Search

Background Purpose Methods Results Conclusions Pneumonia, a major infectious cause of morbidity and mortality among children worldwide, is typically diagnosed using low‐dose pediatric chest X‐ray [CXR (chest radiography)]. In pediatric CXR images, bone occlusion leads to a risk of missed diagnosis. Deep learning–based bone‐suppression networks relying on training data have enabled considerable progress to be achieved in bone suppression in adult CXR images; however, these networks have poor generalizability to pediatric CXR images because of the lack of labeled pediatric CXR images (i.e., bone images vs. soft‐tissue images). Dual‐energy subtraction imaging approaches are capable of producing labeled adult CXR images; however, their application is limited because they require specialized equipment, and they are infrequently employed in pediatric settings. Traditional image processing–based models can be used to label pediatric CXR images, but they are semiautomatic and have suboptimal performance.We developed an efficient labeling approach for fine‐tuning pediatric CXR bone‐suppression networks capable of automatically suppressing bone structures in CXR images for pediatric patients without the need for specialized equipment and technologist training.Three steps were employed to label pediatric CXR images and fine‐tune pediatric bone‐suppression networks: distance transform–based bone‐edge detection, traditional image processing–based bone suppression, and fully automated pediatric bone suppression. In distance transform–based bone‐edge detection, bone edges were automatically detected by predicting bone‐edge distance‐transform images, which were then used as inputs in traditional image processing. In this processing, pediatric CXR images were labeled by obtaining bone images through a series of traditional image processing techniques. Finally, the pediatric bone‐suppression network was fine‐tuned using the labeled pediatric CXR images. This network was initially pretrained on a public adult dataset comprising 240 adult CXR images (A240) and then fine‐tuned and validated on 40 pediatric CXR images (P260_40labeled) from our customized dataset (named P260) through five‐fold cross‐validation; finally, the network was tested on 220 pediatric CXR images (P260_220unlabeled dataset).The distance transform–based bone‐edge detection network achieved a mean boundary distance of 1.029. Moreover, the traditional image processing–based bone‐suppression model obtained bone images exhibiting a relative Weber contrast of 93.0%. Finally, the fully automated pediatric bone‐suppression network achieved a relative mean absolute error of 3.38%, a peak signal‐to‐noise ratio of 35.5 dB, a structural similarity index measure of 98.1%, and a bone‐suppression ratio of 90.1% on P260_40labeled.The proposed fully automated pediatric bone‐suppression network, together with the proposed distance transform–based bone‐edge detection network, can automatically acquire bone and soft‐tissue images solely from CXR images for pediatric patients and has the potential to help diagnose pneumonia in children. [ABSTRACT FROM AUTHOR]

Purpose Acquisition and validation methods Data format and usage notes Potential applications Evaluating deformable image registration (DIR) algorithms is vital for enhancing algorithm performance and gaining clinical acceptance. However, there is a notable lack of dependable DIR benchmark datasets for assessing DIR performance except for lung images. To address this gap, we aim to introduce our comprehensive liver computed tomography (CT) DIR landmark dataset library. This dataset is designed for efficient and quantitative evaluation of various DIR methods for liver CTs, paving the way for more accurate and reliable image registration techniques.Forty CT liver image pairs were acquired from several publicly available image archives and authors’ institutions under institutional review board (IRB) approval. The images were processed with a semi‐automatic procedure to generate landmark pairs: (1) for each case, liver vessels were automatically segmented on one image; (2) landmarks were automatically detected at vessel bifurcations; (3) corresponding landmarks in the second image were placed using two deformable image registration methods to avoid algorithm‐specific biases; (4) a comprehensive validation process based on quantitative evaluation and manual assessment was applied to reject outliers and ensure the landmarks’ positional accuracy. This workflow resulted in an average of ∼56 landmark pairs per image pair, comprising a total of 2220 landmarks for 40 cases. The general landmarking accuracy of this procedure was evaluated using digital phantoms and manual landmark placement. The landmark pair target registration errors (TRE) on digital phantoms were 0.37 ± 0.26 and 0.55 ± 0.34 mm respectively for the two selected DIR algorithms used in our workflow, with 97% of landmark pairs having TREs below 1.5 mm. The distances from the calculated landmarks to the averaged manual placement were 1.27 ± 0.79 mm.All data, including image files and landmark information, are publicly available at Zenodo (https://zenodo.org/records/13738577). Instructions for using our data can be found on our GitHub page at https://github.com/deshanyang/Liver‐DIR‐QA.The landmark dataset generated in this work is the first collection of large‐scale liver CT DIR landmarks prepared on real patient images. This dataset can provide researchers with a dense set of ground truth benchmarks for the quantitative evaluation of DIR algorithms within the liver. [ABSTRACT FROM AUTHOR]

Background: Focusing on the complicated pathological features, such as blurred boundaries, severe scale differences between symptoms, and background noise interference, we aim to enhance the reliability of multiple lesions joint segmentation from medical images. Purpose: Propose a novel reliable multi‐scale wavelet‐enhanced transformer network, which can provide accurate segmentation results with reliability assessment. Methods: Focusing on enhancing the model's capability to capture intricate pathological features in medical images, this work introduces a novel segmentation backbone. The backbone integrates a wavelet‐enhanced feature extractor network and incorporates a multi‐scale transformer module developed within the scope of this work. Simultaneously, to enhance the reliability of segmentation outcomes, a novel uncertainty segmentation head is proposed. This segmentation head is rooted in the SL, contributing to the generation of final segmentation results along with an associated overall uncertainty evaluation score map. Results: Comprehensive experiments are conducted on the public database of AI‐Challenge 2018 for retinal edema lesions segmentation and the segmentation of Thoracic Organs at Risk in CT images. The experimental results highlight the superior segmentation accuracy and heightened reliability achieved by the proposed method in comparison to other state‐of‐the‐art segmentation approaches. Conclusions: Unlike previous segmentation methods, the proposed approach can produce reliable segmentation results with an estimated uncertainty and higher accuracy, enhancing the overall reliability of the model. The code will be release on https://github.com/LooKing9218/ReMultiSeg. [ABSTRACT FROM AUTHOR]

BACKGROUND: This study reports the results of a set of discrimination experiments using simulated images that represent the appearance of subtle lesions in low-dose computed tomography (CT) of the lungs. Noise in these images has a characteristic ramp-spectrum before apodization by noise control filters. We consider three specific diagnostic features that determine whether a lesion is considered malignant or benign, two system-resolution levels, and four apodization levels for a total of 24 experimental conditions. PURPOSE: The goal of the investigation is to better understand how well human observers perform subtle discrimination tasks like these, and the mechanisms of that performance. We use a forced-choice psychophysical paradigm to estimate observer efficiency and classification images. These measures quantify how effectively subjects can read the images, and how they use images to perform discrimination tasks across the different imaging conditions. MATERIALS AND METHODS: The simulated CT images used as stimuli in the psychophysical experiments are generated from high-resolution objects passed through a modulation transfer function (MTF) before down-sampling to the image-pixel grid. Acquisition noise is then added with a ramp noise-power spectrum (NPS), with subsequent smoothing through apodization filters. The features considered are lesion size, indistinct lesion boundary, and a nonuniform lesion interior. System resolution is implemented by an MTF with resolution (10% max.) of 0.47 or 0.58 cyc/mm. Apodization is implemented by a Shepp-Logan filter (Sinc profile) with various cutoffs. Six medically naïve subjects participated in the psychophysical studies, entailing training and testing components for each condition. Training consisted of staircase procedures to find the 80% correct threshold for each subject, and testing involved 2000 psychophysical trials at the threshold value for each subject. Human-observer performance is compared to the Ideal Observer to generate estimates of task efficiency. The significance of imaging factors is assessed using ANOVA. Classification images are used to estimate the linear template weights used by subjects to perform these tasks. Classification-image spectra are used to analyze subject weights in the spatial-frequency domain. RESULTS: Overall, average observer efficiency is relatively low in these experiments (10%-40%) relative to detection and localization studies reported previously. We find significant effects for feature type and apodization level on observer efficiency. Somewhat surprisingly, system resolution is not a significant factor. Efficiency effects of the different features appear to be well explained by the profile of the linear templates in the classification images. Increasingly strong apodization is found to both increase the classification-image weights and to increase the mean-frequency of the classification-image spectra. A secondary analysis of Unapodized classification images shows that this is largely due to observers undoing (inverting) the effects of apodization filters. CONCLUSIONS: These studies demonstrate that human observers can be relatively inefficient at feature-discrimination tasks in ramp-spectrum noise. Observers appear to be adapting to frequency suppression implemented in apodization filters, but there are residual effects that are not explained by spatial weighting patterns. The studies also suggest that the mechanisms for improving performance through the application of noise-control filters may require further investigation.

BackgroundMotion during data acquisition leads to artifacts in computed tomography (CT) reconstructions. In cases such as cardiac imaging, not only is motion unavoidable, but evaluating the motion of the object is of clinical interest. Reducing motion artifacts has typically been achieved by developing systems with faster gantry rotation or via algorithms which measure and/or estimate the displacement. However, these approaches have had limited success due to both physical constraints as well as the challenge of estimating non-rigid, temporally varying, and patient-specific motion fields.PurposeTo develop a novel reconstruction method which generates time-resolved, artifact-free images without estimation or explicit modeling of the motion.MethodsWe describe an analysis-by-synthesis approach which progressively regresses a solution consistent with the acquired sinogram. In our method, we focus on the movement of object boundaries. Not only are the boundaries the source of image artifacts, but object boundaries can simultaneously be used to represent both the object as well as its motion over time without the need for an explicit motion model. We represent the object boundaries via a signed distance function (SDF) which can be efficiently modeled using neural networks. As a result, optimization can be performed under spatial and temporal smoothness constraints without the need for explicit motion estimation.ResultsWe illustrate the utility of DiFiR-CT in three imaging scenarios with increasing motion complexity: translation of a small circle, heart-like change in an ellipse's diameter, and a complex topological deformation. Compared to filtered backprojection, DiFiR-CT provides high quality image reconstruction for all three motions without hyperparameter tuning or change to the architecture. We also evaluate DiFiR-CT's robustness to noise in the acquired sinogram and found its reconstruction to be accurate across a wide range of noise levels. Lastly, we demonstrate how the approach could be used for multi-intensity scenes and illustrate the importance of the initial segmentation providing a realistic initialization. Code and supplemental movies are available at https://kunalmgupta.github.io/projects/DiFiR-CT.html.ConclusionsProjection data can be used to accurately estimate a temporally-evolving scene without the need for explicit motion estimation using a neural implicit representation and analysis-by-synthesis approach.

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]

PurposesPreimplant diagnostic magnetic resonance imaging is the gold standard for image-guided tandem-and-ovoids (T&O) brachytherapy for cervical cancer. However, high dose rate brachytherapy planning is typically done on postimplant CT-based high-risk clinical target volume (HR-CTVCT ) because the transfer of preimplant Magnetic resonance (MR)-based HR-CTV (HR-CTVMR ) to the postimplant planning CT is difficult due to anatomical changes caused by applicator insertion, vaginal packing, and the filling status of the bladder and rectum. This study aims to train a dual-path convolutional neural network (CNN) for automatic segmentation of HR-CTVCT on postimplant planning CT with guidance from preimplant diagnostic MR.MethodsPreimplant T2-weighted MR and postimplant CT images for 65 (48 for training, eight for validation, and nine for testing) patients were retrospectively solicited from our institutional database. MR was aligned to the corresponding CT using rigid registration. HR-CTVCT and HR-CTVMR were manually contoured on CT and MR by an experienced radiation oncologist. All images were then resampled to a spatial resolution of 0.5 × 0.5 × 1.25 mm. A dual-path 3D asymmetric CNN architecture with two encoding paths was built to extract CT and MR image features. The MR was masked by HR-CTVMR contour while the entire CT volume was included. The network put an asymmetric weighting of 18:6 for CT: MR. Voxel-based dice similarity coefficient (DSCV ), sensitivity, precision, and 95% Hausdorff distance (95-HD) were used to evaluate model performance. Cross-validation was performed to assess model stability. The study cohort was divided into a small tumor group (40 cc) based on the HR-CTVCT for model evaluation. Single-path CNN models were trained with the same parameters as those in dual-path models.ResultsFor this patient cohort, the dual-path CNN model improved each of our objective findings, including DSCV , sensitivity, and precision, with an average improvement of 8%, 7%, and 12%, respectively. The 95-HD was improved by an average of 1.65 mm compared to the single-path model with only CT images as input. In addition, the area under the curve for different networks was 0.86 (dual-path with CT and MR) and 0.80 (single-path with CT), respectively. The dual-path CNN model with asymmetric weighting achieved the best performance with DSCV of 0.65 ± 0.03 (0.61-0.70), 0.79 ± 0.02 (0.74-0.85), and 0.75 ± 0.04 (0.68-0.79) for small, medium, and large group. 95-HD were 7.34 (5.35-10.45) mm, 5.48 (3.21-8.43) mm, and 6.21 (5.34-9.32) mm for the three size groups, respectively.ConclusionsAn asymmetric CNN model with two encoding paths from preimplant MR (masked by HR-CTVMR ) and postimplant CT images was successfully developed for automatic segmentation of HR-CTVCT for T&O brachytherapy patients.

PurposeAccurate and robust auto-segmentation of highly deformable organs (HDOs), for example, stomach or bowel, remains an outstanding problem due to these organs' frequent and large anatomical variations. Yet, time-consuming manual segmentation of these organs presents a particular challenge to time-limited modern radiotherapy techniques such as on-line adaptive radiotherapy and high-dose-rate brachytherapy. We propose a machine-assisted interpolation (MAI) that uses prior information in the form of sparse manual delineations to facilitate rapid, accurate segmentation of the stomach from low field magnetic resonance images (MRI) and the bowel from computed tomography (CT) images.MethodsStomach MR images from 116 patients undergoing 0.35T MRI-guided abdominal radiotherapy and bowel CT images from 120 patients undergoing high dose rate pelvic brachytherapy treatment were collected. For each patient volume, the manual delineation of the HDO was extracted from every 8th slice. These manually drawn contours were first interpolated to obtain an initial estimate of the HDO contour. A two-channel 64 × 64 pixel patch-based convolutional neural network (CNN) was trained to localize the position of the organ's boundary on each slice within a five-pixel wide road using the image and interpolated contour estimate. This boundary prediction was then input, in conjunction with the image, to an organ closing CNN which output the final organ segmentation. A Dense-UNet architecture was used for both networks. The MAI algorithm was separately trained for the stomach segmentation and the bowel segmentation. Algorithm performance was compared against linear interpolation (LI) alone and against fully automated segmentation (FAS) using a Dense-UNet trained on the same datasets. The Dice Similarity Coefficient (DSC) and mean surface distance (MSD) metrics were used to compare the predictions from the three methods. Statistically significance was tested using Student's t test.ResultsFor the stomach segmentation, the mean DSC from MAI (0.91 ± 0.02) was 5.0% and 10.0% higher as compared to LI and FAS, respectively. The average MSD from MAI (0.77 ± 0.25 mm) was 0.54 and 3.19 mm lower compared to the two other methods. Only 7% of MAI stomach predictions resulted in a DSC

PurposeThe developments of PET/CT and PET/MR scanners provide opportunities for improving PET image quality by using anatomical information. In this paper, we propose a novel co-learning three-dimensional (3D) convolutional neural network (CNN) to extract modality-specific features from PET/CT image pairs and integrate complementary features into an iterative reconstruction framework to improve PET image reconstruction.MethodsWe used a pretrained deep neural network to represent PET images. The network was trained using low-count PET and CT image pairs as inputs and high-count PET images as labels. This network was then incorporated into a constrained maximum likelihood framework to regularize PET image reconstruction. Two different network structures were investigated for the integration of anatomical information from CT images. One was a multichannel CNN, which treated PET and CT volumes as separate channels of the input. The other one was multibranch CNN, which implemented separate encoders for PET and CT images to extract latent features and fed the combined latent features into a decoder. Using computer-based Monte Carlo simulations and two real patient datasets, the proposed method has been compared with existing methods, including the maximum likelihood expectation maximization (MLEM) reconstruction, a kernel-based reconstruction and a CNN-based deep penalty method with and without anatomical guidance.ResultsReconstructed images showed that the proposed constrained ML reconstruction approach produced higher quality images than the competing methods. The tumors in the lung region have higher contrast in the proposed constrained ML reconstruction than in the CNN-based deep penalty reconstruction. The image quality was further improved by incorporating the anatomical information. Moreover, the liver standard deviation was lower in the proposed approach than all the competing methods at a matched lesion contrast.ConclusionsThe supervised co-learning strategy can improve the performance of constrained maximum likelihood reconstruction. Compared with existing techniques, the proposed method produced a better lesion contrast versus background standard deviation trade-off curve, which can potentially improve lesion detection.

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]

PurposeMissing or discrepant imaging volume is a common challenge in deformable image registration (DIR). To minimize the adverse impact, we train a neural network to synthesize cropped portions of head and neck CT's and then test its use in DIR.MethodsUsing a training dataset of 409 head and neck CT's, we trained a generative adversarial network to take in a cropped 3D image and output an image with synthesized anatomy in the cropped region. The network used a 3D U-Net generator along with Visual Geometry Group (VGG) deep feature losses. To test our technique, for each of the 53 test volumes, we used Elastix to deformably register combinations of a randomly cropped, full, and synthetically full volume to a single cropped, full, and synthetically full target volume. We additionally tested our method's robustness to crop extent by progressively increasing the amount of cropping, synthesizing the missing anatomy using our network, and then performing the same registration combinations. Registration performance was measured using 95% Hausdorff distance across 16 contours.ResultsWe successfully trained a network to synthesize missing anatomy in superiorly and inferiorly cropped images. The network can estimate large regions in an incomplete image, far from the cropping boundary. Registration using our estimated full images was not significantly different from registration using the original full images. The average contour matching error for full image registration was 9.9 mm, whereas our method was 11.6, 12.1, and 13.6 mm for synthesized-to-full, full-to-synthesized, and synthesized-to-synthesized registrations, respectively. In comparison, registration using the cropped images had errors of 31.7 mm and higher. Plotting the registered image contour error as a function of initial preregistered error shows that our method is robust to registration difficulty. Synthesized-to-full registration was statistically independent of cropping extent up to 18.7 cm superiorly cropped. Synthesized-to-synthesized registration was nearly independent, with a -0.04 mm of change in average contour error for every additional millimeter of cropping.ConclusionsDifferent or inadequate in scan extent is a major cause of DIR inaccuracies. We address this challenge by training a neural network to complete cropped 3D images. We show that with image completion, the source of DIR inaccuracy is eliminated, and the method is robust to varying crop extent.

PurposeTo improve image quality and computed tomography (CT) number accuracy of daily cone beam CT (CBCT) through a deep learning methodology with generative adversarial network.MethodsOne hundred and fifty paired pelvic CT and CBCT scans were used for model training and validation. An unsupervised deep learning method, 2.5D pixel-to-pixel generative adversarial network (GAN) model with feature mapping was proposed. A total of 12 000 slice pairs of CT and CBCT were used for model training, while ten-fold cross validation was applied to verify model robustness. Paired CT-CBCT scans from an additional 15 pelvic patients and 10 head-and-neck (HN) patients with CBCT images collected at a different machine were used for independent testing purpose. Besides the proposed method above, other network architectures were also tested as: 2D vs 2.5D; GAN model with vs without feature mapping; GAN model with vs without additional perceptual loss; and previously reported models as U-net and cycleGAN with or without identity loss. Image quality of deep-learning generated synthetic CT (sCT) images was quantitatively compared against the reference CT (rCT) image using mean absolute error (MAE) of Hounsfield units (HU) and peak signal-to-noise ratio (PSNR). The dosimetric calculation accuracy was further evaluated with both photon and proton beams.ResultsThe deep-learning generated sCTs showed improved image quality with reduced artifact distortion and improved soft tissue contrast. The proposed algorithm of 2.5 Pix2pix GAN with feature matching (FM) was shown to be the best model among all tested methods producing the highest PSNR and the lowest MAE to rCT. The dose distribution demonstrated a high accuracy in the scope of photon-based planning, yet more work is needed for proton-based treatment. Once the model was trained, it took 11-12 ms to process one slice, and could generate a 3D volume of dCBCT (80 slices) in less than a second using a NVIDIA GeForce GTX Titan X GPU (12 GB, Maxwell architecture).ConclusionThe proposed deep learning algorithm is promising to improve CBCT image quality in an efficient way, thus has a potential to support online CBCT-based adaptive radiotherapy.

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]

PurposeSegmentation of multiple organs-at-risk (OARs) is essential for magnetic resonance (MR)-only radiation therapy treatment planning and MR-guided adaptive radiotherapy of abdominal cancers. Current practice requires manual delineation that is labor-intensive, time-consuming, and prone to intra- and interobserver variations. We developed a deep learning (DL) technique for fully automated segmentation of multiple OARs on clinical abdominal MR images with high accuracy, reliability, and efficiency.MethodsWe developed Automated deep Learning-based abdominal multiorgan segmentation (ALAMO) technique based on two-dimensional U-net and a densely connected network structure with tailored design in data augmentation and training procedures such as deep connection, auxiliary supervision, and multiview. The model takes in multislice MR images and generates the output of segmentation results. 3.0-Tesla T1 VIBE (Volumetric Interpolated Breath-hold Examination) images of 102 subjects were used in our study and split into 66 for training, 16 for validation, and 20 for testing. Ten OARs were studied, including the liver, spleen, pancreas, left/right kidneys, stomach, duodenum, small intestine, spinal cord, and vertebral bodies. An experienced radiologist manually labeled each OAR, followed by reediting, if necessary, by a senior radiologist, to create the ground-truth. The performance was measured using volume overlapping and surface distance.ResultsThe ALAMO technique generated segmentation labels in good agreement with the manual results. Specifically, among the ten OARs, nine achieved high dice similarity coefficients (DSCs) in the range of 0.87-0.96, except for the duodenum with a DSC of 0.80. The inference completed within 1 min for a three-dimensional volume of 320 × 288 × 180. Overall, the ALAMO model matched the state-of-the-art techniques in performance.ConclusionThe proposed ALAMO technique allows for fully automated abdominal MR segmentation with high accuracy and practical memory and computation time demands.

PurposeFluence-modulated proton computed tomography (FMpCT) using pencil beam scanning aims at achieving task-specific image noise distributions by modulating the imaging proton fluence spot-by-spot based on an object-specific noise model. In this work, we present a method for fluence field optimization and investigate its performance in dose reduction for various phantoms and image variance targets.MethodsThe proposed method uses Monte Carlo simulations of a proton CT (pCT) prototype scanner to estimate expected variance levels at uniform fluence. Using an iterative approach, we calculate a stack of target variance projections that are required to achieve the prescribed image variance, assuming a reconstruction using filtered backprojection. By fitting a pencil beam model to the ratio of uniform fluence variance and target variance, relative weights for each pencil beam can be calculated. The quality of the resulting fluence modulations is evaluated by scoring imaging doses and comparing them to those at uniform fluence, as well as evaluating conformity of the achieved variance with the prescription. For three different phantoms, we prescribed constant image variance as well as two regions-of-interest (ROI) imaging tasks with inhomogeneous image variance. The shape of the ROIs followed typical beam profiles for proton therapy.ResultsPrescription of constant image variance resulted in a dose reduction of 8.9% for a homogeneous water phantom compared to a uniform fluence scan at equal peak variance level. For a more heterogeneous head phantom, dose reduction increased to 16.0% for the same task. Prescribing two different ROIs resulted in dose reductions between 25.7% and 40.5% outside of the ROI at equal peak variance levels inside the ROI. Imaging doses inside the ROI were increased by 9.2% to 19.2% compared to the uniform fluence scan, but can be neglected assuming that the ROI agrees with the therapeutic dose region. Agreement of resulting variance maps with the prescriptions was satisfactory.ConclusionsWe developed a method for fluence field optimization based on a noise model for a real scanner used in pCT. We demonstrated that it can achieve prescribed image variance targets. A uniform fluence field was shown not to be dose optimal and dose reductions achievable with the proposed method for FMpCT were considerable, opening an interesting perspective for image guidance and adaptive therapy.

PurposeTo develop and demonstrate the efficacy of a novel head-and-neck multimodality image registration technique using deep-learning-based cross-modality synthesis.Methods and materialsTwenty-five head-and-neck patients received magnetic resonance (MR) and computed tomography (CT) (CTaligned ) scans on the same day with the same immobilization. Fivefold cross validation was used with all of the MR-CT pairs to train a neural network to generate synthetic CTs from MR images. Twenty-four of 25 patients also had a separate CT without immobilization (CTnon-aligned ) and were used for testing. CTnon-aligned 's were deformed to the synthetic CT, and compared to CTnon-aligned registered to MR. The same registrations were performed from MR to CTnon-aligned and from synthetic CT to CTnon-aligned . All registrations used B-splines for modeling the deformation, and mutual information for the objective. Results were evaluated using the 95% Hausdorff distance among spinal cord contours, landmark error, inverse consistency, and Jacobian determinant of the estimated deformation fields.ResultsWhen large initial rigid misalignment is present, registering CT to MRI-derived synthetic CT aligns the cord better than a direct registration. The average landmark error decreased from 9.8 ± 3.1 mm in MR→CTnon-aligned to 6.0 ± 2.1 mm in CTsynth →CTnon-aligned deformable registrations. In the CT to MR direction, the landmark error decreased from 10.0 ± 4.3 mm in CTnon-aligned →MR deformable registrations to 6.6 ± 2.0 mm in CTnon-aligned →CTsynth deformable registrations. The Jacobian determinant had an average value of 0.98. The proposed method also demonstrated improved inverse consistency over the direct method.ConclusionsWe showed that using a deep learning-derived synthetic CT in lieu of an MR for MR→CT and CT→MR deformable registration offers superior results to direct multimodal registration.

PurposeCoronary x-ray computed tomography angiography (CCTA) continues to develop as a noninvasive method for the assessment of coronary vessel geometry and the identification of physiologically significant lesions. The uncertainty of quantitative lesion diameter measurement due to limited spatial resolution and vessel motion reduces the accuracy of CCTA diagnoses. In this paper, we introduce a new technique called computed tomography (CT)-number-Calibrated Diameter to improve the accuracy of the vessel and stenosis diameter measurements with CCTA.MethodsA calibration phantom containing cylindrical holes (diameters spanning from 0.8 mm through 4.0 mm) capturing the range of diameters found in human coronary vessels was three-dimensional printed. We also printed a human stenosis phantom with 17 tubular channels having the geometry of lesions derived from patient data. We acquired CT scans of the two phantoms with seven different imaging protocols. Calibration curves relating vessel intraluminal maximum voxel value (maximum CT number of a voxel, described in Hounsfield Units, HU) to true diameter, and full-width-at-half maximum (FWHM) to true diameter were constructed for each CCTA protocol. In addition, we acquired scans with a small constant motion (15 mm/s) and used a motion correction reconstruction (Snapshot Freeze) algorithm to correct motion artifacts. We applied our technique to measure the lesion diameter in the 17 lesions in the stenosis phantom and compared the performance of CT-number-Calibrated Diameter to the ground truth diameter and a FWHM estimate.ResultsIn all cases, vessel intraluminal maximum voxel value vs diameter was found to have a simple functional form based on the two-dimensional point spread function yielding a constant maximum voxel value region above a cutoff diameter, and a decreasing maximum voxel value vs decreasing diameter below a cutoff diameter. After normalization, focal spot size and reconstruction kernel were the principal determinants of cutoff diameter and the rate of maximum voxel value reduction vs decreasing diameter. The small constant motion had a significant effect on the CT number calibration; however, the motion-correction algorithm returned the maximum voxel value vs diameter curve to that of stationary vessels. The CT number Calibration technique showed better performance than FWHM estimation of diameter, yielding a high accuracy in the tested range (0.8 mm through 2.5 mm). We found a strong linear correlation between the smallest diameter in each of 17 lesions measured by CT-number-Calibrated Diameter (DC ) and ground truth diameter (Dgt ), (DC  = 0.951 × Dgt  + 0.023 mm, r = 0.998 with a slope very close to 1.0 and intercept very close to 0 mm.ConclusionsComputed tomography-number-Calibrated Diameter is an effective method to enhance the accuracy of the estimate of small vessel diameters and degree of coronary stenosis in CCTA.

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]

PurposeThe purpose of this work was twofold: (a) To provide a robust and accurate method for image segmentation of dedicated breast CT (bCT) volume data sets, and (b) to improve Hounsfield unit (HU) accuracy in bCT by means of a postprocessing method that uses the segmented images to correct for the low-frequency shading artifacts in reconstructed images.MethodsA sequential and iterative application of image segmentation and low-order polynomial fitting to bCT volume data sets was used in the interleaved correction (IC) method. Image segmentation was performed through a deep convolutional neural network (CNN) with a modified U-Net architecture. A total of 45 621 coronal bCT images from 111 patient volume data sets were segmented (using a previously published segmentation algorithm) and used for neural network training, validation, and testing. All patient data sets were selected from scans performed on four different prototype breast CT systems. The adipose voxels for each patient volume data set, segmented using the proposed CNN, were then fit to a three-dimensional low-order polynomial. The polynomial fit was subsequently used to correct for the shading artifacts introduced by scatter and beam hardening in a method termed "flat fielding." An interleaved utilization of image segmentation and flat fielding was repeated until a convergence criterion was satisfied. Mathematical and physical phantom studies were conducted to evaluate the dependence of the proposed algorithm on breast size and the distribution of fibroglandular tissue. In addition, a subset of patient scans (not used in the CNN training, testing or validation) were used to investigate the accuracy of the IC method across different scanner designs and beam qualities.ResultsThe IC method resulted in an accurate classification of different tissue types with an average Dice similarity coefficient > 95%, precision > 97%, recall > 95%, and F1-score > 96% across all tissue types. The flat fielding correction of bCT images resulted in a significant reduction in either cupping or capping artifacts in both mathematical and physical phantom studies as measured by the integral nonuniformity metric with an average reduction of 71% for cupping and 30% for capping across different phantom sizes, and the Uniformity Index with an average reduction of 53% for cupping and 34% for capping.ConclusionThe validation studies demonstrated that the IC method improves Hounsfield Units (HU) accuracy and effectively corrects for shading artifacts caused by scatter contamination and beam hardening. The postprocessing approach described herein is relevant to the broad scope of bCT devices and does not require any modification in hardware or existing scan protocols. The trained CNN parameters and network architecture are available for interested users.

PurposeWith recent substantial improvements in modern computing, interest in quantitative imaging with CT has seen a dramatic increase. As a result, the need to both create and analyze large, high-quality datasets of clinical studies has increased as well. At present, no efficient, widely available method exists to accomplish this. The purpose of this technical note is to describe an open-source high-throughput computational pipeline framework for the reconstruction and analysis of diagnostic CT imaging data to conduct large-scale quantitative imaging studies and to accelerate and improve quantitative imaging research.MethodsThe pipeline consists of two, primary "blocks": reconstruction and analysis. Reconstruction is carried out via a graphics processing unit (GPU) queuing framework developed specifically for the pipeline that allows a dataset to be reconstructed using a variety of different parameter configurations such as slice thickness, reconstruction kernel, and simulated acquisition dose. The analysis portion then automatically analyzes the output of the reconstruction using "modules" that can be combined in various ways to conduct different experiments. Acceleration of analysis is achieved using cluster processing. Efficiency and performance of the pipeline are demonstrated using an example 142 subject lung screening cohort reconstructed 36 different ways and analyzed using quantitative emphysema scoring techniques.ResultsThe pipeline reconstructed and analyzed the 5112 reconstructed datasets in approximately 10 days, a roughly 72× speedup over previous efforts using the scanner for reconstructions. Tightly coupled pipeline quality assurance software ensured proper performance of analysis modules with regard to segmentation and emphysema scoring.ConclusionsThe pipeline greatly reduced the time from experiment conception to quantitative results. The modular design of the pipeline allows the high-throughput framework to be utilized for other future experiments into different quantitative imaging techniques. Future applications of the pipeline being explored are robustness testing of quantitative imaging metrics, data generation for deep learning, and use as a test platform for image-processing techniques to improve clinical quantitative imaging.

PurposeWomen with radiographically dense or texturally complex breasts are at increased risk for interval cancer, defined as cancers diagnosed after a normal screening examination. The purpose of this study was to create masking measures and apply them to identify interval risk in a population of women who experienced either screen-detected or interval cancers after controlling for breast density.MethodsWe examined full-field digital screening mammograms acquired from 2006 to 2015. Examinations associated with 182 interval cancers were matched to 173 screen-detected cancers on age, race, exam date and time since last imaging examination. Local Image Quality Factor (IQF) values were calculated and used to create IQF maps that represented mammographic masking. We used various statistics to define global masking measures of these maps. Association of these masking measures with interval cancer vs screen-detected cancer was estimated using conditional logistic regression in a univariate and adjusted model for Breast Imaging-Reporting and Data System (BI-RADS) density. Receiver operator curves were calculated in each case to compare specificity vs sensitivity, and area under those curves were generated. Proportion of screen-detected cancer was estimated for stratifications of IQF features.ResultsSeveral masking features showed significant association with interval compared to screen-detected cancers after adjusting for BI-RADS density (up to P = 2.52E-6), and the 10th percentile of the IQF value (P = 1.72E-3) showed the strongest improvement in the area under the receiver operator curve, increasing from 0.65 using only BI-RADS density to 0.69. The highest masking group had a 32% proportion of screen-detected cancers while the low masking group had a 69% proportion.ConclusionsWe conclude that computer vision methods using model observers may improve quantifying the probability of breast cancer detection beyond using breast density alone.

PurposeCT ventilation imaging (CTVI) is being used to achieve functional avoidance lung cancer radiation therapy in three clinical trials (NCT02528942, NCT02308709, NCT02843568). To address the need for common CTVI validation tools, we have built the Ventilation And Medical Pulmonary Image Registration Evaluation (VAMPIRE) Dataset, and present the results of the first VAMPIRE Challenge to compare relative ventilation distributions between different CTVI algorithms and other established ventilation imaging modalities.MethodsThe VAMPIRE Dataset includes 50 pairs of 4DCT scans and corresponding clinical or experimental ventilation scans, referred to as reference ventilation images (RefVIs). The dataset includes 25 humans imaged with Galligas 4DPET/CT, 21 humans imaged with DTPA-SPECT, and 4 sheep imaged with Xenon-CT. For the VAMPIRE Challenge, 16 subjects were allocated to a training group (with RefVI provided) and 34 subjects were allocated to a validation group (with RefVI blinded). Seven research groups downloaded the Challenge dataset and uploaded CTVIs based on deformable image registration (DIR) between the 4DCT inhale/exhale phases. Participants used DIR methods broadly classified into B-splines, Free-form, Diffeomorphisms, or Biomechanical modeling, with CT ventilation metrics based on the DIR evaluation of volume change, Hounsfield Unit change, or various hybrid approaches. All CTVIs were evaluated against the corresponding RefVI using the voxel-wise Spearman coefficient rS , and Dice similarity coefficients evaluated for low function lung ( DSClow ) and high function lung ( DSChigh ).ResultsA total of 37 unique combinations of DIR method and CT ventilation metric were either submitted by participants directly or derived from participant-submitted DIR motion fields using the in-house software, VESPIR. The rS and DSC results reveal a high degree of inter-algorithm and intersubject variability among the validation subjects, with algorithm rankings changing by up to ten positions depending on the choice of evaluation metric. The algorithm with the highest overall cross-modality correlations used a biomechanical model-based DIR with a hybrid ventilation metric, achieving a median (range) of 0.49 (0.27-0.73) for rS , 0.52 (0.36-0.67) for DSClow , and 0.45 (0.28-0.62) for DSChigh . All other algorithms exhibited at least one negative rS value, and/or one DSC value less than 0.5.ConclusionsThe VAMPIRE Challenge results demonstrate that the cross-modality correlation between CTVIs and the RefVIs varies not only with the choice of CTVI algorithm but also with the choice of RefVI modality, imaging subject, and the evaluation metric used to compare relative ventilation distributions. This variability may arise from the fact that each of the different CTVI algorithms and RefVI modalities provides a distinct physiologic measurement. Ultimately this variability, coupled with the lack of a "gold standard," highlights the ongoing importance of further validation studies before CTVI can be widely translated from academic centers to the clinic. It is hoped that the information gleaned from the VAMPIRE Challenge can help inform future validation efforts.

PURPOSE: We propose a deep learning-based approach to breast mass classification in sonography and compare it with the assessment of four experienced radiologists employing breast imaging reporting and data system 4th edition lexicon and assessment protocol. METHODS: Several transfer learning techniques are employed to develop classifiers based on a set of 882 ultrasound images of breast masses. Additionally, we introduce the concept of a matching layer. The aim of this layer is to rescale pixel intensities of the grayscale ultrasound images and convert those images to red, green, blue (RGB) to more efficiently utilize the discriminative power of the convolutional neural network pretrained on the ImageNet dataset. We present how this conversion can be determined during fine-tuning using back-propagation. Next, we compare the performance of the transfer learning techniques with and without the color conversion. To show the usefulness of our approach, we additionally evaluate it using two publicly available datasets. RESULTS: Color conversion increased the areas under the receiver operating curve for each transfer learning method. For the better-performing approach utilizing the fine-tuning and the matching layer, the area under the curve was equal to 0.936 on a test set of 150 cases. The areas under the curves for the radiologists reading the same set of cases ranged from 0.806 to 0.882. In the case of the two separate datasets, utilizing the proposed approach we achieved areas under the curve of around 0.890. CONCLUSIONS: The concept of the matching layer is generalizable and can be used to improve the overall performance of the transfer learning techniques using deep convolutional neural networks. When fully developed as a clinical tool, the methods proposed in this paper have the potential to help radiologists with breast mass classification in ultrasound.

PurposeTime of flight (TOF) PET reconstruction is well known to statistically improve the image quality compared to non-TOF PET. Although TOF PET can improve the overall signal to noise ratio (SNR) of the image compared to non-TOF PET, the SNR disparity between separate regions in the reconstructed image using TOF data becomes higher than that using non-TOF data. Using the conventional ordered subset expectation maximization (OS-EM) method, the SNR in the low activity regions becomes significantly lower than in the high activity regions due to the different photon statistics of TOF bins. A uniform recovery across different SNR regions is preferred if it can yield an overall good image quality within small number of iterations in practice. To allow more uniform recovery of regions, a spatially variant update is necessary for different SNR regions.MethodsThis paper focuses on designing a spatially variant step size and proposes a TOF-PET reconstruction method that uses a nonuniform separable quadratic surrogates (NUSQS) algorithm, providing a straightforward control of spatially variant step size. To control the noise, a spatially invariant quadratic regularization is incorporated, which by itself does not theoretically affect the recovery uniformity. The Nesterov's momentum method with ordered subsets (OS) is also used to accelerate the reconstruction speed. To evaluate the proposed method, an XCAT simulation phantom and clinical data from a pancreas cancer patient with full (ground truth) and 6× downsampled counts were used, where a Poisson thinning process was employed for downsampling. We selected tumor and cold regions of interest (ROIs) and compared the proposed method with the TOF-based conventional OS-EM and OS-SQS algorithms with an early stopping criterion.ResultsIn computer simulation, without regularization, hot regions of OS-EM and OS-NUSQS converged similarly, but cold region of OS-EM was noisier than OS-NUSQS after 24 iterations. With regularization, although the overall speeds of OS-EM and OS-NUSQS were similar, recovery ratios of hot and cold regions reconstructed by the OS-NUSQS were more uniform compared to those of the conventional OS-SQS and OS-EM. The OS-NUSQS with Nesterov's momentum converged faster than others while preserving the uniform recovery. In the clinical example, we demonstrated that the OS-NUSQS with Nesterov's momentum provides more uniform recovery ratios of hot and cold ROIs compared to the OS-SQS and OS-EM. Although the cost function of all methods is equivalent, the proposed method has higher structural similarity (SSIM) values of hot and cold regions compared to other methods after 24 iterations. Furthermore, our computing time using graphics processing unit was 80× shorter than the time using quad-core CPUs.ConclusionThis paper proposes a TOF PET reconstruction method using the OS-NUSQS with Nesterov's momentum for uniform recovery of different SNR regions. In particular, the spatially nonuniform step size in the proposed method provides uniform recovery ratios of different SNR regions, and the Nesterov's momentum further accelerates overall convergence while preserving uniform recovery. The computer simulation and clinical example demonstrate that the proposed method converges uniformly across ROIs. In addition, tumor contrast and SSIM of the proposed method were higher than those of the conventional OS-EM and OS-SQS in early iterations.

PurposeSize-specific dose estimates (SSDE) conversion factors have been determined by AAPM Report 204 to adjust CTDIvol to account for patient size but were limited to body CT examinations. The purpose of this work was to determine conversion factors that could be used for an SSDE for helical, head CT examinations for patients of different sizes.MethodsValidated Monte Carlo (MC) simulation methods were used to estimate dose to the center of the scan volume from a routine, helical head examination for a group of patient models representing a range of ages and sizes. Ten GSF/ICRP voxelized phantom models and five pediatric voxelized patient models created from CT image data were used in this study. CT scans were simulated using a Siemens multidetector row CT equivalent source model. Scan parameters were taken from the AAPM Routine Head protocols for a fixed tube current (FTC), helical protocol, and scan lengths were adapted to the anatomy of each patient model. MC simulations were performed using mesh tallies to produce voxelized dose distributions for the entire scan volume of each model. Three tally regions were investigated: (1) a small 0.6 cc volume at the center of the scan volume, (2) 0.8-1.0 cm axial slab at the center of the scan volume, and (3) the entire scan volume. Mean dose to brain parenchyma for all three regions was calculated. Mean bone dose and a mass-weighted average dose, consisting of brain parenchyma and bone, were also calculated for the slab in the central plane and the entire scan volume. All dose measures were then normalized by CTDIvol for the 16 cm phantom (CTDIvol,16 ). Conversion factors were determined by calculating the relationship between normalized doses and water equivalent diameter (Dw ).ResultsCTDIvol,16 -normalized mean brain parenchyma dose values within the 0.6 cc volume, 0.8-1.0 cm central axial slab, and the entire scan volume, when parameterized by Dw , had an exponential relationship with a coefficient of determination (R2 ) of 0.86, 0.84, and 0.88, respectively. There was no statistically significant difference between the conversion factors resulting from these three different tally regions. Exponential relationships between CTDIvol,16 -normalized mean bone doses had R2 values of 0.83 and 0.87 for the central slab and for the entire scan volume, respectively. CTDIvol,16 -normalized mass-weighted average doses had R2 values of 0.39 and 0.51 for the central slab and for the entire scan volume, respectively.ConclusionsConversion factors that describe the exponential relationship between CTDIvol,16 -normalized mean brain dose and a size metric (Dw ) for helical head CT examinations have been reported for two different interpretations of the center of the scan volume. These dose descriptors have been extended to describe the dose to bone in the center of the scan volume as well as a mass-weighted average dose to brain and bone. These may be used, when combined with other efforts, to develop an SSDE dose coefficients for routine, helical head CT examinations.

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]

PURPOSE:Intensity modulated radiation therapy (IMRT) is commonly employed for treating head and neck (H&N) cancer with uniform tumor dose and conformal critical organ sparing. Accurate delineation of organs-at-risk (OARs) on H&N CT images is thus essential to treatment quality. Manual contouring used in current clinical practice is tedious, time-consuming, and can produce inconsistent results. Existing automated segmentation methods are challenged by the substantial inter-patient anatomical variation and low CT soft tissue contrast. To overcome the challenges, we developed a novel automated H&N OARs segmentation method that combines a fully convolutional neural network (FCNN) with a shape representation model (SRM). METHODS:Based on manually segmented H&N CT, the SRM and FCNN were trained in two steps: (a) SRM learned the latent shape representation of H&N OARs from the training dataset; (b) the pre-trained SRM with fixed parameters were used to constrain the FCNN training. The combined segmentation network was then used to delineate nine OARs including the brainstem, optic chiasm, mandible, optical nerves, parotids, and submandibular glands on unseen H&N CT images. Twenty-two and 10 H&N CT scans provided by the Public Domain Database for Computational Anatomy (PDDCA) were utilized for training and validation, respectively. Dice similarity coefficient (DSC), positive predictive value (PPV), sensitivity (SEN), average surface distance (ASD), and 95% maximum surface distance (95%SD) were calculated to quantitatively evaluate the segmentation accuracy of the proposed method. The proposed method was compared with an active appearance model that won the 2015 MICCAI H&N Segmentation Grand Challenge based on the same dataset, an atlas method and a deep learning method based on different patient datasets. RESULTS:An average DSC = 0.870 (brainstem), DSC = 0.583 (optic chiasm), DSC = 0.937 (mandible), DSC = 0.653 (left optic nerve), DSC = 0.689 (right optic nerve), DSC = 0.835 (left parotid), DSC = 0.832 (right parotid), DSC = 0.755 (left submandibular), and DSC = 0.813 (right submandibular) were achieved. The segmentation results are consistently superior to the results of atlas and statistical shape based methods as well as a patch-wise convolutional neural network method. Once the networks are trained off-line, the average time to segment all 9 OARs for an unseen CT scan is 9.5 s. CONCLUSION:Experiments on clinical datasets of H&N patients demonstrated the effectiveness of the proposed deep neural network segmentation method for multi-organ segmentation on volumetric CT scans. The accuracy and robustness of the segmentation were further increased by incorporating shape priors using SMR. The proposed method showed competitive performance and took shorter time to segment multiple organs in comparison to state of the art methods.

PurposeTo evaluate the accuracy of relative stopping power and spatial resolution of images reconstructed with simulated helium CT (HeCT) in comparison to proton CT (pCT).MethodsA Monte Carlo (MC) study with the TOPAS tool was performed to compare the accuracy of relative stopping power (RSP) reconstruction and spatial resolution of low-fluence HeCT to pCT, both using 200 MeV/u particles. An ideal setup consisting of a flat beam source and a totally absorbing energy-range detector was implemented to estimate the theoretically best achievable RSP accuracy for the calibration and reconstruction methods currently used for pCT. The phantoms imaged included a cylindrical water phantom with inserts of different materials, sizes, and positions, a Catphan phantom with a module containing high-contrast line pairs (CTP528) and a module with cylindrical inserts of different RSP (CTP404), as well as a voxelized 10-year-old female phantom. Dose to the cylindrical water phantom was also calculated. The RSP accuracy was studied for all phantoms except the CTP528 module. The latter was used for the estimation of the spatial resolution, evaluated as the modulation transfer function (MTF) at 10%.ResultsAn overall error under 0.5% was achieved for HeCT for the water phantoms with the different inserts, in all cases better than that for pCT, in some cases by a factor 3. The inserts in the CTP404 module were reconstructed with an average RSP accuracy of 0.3% for HeCT and 0.2% for pCT. Anatomic structures (brain, bones, air cavities, etc.) in the digitized head phantom were well recognizable and no artifacts were visible with both HeCT and pCT. The three main tissue materials (soft tissue, brain, and cranium) were well identifiable in the reconstructed RSP-volume distribution with both imaging modalities. Using 360 projection angles, the spatial resolution was 4 lp/cm for HeCT and 3 lp/cm for pCT. Generally, spatial resolution increased with the number of projection angles and was always higher for HeCT than for pCT for the same number of projections. When HeCT and pCT scan were performed to deliver the same dose in the phantom, the resolution for HeCT was higher than pCT.ConclusionMC simulations were used to compare HeCT and pCT image reconstruction. HeCT images had similar or better RSP accuracy and higher spatial resolution compared to pCT. Further investigation of the potential of helium ion imaging is warranted.

PurposeThis study investigates forced localization of targets in simulated images with statistical properties similar to trans-axial sections of x-ray computed tomography (CT) volumes. A total of 24 imaging conditions are considered, comprising two target sizes, three levels of background variability, and four levels of frequency apodization. The goal of the study is to better understand how human observers perform forced-localization tasks in images with CT-like statistical properties.MethodsThe transfer properties of CT systems are modeled by a shift-invariant transfer function in addition to apodization filters that modulate high spatial frequencies. The images contain noise that is the combination of a ramp-spectrum component, simulating the effect of acquisition noise in CT, and a power-law component, simulating the effect of normal anatomy in the background, which are modulated by the apodization filter as well. Observer performance is characterized using two psychophysical techniques: efficiency analysis and classification image analysis. Observer efficiency quantifies how much diagnostic information is being used by observers to perform a task, and classification images show how that information is being accessed in the form of a perceptual filter.ResultsPsychophysical studies from five subjects form the basis of the results. Observer efficiency ranges from 29% to 77% across the different conditions. The lowest efficiency is observed in conditions with uniform backgrounds, where significant effects of apodization are found. The classification images, estimated using smoothing windows, suggest that human observers use center-surround filters to perform the task, and these are subjected to a number of subsequent analyses. When implemented as a scanning linear filter, the classification images appear to capture most of the observer variability in efficiency (r2 = 0.86). The frequency spectra of the classification images show that frequency weights generally appear bandpass in nature, with peak frequency and bandwidth that vary with statistical properties of the images.ConclusionsIn these experiments, the classification images appear to capture important features of human-observer performance. Frequency apodization only appears to have a significant effect on performance in the absence of anatomical variability, where the observers appear to underweight low spatial frequencies that have relatively little noise. Frequency weights derived from the classification images generally have a bandpass structure, with adaptation to different conditions seen in the peak frequency and bandwidth. The classification image spectra show relatively modest changes in response to different levels of apodization, with some evidence that observers are attempting to rebalance the apodized spectrum presented to them.

PurposeTo study the variability in volume change estimates of pulmonary nodules due to segmentation approaches used across several algorithms and to evaluate these effects on the ability to predict nodule malignancy.MethodsWe obtained 100 patient image datasets from the National Lung Screening Trial (NLST) that had a nodule detected on each of two consecutive low dose computed tomography (LDCT) scans, with an equal proportion of malignant and benign cases (50 malignant, 50 benign). Information about the nodule location for the cases was provided by a screen capture with a bounding box and its axial location was indicated. Five participating quantitative imaging network (QIN) institutions performed nodule segmentation using their preferred semi-automated algorithms with no manual correction; teams were allowed to provide additional manually corrected segmentations (analyzed separately). The teams were asked to provide segmentation masks for each nodule at both time points. From these masks, the volume was estimated for the nodule at each time point; the change in volume (absolute and percent change) across time points was estimated as well. We used the concordance correlation coefficient (CCC) to compare the similarity of computed nodule volumes (absolute and percent change) across algorithms. We used Logistic regression model on the change in volume (absolute change and percent change) of the nodules to predict the malignancy status, the area under the receiver operating characteristic curve (AUROC) and confidence intervals were reported. Because the size of nodules was expected to have a substantial effect on segmentation variability, analysis of change in volumes was stratified by lesion size, where lesions were grouped into those with a longest diameter of

PurposeCT Perfusion (CTP) is a widely used clinical imaging modality. However, CTP typically involves the use of substantial radiation dose (CTDIvol ≥~200 mGy). The purpose of this study is to present a low-dose CTP technique using a projection view-sharing reconstruction algorithm originally developed for dynamic MRI - "K-space Weighted Image Contrast" (KWIC).MethodsThe KWIC reconstruction is based on an angle-bisection scheme. In KWIC, a Fourier transform was performed along each projection to form a "k-space"-like CT data space, based on the central-slice theorem. As a projection view-sharing technique, KWIC preserves the spatiotemporal resolution of undersampled CTP data by progressively increasing the number of projection views shared for more distant regions of "k-space". KWIC reconstruction was evaluated on a digital FORBILD head phantom with numerically simulated time-varying objects. The numerically simulated scans were undersampled using the angle-bisection scheme to achieve 50%, 25%, and 12.5% of the original dose (288, 144, and 72 projections, respectively). The area-under-the-curve (AUC), time-to-peak (TTP), and full width half maximum (FWHM) were measured in KWIC recons and compared to fully sampled filtered back projection (FBP) reconstructions. KWIC reconstruction and dose reduction was also implemented for three clinical CTP cases (45 s, 1156 projections per turn, 1 s/turn, CTDIvol 217 mGy). Quantitative perfusion metrics were computed and compared between KWIC reconstructed CTP data and those of standard FBP reconstruction.ResultsThe AUC, TTP, and FWHM in the numerical simzulations were unaffected by the undersampling/dose reduction (down to 12.5% dose) with KWIC reconstruction compared to the fully sampled FBP reconstruction. The normalized root-mean-square-error (NRMSE) of the AUC in the FORBILD head phantom is 0.04, 0.05, and 0.07 for 50%, 25%, and 12.5% KWIC, respectively, as compared to FBP reconstruction. The cerebral blood flow (CBF) and cerebral blood volume had no significant difference between FBP and 50%, 25%, and 12.5% KWIC reconstructions (P > 0.05).ConclusionsThis study demonstrates that KWIC preserves perfusion metrics for CTP with substantially reduced dose. Clinical implementation will require further investigation into methods of rapid switching of a CT x-ray source.

PurposeMicrovascular 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.MethodsAn US system equipped with a linear array transducer was used in a harmonic imaging mode at low transmit power. C57BL/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.ResultsExperimental 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.ConclusionsThis 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.

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]

PurposeTo determine the dependence of the accuracy in reconstruction of relative stopping power (RSP) with proton computerized tomography (pCT) scans on the purity of the proton beam and the technological complexity of the pCT scanner using standard phantoms and a digital representation of a pediatric patient.MethodsThe Monte Carlo method was applied to simulate the pCT scanner, using both a pure proton beam (uniform 200 MeV mono-energetic, parallel beam) and the Northwestern Medicine Chicago Proton Center (NMCPC) clinical beam in uniform scanning mode. The accuracy of the simulation was validated with measurements performed at NMCPC including reconstructed RSP images obtained with a preclinical prototype pCT scanner. The pCT scanner energy detector was then simulated in three configurations of increasing complexity: an ideal totally absorbing detector, a single stage detector and a multi-stage detector. A set of 15 cm diameter water cylinders containing either water alone or inserts of different material, size, and position were simulated at 90 projection angles (4° steps) for the pure and clinical proton beams and the three pCT configurations. A pCT image of the head of a detailed digital pediatric phantom was also reconstructed from the simulated pCT scan with the prototype detector.ResultsThe RSP error increased for all configurations for insert sizes under 7.5 mm in radius, with a sharp increase below 5 mm in radius, attributed to a limit in spatial resolution. The highest accuracy achievable using the current pCT calibration step phantom and reconstruction algorithm, calculated for the ideal case of a pure beam with totally absorbing energy detector, was 1.3% error in RSP for inserts of 5 mm radius or more, 0.7 mm in range for the 2.5 mm radius inserts, or better. When the highest complexity of the scanner geometry was introduced, some artifacts arose in the reconstructed images, particularly in the center of the phantom. Replacing the step phantom used for calibration with a wedge phantom led to RSP accuracy close to the ideal case, with no significant dependence of RSP error on insert location or material. The accuracy with the multi-stage detector and NMCPC beam for the cylindrical phantoms was 2.2% in RSP error for inserts of 5 mm radius or more, 0.7 mm in range for the 2.5 mm radius inserts, or better. The pCT scan of the pediatric phantom resulted in mean RSP values within 1.3% of the reference RSP, with a range error under 1 mm, except in exceptional situations of parallel incidence on a boundary between low and high density.ConclusionsThe pCT imaging technique proved to be a precise and accurate imaging tool, rivaling the current x-rays based techniques, with the advantage of being directly sensitive to proton stopping power rather than photon interaction coefficients. Measured and simulated pCT images were obtained from a wobbled proton beam for the first time. Since the in-silico results are expected to accurately represent the prototype pCT, upcoming measurements using the wedge phantom for calibration are expected to show similar accuracy in the reconstructed RSP.

PurposeThe positive predictive power for malignancy can potentially be improved, if the chemical compositions of suspicious breast lesions can be reliably measured in screening mammography. The purpose of this study is to investigate the feasibility of quantifying breast lesion composition, in terms of water and lipid contents, with spectral mammography.MethodsPhantom and tissue samples were imaged with a spectral mammography system based on silicon-strip photon-counting detectors. Dual-energy calibration was performed for material decomposition, using plastic water and adipose-equivalent phantoms as the basis materials. The step wedge calibration phantom consisted of 20 calibration configurations, which ranged from 2 to 8 cm in thickness and from 0% to 100% in plastic water density. A nonlinear rational fitting function was used in dual-energy calibration of the imaging system. Breast lesion phantoms, made from various combinations of plastic water and adipose-equivalent disks, were embedded in a breast mammography phantom with a heterogeneous background pattern. Lesion phantoms with water densities ranging from 0% to 100% were placed at different locations of the heterogeneous background phantom. The water density in the lesion phantoms was measured using dual-energy material decomposition. The thickness and density of the background phantom were varied to test the accuracy of the decomposition technique in different configurations. In addition, an in vitro study was also performed using mixtures of lean and fat bovine tissue of 25%, 50%, and 80% lean weight percentages as the background. Lesions were simulated by using breast lesion phantoms, as well as small bovine tissue samples, composed of carefully weighed lean and fat bovine tissues. The water densities in tissue samples were measured using spectral mammography and compared to measurement using chemical decomposition of the tissue.ResultsThe thickness of measured and known water contents was compared for various lesion configurations. There was a good linear correlation between the measured and the known values. The root-mean-square errors in water thickness measurements were 0.3 and 0.2 mm for the plastic phantom and bovine tissue backgrounds, respectively.ConclusionsThe results indicate that spectral mammography can be used to accurately characterize breast lesion composition in terms of their equivalent water and lipid contents.

PurposeTo investigate the effects of dose level and reconstruction method on density and texture based features computed from CT lung nodules.MethodsThis study had two major components. In the first component, a uniform water phantom was scanned at three dose levels and images were reconstructed using four conventional filtered backprojection (FBP) and four iterative reconstruction (IR) methods for a total of 24 different combinations of acquisition and reconstruction conditions. In the second component, raw projection (sinogram) data were obtained for 33 lung nodules from patients scanned as a part of their clinical practice, where low dose acquisitions were simulated by adding noise to sinograms acquired at clinical dose levels (a total of four dose levels) and reconstructed using one FBP kernel and two IR kernels for a total of 12 conditions. For the water phantom, spherical regions of interest (ROIs) were created at multiple locations within the water phantom on one reference image obtained at a reference condition. For the lung nodule cases, the ROI of each nodule was contoured semiautomatically (with manual editing) from images obtained at a reference condition. All ROIs were applied to their corresponding images reconstructed at different conditions. For 17 of the nodule cases, repeat contours were performed to assess repeatability. Histogram (eight features) and gray level co-occurrence matrix (GLCM) based texture features (34 features) were computed for all ROIs. For the lung nodule cases, the reference condition was selected to be 100% of clinical dose with FBP reconstruction using the B45f kernel; feature values calculated from other conditions were compared to this reference condition. A measure was introduced, which the authors refer to as Q, to assess the stability of features across different conditions, which is defined as the ratio of reproducibility (across conditions) to repeatability (across repeat contours) of each feature.ResultsThe water phantom results demonstrated substantial variability among feature values calculated across conditions, with the exception of histogram mean. Features calculated from lung nodules demonstrated similar results with histogram mean as the most robust feature (Q ≤ 1), having a mean and standard deviation Q of 0.37 and 0.22, respectively. Surprisingly, histogram standard deviation and variance features were also quite robust. Some GLCM features were also quite robust across conditions, namely, diff. variance, sum variance, sum average, variance, and mean. Except for histogram mean, all features have a Q of larger than one in at least one of the 3% dose level conditions.ConclusionsAs expected, the histogram mean is the most robust feature in their study. The effects of acquisition and reconstruction conditions on GLCM features vary widely, though trending toward features involving summation of product between intensities and probabilities being more robust, barring a few exceptions. Overall, care should be taken into account for variation in density and texture features if a variety of dose and reconstruction conditions are used for the quantification of lung nodules in CT, otherwise changes in quantification results may be more reflective of changes due to acquisition and reconstruction conditions than in the nodule itself.

PurposeBreast elastography is a critical tool for improving the targeted radiotherapy treatment of breast tumors. Current breast radiotherapy imaging protocols only involve prone and supine CT scans. There is a lack of knowledge on the quantitative accuracy with which breast elasticity can be systematically measured using only prone and supine CT datasets. The purpose of this paper is to describe a quantitative elasticity estimation technique for breast anatomy using only these supine/prone patient postures. Using biomechanical, high-resolution breast geometry obtained from CT scans, a systematic assessment was performed in order to determine the feasibility of this methodology for clinically relevant elasticity distributions.MethodsA model-guided inverse analysis approach is presented in this paper. A graphics processing unit (GPU)-based linear elastic biomechanical model was employed as a forward model for the inverse analysis with the breast geometry in a prone position. The elasticity estimation was performed using a gradient-based iterative optimization scheme and a fast-simulated annealing (FSA) algorithm. Numerical studies were conducted to systematically analyze the feasibility of elasticity estimation. For simulating gravity-induced breast deformation, the breast geometry was anchored at its base, resembling the chest-wall/breast tissue interface. Ground-truth elasticity distributions were assigned to the model, representing tumor presence within breast tissue. Model geometry resolution was varied to estimate its influence on convergence of the system. A priori information was approximated and utilized to record the effect on time and accuracy of convergence. The role of the FSA process was also recorded. A novel error metric that combined elasticity and displacement error was used to quantify the systematic feasibility study. For the authors' purposes, convergence was set to be obtained when each voxel of tissue was within 1 mm of ground-truth deformation.ResultsThe authors' analyses showed that a ∼97% model convergence was systematically observed with no-a priori information. Varying the model geometry resolution showed no significant accuracy improvements. The GPU-based forward model enabled the inverse analysis to be completed within 10-70 min. Using a priori information about the underlying anatomy, the computation time decreased by as much as 50%, while accuracy improved from 96.81% to 98.26%. The use of FSA was observed to allow the iterative estimation methodology to converge more precisely.ConclusionsBy utilizing a forward iterative approach to solve the inverse elasticity problem, this work indicates the feasibility and potential of the fast reconstruction of breast tissue elasticity using supine/prone patient postures.

PurposeWith growing interest in quantitative imaging, radiomics, and CAD using CT imaging, the need to explore the impacts of acquisition and reconstruction parameters has grown. This usually requires extensive access to the scanner on which the data were acquired and its workflow is not designed for large-scale reconstruction projects. Therefore, the authors have developed a freely available, open-source software package implementing a common reconstruction method, weighted filtered backprojection (wFBP), for helical fan-beam CT applications.MethodsFreeCT_wFBP is a low-dependency, GPU-based reconstruction program utilizing c for the host code and Nvidia CUDA C for GPU code. The software is capable of reconstructing helical scans acquired with arbitrary pitch-values, and sampling techniques such as flying focal spots and a quarter-detector offset. In this work, the software has been described and evaluated for reconstruction speed, image quality, and accuracy. Speed was evaluated based on acquisitions of the ACR CT accreditation phantom under four different flying focal spot configurations. Image quality was assessed using the same phantom by evaluating CT number accuracy, uniformity, and contrast to noise ratio (CNR). Finally, reconstructed mass-attenuation coefficient accuracy was evaluated using a simulated scan of a FORBILD thorax phantom and comparing reconstructed values to the known phantom values.ResultsThe average reconstruction time evaluated under all flying focal spot configurations was found to be 17.4 ± 1.0 s for a 512 row × 512 column × 32 slice volume. Reconstructions of the ACR phantom were found to meet all CT Accreditation Program criteria including CT number, CNR, and uniformity tests. Finally, reconstructed mass-attenuation coefficient values of water within the FORBILD thorax phantom agreed with original phantom values to within 0.0001 mm(2)/g (0.01%).ConclusionsFreeCT_wFBP is a fast, highly configurable reconstruction package for third-generation CT available under the GNU GPL. It shows good performance with both clinical and simulated data.

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]