artificial intelligence‐based lesion detection and segmentation methods, and the classification of BUS images into pathological classes and BI‐RADS categories. Other potential applications include developing image processing methods like despeckle filtering and contrast enhancement methods to improve image quality and feature engineering for image description. [ABSTRACT FROM AUTHOR]

The last decade has seen a large growth in fluorescence‐guided surgery (FGS) imaging and interventions. With the increasing number of clinical specialties implementing FGS, the range of systems with radically different physical designs, image processing approaches, and performance requirements is expanding. This variety of systems makes it nearly impossible to specify uniform performance goals, yet at the same time, utilization of different devices in new clinical procedures and trials indicates some need for common knowledge bases and a quality assessment paradigm to ensure that effective translation and use occurs. It is feasible to identify key fundamental image quality characteristics and corresponding objective test methods that should be determined such that there are consistent conventions across a variety of FGS devices. This report outlines test methods, tissue simulating phantoms and suggested guidelines, as well as personnel needs and professional knowledge bases that can be established. This report frames the issues with guidance and feedback from related societies and agencies having vested interest in the outcome, coming from an independent scientific group formed from academics and international federal agencies for the establishment of these professional guidelines. [ABSTRACT FROM AUTHOR]

Background: 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. Purpose: 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. Methods: 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. Results: 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. Conclusions: 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. [ABSTRACT FROM AUTHOR]

Background: Peripheral arterial disease (PAD) is a chronic occlusive disease that restricts blood flow in the lower limbs, causing partial or complete blockages of the blood flow. While digital subtraction angiography (DSA) has traditionally been the preferred method for assessing blood flow in the lower limbs, advancements in wide beam Computed Tomography (CT), allowing successive acquisition at high frame rate, might enable hemodynamic measurements. Purpose: To quantify the arterial blood flow in stenotic below‐the‐knee (BTK) arteries. To this end, we propose a novel method for contrast bolus tracking and assessment of quantitative hemodynamic parameters in stenotic arteries using 4D‐CT. Methods: Fifty patients with suspected PAD underwent 4D‐CT angiography in addition to the clinical run‐off computed tomography angiography (CTA). From these dynamic acquisitions, the BTK arteries were segmented and the region of maximum blood flow was extracted. Time attenuation curves (TAC) were estimated using 2D spatio‐temporal B‐spline regression, enforcing both spatial and temporal smoothness. From these curves, quantitative hemodynamic parameters, describing the shape of the propagating contrast bolus were automatically extracted. We evaluated the robustness of the proposed TAC fitting method with respect to interphase delay and imaging noise and compared it to commonly used approaches. Finally, to illustrate the potential value of 4D‐CT, we assessed the correlation between the obtained hemodynamic parameters and the presence of PAD. Results: 280 out of 292 arteries were successfully segmented, with failures mainly due to a delayed contrast arrival. The proposed method led to physiologically plausible hemodynamic parameters and was significantly more robust compared to 1D temporal regression. A significant correlation between the presence of proximal stenoses and several hemodynamic parameters was found. Conclusions: The proposed method based on spatio‐temporal bolus tracking was shown to lead to stable and physiologically plausible estimation of quantitative hemodynamic parameters, even in the case of stenotic arteries. These parameters may provide valuable information in the evaluation of PAD and contribute to its diagnosis. [ABSTRACT FROM AUTHOR]

Purpose: To propose a markerless beam's eye view (BEV) motion monitoring algorithm, which works with the inferior quality megavolt (MV) images with multi‐leaf collimator (MLC) occlusion‐compatible. Methods: A thorax phantom was used to verify the accuracy of the algorithm. Lung tumor quality assurance (QA) plans were generated for the phantom, and delivered 10 times on the linear accelerator with manually treatment offsets in various directions. The algorithm was used to register 753 electronic portal imaging device (EPID) images with the appropriate digitally reconstructed radiograph (DRR), calculating a registration offset that was compared with the actual offset to determine the monitoring errors. Image similarity measure was used as an independent check. Additionally, patient data of 21 lung tumor treatment plans were gathered. A total of 533 pairs of patient images were acquired for motion monitoring study, to offer quantifiable data of the tumor position change during treatment. Results: The monitoring algorithm can process various degrees (10%–80%) of image loss, and performs well when dealing with non‐rigid registration for partial data images. About 86.8% of the monitoring errors are less than 3 mm in the algorithm verification of the phantom study, and about 80% of the errors are under than 2 mm. Normalized Mutual Information (NMI) of phantom images changes from 1.182 ± 0.026 to 1.202 ± 0.027, with p < 0.005, and the Hausdorff‐Distance (HD) changes from 3.506 ± 0.417 mm to 3.466 ± 0.473 mm, with p < 0.005. Translation with a displacement range of ‐6.0 mm to 6.2 mm is the predominant change of the patient target during treatment. NMI of patient images changes from 1.216 ± 0.031 to 1.225 ± 0.031, with p < 0.005, and HD changes from 3.131 ± 0.876 mm to 3.118 ± 0.038 mm, with p < 0.005. The dice index of target before and after registration is 0.264 ± 0.336, indicating the presence of non‐negligible non‐rigid deformation. Conclusions: The study provides a robust markerless motion monitoring algorithm for multi‐modal, partial data and inferior quality image processing. [ABSTRACT FROM AUTHOR]

Background: Voxel-Based (VB) analysis embraces a multifaceted ensemble of sophisticated techniques, lying at the boundary between image processing and statistical modeling, that allow for a frequentist inference of pathophysiological properties anchored to an anatomical reference. VB methods has been widely adopted in neuroimaging studies and, more recently, they are gaining momentum in radiation oncology research. However, the price for the power of VB analysis is the complexity of the underlying mathematics and algorithms. Purpose: In this paper, we present the Multi-pAradigM voxel-Based Analysis (MAMBA) toolbox, which is intended for a flexible application of VB analysis in a wide variety of scenarios in medical imaging and radiation oncology. Methods: The MAMBA toolbox is implemented in Matlab. It provides opensource functions to compute VB statistical models of the input data,according to a great variety of regression schemes, and to derive VB maps of the observed significance level, performing a non-parametric permutation inference. The toolbox allows for including VB and global outcomes, as well as an arbitrary amount of VB and global Explanatory Variables (EVs). In addition, the Matlab Parallel Computing Toolbox is exploited to take advantage of the perfect parallelizability of most workloads. Results: The use of MAMBA was demonstrated by means of several realistic examples on a synthetic dataset mimicking a radiation oncology scenario. Conclusion: MAMBA is an open-source toolbox, freely available for academic and non-commercial purposes. It is designed to make state-of -the-art VB analysis accessible to research scientists without the programming resources needed to build from scratch their own software solutions. At the same time, the source code is handed out for more experienced users to complement their own tools, also customizing user-defined models. MAMBA guarantees high generality and flexibility in the design of the statistical models, significantly expanding on the features of available free tools for VB analysis. The presented toolbox aims at increasing the reach of VB studies as well as the sharing of research results. [ABSTRACT FROM AUTHOR]

Purpose: Early screening is crucial to improve the survival rate and recovery rate of lung cancer patients. Computer-aided diagnosis system (CAD) is a powerful tool to assist clinicians in early diagnosis. Lung nodules are characterized by spatial heterogeneity.However,many attempts use the two-dimensional multiview (MV) framework to learn and simply integrate multiple view features.These methods suffer from the problems of not capturing the spatial characteristics effectively and ignoring the variability of multiple views.In this paper,we propose a three-dimensional MV convolutional neural network (3D MVCNN) framework and embed the squeeze-and-excitation (SE) module in it to further address the variability of each view in the MV framework. Methods: First, the 3D multiple view samples of lung nodules are extracted by the spatial sampling method, and a 3D CNN is established to extract 3D abstract features. Second, build a 3D MVCNN framework according to the 3D multiple view samples and 3D CNN. This framework can learn more features of different views of lung nodules,taking into account the characteristics of spatial heterogeneity of lung nodules. Finally, to further address the variability of each view in the MV framework, a 3D MVSECNN model is constructed by introducing a SE module in the feature fusion stage. For training and testing purposes we used independent subsets of the public LIDC-IDRI dataset. Results: For the LIDC-IDRI dataset, this study achieved 96.04% accuracy and 98.59% sensitivity in the binary classification, and 87.76% accuracy in the ternary classification, which was higher than other state-of -the-art studies. The consistency score of 0.948 between the model predictions and pathological diagnosis was significantly higher than that between the clinician’s annotations and pathological diagnosis. Conclusions: The results show that our proposed method can effectively learn the spatial heterogeneity of nodules and solve the problem of multiple view variability.Moreover,the consistency analysis indicates that our method can provide clinicians with more accurate results of benign-malignant lung nodule classification for auxiliary diagnosis,which is important for assisting clinicians in clinical diagnosis. [ABSTRACT FROM AUTHOR]

Background: With the increasing use of computed tomography (CT) in clinical practice, limiting CT radiation exposure to reduce potential cancer risks has become one of the important directions of medical imaging research. As the dose decreases, the reconstructed CT image will be severely degraded by projection noise. Purpose: As an important method of image processing, supervised deep learning has been widely used in the restoration of low-dose CT (LDCT) in recent years. However, the normal-dose CT (NDCT) corresponding to a specific LDCT (it is regarded as the label of the LDCT, which is necessary for supervised learning) is very difficult to obtain so that the application of supervised learning methods in LDCT reconstruction is limited. It is necessary to construct a unsupervised deep learning framework for LDCT reconstruction that does not depend on paired LDCT-NDCT datasets. Methods: We presented an unsupervised learning framework for the transferring from the identity mapping to the low-dose reconstruction task, called marginal distribution adaptation in multiscale (MDAM). For NDCTs as source domain data, MDAM is an identity map with two parts: firstly, it establishes a dimensionality reduction mapping, which can obtain the same feature distribution from NDCTs and LDCTs; and then NDCTs is retrieved by reconstructing the image overview and details from the low-dimensional features. For the purpose of the feature transfer between source domain and target domain (LDCTs), we introduce the multiscale feature extraction in the MDAM, and then eliminate differences in probability distributions of these multiscale features between NDCTs and LDCTs through wavelet decomposition and domain adaptation learning. Results: Image quality evaluation metrics and subjective quality scores show that, as an unsupervised method, the performance of the MDAM approaches or even surpasses some state-of -the-art supervised methods. Especially, MDAM has been favorably evaluated in terms of noise suppression, structural preservation, and lesion detection. Conclusions: We demonstrated that, the MDAM framework can reconstruct corresponding NDCTs from LDCTs with high accuracy, and without relying on any labeles. Moreover, it is more suitable for clinical application compared with supervised learning methods. [ABSTRACT FROM AUTHOR]

Background: Digital radiography (DR) still presentsmany challenges and could have complex imaging acquisition and processing patterns in a clinical practice hindering quality standardization. Purpose: This technical note aims to report the 4-year experience with utilizing a custom DICOM metadata analytics program in clinical DR at a large institution. Methods: Thirty-eight DR systems of three vendors at multiple locations were configured to automatically send clinical DICOM images to a DICOM receiver.A suite of custom MATLAB programs was established to extract and store public and private header data weekly. Specific use cases are provided for systematic image acquisition investigation, image processing harmonization, exposure index (EI) longitudinal monitoring and EI target optimization. Results: For systematic acquisition investigation, an example of adult lumbar spine exam analysis was provided with statistics on manual acquisition versus the use of automatic exposure control (AEC, including AEC dose level, active cell, and backup timer), grid usage, and collimation for various projections. For processing harmonization, up to 12.6% of protocols were revealed to have processing parameter differences in an example of a mobile radiography fleet. In addition, inconsistent use of a post-acquisition image size function was also demonstrated, which resulted in anatomy size display variations. Bimonthly monitoring of median EI values showed expected trends, including changes after an AEC dose level adjustment for adult posterior-anterior chest imaging on a scanner system. An example of adult axillary shoulder EI target refinement was shared using the median value, eµ, based on the lognormal EI data distribution after parsing down to acquisitions with appropriate techniques. Conclusions: This analytics program enables systematic analysis of image acquisition and processing details. The information provides invaluable insights into real practice patterns, which can support data-driven quality standardization and optimization. [ABSTRACT FROM AUTHOR]

Background: Cone‐beam computed tomography (CBCT) systems acquire volumetric data more efficiently than fan‐beam or multislice CT, particularly when the anatomy of interest resides within the axial field‐of‐view of the detector and data can be acquired in one rotation. For such systems, scattered radiation remains a source of image quality degradation leading to increased noise, image artifacts, and CT number inaccuracies. Purpose: Recent advances in metal additive manufacturing allow the production of highly focused antiscatter grids (2D‐ASGs) that can be used to reduce scatter intensity, while preserving primary radiation transmission. We present the first implementation of a large‐area, 2D‐ASG for flat‐panel CBCT, including grid‐line artifact removal and related improvements in image quality. Methods: A 245 × 194 × 10 mm 2D‐ASG was manufactured from chrome–cobalt alloy using laser powder‐bed fusion (LPBF) (AM‐400; Renishaw plc, New Mills Wotton‐under‐Edge, UK). The 2D‐ASG had a square profile with a pitch of 9.09 lines/cm and 10:1 grid‐ratio. The nominal 0.1 mm grid septa were focused to a 732 mm x‐ray source to optimize primary x‐ray transmission and reduce grid‐line shadowing at the detector. Powder‐bed fusion ensured the structural stability of the ASG with no need for additional interseptal support. The 2D‐ASG was coupled to a 0.139‐mm element pitch flat‐panel detector (DRX 3543, Carestream Health) and proper alignment was confirmed by consistent grid‐line shadow thickness across the whole detector array. A 154‐mm diameter CBCT image‐quality‐assurance phantom was imaged using a rotary stage and a ceiling‐mounted, x‐ray unit (Proteus XR/a, GE Medical Systems, 80kVp, 0.5mAs). Grid‐line artifacts were removed using a combination of exposure‐dependent gain correction and spatial‐frequency, Fourier filtering. Projections were reconstructed using a Parker‐weighted, FDK algorithm and voxels were spatially averaged to 357 × 357 × 595 µm to improve the signal‐to‐noise characteristics of the CBCT reconstruction. Finally, in order to compare image quality with and without scatter, the phantom was scanned again under the same CBCT conditions but with no 2D‐ASG. No additional antiscatter (i.e., air‐gap, bowtie filtration) strategies were used to evaluate the effects in image quality caused by the 2D‐ASG alone. Results: The large‐area, 2D‐ASG prototype was successfully designed and manufactured using LPBF. CBCT image‐quality improvements using the 2D‐ASG included: an overall 14.5% CNR increase across the volume; up to 48.8% CNR increase for low‐contrast inserts inside the contrast plate of the QA phantom; and a 65% reduction of cupping artifact in axial profiles of water‐filled cross sections of the phantom. Advanced image processing strategies to remove grid line artifacts did not affect the spatial resolution or geometric accuracy of the system. Conclusions: LPBF can be used to manufacture highly efficient, 2D‐focused ASGs that can be easily coupled to clinical, flat‐panel detectors. The implementation of ASGs in CBCT leads to reduced scatter‐related artifacts, improved CT number accuracy, and enhanced CNR with no increased equivalent dose to the patient. Further improvements to image quality might be achieved with a combination of scatter‐correction algorithms and iterative‐reconstruction strategies. Finally, clinical applications where other scatter removal strategies are unfeasible might now achieve superior soft‐tissue visualization and quantitative capabilities. [ABSTRACT FROM AUTHOR]

Background: Auxiliary diagnosis and monitoring of lung diseases based on lung ultrasound (LUS) images is important clinical research. A‐line is one of the most common indicators of LUS that can offer support for the assessment of lung diseases. A traditional A‐line detection method mainly relies on experienced clinicians, which is inefficient and cannot meet the needs of these areas with backward medical level. Therefore, how to realize the automatic detection of A‐line in LUS image is important. Purpose: In order to solve the disadvantages of traditional A‐line detection methods, realize automatic and accurate detection, and provide theoretical support for clinical application, we proposed a novel A‐line detection method for LUS images with different probe types in this paper. Methods: First, the improved Faster R‐CNN model with a selection strategy of localization box was designed to accurately locate the pleural line. Then, the LUS image below the pleural line was segmented for independent analysis excluding the influence of other similar structures. Next, image‐processing methods based on total variation, matched filter, and gray difference were applied to achieve the automatic A‐line detection. Finally, the "depth" index was designed to verify the accuracy by judging whether the automatic measurement results belong to corresponding manual results (±5%). In experiments, 3000 convex array LUS images were used for training and validating the improved pleural line localization model by five‐fold cross validation. 850 convex array LUS images and 1080 linear array LUS images were used for testing the trained pleural line localization model and the proposed image‐processing‐based A‐line detection method. The accuracy analysis, error statistics, and Harsdorff distance were employed to evaluate the experimental results. Results: After 100 epochs, the mean loss value of training and validation set of improved Faster R‐CNN model reached 0.6540 and 0.7882, with the validation accuracy of 98.70%. The trained pleural line localization model was applied in the testing set of convex and linear probes and reached the accuracy of 97.88% and 97.11%, respectively, which were 3.83% and 8.70% higher than the original Faster R‐CNN model. The accuracy, sensitivity, and specificity of A‐line detection reached 95.41%, 0.9244%, 0.9875%, and 94.63%, 0.9230%, and 0.9766% for convex and linear probes, respectively. Compared to the experienced clinicians' results, the mean value and p value of depth error were 1.5342 ± 1.2097 and 0.9021, respectively, and the Harsdorff distance was 5.7305 ± 1.8311. In addition, the accumulated accuracy of the two‐stage experiment (pleural line localization and A‐line detection) was calculated as the final accuracy of the whole A‐line detection system. They were 93.39% and 91.90% for convex and linear probes, respectively, which were higher than these previous methods. Conclusions: The proposed method combining image processing and deep learning can automatically and accurately detect A‐line in LUS images with different probe types, which has important application value for clinical diagnosis. [ABSTRACT FROM AUTHOR]

Background: The morbidity of pulmonary embolism (PE) is only lower than that of coronary heart disease and hypertension. Early detection, early diagnosis, and timely treatment are the keys to effectively reduce the risk of death. Nevertheless, PE segmentation is still a challenging task at present. The automatic segmentation of PE is particularly important. On the one hand, manual segmentation of PE from a computed tomography (CT) sequence is very time‐consuming and prone to misdiagnose. On the other hand, an accurate contour of the location, volume, and shape of PE can help radiotherapists carry out targeted treatment and thus greatly increase the survival rate of patients. Therefore, developing an automatic and efficient PE segmentation approach is an urgent demand in clinical diagnosis. Purpose: An accurate segmentation of PE is critical for the diagnosis of PE. However, it remains a difficult and relevant problem in the field of medical image processing due to factors like incongruent sizes and shapes of emboli regions, and low contrast between embolisms and other tissues. To address this conundrum, in this study, a deep neural network (CAM‐Wnet) that incorporates coordinate attention (CA) mechanisms and pyramid pooling modules (PPMs) is proposed to end‐to‐end segment PE from CT image. Methods: CAM‐Wnet architecture is composed of coarse U‐Net and subdivision U‐Net stacked on top of each other. First, the coarse U‐Net uses a pretrained VGG‐19 as an encoder, which can transfer the features learned from ImageNet to other tasks. At the same time, CA residual blocks (CARBs) are introduced into the decoder of the coarse network to obtain a wider range of semantic information and find out the correlation between channels. Then, the multiplied results of input image and preliminary mask are put into the subdivision U‐Net for secondary feature distillation, and the encoder and decoder of the subdivision U‐Net are both constructed from CARBs, too. The PPMs are used between the encoder and the decoder of two U‐Net architectures to utilize global context information and further enhance the feature extraction effect. Finally, the improved focal loss function is used to train the network to further improve the segmentation effect. Results: In this study, we used the doctors' manual contours of the China‐Japan Friendship Hospital dataset to test the proposed architecture. We calculated the Precision, Recall, IoU, and F1‐score to evaluate the accuracy of the architecture for PE segmentation. The segmentation Precision for PE was found to be 0.9703, Recall was 0.963, IoU was 0.9353, and F1‐score was 0.9665. The experimental results show the effectiveness of the proposed method to automatically and accurately segment embolism in lung CT images. Furthermore, we also test the performance of our method on the liver tumor segmentation public dataset, which demonstrates the effectiveness and generalization ability of our method. Conclusions: CAM‐Wnet obtained more global information and semantic information with the introduction of multiscale pooling and attention mechanisms. Experimental results showed that the proposed method effectively improved the segmentation effect of PE in lung CT images and could be applied to assist doctors in clinical treatment. [ABSTRACT FROM AUTHOR]

Purpose: The liver is a common site for metastatic disease, which is a challenging and life‐threatening condition with a grim prognosis and outcome. We propose a standardized workflow for the diagnosis of oligometastatic disease (OMD), as a gold standard workflow has not been established yet. The envisioned workflow comprises the acquisition of a multimodal image data set, novel image processing techniques, and cone beam computed tomography (CBCT)‐guided biopsy for subsequent molecular subtyping. By combining morphological, molecular, and functional information about the tumor, a patient‐specific treatment planning is possible. We designed and manufactured an abdominal liver phantom that we used to demonstrate multimodal image acquisition, image processing, and biopsy of the OMD diagnosis workflow. Methods: The anthropomorphic abdominal phantom contains a rib cage, a portal vein, lungs, a liver with six lesions, and a hepatic vessel tree. This phantom incorporates three different lesion types with varying visibility under computed tomography (CT), magnetic resonance imaging (MRI), and positron emission tomography CT (PET‐CT), which reflects clinical reality. The phantom is puncturable and the size of the corpus and the organs is comparable to those of a real human abdomen. By using several modern additive manufacturing techniques, the manufacturing process is reproducible and allows to incorporate patient‐specific anatomies. As a first step of the OMD diagnosis workflow, a preinterventional CT, MRI, and PET‐CT data set of the phantom was acquired. The image information was fused using image registration and organ information was extracted via image segmentation. A CBCT‐guided needle puncture experiment was performed, where all six liver lesions were punctured with coaxial biopsy needles. Results: Qualitative observation of the image data and quantitative evaluation using contrast‐to‐noise ratio (CNR) confirms that one lesion type is visible only in MRI and not CT. The other two lesion types are visible in CT and MRI. The CBCT‐guided needle placement was performed for all six lesions, including those visible only in MRI and not CBCT. This was possible by successfully merging multimodal preinterventional image data. Lungs, bones, and liver vessels serve as realistic inhibitions during needle path planning. Conclusions: We have developed a reusable abdominal phantom that has been used to validate a standardized OMD diagnosis workflow. Utilizing the phantom, we have been able to show that a multimodal imaging pipeline is advantageous for a comprehensive detection of liver lesions. In a CBCT‐guided needle placement experiment we have punctured lesions that are invisible in CBCT using registered preinterventional MRI scans for needle path planning. [ABSTRACT FROM AUTHOR]

Purpose: Gadolinium‐based contrast agents (GBCAs) have been successfully applied in magnetic resonance (MR) imaging to facilitate better lesion visualization. However, gadolinium deposition in the human brain raised widespread concerns recently. On the other hand, although high‐resolution three‐dimensional (3D) MR images are more desired for most existing medical image processing algorithms, their long scan duration and high acquiring costs make 2D MR images still much more common clinically. Therefore, developing alternative solutions for 3D contrast‐enhanced MR image synthesis to replace GBCAs injection becomes an urgent requirement. Methods: This study proposed a deep learning framework that produces 3D isotropic full‐contrast T2Flair images from 2D anisotropic noncontrast T2Flair image stacks. The super‐resolution (SR) and contrast‐enhanced (CE) synthesis tasks are completed in sequence by using an identical generative adversarial network (GAN) with the same techniques. To solve the problem that intramodality datasets from different scanners have specific combinations of orientations, contrasts, and resolutions, we conducted a region‐based data augmentation technique on the fly during training to simulate various imaging protocols in the clinic. We further improved our network by introducing atrous spatial pyramid pooling, enhanced residual blocks, and deep supervision for better quantitative and qualitative results. Results: Our proposed method achieved superior CE‐synthesized performance in quantitative metrics and perceptual evaluation. In detail, the PSNR, structural‐similarity‐index, and AUC are 32.25 dB, 0.932, and 0.991 in the whole brain and 24.93 dB, 0.851, and 0.929 in tumor regions. The radiologists' evaluations confirmed that our proposed method has high confidence in the diagnosis. Analysis of the generalization ability showed that benefiting from the proposed data augmentation technique, our network can be applied to "unseen" datasets with slight drops in quantitative and qualitative results. Conclusion: Our work demonstrates the clinical potential of synthesizing diagnostic 3D isotropic CE brain MR images from a single 2D anisotropic noncontrast sequence. [ABSTRACT FROM AUTHOR]

Purpose: Breast cancer is the most common malignancy in women. Unfortunately, current breast imaging techniques all suffer from certain limitations: they are either not fully three dimensional, have an insufficient resolution or low soft‐tissue contrast. Grating interferometry breast computed tomography (GI‐BCT) is a promising X‐ray phase contrast modality that could overcome these limitations by offering high soft‐tissue contrast and excellent three‐dimensional resolution. To enable the transition of this technology to clinical practice, dedicated data‐processing algorithms must be developed in order to effectively retrieve the signals of interest from the measured raw data. Methods: This article proposes a novel denoising algorithm that can cope with the high‐noise amplitudes and heteroscedasticity which arise in GI‐BCT when operated in a low‐dose regime to effectively regularize the ill‐conditioned GI‐BCT inverse problem. We present a data‐driven algorithm called INSIDEnet, which combines different ideas such as multiscale image processing, transform‐domain filtering, transform learning, and explicit orthogonality to build an Interpretable NonexpanSIve Data‐Efficient network (INSIDEnet). Results: We apply the method to simulated breast phantom datasets and to real data acquired on a GI‐BCT prototype and show that the proposed algorithm outperforms traditional state‐of‐the‐art filters and is competitive with deep neural networks. The strong inductive bias given by the proposed model's architecture allows to reliably train the algorithm with very limited data while providing high model interpretability, thus offering a great advantage over classical convolutional neural networks (CNNs). Conclusions: The proposed INSIDEnet is highly data‐efficient, interpretable, and outperforms state‐of‐the‐art CNNs when trained on very limited training data. We expect the proposed method to become an important tool as part of a dedicated plug‐and‐play GI‐BCT reconstruction framework, needed to translate this promising technology to the clinics. [ABSTRACT FROM AUTHOR]

Purpose: Involuntary patient movement results in data discontinuities during computed tomography (CT) scans which lead to a serious degradation in the image quality. In this paper, we specifically address artifacts induced by patient motion during a head scan. Method: Instead of trying to solve an inverse problem, we developed a motion simulation algorithm to synthesize images with motion‐induced artifacts. The artifacts induced by rotation, translation, oscillation and any possible combination are considered. Taking advantage of the powerful learning ability of neural networks, we designed a novel 3D network structure with both a large reception field and a high image resolution to map the artifact‐free images from artifact‐contaminated images. Quantitative results of the proposed method were evaluated against the results of U‐Net and proposed networks without dilation structure. Thirty sets of motion contaminated images from two hospitals were selected to do a clinical evaluation. Result: Facilitating the training dataset with artifacts induced by variable motion patterns and the neural network, the artifact can be removed with good performance. Validation dataset with simulated random motion pattern showed outperformed image correction, and quantitative results showed the proposed network had the lowest normalized root‐mean‐square error, highest peak signal‐to‐noise ratio and structure similarity, indicating our network gave the best approximation of gold standard. Clinical image processing results further confirmed the effectiveness of our method. Conclusion: We proposed a novel deep learning‐based algorithm to eliminate motion artifacts. The convolutional neural networks trained with synthesized image pairs achieved promising results in artifacts reduction. The corrected images increased the diagnostic confidence compared with artifacts contaminated images. We believe that the correction method can restore the ability to successfully diagnose and avoid repeated CT scans in certain clinical circumstances. [ABSTRACT FROM AUTHOR]

Purpose: To create a network which fully utilizes multi‐sequence MRI and compares favorably with manual human contouring. Methods: We retrospectively collected 89 MRI studies of the pelvic cavity from patients with prostate cancer and cervical cancer. The dataset contained 89 samples from 87 patients with a total of 84 valid samples. MRI was performed with T1‐weighted (T1), T2‐weighted (T2), and Enhanced Dixon T1‐weighted (T1DIXONC) sequences. There were two cohorts. The training cohort contained 55 samples and the testing cohort contained 29 samples. The MRI images in the training cohort contained contouring data from radiotherapist α. The MRI images in the testing cohort contained contouring data from radiotherapist α and contouring data from another radiotherapist: radiotherapist β. The training cohort was used to optimize the convolution neural networks, which included the attention mechanism through the proposed activation module and the blended module into multiple MRI sequences, to perform autodelineation. The testing cohort was used to assess the networks' autodelineation performance. The contoured organs at risk (OAR) were the anal canal, bladder, rectum, femoral head (L), and femoral head (R). Results: We compared our proposed network with UNet and FuseUNet using our dataset. When T1 was the main sequence, we input three sequences to segment five organs and evaluated the results using four metrics: the DSC (Dice similarity coefficient), the JSC (Jaccard similarity coefficient), the ASD (average mean distance), and the 95% HD (robust Hausdorff distance). The proposed network achieved improved results compared with the baselines among all metrics. The DSC were 0.834±0.029, 0.818±0.037, and 0.808±0.050 for our proposed network, FuseUNet, and UNet, respectively. The 95% HD were 7.256±2.748 mm, 8.404±3.297 mm, and 8.951±4.798 mm for our proposed network, FuseUNet, and UNet, respectively. Our proposed network also had superior performance on the JSC and ASD coefficients. Conclusion: Our proposed activation module and blended module significantly improved the performance of FuseUNet for multi‐sequence MRI segmentation. Our proposed network integrated multiple MRI sequences efficiently and autosegmented OAR rapidly and accurately. We also discovered that three‐sequence fusion (T1‐T1DIXONC‐T2) was superior to two‐sequence fusion (T1‐T2 and T1‐T1DIXONC, respectively). We infer that the more MRI sequences fused, the better the automatic segmentation results. [ABSTRACT FROM AUTHOR]

Purpose: X‐ray phase and dark‐field (DF) imaging have been shown to improve the diagnostic capabilities of X‐ray systems. However, these methods have found limited clinical use due to the need for multiple precision gratings with limited field of view or requirements on X‐ray coherence that may not be easily translated to clinical practice. This work aims to develop a practicable X‐ray phase and DF imaging system that could be translated and practiced in the clinic. Methods: This work employs a conventional source to create structured illumination with a simple wire mesh. A mesh‐shifting algorithm is used to allow wider Fourier windowing to enhance resolution. Deconvolution of the source spot width and camera resolution improves accuracy. Polycapillary optics are employed to enhance coherence. The effects of incorporating optics with two different focal lengths are compared. Information apparent in enhanced absorption images, phase images, and DF images of fat embedded phantoms were compared and subjected to a limited receiver operator characteristic (ROC) study. The DF images of the moist and dry porous object (sponges) were compared. Results: The mesh‐based phase and DF imaging system constructs images with three different information types: scatter‐free absorption images, differential phase images, and scatter magnitude/DF images, simultaneously from the same original image. The polycapillary optic enhances the coherence of the beam. The deblurring technique corrects the phase signal error due to geometrical blur and the limitation of the camera modulation transfer function (MTF) and removes image artifacts to improve the resolution in a single shot. The mesh‐shifting method allows the use of a wider Fourier processing window, which gives even higher resolution, at the expense of an increased dose. The limited ROC study confirms the efficacy of the system over the conventional system. DF images of moist and dry porous object show the significance of the system in the imaging of lung infections. Conclusion: The mesh‐based X‐ray phase and DF imaging system is an inexpensive and easy setup in terms of alignment and data acquisition and can produce phase and DF images in a single shot with wide field of view. The system shows significant potential for use in diagnostic imaging in a clinical setting. [ABSTRACT FROM AUTHOR]

Purpose: Deep learning has achieved impressive performance across a variety of tasks, including medical image processing. However, recent research has shown that deep neural networks (DNNs) are susceptible to small adversarial perturbations in the image, which raise safety concerns about the deployment of these systems in clinical settings. Methods: To improve the defense of the medical imaging system against adversarial examples, we propose a new model‐based defense framework for medical image DNNs model equipped with pruning and attention mechanism module based on the analysis of the reason why existing medical image DNNs models are vulnerable to attacks from adversarial examples is that complex biological texture of medical imaging and overparameterized medical image DNNs model. Results: Three benchmark medical image datasets have verified the effectiveness of our method in improving the robustness of medical image DNNs models. In the chest X‐ray datasets, our defending method can even achieve up 77.18% defense rate for projected gradient descent attack and 69.49% defense rate for DeepFool attack. And through ablation experiments on the pruning module and the attention mechanism module, it is verified that the use of pruning and attention mechanism can effectively improve the robustness of the medical image DNNs model. Conclusions: Compared with the existing model‐based defense methods proposed for natural images, our defense method is more suitable for medical images. Our method can be a general strategy to approach the design of more explainable and secure medical deep learning systems, and can be widely used in various medical image tasks to improve the robustness of medical models. [ABSTRACT FROM AUTHOR]

Purpose: Megavoltage computed tomography (MVCT) offers an opportunity for adaptive helical tomotherapy. However, high noise and reduced contrast in the MVCT images due to a decrease in the imaging dose to patients limits its usability. Therefore, we propose an algorithm to improve the image quality of MVCT. Methods: The proposed algorithm generates kilovoltage CT (kVCT)‐like images from MVCT images using a cycle‐consistency generative adversarial network (cycleGAN)‐based image synthesis model. Data augmentation using an affine transformation was applied to the training data to overcome the lack of data diversity in the network training. The mean absolute error (MAE), root‐mean‐square error (RMSE), peak signal‐to‐noise ratio (PSNR), and structural similarity index measure (SSIM) were used to quantify the correction accuracy of the images generated by the proposed algorithm. The proposed method was validated by comparing the images generated with those obtained from conventional and deep learning‐based image processing method through non‐augmented datasets. Results: The average MAE, RMSE, PSNR, and SSIM values were 18.91 HU, 69.35 HU, 32.73 dB, and 95.48 using the proposed method, respectively, whereas cycleGAN with non‐augmented data showed inferior results (19.88 HU, 70.55 HU, 32.62 dB, 95.19, respectively). The voxel values of the image obtained by the proposed method also indicated similar distributions to those of the kVCT image. The dose‐volume histogram of the proposed method was also similar to that of electron density corrected MVCT. Conclusions: The proposed algorithm generates synthetic kVCT images from MVCT images using cycleGAN with small patient datasets. The image quality achieved by the proposed method was correspondingly improved to the level of a kVCT image while maintaining the anatomical structure of an MVCT image. The evaluation of dosimetric effectiveness of the proposed method indicates the applicability of accurate treatment planning in adaptive radiation therapy. [ABSTRACT FROM AUTHOR]

Purpose: Positron emission tomography (PET) imaging with various tracers is increasingly used in Alzheimer's disease (AD) studies. However, access to PET scans using new or less‐available tracers with sophisticated synthesis and short half‐life isotopes may be very limited. Therefore, it is of great significance and interest in AD research to assess the feasibility of generating synthetic PET images of less‐available tracers from the PET image of another common tracer, in particular 18F‐FDG. Methods: We implemented advanced deep learning methods using the U‐Net model to predict 11C‐UCB‐J PET images of synaptic vesicle protein 2A (SV2A), a surrogate of synaptic density, from 18F‐FDG PET data. Dynamic 18F‐FDG and 11C‐UCB‐J scans were performed in 21 participants with normal cognition (CN) and 33 participants with Alzheimer's disease (AD). Cerebellum was used as the reference region for both tracers. For 11C‐UCB‐J image prediction, four network models were trained and tested, which included 1) 18F‐FDG SUV ratio (SUVR) to 11C‐UCB‐J SUVR, 2) 18F‐FDG Ki ratio to 11C‐UCB‐J SUVR, 3) 18F‐FDG SUVR to 11C‐UCB‐J distribution volume ratio (DVR), and 4) 18F‐FDG Ki ratio to 11C‐UCB‐J DVR. The normalized root mean square error (NRMSE), structure similarity index (SSIM), and Pearson's correlation coefficient were calculated for evaluating the overall image prediction accuracy. Mean bias of various ROIs in the brain and correlation plots between predicted images and true images were calculated for ROI‐based prediction accuracy. Following a similar training and evaluation strategy, 18F‐FDG SUVR to 11C‐PiB SUVR network was also trained and tested for 11C‐PiB static image prediction. Results: The results showed that all four network models obtained satisfactory 11C‐UCB‐J static and parametric images. For 11C‐UCB‐J SUVR prediction, the mean ROI bias was −0.3% ± 7.4% for the AD group and −0.5% ± 7.3% for the CN group with 18F‐FDG SUVR as the input, −0.7% ± 8.1% for the AD group, and −1.3% ± 7.0% for the CN group with 18F‐FDG Ki ratio as the input. For 11C‐UCB‐J DVR prediction, the mean ROI bias was −1.3% ± 7.5% for the AD group and −2.0% ± 6.9% for the CN group with 18F‐FDG SUVR as the input, −0.7% ± 9.0% for the AD group, and −1.7% ± 7.8% for the CN group with 18F‐FDG Ki ratio as the input. For 11C‐PiB SUVR image prediction, which appears to be a more challenging task, the incorporation of additional diagnostic information into the network is needed to control the bias below 5% for most ROIs. Conclusions: It is feasible to use 3D U‐Net‐based methods to generate synthetic 11C‐UCB‐J PET images from 18F‐FDG images with reasonable prediction accuracy. It is also possible to predict 11C‐PiB SUVR images from 18F‐FDG images, though the incorporation of additional non‐imaging information is needed. [ABSTRACT FROM AUTHOR]

Purpose: To compare a novel thick‐slab projection technique for ultra‐low dose computed tomography (CT; thoracic tomogram) with conventional chest x ray with respect to 13 diagnostic categories. Methods: With the approval of the institutional ethics board, a dataset was retrospectively collected of 22 consecutive patients who had undergone a clinically requested emergency room conventional chest x ray (CXR) and a same‐day standard‐of‐care non‐contrast CT. Scanner specific noise was added to the CT images to simulate a target dose of 0.18 mSv. A novel algorithm was used to post‐process CT images as coronal isotropic reformats by applying a voxel‐based, locally normalized weighted‐intensity projection to generate 2 cm thick slabs with 1 cm overlap. Three chest radiologists with no prior training for the study reviewed the CXR and thoracic tomogram for each case and assessed each diagnostic category (pneumonic infiltrates, pulmonary edema, interstitial lung disease, nodules > 5 mm, nodules < 5 mm, pleural effusion, pericardial effusion, heart size, acute bone fractures, foreign bodies, pneumothorax, mediastinal vessel diameter, free abdominal air) on a Likert scale from −4 (definitely absent/normal) to +4 (definitely present/abnormal). MRMC ROC curves were generated for each category. Time for interpretation and subjective image quality score (0–10) were also assessed. Results: For focal lung disease (pneumonic infiltrates, nodules < 5 mm, nodules > 5mm), the area under the ROC curve (AUC) was significantly higher for thoracic tomograms than CXR (0.803 vs 0.648, respectively, P = 0.02). For non‐focal lung disease (pulmonary edema, interstitial lung disease) and effusions (pulmonary, pericardial), the AUC was larger for thoracic tomograms than CXR but the difference did not reach significance (0.870 vs 0.833, P = 0.141; and 0.823 vs 0.752, P = 0.296, respectively). For acute bone fractures and foreign bodies, the AUC was smaller for thoracic tomograms than CXR, the difference was however not significant (0.491 vs 0.532, P = 0.42; and 0.871 vs 0.971, P = 0.39, respectively). Other diagnostic categories had no true positive cases in the dataset. The mean time for interpretation for each was 36.9 and 24.0 s with standard deviations of 0.857 and 5.977. The image quality score for each was 8.2 and 7.8 with standard deviations of 0.970 and 1.614. Conclusion: Thoracic tomograms were found to be diagnostically superior to CXR for focal lung disease, at no increased radiation dose. The thoracic tomogram presents an opportunity to improve the standard‐of‐care for patients who would otherwise receive a conventional CXR. [ABSTRACT FROM AUTHOR]

Purpose: One of the biggest challenges in applying megavoltage (MV) treatment beam imaging for monitoring spine motion in stereotactic body radiotherapy (SBRT) is the small beam apertures in the images due to strong beam modulations in IMRT planning. The purpose of this study is to investigate the feasibility of a markerless motion tracking method in spine SBRT delivery using a novel enhanced synthetic treatment beam (ESTB) imaging technique. Methods: Three clinical spine SBRT plans using 6XFFF beams and sliding window IMRT technique were transferred to a thorax phantom and delivered by a TrueBeam machine. Before delivery, the phantom was aligned to the plan isocenter using CBCT setup and verified with a second CBCT, and then, 2 mm shifts were introduced in both the craniocaudal (CC) and the left–right (LR) directions with the couch. During beam delivery, MV images were continuously taken with an electronic portal imaging device (EPID) and automatically grabbed by Varian iTools Capture software with a frame rate of 11.6 Hz. After preprocessing for scatter correction and beam intensity compensation, every 50 frames of MV images were combined to generate a series of ESTB images for each beam. The ESTB images were then registered to the projections of the verification CBCT at the matched beam angles to detect the 2 mm shifts. Results: Compared to snapshot MV images, the ESTB images had significantly enlarged fields of view (FOVs) and improved image quality. Based on two‐dimensional (2D) rigid registration, the ESTB image to CBCT projection matching showed submillimeter accuracy in detecting motion. Specifically, the root mean square errors in detecting the LR/CC shifts were 0.35/0.28, 0.32/0.35, 0.63/0.44, 0.55/0.51, and 0.69/0.42 mm at gantry angles 180, 160, 140, 120, and 100, respectively. Conclusion: Our results in the phantom study suggest that ESTB images from a sliding window IMRT plan can be used to detect spine motion, with submillimeter precision in the 2D plane perpendicular to the beam. [ABSTRACT FROM AUTHOR]

Purpose: Myocardial perfusion imaging using computed tomography (MPI‐CT) and coronary CT angiography (CTA) have the potential to make CT an ideal noninvasive imaging gatekeeper exam for invasive coronary angiography. However, beam hardening can prevent accurate blood flow estimation in dynamic MPI‐CT and can create artifacts that resemble flow deficits in single‐shot MPI‐CT. In this work, we compare four automatic beam hardening correction algorithms (ABHCs) applied to CT images, for their ability to produce accurate single images of contrast and accurate MPI flow maps using images from conventional CT systems, without energy sensitivity. Methods: Previously, we reported a method, herein called ABHC‐1, where we iteratively optimized a cost function sensitive to beam hardening artifacts in MPI‐CT images and used a low order polynomial correction on projections of segmentation‐processed CT images. Here, we report results from two new algorithms with higher order polynomial corrections, ABHC‐2 and ABHC‐3 (with three and seven free parameters, respectively), having potentially better correction but likely reduced estimability. Additionally, we compared results to an algorithm reported by others in the literature (ABHC‐NH). Comparisons were made on a digital static phantom with simulated water, bone, and iodine regions; on a digital dynamic anthropomorphic phantom, with simulated blood flow; and on preclinical porcine experiments. We obtained CT images on a prototype spectral detector CT (Philips Healthcare) scanner that provided both conventional and virtual keV images, allowing us to quantitatively compare corrected CT images to virtual keV images. To test these methods' parameter optimization sensitivity to noise, we evaluated results on images obtained using different mAs. Results: In images of the static phantom, ABHC‐2 reduced beam hardening artifacts better than our previous ABHC‐1 algorithm, giving artifacts smaller than 1.8 HU, even in the presence of high noise which should affect parameter optimization. Taken together, the quality of static phantom results ordered ABHC‐2> ABHC‐3> ABHC‐1>> ABHC‐NH. In an anthropomorphic MPI‐CT simulator with homogeneous myocardial blood flow of 100 ml⋅min−1⋅100 g−1, blood flow estimation results were 122 ± 24 (FBP), 135 ± 24 (ABHC‐NH), 104 ± 14 (ABHC‐1), 100 ± 12 (ABHC‐2), and 108 ± 18 (ABHC‐3) ml⋅min−1⋅100 g−1, showing ABHC‐2 as a clear winner. Visual and quantitative evaluations showed much improved homogeneity of myocardial flow with ABHC‐2, nearly eliminating substantial artifacts in uncorrected flow maps which could be misconstrued as flow deficits. ABHC‐2 performed universally better than ABHC‐1, ABHC‐3, and ABHC‐NH in simulations with different acquisitions (varying noise and kVp values). In the presence of a simulated flow deficit, all ABHC methods retained the flow deficit, and ABHC‐2 gave the most accurate flow ratio and homogeneity. ABHC‐3 corrected phantom flow values were slightly better than ABHC‐2, in noiseless images, suggesting that reduced quality in noisy images was due to reduced estimability. In an experiment with a pig expected to have uniform flow, ABHC‐2 applied to conventional images improved flow maps to compare favorably to those from 70keV images. Conclusion: The automated algorithm can be used with different parametric BH correction models. ABHC‐2 improved MPI‐CT blood flow estimation as compared to other approaches and was robust to noisy images. In simulation and preclinical experiments, ABHC‐2 gave results approaching gold standard 70 keV measurements. [ABSTRACT FROM AUTHOR]

Purpose: To propose and evaluate a fully automated technique for quantification of fibroglandular tissue (FGT) and background parenchymal enhancement (BPE) in breast MRI. Methods: We propose a fully automated method, where after preprocessing, FGT is segmented in T1‐weighted, nonfat‐saturated MRI. Incorporating an anatomy‐driven prior probability for FGT and robust texture descriptors against intensity variations, our method effectively addresses major image processing challenges, including wide variations in breast anatomy and FGT appearance among individuals. Our framework then propagates this segmentation to dynamic contrast‐enhanced (DCE)‐MRI to quantify BPE within the segmented FGT regions. Axial and sagittal image data from 40 cancer‐unaffected women were used to evaluate our proposed method vs a manually annotated reference standard. Results: High spatial correspondence was observed between the automatic and manual FGT segmentation (mean Dice similarity coefficient 81.14%). The FGT and BPE quantifications (denoted FGT% and BPE%) indicated high correlation (Pearson's r = 0.99 for both) between automatic and manual segmentations. Furthermore, the differences between the FGT% and BPE% quantified using automatic and manual segmentations were low (mean differences: −0.66 ± 2.91% for FGT% and −0.17 ± 1.03% for BPE%). When correlated with qualitative clinical BI‐RADS ratings, the correlation coefficient for FGT% was still high (Spearman's ρ = 0.92), whereas that for BPE was lower (ρ = 0.65). Our proposed approach also performed significantly better than a previously validated method for sagittal breast MRI. Conclusions: Our method demonstrated accurate fully automated quantification of FGT and BPE in both sagittal and axial breast MRI. Our results also suggested the complexity of BPE assessment, demonstrating relatively low correlation between segmentation and clinical rating. [ABSTRACT FROM AUTHOR]

Purpose: One of the main sources of uncertainty in proton therapy is the conversion of the Hounsfield Units of the planning CT to (relative) proton stopping powers. Proton radiography provides range error maps but these can be affected by other sources of errors as well as the CT conversion (e.g., residual misalignment). To better understand and quantify range uncertainty, it is desirable to measure the individual contributions and particularly those associated to the CT conversion. Methods: A workflow is proposed to carry out an assessment of the CT conversion solely on the basis of proton radiographs of real tissues measured with a multilayer ionization chamber (MLIC). The workflow consists of a series of four stages: (a) CT and proton radiography acquisitions, (b) CT and proton radiography registration in postprocessing, (c) sample‐specific validation of the semi‐empirical model both used in the registration and to estimate the water equivalent path length (WEPL), and (d) WEPL error estimation. The workflow was applied to a pig head as part of the validation of the CT calibration of the proton therapy center PARTICLE at UZ Leuven, Belgium. Results: The CT conversion‐related uncertainty computed based on the well‐established safety margin rule of 1.2 mm + 2.4% were overestimated by 71% on the pig head. However, the range uncertainty was very much underestimated where cavities were encountered by the protons. Excluding areas with cavities, the overestimation of the uncertainty was 500%. A correlation was found between these localized errors and HUs between −1000 and −950, suggesting that the underestimation was not a consequence of an inaccurate conversion but was probably rather due to the resolution of the CT leading to material mixing at interfaces. To reduce these errors, the CT calibration curve was adapted by increasing the HU interval corresponding to the air up to −950. Conclusion: The application of the workflow as part of the validation of the CT conversion to RSPs showed an overall overestimation of the expected uncertainty. Moreover, the largest WEPL errors were found to be related to the presence of cavities which nevertheless are associated with low WEPL values. This suggests that the use of this workflow on patients or in a generalized study on different types of animal tissues could shed sufficient light on how the contributions to the CT conversion‐related uncertainty add up to potentially reduce up to several millimeters the uncertainty estimations taken into account in treatment planning. All the algorithms required to perform the workflow were implemented in the computational tool named openPR which is part of openREGGUI, an open‐source image processing platform for adaptive proton therapy. [ABSTRACT FROM AUTHOR]

Purpose: High quality radiotherapy is challenging in cases where multiple targets with independent motion are simultaneously treated. A real‐time tumor tracking system that can simultaneously account for the motion of two targets was developed and characterized. Methods: The multitarget tracking system was implemented on a magnetic resonance imaging (MRI)‐linac and utilized multi‐leaf collimator (MLC) tracking to adapt the radiation beam to phantom targets reproducing motion with prostate and lung motion traces. Multitarget tracking consisted of three stages: (a) pretreatment aperture segmentation where the treatment aperture was divided into segments corresponding to each target, (b) MR imaging where the positions of the two targets were localized, and (c) MLC tracking where an updated treatment aperture was calculated. Electronic portal images (EPID) acquired during irradiation were analyzed to characterize geometric uncertainty and tracking latency. Results: Multitarget MLC tracking effectively accounted for the motion of both targets during treatment. The root‐mean‐square error between the centers of the targets and the centers of the corresponding MLC leaves were reduced from 5.5 mm without tracking to 2.7 mm with tracking for lung motion traces and reduced from 4.2 to 1.4 mm for prostate motion traces. The end‐to‐end latency of tracking was measured to be 328 ± 44 ms. Conclusions: We have demonstrated the first experimental implementation of MLC tracking for multiple targets having independent motion. This technology takes advantage of the imaging capabilities of MRI‐linacs and would allow treatment margins to be reduced in cases where multiple targets are simultaneously treated. [ABSTRACT FROM AUTHOR]

Purpose: To investigate the utility of gradient dose segmented analysis (GDSA) in combination with in vivo electronic portal imaging device (EPID) images to predict changes in the PTV mean dose for patient cases. Also, we use the GDSA to retrospectively analyze patients treated in our clinic to assess deviations for different treatment sites and use time‐series data to observe any day‐to‐day changes. Methods: In vivo EPID transit images acquired on the Varian Halcyon were analyzed for simulated errors in a phantom, including gas bubbles, weight loss, patient shifts, and an arm erroneously in the field. GDSA threshold parameters were tuned to maximize the coefficient of determination (R2) between GDSA metrics and the change in the PTV mean dose (Dmean) as estimated in a treatment planning system (TPS). Similarly for a gamma analysis, the gamma criteria were adjusted to maximize R2 between gamma pass rate and the change in the PTV Dmean from the TPS. The predictive accuracy of these models was tested on patient data measuring the mean and standard deviation of the difference in the predicted change in PTV Dmean and the change in PTV Dmean measured in the TPS. This analysis was extended retrospectively for every patient treated over a 23‐month period (n = 852 patients) to assess the range of expected deviations that occurred during routine clinical operation, as well as to assess any differences between treatment sites. Grouping patients treated on the same day, a time‐series analysis was performed to determine if GDSA metrics could add value in tracking machine behavior over time. Results: For the phantom data, analyzing the errors, except for shifts, and comparing the change in PTV Dmean and GDSA mean, a maximal R2 = 0.90 was found for a dose threshold of 5% and gradient threshold of 3 mm. For the gamma approach a linear fit between the gamma pass rate for change in the PTV Dmean was assessed for different criteria, using the same image data. A maximal, R2 = 0.84 was found for a gamma criteria of 3%/3 mm, 45% lower dose threshold. For patient data, the predictive accuracy of the change in the PTV Dmean using the GDSA approach and the gamma approach was 0.09 ± 0.98 % and − 0.65 ± 2.21%, respectively. Comparing the two approaches the accuracy did not significantly differ (P = 0.38), whereas the precision of the GDSA prediction is significantly less (P < 0.001). The dosimetric impact of shifts was not detectable with either the GDSA or gamma approach. Analysis of all patients treated over 23 months showed that over 95% of fractions treated deviated from the first fraction by 2% or less. Deviations> 2% occurred most frequently for the later fractions of head‐and‐neck and lung treatments. Additionally, averaging the GDSA mean metric over all patients on a given treatment day showed that changes in the machine output on the order of 1% could be identified. Conclusions: GDSA of in vivo EPID images is a useful technique for monitoring patient changes during the course of treatment, particularly weight loss and tumor shrinkage. The GDSA mean provides a quantitative estimate of the change in the PTV Dmean, giving a simple, quantitative metric by which to flag patients with clinically meaningful deviations in treatment. Averaging the GDSA metric over all patients treated on a given day and tracking daily variations can also provide a flag for any systematic deviations in treatment due to machine performance. [ABSTRACT FROM AUTHOR]

Purpose: Body composition is known to be associated with many diseases including diabetes, cancers, and cardiovascular diseases. In this paper, we developed a fully automatic body tissue decomposition procedure to segment three major compartments that are related to body composition analysis — subcutaneous adipose tissue (SAT), visceral adipose tissue (VAT), and muscle. Three additional compartments — the ventral cavity, lung, and bones — were also segmented during the segmentation process to assist segmentation of the major compartments. Methods: A convolutional neural network (CNN) model with densely connected layers was developed to perform ventral cavity segmentation. An image processing workflow was developed to segment the ventral cavity in any patient's computed tomography (CT) using the CNN model, then further segment the body tissue into multiple compartments using hysteresis thresholding followed by morphological operations. It is important to segment ventral cavity firstly to allow accurate separation of compartments with similar Hounsfield unit (HU) inside and outside the ventral cavity. Results: The ventral cavity segmentation CNN model was trained and tested with manually labeled ventral cavities in 60 CTs. Dice scores (mean ± standard deviation) for ventral cavity segmentation were 0.966 ± 0.012. Tested on CT datasets with intravenous (IV) and oral contrast, the Dice scores were 0.96 ± 0.02, 0.94 ± 0.06, 0.96 ± 0.04, 0.95 ± 0.04, and 0.99 ± 0.01 for bone, VAT, SAT, muscle, and lung, respectively. The respective Dice scores were 0.97 ± 0.02, 0.94 ± 0.07, 0.93 ± 0.06, 0.91 ± 0.04, and 0.99 ± 0.01 for non‐contrast CT datasets. Conclusion: A body tissue decomposition procedure was developed to automatically segment multiple compartments of the ventral body. The proposed method enables fully automated quantification of three‐dimensional (3D) ventral body composition metrics from CT images. [ABSTRACT FROM AUTHOR]

This manuscript describes a dataset of thoracic cavity segmentations and discrete pleural effusion segmentations we have annotated on 402 computed tomography (CT) scans acquired from patients with non‐small cell lung cancer. The segmentation of these anatomic regions precedes fundamental tasks in image analysis pipelines such as lung structure segmentation, lesion detection, and radiomics feature extraction. Bilateral thoracic cavity volumes and pleural effusion volumes were manually segmented on CT scans acquired from The Cancer Imaging Archive "NSCLC Radiomics" data collection. Four hundred and two thoracic segmentations were first generated automatically by a U‐Net based algorithm trained on chest CTs without cancer, manually corrected by a medical student to include the complete thoracic cavity (normal, pathologic, and atelectatic lung parenchyma, lung hilum, pleural effusion, fibrosis, nodules, tumor, and other anatomic anomalies), and revised by a radiation oncologist or a radiologist. Seventy‐eight pleural effusions were manually segmented by a medical student and revised by a radiologist or radiation oncologist. Interobserver agreement between the radiation oncologist and radiologist corrections was acceptable. All expert‐vetted segmentations are publicly available in NIfTI format through The Cancer Imaging Archive at https://doi.org/10.7937/tcia.2020.6c7y‐gq39. Tabular data detailing clinical and technical metadata linked to segmentation cases are also available. Thoracic cavity segmentations will be valuable for developing image analysis pipelines on pathologic lungs — where current automated algorithms struggle most. In conjunction with gross tumor volume segmentations already available from "NSCLC Radiomics," pleural effusion segmentations may be valuable for investigating radiomics profile differences between effusion and primary tumor or training algorithms to discriminate between them. [ABSTRACT FROM AUTHOR]

Purpose: In real‐time electronic portal imaging device (EPID) dosimetry applications where on‐treatment measured transmission images are compared to an ideal predicted image, ideally a tight tolerance should be set on the quantitative image comparison in order to detect a wide variety of possible delivery errors. However, this is currently not possible due to the appearance of banding artifacts in individual frames of the measured EPID image sequences. The purpose of this work was to investigate simulating banding artifacts in our cine‐EPID predicted image sequences to improve matching of individual image frames to the acquired image sequence. Increased sensitivity of this method to potential treatment delivery errors would represent an improvement in patient safety and treatment accuracy. Methods: A circuit board was designed and built to capture the target current (TARG‐I) and forward power signals produced by the linac to help model the discrete beam‐formation process of the linac. To simulate the temporal–spatial nature of the EPID readout, a moving read out mask was applied with the timing of the application of the readout mask synchronized to the TARG‐I pulses. Since identifying the timing of the first TARG‐I pulse affected the location of the banding artifacts throughout the image sequence, and furthermore the first several TARG‐I pulses at the beginning of "beam on" are not at full height yet (i.e., dose rate is ramping up), the forward‐power signal was also used to assist in reliable detection of the first radiation pulse of the beam delivery. The predicted EPID cine‐image sequence obtained using a comprehensive physics‐based model was modified to incorporate the discrete nature of the EPID frame readout. This modified banding predicted EPID (MBP‐EPID) image sequence was then compared to its corresponding measured EPID cine‐image sequence on a frame‐by‐frame basis. The EPID was mounted on a Clinac 2100ix linac (Varian Medical Systems, Palo Alto, CA). The field size was set to 21.4 × 28.6 cm2 with no MLC modulation, beam energy of 6 MV, dose rate of 600 MU/min, and 700 MU were delivered for each clockwise (CW) and counter‐clockwise (CCW) arc. No phantoms were placed in the beam. Results: The dose rate ramp up effect was observed at the beginning irradiations, and the identification and timing of the radiation pulses, even during the dose rate ramp up, were able to be quantified using the TARG‐I and forward power signals. The approach of capturing individual dose pulses and synchronizing with the mask image applied to the original predicted EPID image sequence was demonstrated to model the actual EPID readout. The MBP‐EPID image sequences closely reproduced the location and magnitude of the banding features observed in the acquired (i.e., measured) image sequence, for all test irradiations examined here. Conclusions: The banding artifacts observed in the measured EPID cine‐frame sequences were reproduced in the predicted EPID cine‐frames by simulating the discrete temporal–spatial nature of the EPID read out. The MBP‐EPID images showed good agreement qualitatively to the corresponding measured EPID frame sequence of a simple square test field, without any phantom in the beam. This approach will lead to improved image comparison tolerances for real‐time patient dosimetry applications. [ABSTRACT FROM AUTHOR]

Purpose: To develop an image processing methodology for noninvasive three‐dimensional (3D) quantification of microwave thermal ablation zones in vivo using x‐ray computed tomography (CT) imaging without injection of a contrast enhancing material. Methods: Six microwave (MW) thermal ablation procedures were performed in three pigs. The ablations were performed with a constant heating duration of 8 min and power level of 30 W. During the procedure images from sixty 1 mm thick slices were acquired every 30 s. At the end of all ablation procedures for each pig, a contrast‐enhanced scan was acquired for reference. Special algorithms for addressing challenges stemming from the 3D in vivo setup and processing the acquired images were prepared. The algorithms first rearranged the data to account for the oblique needle orientation and for breathing motion. Then, the gray level variance changes were analyzed, and optical flow analysis was applied to the treated volume in order to obtain the ablation contours and reconstruct the ablation zone in 3D. The analysis also included a special correction algorithm for eliminating artifacts caused by proximal major blood vessels and blood flow. Finally, 3D reference reconstructions from the contrast‐enhanced scan were obtained for quantitative comparison. Results: For four ablations located >3 mm from a large blood vessel, the mean dice similarity coefficient (DSC) and the mean absolute radial discrepancy between the contours obtained from the reference contrast‐enhanced images and the contours produced by the algorithm were 0.82 ± 0.03 and 1.92 ± 1.47 mm, respectively. In two cases of ablation adjacent to large blood vessels, the average DSC and discrepancy were: 0.67 ± 0.6 and 2.96 ± 2.15 mm, respectively. The addition of the special correction algorithm utilizing blood vessels mapping improved the mean DSC and the mean absolute discrepancy to 0.85 ± 0.02 and 1.19 ± 1.00 mm, respectively. Conclusions: The developed algorithms provide highly accurate detailed contours in vivo (average error < 2.5 mm) and cope well with the challenges listed above. Clinical implementation of the developed methodology could potentially provide real time noninvasive 3D accurate monitoring of MW thermal ablation in‐vivo, provided that the radiation dose can be reduced. [ABSTRACT FROM AUTHOR]

Background: Dosimetry in nuclear medicine often relies on estimating pharmacokinetics based on sparse temporal data. As analysis methods move toward image‐based three‐dimensional computation, it becomes important to interpolate and extrapolate these data without requiring manual intervention; that is, in a manner that is highly efficient and reproducible. Iterative least‐squares solvers are poorly suited to this task because of the computational overhead and potential to optimize to local minima without applying tight constraints at the outset. Methodology: This work describes a fully analytical method for solving three‐phase exponential time‐activity curves based on three measured time points in a manner that may be readily employed by image‐based dosimetry tools. The methodology uses a series of conditional statements and a piecewise approach for solving exponential slope directly through measured values in most instances. The proposed algorithm is tested against a purpose‐designed iterative fitting technique and linear piecewise method followed by single exponential in a cohort of ten patients receiving 177Lu‐DOTA‐Octreotate therapy. Results: Tri‐exponential time‐integrated values are shown to be comparable to previously published methods with an average difference between organs when computed at the voxel level of 9.8 ± 14.2% and −3.6 ± 10.4% compared to iterative and interpolated methods, respectively. Of the three methods, the proposed tri‐exponential algorithm was most consistent when regional time‐integrated activity was evaluated at both voxel‐ and whole‐organ levels. For whole‐body SPECT imaging, it is possible to compute 3D time‐integrated activity maps in <5 min processing time. Furthermore, the technique is able to predictably and reproducibly handle artefactual measurements due to noise or spatial misalignment over multiple image times. Conclusions: An efficient, analytical algorithm for solving multiphase exponential pharmacokinetics is reported. The method may be readily incorporated into voxel‐dose routines by combining with widely available image registration and radiation transport tools. [ABSTRACT FROM AUTHOR]

Purpose: To accomplish novel radiation treatment techniques in preclinical radiation research, small animal image‐guided radiotherapy systems have been increasingly integrated into preclinical radiation research over the last decade. Although such systems have sophisticated tools (such as cone‐beam computed tomography–based image guidance, robotic couch, treatment planning system (TPS), and electronic portal imaging devices [EPIDs]). To our knowledge, no established technique can perform independent and online verification of the delivered dose during radiotherapy. In this study, we implement an online EPID dosimetry for each administered SA‐IGRT fraction in a rat orthotopic model of prostate cancer. Methods: To verify the accuracy of delivered dose to rat, we compared the two‐dimensional (2D) calculated dose distribution by TPS as the planned dose, with online dose distribution estimated using an EPID as the delivered dose. Since image acquisition software was not capable of acquiring integrated images over a long period of time, we used the EPID to estimate dose rate rather than dose. The central axis (CAX) dose rate values at the beam's exit surface were compared. In addition, 2D dose distributions were also compared under different gamma criteria. To verify the accuracy of our EPID dosimetry technique, we measured transit and exit doses with film during animal treatment. In this study, 20‐mm cone was used to collimate beam. We previously observed that the EPID response was independent of collimator size for collimator size ≥15‐mm, we did not apply for additional correction factor. Results: Comparison of exit CAX dose rate values of TPS‐calculated and EPID‐estimated showed that the average difference was 3.1%, with a maximum of 9.3%. Results of gamma analysis for 2D comparison indicated an average of 90% passing rate with global gamma criterion of 2 mm/5%. We observed that TPS could not calculate dose accurately in peripheral regions in which the penumbra effect was dominant. Dose rate values estimated by EPID were within 2.1% agreement with film at both the imager plane and the beam's exit surface for 4 randomly selected animals for which film measurement verification was performed. Conclusions: The small animal radiation research platform (SARRP) system's built‐in EPID was utilized to estimate dose delivered to rats at kilovoltage energy x‐rays. The results of this study suggest that the EPID is an invaluable tool for verifying delivered dose to small animal to help validate conclusions made from preclinical radiation research. [ABSTRACT FROM AUTHOR]

Purpose: To present a novel method, based on convolutional neural networks (CNN), to automate weighted log subtraction (WLS) for dual‐energy (DE) fluoroscopy to be used in conjunction with markerless tumor tracking (MTT). Methods: A CNN was developed to automate WLS (aWLS) of DE fluoroscopy to enhance soft tissue visibility. Briefly, this algorithm consists of two phases: training a CNN architecture to predict pixel‐wise weighting factors followed by application of WLS subtraction to reduce anatomical noise. To train the CNN, a custom phantom was built consisting of aluminum (Al) and acrylic (PMMA) step wedges. Per‐pixel ground truth (GT) weighting factors were calculated by minimizing the contrast of Al in the step wedge phantom to train the CNN. The pretrained model was then utilized to predict pixel‐wise weighting factors for use in WLS. For comparison, the weighting factor was manually determined in each projection (mWLS). A thorax phantom with five simulated spherical targets (5–25 mm) embedded in a lung cavity, was utilized to assess aWLS performance. The phantom was imaged with fast‐kV dual‐energy (120 and 60 kVp) fluoroscopy using the on‐board imager of a commercial linear accelerator. DE images were processed offline to produce soft tissue images using both WLS methods. MTT was compared using soft tissue images produced with both mWLS and aWLS techniques. Results: Qualitative evaluation demonstrated that both methods achieved soft tissue images with similar quality. The use of aWLS increased the number of tracked frames by 1–5% compared to mWLS, with the largest increase observed for the smallest simulated tumors. The tracking errors for both methods produced agreement to within 0.1 mm. Conclusions: A novel method to perform automated WLS for DE fluoroscopy was developed. Having similar soft tissue quality as well as bone suppression capability as mWLS, this method allows for real‐time processing of DE images for MTT. [ABSTRACT FROM AUTHOR]

Purpose: Filtering measured projections with a particular convolutional kernel is an essential step in analytic reconstruction of computed tomography (CT) images. A tradeoff between noise and spatial resolution exists for different choices of reconstruction kernel. In a clinical setting, this often requires producing multiple images reconstructed with different kernels for a single CT exam, which increases the burden of computation, networking, archival, and reading. We address this problem by training a deep convolutional neural network (CNN) to synthesize multiple input images into a single output image which exhibits low noise while also preserving features in images reconstructed with the sharpest kernels. Methods: A CNN architecture consisting of repeated blocks of residual units containing a total of 20 convolutional layers was used to combine features. The CNN inputs consisted of two images produced with different reconstruction kernels, one smooth and one sharp, which were stacked in the channel dimension. The network was trained using supervised learning with both full‐dose and simulated quarter‐dose abdominal CT images. After training, the performance was evaluated using a reserved set of full‐dose scans that were not used for network optimization. Noise reduction performance was measured by comparing root mean square (RMS) measurements in uniform regions. Spatial resolution was compared using line profiles of anatomic features. Results: For the regions tested, the synthetic images feature noise levels slightly below those of the smooth input images, while maintaining the resolution of anatomic details found in the sharp input images. Conclusions: A deep CNN can be used combine features from CT images reconstructed with different kernels to produce a single synthesized image series that exhibits both low noise and high spatial resolution. This approach has implications for improving image quality, reducing radiation dose, and simplifying the clinical workflow for CT imaging. [ABSTRACT FROM AUTHOR]

Purpose: Measurement of the modulation transfer function (MTF) is performed by evaluating the response of an imaging system to a predefined input. To obtain accurate results when using an edge phantom, the detector input signal must resemble an ideal step function. The MTF of megavoltage (MV) imagers used in radiotherapy has been measured with highly absorbing edge phantoms fabricated from thick metal blocks. This study investigates the influence of the edge phantom design on the accuracy of the resulting MTF. Methods: The MTF of an electronic portal imaging device (EPID) was measured at 6 MV beam quality with four edge phantoms made of lead with 1.3, 3.3, 5.0, and 10.0 cm thickness. Monte Carlo simulations were carried out for these and a selection of tungsten phantoms to determine the photon fluence at the imaging plane and quantify the systematic error in the MTF introduced by the edge phantom design. Results: The measured MTF depends on the design of the edge phantom. The detector input signal of a thin phantom is affected by secondary radiation from the phantom itself, causing an overestimation of the MTF. The amount of secondary radiation can be reduced by increasing the phantom thickness or introducing an air gap between the phantom and the detector. Both methods introduce geometric unsharpness, which can result in an underestimation of the true MTF. Edge phantoms made from 4.0 cm thick tungsten or 5.0 cm thick lead induce comparatively small systematic errors of below 3% or 5%, respectively. Conclusions: When MTF measurements are conducted at MV energies, even a highly absorbing edge phantom will introduce a systematic error of several percent. Direct comparison of MTFs obtained with different edge phantoms should therefore be treated with caution. [ABSTRACT FROM AUTHOR]

Purpose: Early detection of carotid atherosclerosis on the vessel wall (VW) magnetic resonance imaging (MRI) (VW‐MRI) images can prevent the progression of cardiovascular disease. However, the manual inspection process of the VW‐MRI images is cumbersome and has low reproducibility. Therefore in this paper, by using the convolutional neural networks (CNNs), we develop a deep morphology aided diagnosis (DeepMAD) network for automated segmentation of the VW of carotid artery and for automated diagnosis of the carotid atherosclerosis with the black‐blood (BB) VW‐MRI (i.e., the T1‐weighted MRI) in a slice‐by‐slice manner. Methods: The proposed DeepMAD network consists of a segmentation subnetwork and a diagnosis subnetwork for performing the segmentation and diagnosis tasks on the BB‐VW‐MRI images, where the manual labeled lumen area, the manual labeled outer wall area and the manual labeled lesion Types based on the modified American Heart Association (AHA) criteria are used as the ground‐truth. Specifically, a deep U‐shape CNN with a weighted fusion layer is designed as the segmentation subnetwork, where the lumen area and the outer wall area can be simultaneously segmented under the supervision of the triple Dice loss to provide the vessel wall map as morphological information. Then, the image stream from the BB‐VWMRI image and the morphology stream from the obtained vessel wall map are extracted from two deep CNNs and combined to obtain the diagnosis results of atherosclerosis in the diagnosis subnetwork. In addition, the triple input set is formed by three carotid regions of interest (ROIs) from three consecutive slices of the MRI sequence and input to the DeepMAD network, where the first and last slices used as additional adjacent slices to provide 2.5D spatial information along the carotid artery centerline for the intermediate slice, which is the target slice for segmentation and diagnosis in the study. Results: Compared to other existing methods, the DeepMAD network can achieve promising segmentation performances (0.9594 Dice for the lumen and 0.9657 Dice for the outer wall) and better diagnosis Accuracy of the carotid atherosclerosis (0.9503 AUC and 0.8916 Accuracy) in the test dataset (including invisible subjects) from same source as the training dataset. In addition, the trained DeepMAD model can be successfully transferred to another test dataset for segmentation and diagnosis tasks with remarkable performance (0.9475 Dice for the lumen and 0.9542 Dice for the outer wall, 0. 9227 AUC and 0.8679 Accuracy for diagnosis). Conclusions: Even without the intervention of reviewers required for previous works, the proposed DeepMAD network automatically segments the lumen and the outer wall together and diagnoses the carotid atherosclerosis with high performances. The DeepMAD network can be used in clinical trials to help radiologists get rid of tedious reading tasks, such as screening review to separate the normal carotid from the atherosclerotic arteries and outlining the vessel wall contours. [ABSTRACT FROM AUTHOR]

Purpose: Cerenkov photons are generated by high‐energy radiation used in external beam radiation therapy (EBRT). This study expands upon the Cerenkov light dosimetry formula previously developed to relate an image of Cerenkov photons to the primary beam fluence. Extension of this formulation allows for deconvolution to be performed on images acquired from curved geometries. Methods: The integral equation, which represented the formation of Cerenkov photon image from an incident high‐energy photon beam, was expanded to allow for space‐variance of the convolution kernel called as the Cerenkov scatter function (CSF). The GAMOS (Geant4‐based Architecture for Medicine‐Oriented Simulations) Monte Carlo (MC) particle simulation software was used to obtain the CSF for different incident beam angles. The image of a curved surface was first projected to a flat plane by using a perspective correction method. Then, the planar image was partitioned into small segments (or blocks), where a CSF corresponding to a specific beam incident angle was applied for deconvolution. The block size and the margin around the block were optimized by studying the effects of those parameters on the deconvolution accuracy for a test image. We evaluated three deconvolution techniques: Richardson–Lucy, Blind, and Total Variation minimization (TV/L2) algorithms, to select the most accurate method for the current applications. Results: Analysis of deconvolution algorithms showed that the TV/L2 method provided the most accurate solution to the deconvolution problem for Cerenkov imaging. Optimization of space‐variant deconvolution parameters showed that including a margin that is at least 42.9% of the image width provided the most accurate product image. There was no optimal size for the deconvolution area and should be chosen based on the presence of unique CSF kernels within an image. Space‐variant deconvolution improved measured field size in Cerenkov photon images by 7.37%, as compared with 1.74% by space‐invariant deconvolution. Space‐variant deconvolution improved measured penumbra by 99.3%, as compared with 76.7% by space‐invariant deconvolution. Space‐variant deconvolution introduced artifacts in flat regions of the beam. Artifacts were avoided through selective space‐variant deconvolution in only the penumbra region. Conclusions: Primary photon fluence distributions of a curved surface can be obtained by using space‐variant deconvolution methods in Cerenkov light dosimetry. The TV/L2 algorithm is the best method for deconvolution of Cerenkov photon images from an open‐field beam derived from either a flat or curved surface. The partition size chosen for space‐variant deconvolution should be at least six times the full width at half maximum (FWHM) of the corresponding scatter kernel used in deconvolution. Space‐variant deconvolution is necessary if the incident beam angle difference is larger than 6∘ between regions of an image. [ABSTRACT FROM AUTHOR]

Purpose: At our institute, in vivo patient dose distributions are reconstructed for all treatments delivered using conventional linacs from electronic portal imaging device (EPID) transit images acquired during treatment using a simple back‐projection model. Currently, the clinical implementation of MRI‐guided radiotherapy systems, which aims for online and real‐time adaptation of the treatment plan, is progressing. In our department, the MR‐linac (Unity, Elekta AB, Stockholm, Sweden) is now in clinical use. The aim of this work is to demonstrate the feasibility of two‐dimensional (2D) EPID dosimetric verification for the magnetic resonance (MR)‐linac by comparing back‐projected EPID doses to ionization chamber (IC) array dose distributions. Materials and methods: Our conventional back‐projection algorithm was adapted for the MR‐linac. The most important changes involve modeling of the attenuation by and scatter from the cryostat. The commissioning process involved the acquisition of square field EPID measurements using various phantom setups (varying SSD, phantom thickness, and field size). Commissioning models were created for gantry 0°, 90°, and 180° and verified by comparing EPID‐reconstructed 2D dose distributions to measurements made with the OCTAVIUS 1500 IC array (PTW, Freiburg, Germany) for two prostate and one rectum IMRT plans (25 beams total). The average of the γ parameters (y‐mean and y‐pass rate) and the dose difference at a reference point were reported. Due to their construction, the attenuation of couch, bridge, and cryostat shows a much stronger dependence on gantry angle in the MR‐linac compared to conventional linacs. We present a method to correct for these effects. This method is validated by dose reconstruction of the 25 intensity‐modulated radiation therapy beams recorded at a certain gantry angle using the model of another gantry angle, combined with the correction method. Results: For dose verification performed at a gantry angle identical to the commissioned model, the average y‐mean and y‐pass rate values (3% global dose, 2 mm, 10% isodose) were 0.37 ± 0.07 and 98.1, 95% CI [98.1 ± 2.4], respectively. The average dose difference at the reference point was −0.5% ± 1.8%. Verification at gantry angles different from the commissioned model (i.e., using the gantry angle dependent correction) reported 0.39 ± 0.08 and 97.6, 95% CI [96.9, 98.3] average y‐mean and y‐pass rate values. The average dose difference at the reference point was −0.1% ± 1.8%. Conclusion: The EPID dosimetry back‐projection model was successfully adapted for the MR‐linac at gantry 0°, 90°, and 180°, accounting for the presence of the MRI housing between phantom (or patient) and the EPID. A method to account for the gantry angle dependence was also tested reporting similar results. [ABSTRACT FROM AUTHOR]

Purpose: The mechanical and geometric accuracy of small animal image‐guided radiotherapy (SA‐IGRT) systems is critical and is affected by a number of system‐related factors. Because of the small dimensions involved in preclinical radiotherapy research, such factors can individually and/or cumulatively contribute to significant errors in the small animal radiation research. In this study, we developed and implemented a comprehensive quality assurance (QA) framework for characterizing the mechanical and geometric constancy and accuracy of the small animal radiation research platform (SARRP) system. Methods: We quantified the accuracy of gantry and stage rotation isocentricity and positional stage translations. We determined the accuracy and symmetry of field sizes formed by collimators. We evaluated collimator assembly system performance by characterization of collimator axis alignment along the beam axis during gantry rotation. Furthermore, we quantified the end‐to‐end precision and accuracy of image‐guided delivery by examining the congruence of intended (e.g., imaging) and actual delivery (measured during experiment) isocenters. Results: The fine and broad beams showed different central axes. The center of the beam was offset toward the cathode (0.22 ± 0.05 mm) when switching the beam from a fine to a broad focus. Larger (custom‐made) collimators were more symmetrically centered than smaller (standard) collimators. The field formed by a 1‐mm circular collimator was found to deviate from the circular shape, measuring 1.55 mm and 1.25 mm in the X and Y directions, respectively. The 40‐mm collimator showed a field that was 1.65 (4.13%) and 1.3 (3.25%) mm smaller than nominal values in the X and Y directions, respectively, and the 30‐mm collimator field was smaller by 0.75 mm (2.5%) in the X direction. Results showed that fields formed by other collimators were accurate in both directions and had ≤2% error. The size of the gantry rotation isocenter was 1.45 ± 0.15 mm. While the gantry rotated, lateral and longitudinal isocenter displacements ranged from 0 to −0.34 and −0.44 to 0.33 mm, respectively. Maximum lateral and longitudinal displacements were found at obliques gantry angles of −135° and 45°, respectively. The stage translational accuracies were 0.015, 0.010, and 0 mm in the X, Y, and Z directions, respectively. The size of the stage rotation runout was 2.73 ± 0.3 mm. Maximum displacements of the stage rotational axis were −0.38 (X direction) and −0.26 (Y direction) mm at stage angles of −45° and −135°, respectively. We found that displacements of intended and actual delivery isocenters were 0.24 ± 0.10, 0.12 ± 0.62, and 0.12 ± 0.42 mm in the X, Y, and Z directions, respectively. Conclusion: We used the SARRP built‐in electronic portal imaging device (EPID) to perform most of the geometric QA tests, demonstrating the utility of the EPID for characterizing the geometric accuracy and precision of the SA‐IGRT system. However, in principle, the methodology and tests developed here are applicable to any digital imaging detector available in SA‐IGRT systems or film. The flexibility of film allows these tests to be adapted for QA of non‐IGRT, cabinet irradiators, which make up many of preclinical small animal irradiators. [ABSTRACT FROM AUTHOR]

Purpose: To study the feasibility of a channelized Hotelling observer (CHO) to predict human observer performance in detecting calcification‐like signals in mammography images of an anthropomorphic breast phantom, as part of a quality control (QC) framework. Methods: A prototype anthropomorphic breast phantom with inserted gold disks of 0.25 mm diameter was imaged with two different digital mammography x‐ray systems at four different dose levels. Regions of interest (ROIs) were extracted from the acquired processed and unprocessed images, signal‐present and signal‐absent. The ROIs were evaluated by a CHO using four different formulations of the difference of Gaussian (DoG) channel sets. Three human observers scored the ROIs in a two‐alternative forced‐choice experiment. We compared the human and the CHO performance on the simple task to detect calcification‐like disks in ROIs with and without postprocessing. The proportion of correct responses of the human reader (PCH) and the CHO (PCCHO) was calculated and the correlation between the two was analyzed using a mixed‐effect regression model. To address the signal location uncertainty, the impact of shifting the DoG channel sets in all directions up to two pixels was evaluated. Correlation results including the goodness of fit (r2) of PCH and PCCHO for all different parameters were evaluated. Results: Subanalysis by system yielded strong correlations between PCH and PCCHO, with r2 between PCH and PCCHO was found to be between 0.926 and 0.958 for the unshifted and between 0.759 and 0.938 for the shifted channel sets, respectively. However, the linear fit suggested a slight system dependence. PCCHO with shifted channel sets increased CHO performance but the correlation with humans was decreased. These correlations were not considerably affected by of the DoG channel set used. Conclusions: There is potential for the CHO to be used in QC for the evaluation of detectability of calcification‐like signals. The CHO can predict the PC of humans in images of calcification‐like signals of two different systems. However, a global model to be used for all systems requires further investigation. [ABSTRACT FROM AUTHOR]

Purpose: We evaluate the lesion detectability using human and model observer studies in single‐slice and multislice cone beam computed tomography (CBCT) images with a breast anatomical background. The purposes of this work are (a) to compare human observer detectability between single‐slice and multislice images for different signal sizes and noise structures, (b) to investigate the effect of different multislice viewing modes (i.e., sequential and simultaneous) on the detectability by a human observer, and (c) to predict the detectability by a human observer in single‐slice and multislice images using single‐slice channelized Hotelling observer (ssCHO) and multislice CHO (msCHO), respectively. Methods: Breast anatomical background is modeled using a power law spectrum of mammograms and the lesion is modeled with a spherical signal. We conduct signal‐known‐exactly and background‐known‐statistically detection tasks on transverse and longitudinal images reconstructed using the Feldkamp–Davis–Kress algorithm with Hanning and Ram‐Lak weighted ramp filters. The human observer study is conducted on three different viewing modes: single‐slice, and sequential and simultaneous multislice. To predict the detectability by a human observer, we use ssCHO and msCHO with anthropomorphic channels (i.e., dense difference‐of‐Gaussian (D‐DOG) and Gabor channels) and internal noise. Results: The detectability by a human observer increases for multislice images compared to single‐slice images. For multislice images, the sequential viewing mode yields higher detectability than the simultaneous viewing mode. However, the relative rank of detectability by a human observer for different signal sizes, image planes, and reconstruction filters is not much different between the viewing modes. Detectability by CHO with internal noise shows good correlation with that of the human observer for all viewing modes. Conclusions: Detectability by a human observer in CBCT images with breast anatomical background is affected by the image viewing mode, and the effect of the viewing mode depends on the signal size and noise structure. D‐DOG and Gabor CHO with internal noise predict the detectability by a human observer well for both the single‐slice and multislice image viewing modes. [ABSTRACT FROM AUTHOR]

Purpose: Intraductal papillary mucinous neoplasms (IPMNs) are radiographically visible precursor lesions of pancreatic cancer. Despite standard criteria for assessing risk, only 18% of cysts are malignant at resection. Thus, a large number of patients undergo unnecessary invasive surgery for benign disease. The ability to identify IPMNs with low or high risk of transforming into invasive cancer would optimize patient selection and improve surgical decision‐making. The purpose of this study was to investigate quantitative CT imaging features as markers for objective assessment of IPMN risk. Methods: This retrospective study analyzed pancreatic cyst and parenchyma regions extracted from CT scans in 103 patients to predict IPMN risk. Patients who underwent resection between 2005 and 2015 with pathologically proven branch duct (BD)‐IPMN and a preoperative CT scan were included in the study. Expert pathologists categorized IPMNs as low or high risk following resection as part of routine clinical care. We extracted new radiographically inspired features as well as standard texture features and designed prediction models for the categorization of high‐ and low‐risk IPMNs. Five clinical variables were also combined with imaging features to design prediction models. Results: Using images from 103 patients and tenfold cross‐validation technique, the novel radiographically inspired imaging features achieved an area under the receiver operating characteristic curve (AUC) of 0.77, demonstrating their predictive power. The combination of these features with clinical variables obtained the best performance (AUC = 0.81). Conclusion: The present study demonstrates that features extracted from pretreatment CT images can predict the risk of IPMN. Development of a preoperative model to discriminate between low‐risk and high‐risk IPMN will improve surgical decision‐making. [ABSTRACT FROM AUTHOR]

Purpose: Four‐dimensional (4D) cone‐beam computed tomography (CBCT) of the lung is an effective tool for motion management in radiotherapy but presents a challenge because of slow gantry rotation times. Sorting the individual projections by breathing phase and using an established technique such as Feldkamp–Davis–Kress (FDK) to generate corresponding phase‐correlated (PC) three‐dimensional (3D) images results in reconstructions (FDK‐PC) that often contain severe streaking artifacts due to the sparse angular sampling distributions. These can be reduced by further slowing down the gantry at the expense of incurring unwanted increases in scan times and dose. A computationally efficient alternative is the McKinnon‐Bates (MKB) reconstruction algorithm that has shown promise in reducing view aliasing‐induced streaking but can produce ghosting artifacts that reduce contrast and impede the determination of motion trajectories. The purpose of this work was to identify and correct shortcomings in the MKB algorithm. Methods: In the general MKB approach, a time‐averaged 3D prior image is first reconstructed. The prior is then forward‐projected at the same angles as the original projection data creating time‐averaged reprojections. These reprojections are subsequently subtracted from the original (unblurred) projections to create motion‐encoded difference projections. The difference projections are reconstructed into PC difference images that are added to the well‐sampled 3D prior to create the higher quality 4D image. The cause of the ghosting in the traditional 4D MKB images was studied and traced to motion‐induced streaking in the prior that, when reprojected, has the undesirable effect of re‐encoding for motion in what should be a purely time‐averaged reprojection. A new method, designated as the modified McKinnon‐Bates (mMKB) algorithm, was developed based on destreaking the prior. This was coupled with a postprocessing 4D bilateral filter for noise suppression and edge preservation (mMKBbf). The algorithms were tested with the 4D XCAT phantom using four simulated scan times (57, 60, 120, 180 s) and with two in vivo thorax studies (acquisition time of 60 and 90 s). Contrast‐to‐noise ratios (CNRs) of the target lesions and overall visual quality of the images were assessed. Results: Prior destreaking (mMKB algorithm) reduced ghosting artifacts and increased CNRs for all cases, with the biggest impacts seen in the end inhale (EI) and end exhale (EE) phases of the respiratory cycle. For the XCAT phantom, mMKB lesion CNR was 44% higher than the MKB lesion CNR and was 81% higher than the FDK‐PC lesion CNR (EI and EE phases). The bilateral filter provided a further average CNR improvement of 87% with the highest increases associated with longer scan times. Across all phases and scan times, the maximum mMKBbf‐to‐FDK‐PC CNR improvement was over 300%. In vivo results agreed with XCAT results. Significantly less ghosting was observed throughout the mMKB images including near the lesions‐of‐interest and the diaphragm allowing for, in one case, visualization of a small tumor with nearly 30 mm of motion. The maximum FDK‐PC‐to‐MKBbf CNR improvement for Patient 1's lesion was 261% and for Patient 2's lesion was 318%. Conclusions: The 4D mMKB algorithm yields good quality coronal and sagittal images in the thorax that may provide sufficient information for patient verification. [ABSTRACT FROM AUTHOR]

Purpose: The automatic detection of pulmonary nodules using CT scans improves the efficiency of lung cancer diagnosis, and false‐positive reduction plays a significant role in the detection. In this paper, we focus on the false‐positive reduction task and propose an effective method for this task. Methods: We construct a deep 3D residual CNN (convolution neural network) to reduce false‐positive nodules from candidate nodules. The proposed network is much deeper than the traditional 3D CNNs used in medical image processing. Specifically, in the network, we design a spatial pooling and cropping (SPC) layer to extract multilevel contextual information of CT data. Moreover, we employ an online hard sample selection strategy in the training process to make the network better fit hard samples (e.g., nodules with irregular shapes). Results: Our method is evaluated on 888 CT scans from the dataset of the LUNA16 Challenge. The free‐response receiver operating characteristic (FROC) curve shows that the proposed method achieves a high detection performance. Conclusions: Our experiments confirm that our method is robust and that the SPC layer helps increase the prediction accuracy. Additionally, the proposed method can easily be extended to other 3D object detection tasks in medical image processing. [ABSTRACT FROM AUTHOR]

Purpose Microvascular processes play key roles in many diseases including diabetes. Improved understanding of the microvascular changes involved in disease development could offer crucial insight into the relationship of these changes to disease pathogenesis. Super-resolution ultrasound ( SR- US) imaging has showed the potential to visualize microvascular detail down to the capillary level (i.e., subwavelength resolution), but optimization is still necessary. The purpose of this study was to investigate in vivo SR- US imaging of skeletal muscle microvascularity using microbubble ( MB) contrast agents of various size and concentration while evaluating different ultrasound ( US) system level parameters such as imaging frame rate and image acquisition length. Methods An US system equipped with a linear array transducer was used in a harmonic imaging mode at low transmit power. C57 BL/6J mice fed a normal diet were used in this study. An assortment of size-selected MB contrast agents (1-2 μm, 3-4 μm, and 5-8 μm in diameter) were slowly infused in the tail vein at various doses (1.25 × 107, 2.5 × 107, or 5 × 107 MBs). US image data were collected before MB injection and thereafter for 10 min at 30 frames per s (fps). The US transducer was fixed throughout and between each imaging period to help capture microvascular patterns along the same image plane. An adaptive SR- US image processing technique was implemented using custom Matlab software. Results Experimental findings illustrate the use of larger MB results in better SR- US images in terms of skeletal muscle microvascular detail. A dose of 2.5 × 107 MBs resulted in SR- US images with optimal spatial resolution. An US imaging rate of at least 20 fps and image acquisition length of at least 8 min also resulted in SR- US images with pronounced microvascular detail. Conclusions This study indicates that MB size and dose and US system imaging rate and data acquisition length have significant impact on the quality of in vivo SR- US images of skeletal muscle microvascularity. [ABSTRACT FROM AUTHOR]

Purpose Precise delineation of organs at risk is a crucial task in radiotherapy treatment planning for delivering high doses to the tumor while sparing healthy tissues. In recent years, automated segmentation methods have shown an increasingly high performance for the delineation of various anatomical structures. However, this task remains challenging for organs like the esophagus, which have a versatile shape and poor contrast to neighboring tissues. For human experts, segmenting the esophagus from CT images is a time-consuming and error-prone process. To tackle these issues, we propose a random walker approach driven by a 3D fully convolutional neural network (CNN) to automatically segment the esophagus from CT images. Methods First, a soft probability map is generated by the CNN. Then, an active contour model (ACM) is fitted to the CNN soft probability map to get a first estimation of the esophagus location. The outputs of the CNN and ACM are then used in conjunction with a probability model based on CT Hounsfield (HU) values to drive the random walker. Training and evaluation were done on 50 CTs from two different datasets, with clinically used peer-reviewed esophagus contours. Results were assessed regarding spatial overlap and shape similarity. Results The esophagus contours generated by the proposed algorithm showed a mean Dice coefficient of 0.76 ± 0.11, an average symmetric square distance of 1.36 ± 0.90 mm, and an average Hausdorff distance of 11.68 ± 6.80, compared to the reference contours. These results translate to a very good agreement with reference contours and an increase in accuracy compared to existing methods. Furthermore, when considering the results reported in the literature for the publicly available Synapse dataset, our method outperformed all existing approaches, which suggests that the proposed method represents the current state-of-the-art for automatic esophagus segmentation. Conclusion We show that a CNN can yield accurate estimations of esophagus location, and that the results of this model can be refined by a random walk step taking pixel intensities and neighborhood relationships into account. One of the main advantages of our network over previous methods is that it performs 3D convolutions, thus fully exploiting the 3D spatial context and performing an efficient volume-wise prediction. The whole segmentation process is fully automatic and yields esophagus delineations in very good agreement with the gold standard, showing that it can compete with previously published methods. [ABSTRACT FROM AUTHOR]

Purpose: We have demonstrated the ability to identify coronary calcium, a reliable biomarker of coronary artery disease, using nongated, 2-shot, dual energy (DE) chest x-ray imaging. Here we will use digital simulations, backed up by measurements, to characterize DE calcium signals and the role of potential confounds such as beam hardening, x-ray scatter, cardiac motion, and pulmonary artery pulsation. For the DE calcium signal, we will consider quantification, as compared to CT calcium score, and visualization. Methods: We created stylized and anatomical digital 3D phantoms including heart, lung, coronary calcium, spine, ribs, pulmonary artery, and adipose. We simulated high and low kVp x-ray acquisitions with x-ray spectra, energy dependent attenuation, scatter, ideal detector, and automatic exposure control (AEC). Phantoms allowed us to vary adipose thickness, cardiac motion, etc. We used specialized dual energy coronary calcium (DECC) processing that includes corrections for scatter and beam hardening. Results: Beam hardening over a wide range of adipose thickness (0–30 cm) reduced the change in intensity of a coronary artery calcification (ΔICAC) by < 3% in DECC images. Scatter correction errors of ±50% affected the calcium signal (ΔICAC) in DECC images ±9%. If a simulated pulmonary artery fills with blood between exposures, it can give rise to a residual signal in DECC images, explaining pulmonary artery visibility in some clinical images. Residual misregistration can be mostly compensated by integrating signals in an enlarged region encompassing registration artifacts. DECC calcium score compared favorably to CT mass and volume scores over a number of phantom perturbations. Conclusion: Simulations indicate that proper DECC processing can faithfully recover coronary calcium signals. Beam hardening, errors in scatter estimation, cardiac motion, calcium residual misregistration etc., are all manageable. Simulations are valuable as we continue to optimize DE coronary calcium image processing and quantitative analysis. [ABSTRACT FROM AUTHOR]