Search

Explainability of convolutional neural networks (CNNs) is integral for their adoption into radiological practice. Commonly used attribution methods localize image areas important for CNN prediction but do not characterize relevant imaging features underlying these areas, acting as a barrier to the adoption of CNNs for clinical use. We therefore propose Semantic Exploration and Explainability using a Style-based Generative Adversarial Autoencoder Network (SEE-GAAN), an explainability framework that uses latent space manipulation to generate a sequence of synthetic images that semantically visualizes how clinical and CNN features manifest within medical images. Visual analysis of changes in these sequences then facilitates the interpretation of features, thereby improving explainability. SEE-GAAN was first developed on a cohort of 26,664 chest radiographs across 15,409 patients from our institution. SEE-GAAN sequences were then generated across several clinical features and CNN predictions of NT-pro B-type natriuretic peptide (BNPP) as a proxy for acute heart failure. Radiological interpretations indicated SEE-GAAN sequences captured relevant changes in anatomical and pathological morphology associated with clinical and CNN predictions and clarified ambiguous areas highlighted by commonly used attribution methods. Our study demonstrates SEE-GAAN can facilitate our understanding of clinical features for imaging biomarker exploration and improve CNN transparency over commonly used explainability methods.

The diagnosis of a common cause of chronic pelvic pain can be made by visualizing reflux in the ovarian veins. Fluoroscopic venography is the gold standard for diagnosing ovarian vein reflux, but it is an invasive technique that exposes patients to ionizing radiation. MRI, with its lack of ionizing radiation and capability of high-temporal and spatial-resolution vascular imaging, has the potential to provide similar diagnostic information. This retrospective report describes and assesses the utility of a dynamic contrast-enhanced MRI technique based on Differential Subsampling with Cartesian Ordering (DISCO)-MRI in 30 patients with chronic pelvic pain. Among the 14 patients who underwent both DISCO-MRI and fluoroscopic venograms, 11 (78.6%) exhibited concordant results, while 3 patients (21.4%) had discordant findings. These results suggest the potential of multiphasic contrast-enhanced DISCO-MRI as a non-invasive diagnostic tool for evaluating chronic pelvic pain.

PurposeTo assess if a novel automated method to spatially delineate and quantify the extent of hypoperfusion on multienergy CT angiograms can aid the evaluation of chronic thromboembolic pulmonary hypertension (CTEPH) disease severity.Materials and methodsMultienergy CT angiograms obtained between January 2018 and December 2020 in 51 patients with CTEPH (mean age, 47 years ± 17 [SD]; 27 women) were retrospectively compared with those in 110 controls with no imaging findings suggestive of pulmonary vascular abnormalities (mean age, 51 years ± 16; 81 women). Parenchymal iodine values were automatically isolated using deep learning lobar lung segmentations. Low iodine concentration was used to delineate areas of hypoperfusion and calculate hypoperfused lung volume (HLV). Receiver operating characteristic curves, correlations with preoperative and postoperative changes in invasive hemodynamics, and comparison with visual assessment of lobar hypoperfusion by two expert readers were evaluated.ResultsGlobal HLV correctly separated patients with CTEPH from controls (area under the receiver operating characteristic curve = 0.84; 10% HLV cutoff: 90% sensitivity, 72% accuracy, and 64% specificity) and correlated moderately with hemodynamic severity at time of imaging (pulmonary vascular resistance [PVR], ρ = 0.67; P < .001) and change after surgical treatment (∆PVR, ρ = -0.61; P < .001). In patients surgically classified as having segmental disease, global HLV correlated with preoperative PVR (ρ = 0.81) and postoperative ∆PVR (ρ = -0.70). Lobar HLV correlated moderately with expert reader lobar assessment (ρHLV = 0.71 for reader 1; ρHLV = 0.67 for reader 2).ConclusionAutomated quantification of hypoperfused areas in patients with CTEPH can be performed from clinical multienergy CT examinations and may aid clinical evaluation, particularly in patients with segmental-level disease.Keywords: CT-Spectral Imaging (Multienergy), Pulmonary, Pulmonary Arteries, Embolism/Thrombosis, Chronic Thromboembolic Pulmonary Hypertension, Multienergy CT, Hypoperfusion© RSNA, 2023.

Background phase errors in 4D Flow MRI may negatively impact blood flow quantification. In this study, we assessed their impact on cerebrovascular flow volume measurements, evaluated the benefit of manual image-based correction, and assessed the potential of a convolutional neural network (CNN), a form of deep learning, to directly infer the correction vector field. With IRB waiver of informed consent, we retrospectively identified 96 MRI exams from 48 patients who underwent cerebrovascular 4D Flow MRI from October 2015 to 2020. Flow measurements of the anterior, posterior, and venous circulation were performed to assess inflow-outflow error and the benefit of manual image-based phase error correction. A CNN was then trained to directly infer the phase-error correction field, without segmentation, from 4D Flow volumes to automate correction, reserving from 23 exams for testing. Statistical analyses included Spearman correlation, Bland-Altman, Wilcoxon-signed rank (WSR) and F-tests. Prior to correction, there was strong correlation between inflow and outflow (ρ = 0.833-0.947) measurements with the largest discrepancy in the venous circulation. Manual phase error correction improved inflow-outflow correlation (ρ = 0.945-0.981) and decreased variance (p

PurposeTo assess the feasibility of a newly developed algorithm, called deep learning synthetic strain (DLSS), to infer myocardial velocity from cine steady-state free precession (SSFP) images and detect wall motion abnormalities in patients with ischemic heart disease.Materials and methodsIn this retrospective study, DLSS was developed by using a data set of 223 cardiac MRI examinations including cine SSFP images and four-dimensional flow velocity data (November 2017 to May 2021). To establish normal ranges, segmental strain was measured in 40 individuals (mean age, 41 years ± 17 [SD]; 30 men) without cardiac disease. Then, DLSS performance in the detection of wall motion abnormalities was assessed in a separate group of patients with coronary artery disease, and these findings were compared with consensus results of four independent cardiothoracic radiologists (ground truth). Algorithm performance was evaluated by using receiver operating characteristic curve analysis.ResultsMedian peak segmental radial strain in individuals with normal cardiac MRI findings was 38% (IQR: 30%-48%). Among patients with ischemic heart disease (846 segments in 53 patients; mean age, 61 years ± 12; 41 men), the Cohen κ among four cardiothoracic readers for detecting wall motion abnormalities was 0.60-0.78. DLSS achieved an area under the receiver operating characteristic curve of 0.90. Using a fixed 30% threshold for abnormal peak radial strain, the algorithm achieved a sensitivity, specificity, and accuracy of 86%, 85%, and 86%, respectively.ConclusionThe deep learning algorithm had comparable performance with subspecialty radiologists in inferring myocardial velocity from cine SSFP images and identifying myocardial wall motion abnormalities at rest in patients with ischemic heart disease.Keywords: Neural Networks, Cardiac, MR Imaging, Ischemia/Infarction Supplemental material is available for this article. © RSNA, 2023.

Cardiac shape modeling is a useful computational tool that has provided quantitative insights into the mechanisms underlying dysfunction in heart disease. The manual input and time required to make cardiac shape models, however, limits their clinical utility. Here we present an end-to-end pipeline that uses deep learning for automated view classification, slice selection, phase selection, anatomical landmark localization, and myocardial image segmentation for the automated generation of three-dimensional, biventricular shape models. With this approach, we aim to make cardiac shape modeling a more robust and broadly applicable tool that has processing times consistent with clinical workflows. Cardiovascular magnetic resonance (CMR) images from a cohort of 123 patients with repaired tetralogy of Fallot (rTOF) from two internal sites were used to train and validate each step in the automated pipeline. The complete automated pipeline was tested using CMR images from a cohort of 12 rTOF patients from an internal site and 18 rTOF patients from an external site. Manually and automatically generated shape models from the test set were compared using Euclidean projection distances, global ventricular measurements, and atlas-based shape mode scores. The mean absolute error (MAE) between manually and automatically generated shape models in the test set was similar to the voxel resolution of the original CMR images for end-diastolic models (MAE = 1.9 ± 0.5 mm) and end-systolic models (MAE = 2.1 ± 0.7 mm). Global ventricular measurements computed from automated models were in good agreement with those computed from manual models. The average mean absolute difference in shape mode Z-score between manually and automatically generated models was 0.5 standard deviations for the first 20 modes of a reference statistical shape atlas. Using deep learning, accurate three-dimensional, biventricular shape models can be reliably created. This fully automated end-to-end approach dramatically reduces the manual input required to create shape models, thereby enabling the rapid analysis of large-scale datasets and the potential to deploy statistical atlas-based analyses in point-of-care clinical settings. Training data and networks are available from cardiacatlas.org.

Management of chronic thromboembolic pulmonary hypertension (CTEPH) should be determined by a multidisciplinary team, ideally at a specialized CTEPH referral center. Radiologists contribute to this multidisciplinary process by helping to confirm the diagnosis of CTEPH and delineating the extent of disease, both of which help determine a treatment decision. Preoperative assessment of CTEPH usually employs multiple imaging modalities, including ventilation-perfusion (V/Q) scanning, echocardiography, CT pulmonary angiography (CTPA), and right heart catheterization with pulmonary angiography. Accurate diagnosis or exclusion of CTEPH at imaging is imperative, as this remains the only form of pulmonary hypertension that is curative with surgery. Unfortunately, CTEPH is often misdiagnosed at CTPA, which can be due to technical factors, patient-related factors, radiologist-related factors, as well as a host of disease mimics including acute pulmonary embolism, in situ thrombus, vasculitis, pulmonary artery sarcoma, and fibrosing mediastinitis. Although V/Q scanning is thought to be substantially more sensitive for CTEPH compared with CTPA, this is likely due to lack of recognition of CTEPH findings rather than a modality limitation. Preoperative evaluation for pulmonary thromboendarterectomy (PTE) includes assessment of technical operability and surgical risk stratification. While the definitive therapy for CTEPH is PTE, other minimally invasive or noninvasive therapies also lead to clinical improvements including greater survival. Complications of PTE that can be identified at postoperative imaging include infection, reperfusion edema or injury, pulmonary hemorrhage, pericardial effusion or hemopericardium, and rethrombosis. ©RSNA, 2022 Online supplemental material is available for this article.

Mitral valve prolapse (MVP) is the most common nonischemic mitral regurgitation etiology and mitral abnormality requiring surgery in the Western world. There is an increasing awareness that pathological findings in MVP are not confined to the valve tissue; rather, it is a complex disease, involving the mitral valve apparatus, cardiac hemodynamics, and cardiac structure. Imaging has played a fundamental role in the understanding of the diagnosis, prevalence, and consequences of MVP. The diagnosis of MVP by imaging is based upon demonstrating valve leaflets ascending into the left atrium through the saddle-shaped annulus. Transthoracic and transesophageal echocardiography are the primary modalities in the diagnosis and assessment of MVP patients and must include careful assessment of the leaflets, annulus, chords, and papillary muscles. High-spatial-resolution imaging modalities such as cardiac magnetic resonance images and cardiac computed tomography play a secondary role in this regard and can demonstrate the anatomical relation between the mitral valve annulus and leaflet excursion for appropriate diagnosis. Ongoing development of new methods of cardiac imaging can help us to accurately understand the mechanism, diagnose the disease, develop an appropriate treatment plan, and estimate the risk for sudden death. Recently, several new observations with respect to prolapse have been derived from cardiac imaging including three-dimensional echocardiography and tissue-Doppler imaging. The aim of this article is to present these new imaging-derived insights for the diagnosis, risk assessment, treatment, and follow-up of patients with MVP.

Supplemental material is available for this article.

Scoliosis is a condition of abnormal lateral spinal curvature affecting an estimated 2 to 3% of the US population, or seven million people. The Cobb angle is the standard measurement of spinal curvature in scoliosis but is known to have high interobserver and intraobserver variability. Thus, the objective of this study was to build and validate a system for automatic quantitative evaluation of the Cobb angle and to compare AI generated and human reports in the clinical setting. After IRB was obtained, we retrospectively collected 2150 frontal view scoliosis radiographs at a tertiary referral center (January 1, 2019, to January 1, 2021, ≥ 16 years old, no hardware). The dataset was partitioned into 1505 train (70%), 215 validation (10%), and 430 test images (20%). All thoracic and lumbar vertebral bodies were segmented with bounding boxes, generating approximately 36,550 object annotations that were used to train a Faster R-CNN Resnet-101 object detection model. A controller algorithm was written to localize vertebral centroid coordinates and derive the Cobb properties (angle and endplate) of dominant and secondary curves. AI-derived Cobb angle measurements were compared to the clinical report measurements, and the Spearman rank-order demonstrated significant correlation (0.89, p

BackgroundPostural orthostatic tachycardia syndrome (POTS), Ehlers-Danlos syndrome (EDS), and May-Thurner syndrome (MTS) are three syndromes that are often misdiagnosed or underdiagnosed. The true prevalence of these syndromes may be higher than currently reported. The following case series is the first to report a three-way association between POTS, EDS, and MTS.Case summaryWe describe three patients with concomitant POTS, EDS, and MTS. Although abdominopelvic vasculature evaluation can be difficult via conventional imaging techniques, we present the use of novel dynamic contrast-enhanced magnetic resonance angiography with Differential Subsampling with Cartesian Ordering (DISCO) and four-dimensional flow magnetic resonance imaging to aid vasculature evaluation and the diagnosis of MTS. Two patients underwent left common iliac vein stenting to treat MTS, experiencing significant improvement in their POTS symptoms and quality of life.DiscussionEhlers-Danlos syndrome, POTS, and MTS may interact synergistically to exacerbate symptoms. Patients with EDS should be evaluated for possible POTS and pelvic venous complications. Left common iliac vein stenting for MTS can mitigate POTS symptoms by decreasing lower extremity venous pooling and should be considered in this patient population. Further research is needed to understand the exact mechanism and intricacies of this syndrome triad.

PurposeTo assess the potential of a transfer learning strategy leveraging radiologist supervision to enhance convolutional neural network-based (CNN) localization of pneumonia on radiographs and to further assess the prognostic value of CNN severity quantification on patients evaluated for COVID-19 pneumonia, for whom severity on the presenting radiograph is a known predictor of mortality and intubation.Materials and methodsWe obtained an initial CNN previously trained to localize pneumonia along with 25,684 radiographs used for its training. We additionally curated 1466 radiographs from patients who had a computed tomography (CT) performed on the same day. Regional likelihoods of pneumonia were then annotated by cardiothoracic radiologists, referencing these CTs. Combining data, a preexisting CNN was fine-tuned using transfer learning. Whole-image and regional performance of the updated CNN was assessed using receiver-operating characteristic area under the curve and Dice. Finally, the value of CNN measurements was assessed with survival analysis on 203 patients with COVID-19 and compared against modified radiographic assessment of lung edema (mRALE) score.ResultsPneumonia detection area under the curve improved on both internal (0.756 to 0.841) and external (0.864 to 0.876) validation data. Dice overlap also improved, particularly in the lung bases (R: 0.121 to 0.433, L: 0.111 to 0.486). There was strong correlation between radiologist mRALE score and CNN fractional area of involvement (ρ=0.85). Survival analysis showed similar, strong prognostic ability of the CNN and mRALE for mortality, likelihood of intubation, and duration of hospitalization among patients with COVID-19.ConclusionsRadiologist-supervised transfer learning can enhance the ability of CNNs to localize and quantify the severity of disease. Closed-loop systems incorporating radiologists may be beneficial for continued improvement of artificial intelligence algorithms.

Quantitative imaging measurements can be facilitated by artificial intelligence (AI) algorithms, but how they might impact decision-making and be perceived by radiologists remains uncertain. After creation of a dedicated inspiratory-expiratory CT examination and concurrent deployment of a quantitative AI algorithm for assessing air trapping, five cardiothoracic radiologists retrospectively evaluated severity of air trapping on 17 examination studies. Air trapping severity of each lobe was evaluated in three stages: qualitatively (visually); semiquantitatively, allowing manual region-of-interest measurements; and quantitatively, using results from an AI algorithm. Readers were surveyed on each case for their perceptions of the AI algorithm. The algorithm improved interreader agreement (intraclass correlation coefficients: visual, 0.28; semiquantitative, 0.40; quantitative, 0.84; P < .001) and improved correlation with pulmonary function testing (forced expiratory volume in 1 second-to-forced vital capacity ratio) (visual r = -0.26, semiquantitative r = -0.32, quantitative r = -0.44). Readers perceived moderate agreement with the AI algorithm (Likert scale average, 3.7 of 5), a mild impact on their final assessment (average, 2.6), and a neutral perception of overall utility (average, 3.5). Though the AI algorithm objectively improved interreader consistency and correlation with pulmonary function testing, individual readers did not immediately perceive this benefit, revealing a potential barrier to clinical adoption. Keywords: Technology Assessment, Quantification © RSNA, 2021.

CT pulmonary angiography (CTPA) is the first-line imaging test for evaluation of acute pulmonary emboli. However, diagnostic quality is heterogeneous across institutions and is frequently limited by suboptimal pulmonary artery (PA) contrast enhancement. In this retrospective study, a deep learning algorithm for measuring enhancement of the central PAs was developed and assessed for feasibility of its use in quality improvement of CTPA. In a convenience sample of 450 patients, automated measurement of CTPA enhancement showed high agreement with manual radiologist measurement (r = 0.996). Using a threshold of less than 250 HU for suboptimal enhancement, the sensitivity and specificity of the automated classification were 100% and 99.5%, respectively. The algorithm was further evaluated in a random sampling of 3195 CTPA examinations from January 2019 through May 2021. Beginning in January 2021, the scanning protocol was transitioned from bolus tracking to a timing bolus strategy. Automated analysis of these examinations showed that most suboptimal examinations following the change in protocol were performed using one scanner, highlighting the potential value of deep learning algorithms for quality improvement in the radiology department. Keywords: CT Angiography, Pulmonary Arteries © RSNA, 2022.

Background Four-dimensional (4D) flow MRI has the potential to provide hemodynamic insights for a variety of abdominopelvic vascular diseases, but its clinical utility is currently impaired by background phase error, which can be challenging to correct. Purpose To assess the feasibility of using deep learning to automatically perform image-based background phase error correction in 4D flow MRI and to compare its effectiveness relative to manual image-based correction. Materials and Methods A convenience sample of 139 abdominopelvic 4D flow MRI acquisitions performed between January 2016 and July 2020 was retrospectively collected. Manual phase error correction was performed using dedicated imaging software and served as the reference standard. After reserving 40 examinations for testing, the remaining examinations were randomly divided into training (86% [85 of 99]) and validation (14% [14 of 99]) data sets to train a multichannel three-dimensional U-Net convolutional neural network. Flow measurements were obtained for the infrarenal aorta, common iliac arteries, common iliac veins, and inferior vena cava. Statistical analyses included Pearson correlation, Bland-Altman analysis, and F tests with Bonferroni correction. Results A total of 139 patients (mean age, 47 years ± 14 [standard deviation]; 108 women) were included. Inflow-outflow correlation improved after manual correction (ρ = 0.94, P < .001) compared with that before correction (ρ = 0.50, P < .001). Automated correction showed similar results (ρ = 0.91, P < .001) and demonstrated very strong correlation with manual correction (ρ = 0.98, P < .001). Both correction methods reduced inflow-outflow variance, improving mean difference from -0.14 L/min (95% limits of agreement: -1.61, 1.32) (uncorrected) to 0.05 L/min (95% limits of agreement: -0.32, 0.42) (manually corrected) and 0.05 L/min (95% limits of agreement: -0.38, 0.49) (automatically corrected). There was no significant difference in inflow-outflow variance between manual and automated correction methods (P = .10). Conclusion Deep learning automated phase error correction reduced inflow-outflow bias and variance of volumetric flow measurements in four-dimensional flow MRI, achieving results comparable with manual image-based phase error correction. © RSNA, 2021 See also the editorial by Roldán-Alzate and Grist in this issue.

Understanding of coronary sinus (CS) anatomy and abnormalities is of critical importance due to their use in interventional procedures. Herein, the authors report a rare case of an asymptomatic 72-year-old man with a left circumflex coronary artery-to-CS fistula, together with CS ostial atresia and persistent left superior vena cava. These findings are described using both cardiac CT angiography and MRI with four-dimensional flow for anatomic and functional assessment. Keywords: Cardiac, Coronary Sinus, Aneurysms, Fistula, CT Angiography, MR Imaging Supplemental material is available for this article. © RSNA, 2022.

PurposeTo develop a convolutional neural network (CNN)-based deformable lung registration algorithm to reduce computation time and assess its potential for lobar air trapping quantification.Materials and methodsIn this retrospective study, a CNN algorithm was developed to perform deformable registration of lung CT (LungReg) using data on 9118 patients from the COPDGene Study (data collected between 2007 and 2012). Loss function constraints included cross-correlation, displacement field regularization, lobar segmentation overlap, and the Jacobian determinant. LungReg was compared with a standard diffeomorphic registration (SyN) for lobar Dice overlap, percentage voxels with nonpositive Jacobian determinants, and inference runtime using paired t tests. Landmark colocalization error (LCE) across 10 patients was compared using a random effects model. Agreement between LungReg and SyN air trapping measurements was assessed using intraclass correlation coefficient. The ability of LungReg versus SyN emphysema and air trapping measurements to predict Global Initiative for Chronic Obstructive Lung Disease (GOLD) stages was compared using area under the receiver operating characteristic curves.ResultsAverage performance of LungReg versus SyN showed lobar Dice overlap score of 0.91-0.97 versus 0.89-0.95, respectively (P < .001); percentage voxels with nonpositive Jacobian determinant of 0.04 versus 0.10, respectively (P < .001); inference run time of 0.99 second (graphics processing unit) and 2.27 seconds (central processing unit) versus 418.46 seconds (central processing unit) (P < .001); and LCE of 7.21 mm versus 6.93 mm (P < .001). LungReg and SyN whole-lung and lobar air trapping measurements achieved excellent agreement (intraclass correlation coefficients > 0.98). LungReg versus SyN area under the receiver operating characteristic curves for predicting GOLD stage were not statistically different (range, 0.88-0.95 vs 0.88-0.95, respectively; P = .31-.95).ConclusionCNN-based deformable lung registration is accurate and fully automated, with runtime feasible for clinical lobar air trapping quantification, and has potential to improve diagnosis of small airway diseases.Keywords: Air Trapping, Convolutional Neural Network, Deformable Registration, Small Airway Disease, CT, Lung, Semisupervised Learning, Unsupervised Learning Supplemental material is available for this article. © RSNA, 2021 An earlier incorrect version of this article appeared online. This article was corrected on December 22, 2021.

Stereotactic radiosurgery planning for cerebral arteriovenous malformations (AVM) is complicated by the variability in appearance of an AVM nidus across different imaging modalities. We developed a deep learning approach to automatically segment cerebrovascular-anatomical maps from multiple high-resolution magnetic resonance imaging/angiography (MRI/MRA) sequences in AVM patients, with the goal of facilitating target delineation. Twenty-three AVM patients who were evaluated for radiosurgery and underwent multi-parametric MRI/MRA were included. A hybrid semi-automated and manual approach was used to label MRI/MRAs with arteries, veins, brain parenchyma, cerebral spinal fluid (CSF), and embolized vessels. Next, these labels were used to train a convolutional neural network to perform this task. Imaging from 17 patients (6362 image slices) was used for training, and 6 patients (1224 slices) for validation. Performance was evaluated by Dice Similarity Coefficient (DSC). Classification performance was good for arteries, veins, brain parenchyma, and CSF, with DSCs of 0.86, 0.91, 0.98, and 0.91, respectively in the validation image set. Performance was lower for embolized vessels, with a DSC of 0.75. This demonstrates the proof of principle that accurate, high-resolution cerebrovascular-anatomical maps can be generated from multiparametric MRI/MRA. Clinical validation of their utility in radiosurgery planning is warranted.

Stereotactic radiosurgery (SRS) is used to treat cerebral arteriovenous malformations (AVMs). However, early evaluation of efficacy is difficult as structural magnetic resonance imaging (MRI)/magnetic resonance angiography (MRA) often does not demonstrate appreciable changes within the first 6 months. The aim of this study was to evaluate the use of four-dimensional (4D) flow MRI to quantify hemodynamic changes after SRS as early as 2 months. This was a retrospective observational study, which included 14 patients with both pre-SRS and post-SRS imaging obtained at multiple time points from 1 to 27 months after SRS. A 3T MRI Scanner was used to obtain T2 single-shot fast spin echo, time-of-flight MRA, and postcontrast 4D flow with three-dimensional velocity encoding between 150 and 200 cm/s. Post-hoc two-dimensional cross-sectional flow was measured for the dominant feeding artery, the draining vein, and the corresponding contralateral artery as a control. Measurements were performed by two independent observers, and reproducibility was assessed. Wilcoxon signed-rank tests were used to compare differences in flow, circumference, and pulsatility between the feeding artery and the contralateral artery both before and after SRS; and differences in nidus size and flow and circumference of the feeding artery and draining vein before and after SRS. Arterial flow (L/min) decreased in the primary feeding artery (mean: 0.1 ± 0.07 vs. 0.3 ± 0.2; p

BackgroundNon-invasive assessment of the hemodynamic changes of cirrhosis might help guide management of patients with liver disease but are currently limited.PurposeTo determine whether free-breathing 4D flow MRI can be used to quantify the hemodynamic effects of cirrhosis and introduce hydraulic circuit indexes of severity.Study typeRetrospective.PopulationForty-seven patients including 26 with cirrhosis.Field strength/sequence3 T/free-breathing 4D flow MRI with soft gating and golden-angle view ordering.AssessmentMeasurements of the supra-celiac abdominal aorta, supra-renal abdominal aorta (SRA), celiac trunk (CeT), superior mesenteric artery (SMA), splenic artery (SpA), common hepatic artery (CHA), portal vein (PV), and supra-renal inferior vena cava (IVC) were made by two radiologists. Measures of hepatic vascular resistance (hepatic arterial relative resistance [HARR]; portal resistive index [PRI]) were proposed and calculated.Statistical analysisBland-Altman, Pearson's correlation, Tukey's multiple comparison, and Cohen's kappa. P

PurposeTo compare invasive right heart catheterization with four-dimensional (4D) flow MRI for estimating shunt fraction in patients with intracardiac and extracardiac shunts.Materials and methodsIn this retrospective study, patients who underwent 4D flow MRI and invasive right heart catheterization with a shunt run between August 2015 and November 2018 were included. The primary objective was comparison of estimated shunt fraction (ratio of pulmonary-to-systemic flow, Qp/Qs) at 4D flow and catheterization. Secondary objectives included comparison of the right ventricular-to-left ventricular stroke volume ratio (RVSV/LVSV) to shunt fraction (for those with applicable shunts) and comparison of cardiac output between 4D flow and catheterization. Statistical analysis included Pearson correlation and Bland-Altman plots.ResultsA total of 33 patients met inclusion criteria (mean age, 49 years ± 16 [standard deviation]; 24 women). 4D flow measurements of Qp/Qs strongly correlated with those at catheterization (r = 0.938), and there was no bias. RVSV/LVSV correlated strongly with Qp/Qs from 4D flow (r = 0.852) and catheterization (r = 0.842). Measurements of left ventricle (Qs) and right ventricle (QP) cardiac output from 4D flow and catheterization (Fick) correlated moderately overall (r = 0.673 [Qp] and r = 0.750 [Qs]).ConclusionShunt fraction measurement using 4D flow MRI compares well with that using invasive cardiac catheterization.Supplemental material is available for this article.© RSNA, 2021.

PurposeTo develop a deep learning-based algorithm to stage the severity of chronic obstructive pulmonary disease (COPD) through quantification of emphysema and air trapping on CT images and to assess the ability of the proposed stages to prognosticate 5-year progression and mortality.Materials and methodsIn this retrospective study, an algorithm using co-registration and lung segmentation was developed in-house to automate quantification of emphysema and air trapping from inspiratory and expiratory CT images. The algorithm was then tested in a separate group of 8951 patients from the COPD Genetic Epidemiology study (date range, 2007-2017). With measurements of emphysema and air trapping, bivariable thresholds were determined to define CT stages of severity (mild, moderate, severe, and very severe) and were evaluated for their ability to prognosticate disease progression and mortality using logistic regression and Cox regression.ResultsOn the basis of CT stages, the odds of disease progression were greatest among patients with very severe disease (odds ratio [OR], 2.67; 95% CI: 2.02, 3.53; P < .001) and were elevated in patients with moderate disease (OR, 1.50; 95% CI: 1.22, 1.84; P = .001). The hazard ratio of mortality for very severe disease at CT was 2.23 times the normal ratio (95% CI: 1.93, 2.58; P < .001). When combined with Global Initiative for Chronic Obstructive Lung Disease (GOLD) staging, patients with GOLD stage 2 disease had the greatest odds of disease progression when the CT stage was severe (OR, 4.48; 95% CI: 3.18, 6.31; P < .001) or very severe (OR, 4.72; 95% CI: 3.13, 7.13; P < .001).ConclusionAutomated CT algorithms can facilitate staging of COPD severity, have diagnostic performance comparable with that of spirometric GOLD staging, and provide further prognostic value when used in conjunction with GOLD staging.Supplemental material is available for this article.© RSNA, 2021See also commentary by Kalra and Ebrahimian in this issue.

Imaging is key to nearly all aspects of chronic thromboembolic pulmonary hypertension including management for screening, assessing eligibility for pulmonary endarterectomy, and post-operative follow-up. While ventilation/perfusion scintigraphy, the gold standard technique for chronic thromboembolic pulmonary hypertension screening, can have excellent sensitivity, it can be confounded by other etiologies of pulmonary malperfusion, and does not provide structural information to guide operability assessment. Conventional computed tomography pulmonary angiography has high specificity, though findings of chronic thromboembolic pulmonary hypertension can be visually subtle and unrecognized. In addition, computed tomography pulmonary angiography can provide morphologic information to aid in pre-operative workup and assessment of other structural abnormalities. Advances in computed tomography imaging techniques, including dual-energy computed tomography and spectral-detector computed tomography, allow for improved sensitivity and specificity in detecting chronic thromboembolic pulmonary hypertension, comparable to that of ventilation/perfusion scans. Furthermore, these advanced computed tomography techniques, compared with conventional computed tomography, provide additional physiologic data from perfused blood volume maps and improved resolution to better visualize distal chronic thromboembolic pulmonary hypertension, an important consideration for balloon pulmonary angioplasty for inoperable patients. Electrocardiogram-synchronized techniques in electrocardiogram-gated computed tomography can also show further information regarding right ventricular function and structure. While the standard of care in the workup of chronic thromboembolic pulmonary hypertension includes a ventilation/perfusion scan, computed tomography pulmonary angiography, direct catheter angiography, echocardiogram, and coronary angiogram, in the future an electrocardiogram-gated dual-energy computed tomography angiography scan may enable a "one-stop" imaging study to guide diagnosis, operability assessment, and treatment decisions with less radiation exposure and cost than traditional chronic thromboembolic pulmonary hypertension imaging modalities.

ObjectiveThe coronavirus disease 2019 pandemic has inspired new innovations in diagnosing, treating, and dispositioning patients during high census conditions with constrained resources. Our objective is to describe first experiences of physician interaction with a novel artificial intelligence (AI) algorithm designed to enhance physician abilities to identify ground-glass opacities and consolidation on chest radiographs.MethodsDuring the first wave of the pandemic, we deployed a previously developed and validated deep-learning AI algorithm for assisted interpretation of chest radiographs for use by physicians at an academic health system in Southern California. The algorithm overlays radiographs with "heat" maps that indicate pneumonia probability alongside standard chest radiographs at the point of care. Physicians were surveyed in real time regarding ease of use and impact on clinical decisionmaking.ResultsOf the 5125 total visits and 1960 chest radiographs obtained in the emergency department (ED) during the study period, 1855 were analyzed by the algorithm. Among these, emergency physicians were surveyed for their experiences on 202 radiographs. Overall, 86% either strongly agreed or somewhat agreed that the intervention was easy to use in their workflow. Of the respondents, 20% reported that the algorithm impacted clinical decisionmaking.ConclusionsTo our knowledge, this is the first published literature evaluating the impact of medical imaging AI on clinical decisionmaking in the emergency department setting. Urgent deployment of a previously validated AI algorithm clinically was easy to use and was found to have an impact on clinical decision making during the predicted surge period of a global pandemic.

PurposePneumonia is a common clinical diagnosis for which chest radiographs are often an important part of the diagnostic workup. Deep learning has the potential to expedite and improve the clinical interpretation of chest radiographs. While earlier approaches have emphasized the feasibility of "binary classification" to accomplish this task, alternative strategies may be possible. We explore the feasibility of a "semantic segmentation" deep learning approach to highlight the potential foci of pneumonia on frontal chest radiographs.Materials and methodsIn this retrospective study, we trained a U-net convolutional neural network (CNN) to predict pixel-wise probability maps for pneumonia using a public data set provided by the Radiological Society of North America (RSNA) comprised of 22,000 radiographs and radiologist-defined bounding boxes. We reserved 3684 radiographs as an independent validation data set and assessed overall performance for localization using Dice overlap and classification performance using the area under the receiver-operator characteristic curve.ResultsFor classification/detection of pneumonia, area under the receiver-operator characteristic curve on frontal radiographs was 0.854 with a sensitivity of 82.8% and specificity of 72.6%. Using this strategy of neural network training, probability maps localized pneumonia to lung parenchyma for essentially all validation cases. For segmentation of pneumonia for positive cases, predicted probability maps had a mean Dice score (±SD) of 0.603±0.204, and 60.0% of these had a Dice score >0.5.ConclusionsA "semantic segmentation" deep learning approach can provide a probabilistic map to assist in the diagnosis of pneumonia. In combination with the patient's history, clinical findings and other imaging, this strategy may help expedite and improve diagnosis.

PurposeTo evaluate the performance of a deep learning (DL) algorithm for clinical measurement of right and left ventricular volume and function across cardiac MR images obtained for a range of clinical indications and pathologies.Materials and methodsA retrospective, Health Insurance Portability and Accountability Act-compliant study was conducted using the first 200 noncongenital clinical cardiac MRI examinations from June 2015 to June 2017 for which volumetry was available. Images were analyzed using commercially available software for automated DL-based and manual contouring of biventricular volumes. Fully automated measurements were compared using Pearson correlations, relative volume errors, and Bland-Altman analyses. Manual, automated, and expert revised contours for 50 MR images were examined by comparing regional Dice coefficients at the base, midventricle, and apex to further analyze the contour quality.ResultsFully automated and manual left ventricular volumes were strongly correlated for end-systolic volume (ESV: Pearson r = 0.99, P < .001), end-diastolic volume (EDV: r = 0.97, P < .001), and ejection fraction (EF: r = 0.94, P < .001). Right ventricular measurements were also correlated for ESV (r = 0.93, P < .001), EDV (r = 0.92, P < .001), and EF (r = 0.73, P < .001). Visual inspection of segmentation quality showed most errors (73%) occurred at the cardiac base. Mean Dice coefficients between manual, automated, and expert revised contours ranged from 0.92 to 0.95, with greatest variance at the base and apex.ConclusionFully automated ventricular segmentation by the tested algorithm provides contours and ventricular volumes that could be used to aid expert segmentation, but can benefit from expert supervision, particularly to resolve errors at the basal and apical slices. Supplemental material is available for this article. © RSNA, 2020.

Background Cardiac MRI is limited by long acquisition times, yet faster acquisition of smaller-matrix images reduces spatial detail. Deep learning (DL) might enable both faster acquisition and higher spatial detail via super-resolution. Purpose To explore the feasibility of using DL to enhance spatial detail from small-matrix MRI acquisitions and evaluate its performance against that of conventional image upscaling methods. Materials and Methods Short-axis cine cardiac MRI examinations performed between January 2012 and December 2018 at one institution were retrospectively collected for algorithm development and testing. Convolutional neural networks (CNNs), a form of DL, were trained to perform super resolution in image space by using synthetically generated low-resolution data. There were 70%, 20%, and 10% of examinations allocated to training, validation, and test sets, respectively. CNNs were compared against bicubic interpolation and Fourier-based zero padding by calculating the structural similarity index (SSIM) between high-resolution ground truth and each upscaling method. Means and standard deviations of the SSIM were reported, and statistical significance was determined by using the Wilcoxon signed-rank test. For evaluation of clinical performance, left ventricular volumes were measured, and statistical significance was determined by using the paired Student t test. Results For CNN training and retrospective analysis, 400 MRI scans from 367 patients (mean age, 48 years ± 18; 214 men) were included. All CNNs outperformed zero padding and bicubic interpolation at upsampling factors from two to 64 (P < .001). CNNs outperformed zero padding on more than 99.2% of slices (9828 of 9907). In addition, 10 patients (mean age, 51 years ± 22; seven men) were prospectively recruited for super-resolution MRI. Super-resolved low-resolution images yielded left ventricular volumes comparable to those from full-resolution images (P > .05), and super-resolved full-resolution images appeared to further enhance anatomic detail. Conclusion Deep learning outperformed conventional upscaling methods and recovered high-frequency spatial information. Although training was performed only on short-axis cardiac MRI examinations, the proposed strategy appeared to improve quality in other imaging planes. © RSNA, 2020 Online supplemental material is available for this article.

PurposeTo develop and evaluate the performance of a fully-automated convolutional neural network (CNN)-based algorithm to evaluate hepatobiliary phase (HBP) adequacy of gadoxetate disodium (EOB)-enhanced MRI. Secondarily, we explored the potential of the proposed CNN algorithm to reduce examination length by applying it to EOB-MRI examinations.MethodsWe retrospectively identified EOB-enhanced MRI-HBP series from examinations performed 2011-2018 (internal and external datasets). Our algorithm, comprising a liver segmentation and classification CNN, produces an adequacy score. Two abdominal radiologists independently classified series as adequate or suboptimal. The consensus determination of HBP adequacy was used as ground truth for CNN model training and validation. Reader agreement was evaluated with Cohen's kappa. Performance of the algorithm was assessed by receiver operating characteristics (ROC) analysis and computation of the area under the ROC curve (AUC). Potential examination duration reduction was evaluated descriptively.Results1408 HBP series from 484 patients were included. Reader kappa agreement was 0.67 (internal dataset) and 0.80 (external dataset). AUCs were 0.97 (0.96-0.99) for internal and 0.95 (0.92-96) for external and were not significantly different from each other (p = 0.24). 48 % (50/105) examinations could have been shorter by applying the algorithm.ConclusionA proposed CNN-based algorithm achieves higher than 95 % AUC for classifying HBP images as adequate versus suboptimal. The application of this algorithm could potentially shorten examination time and aid radiologists in recognizing technically suboptimal images, avoiding diagnostic pitfalls.

PurposeTo evaluate whether automated vorticity mapping four-dimensional (4D) flow MRI can identify regions of quantitative flow inconsistency.Materials and methodsIn this retrospective study, 35 consecutive patients who underwent MR angiography with 4D flow MRI at 3.0 T from December 2017 to October 2018 were analyzed using a λ 2-based technique for vorticity visualization and quantification. The patients were aged 58.6 years ± 14.4 (standard deviation), 12 were women, 18 had ascending aortic aneurysms (maximal diameter > 4.0 cm), and 10 had bicuspid aortic valves. Flow measurements were made in the ascending aorta (aAo), mid-descending aorta, main pulmonary artery, and superior vena cava. Statistical tests included t tests and F tests with a type I error threshold (α) of .05.ResultsThe 35 patients were visually classified as having no (n = 9), mild (n = 8), moderate (n = 11), or severe vorticity (n = 7). Across all patients, standard deviation of cardiac output in the aAo (0.58 L/min ± 0.45) was significantly (P < .001) higher than in the pulmonary arteries (0.15 L/min ± 0.10) and descending aorta and superior vena cava (0.14 L/min ± 0.12). The standard deviation of cardiac output observed in the aAo was significantly greater (P < .01) in patients with moderate or severe vorticity (0.73 L/min ± 0.55) than in those with none or mild vorticity (0.44 L/min ± 0.26).ConclusionCardiac output and blood flow are essential MRI measurements in the evaluation of structural heart disease. Vorticity visualization may be used to help guide optimal location for flow quantification.© RSNA, 2020See also the commentary by Markl in this issue.

We present a case of a 20-year-old man with tetralogy of Fallot status post-repair with a transannular patch and left pulmonary artery reconstruction at age 1 and subsequent pulmonary valve replacement with bioprosthetic valve 3 years ago. A transthoracic echocardiogram revealed severe bioprosthetic valve insufficiency. Cardiac magnetic resonance with 4-dimensional flow revealed a paravalvular leak. (Level of Difficulty: Intermediate.).

BackgroundLiver alignment between series/exams is challenged by dynamic morphology or variability in patient positioning or motion. Image registration can improve image interpretation and lesion co-localization. We assessed the performance of a convolutional neural network algorithm to register cross-sectional liver imaging series and compared its performance to manual image registration.MethodsThree hundred fourteen patients, including internal and external datasets, who underwent gadoxetate disodium-enhanced magnetic resonance imaging for clinical care from 2011 to 2018, were retrospectively selected. Automated registration was applied to all 2,663 within-patient series pairs derived from these datasets. Additionally, 100 within-patient series pairs from the internal dataset were independently manually registered by expert readers. Liver overlap, image correlation, and intra-observation distances for manual versus automated registrations were compared using paired t tests. Influence of patient demographics, imaging characteristics, and liver uptake function was evaluated using univariate and multivariate mixed models.ResultsCompared to the manual, automated registration produced significantly lower intra-observation distance (p < 0.001) and higher liver overlap and image correlation (p < 0.001). Intra-exam automated registration achieved 0.88 mean liver overlap and 0.44 mean image correlation for the internal dataset and 0.91 and 0.41, respectively, for the external dataset. For inter-exam registration, mean overlap was 0.81 and image correlation 0.41. Older age, female sex, greater inter-series time interval, differing uptake, and greater voxel size differences independently reduced automated registration performance (p ≤ 0.020).ConclusionA fully automated algorithm accurately registered the liver within and between examinations, yielding better liver and focal observation co-localization compared to manual registration.

PurposeTo develop and evaluate a system to prescribe imaging planes for cardiac MRI based on deep learning (DL)-based localization of key anatomic landmarks.Materials and methodsAnnotated landmarks on 892 long-axis (LAX) and 493 short-axis (SAX) cine steady-state free precession series from cardiac MR images were retrospectively collected between February 2012 and June 2017. U-Net-based heatmap regression was used for localization of cardiac landmarks, which were used to compute cardiac MRI planes. Performance was evaluated by comparing localization distances and plane angle differences between DL predictions and ground truth. The plane angulations from DL were compared with those prescribed by the technologist at the original time of acquisition. Data were split into 80% for training and 20% for testing, and results confirmed with fivefold cross-validation.ResultsOn LAX images, DL localized the apex within mean 12.56 mm ± 19.11 (standard deviation) and the mitral valve (MV) within 7.68 mm ± 6.91. On SAX images, DL localized the aortic valve within 5.78 mm ± 5.68, MV within 5.90 mm ± 5.24, pulmonary valve within 6.55 mm ± 6.39, and tricuspid valve within 6.39 mm ± 5.89. On the basis of these localizations, average angle bias and mean error of DL-predicted imaging planes relative to ground truth annotations were as follows: SAX, -1.27° ± 6.81 and 4.93° ± 4.86; four chambers, 0.38° ± 6.45 and 5.16° ± 3.80; three chambers, 0.13° ± 12.70 and 9.02° ± 8.83; and two chamber, 0.25° ± 9.08 and 6.53° ± 6.28, respectively.ConclusionDL-based anatomic localization is a feasible strategy for planning cardiac MRI planes. This approach can produce imaging planes comparable to those defined by ground truth landmarks.© RSNA, 2019 Supplemental material is available for this article.

BackgroundChronic obstructive pulmonary disease (COPD) remains a major cause of morbidity and mortality. Present-day diagnostic criteria are largely based solely on spirometric criteria. Accumulating evidence has identified a substantial number of individuals without spirometric evidence of COPD who suffer from respiratory symptoms and/or increased morbidity and mortality. There is a clear need for an expanded definition of COPD that is linked to physiologic, structural (computed tomography [CT]) and clinical evidence of disease. Using data from the COPD Genetic Epidemiology study (COPDGene®), we hypothesized that an integrated approach that includes environmental exposure, clinical symptoms, chest CT imaging and spirometry better defines disease and captures the likelihood of progression of respiratory obstruction and mortality.MethodsFour key disease characteristics - environmental exposure (cigarette smoking), clinical symptoms (dyspnea and/or chronic bronchitis), chest CT imaging abnormalities (emphysema, gas trapping and/or airway wall thickening), and abnormal spirometry - were evaluated in a group of 8784 current and former smokers who were participants in COPDGene® Phase 1. Using these 4 disease characteristics, 8 categories of participants were identified and evaluated for odds of spirometric disease progression (FEV1 > 350 ml loss over 5 years), and the hazard ratio for all-cause mortality was examined.ResultsUsing smokers without symptoms, CT imaging abnormalities or airflow obstruction as the reference population, individuals were classified as Possible COPD, Probable COPD and Definite COPD. Current Global initiative for obstructive Lung Disease (GOLD) criteria would diagnose 4062 (46%) of the 8784 study participants with COPD. The proposed COPDGene® 2019 diagnostic criteria would add an additional 3144 participants. Under the new criteria, 82% of the 8784 study participants would be diagnosed with Possible, Probable or Definite COPD. These COPD groups showed increased risk of disease progression and mortality. Mortality increased in patients as the number of their COPD characteristics increased, with a maximum hazard ratio for all cause-mortality of 5.18 (95% confidence interval [CI]: 4.15-6.48) in those with all 4 disease characteristics.ConclusionsA substantial portion of smokers with respiratory symptoms and imaging abnormalities do not manifest spirometric obstruction as defined by population normals. These individuals are at significant risk of death and spirometric disease progression. We propose to redefine the diagnosis of COPD through an integrated approach using environmental exposure, clinical symptoms, CT imaging and spirometric criteria. These expanded criteria offer the potential to stimulate both current and future interventions that could slow or halt disease progression in patients before disability or irreversible lung structural changes develop.

Small-bowel obstruction (SBO) is a common and important disease, for which machine learning tools have yet to be developed. Image annotation is a critical first step for development of such tools. This study assesses whether image annotation by eye tracking is sufficiently accurate and precise to serve as a first step in the development of machine learning tools for detection of SBO on CT. Seven subjects diagnosed with SBO by CT were included in the study. For each subject, an obstructed segment of bowel was chosen. Three observers annotated the centerline of the segment by manual fiducial placement and by visual fiducial placement using a Tobii 4c eye tracker. Each annotation was repeated three times. The distance between centerlines was calculated after alignment using dynamic time warping (DTW) and statistically compared to clinical thresholds for diagnosis of SBO. Intra-observer DTW distance between manual and visual centerlines was calculated as a measure of accuracy. These distances were 1.1 ± 0.2, 1.3 ± 0.4, and 1.8 ± 0.2 cm for the three observers and were less than 1.5 cm for two of three observers (P

Advances in technology have always had the potential and opportunity to shape the practice of medicine, and in no medical specialty has technology been more rapidly embraced and adopted than radiology. Machine learning and deep neural networks promise to transform the practice of medicine, and, in particular, the practice of diagnostic radiology. These technologies are evolving at a rapid pace due to innovations in computational hardware and novel neural network architectures. Several cutting-edge postprocessing analysis applications are actively being developed in the fields of thoracic and cardiovascular imaging, including applications for lesion detection and characterization, lung parenchymal characterization, coronary artery assessment, cardiac volumetry and function, and anatomic localization. Cardiothoracic and cardiovascular imaging lies at the technological forefront of radiology due to a confluence of technical advances. Enhanced equipment has enabled computed tomography and magnetic resonance imaging scanners that can safely capture images that freeze the motion of the heart to exquisitely delineate fine anatomic structures. Computing hardware developments have enabled an explosion in computational capabilities and in data storage. Progress in software and fluid mechanical models is enabling complex 3D and 4D reconstructions to not only visualize and assess the dynamic motion of the heart, but also quantify its blood flow and hemodynamics. And now, innovations in machine learning, particularly in the form of deep neural networks, are enabling us to leverage the increasingly massive data repositories that are prevalent in the field. Here, we discuss developments in machine learning techniques and deep neural networks to highlight their likely role in future radiologic practice, both in and outside of image interpretation and analysis. We discuss the concepts of validation, generalizability, and clinical utility, as they pertain to this and other new technologies, and we reflect upon the opportunities and challenges of bringing these into daily use.

PurposeDelayed enhancement imaging is an essential component of cardiac MRI, which is used widely for the evaluation of myocardial scar and viability. The selection of an optimal inversion time (TI) or null point (TINP ) to suppress the background myocardial signal is required. The purpose of this study was to assess the feasibility of automated selection of TINP using a convolutional neural network (CNN). We hypothesized that a CNN may use spatial and temporal imaging characteristics from an inversion-recovery scout to select TINP , without the aid of a human observer.MethodsWe retrospectively collected 425 clinically acquired cardiac MRI exams performed at 1.5 T that included inversion-recovery scout acquisitions. We developed a VGG19 classifier ensembled with long short-term memory to identify the TINP . We compared the performance of the ensemble CNN in predicting TINP against ground truth, using linear regression analysis. Ground truth was defined as the expert physician annotation of the optimal TI. In a backtrack approach, saliency maps were generated to interpret the classification outcome and to increase the model's transparency.ResultsPrediction of TINP from our ensemble VGG19 long short-term memory closely matched with expert annotation (ρ = 0.88). Ninety-four percent of the predicted TINP were within ±36 ms, and 83% were at or after expert TI selection.ConclusionIn this study, we show that a CNN is capable of automated prediction of myocardial TI from an inversion-recovery experiment. Merging the spatial and temporal characteristics of the VGG-19 and long short-term-memory CNN structures appears to be sufficient to predict myocardial TI from TI scout.

PurposeWith the hypothesis that 4D flow can be used in evaluation of cardiac shunts, we seek to evaluate the multilevel and interreader reproducibility of measurements of the blood flow, shunt fraction and shunt volume in patients with atrial septum defect (ASD) in practice at multiple clinical sites.Materials and methodsFour-dimensional flow MRI examinations were performed at four institutions across Europe and the US. Twenty-nine patients (mean age, 43 years; 11 male) were included in the study. Flow measurements were performed at three levels (valve, main artery and periphery) in both the pulmonary and systemic circulation by two independent readers and compared against stroke volumes from 4D flow anatomic data. Further, the shunt ratio (Qp/Qs) was calculated. Additionally, shunt volume was quantified at the atrial level by tracking the atrial septum.ResultsMeasurements of the pulmonary blood flow at multiple levels correlate well whether measuring at the valve, main pulmonary artery or branch pulmonary arteries (r = 0.885-0.886). Measurements of the systemic blood flow show excellent correlation, whether measuring at the valve, ascending aorta or sum of flow from the superior vena cava (SVC) and descending aorta (r = 0.974-0.991). Intraclass agreement between the two observers for the flow measurements varies between 0.96 and 0.99. Compared with stroke volume, pulmonic flow is underestimated with 0.26 l/min at the main pulmonary artery level, and systemic flow is overestimated with 0.16 l/min at the ascending aorta level. Direct measurements of ASD flow are feasible in 20 of 29 (69%) patients.ConclusionBlood flow and shunt quantification measured at multiple levels and performed by different readers are reproducible and consistent with 4D flow MRI.

PurposeTo assess feasibility of training a convolutional neural network (CNN) to automate liver segmentation across different imaging modalities and techniques used in clinical practice and apply this to enable automation of liver biometry.MethodsWe trained a 2D U-Net CNN for liver segmentation in two stages using 330 abdominal MRI and CT exams acquired at our institution. First, we trained the neural network with non-contrast multi-echo spoiled-gradient-echo (SGPR)images with 300 MRI exams to provide multiple signal-weightings. Then, we used transfer learning to generalize the CNN with additional images from 30 contrast-enhanced MRI and CT exams.We assessed the performance of the CNN using a distinct multi-institutional data set curated from multiple sources (n = 498 subjects). Segmentation accuracy was evaluated by computing Dice scores. Utilizing these segmentations, we computed liver volume from CT and T1-weighted (T1w) MRI exams, and estimated hepatic proton- density-fat-fraction (PDFF) from multi-echo T2*w MRI exams. We compared quantitative volumetry and PDFF estimates between automated and manual segmentation using Pearson correlation and Bland-Altman statistics.ResultsDice scores were 0.94 ± 0.06 for CT (n = 230), 0.95 ± 0.03 (n = 100) for T1w MR, and 0.92 ± 0.05 for T2*w MR (n = 169). Liver volume measured by manual and automated segmentation agreed closely for CT (95% limit-of-agreement (LoA) = [-298 mL, 180 mL]) and T1w MR (LoA = [-358 mL, 180 mL]). Hepatic PDFF measured by the two segmentations also agreed closely (LoA = [-0.62%, 0.80%]).ConclusionsUtilizing a transfer-learning strategy, we have demonstrated the feasibility of a CNN to be generalized to perform liver segmentations across different imaging techniques and modalities. With further refinement and validation, CNNs may have broad applicability for multimodal liver volumetry and hepatic tissue characterization.

BackgroundIn patients with mitral or tricuspid valve regurgitation, evaluation of regurgitant severity is essential for determining the need for surgery. While transthoracic echocardiography is widely accessible, it has limited reproducibility for grading inlet valve regurgitation. Multiplanar cardiac MRI is the quantitative standard but requires specialized local expertise, and is thus not widely available. Volumetric 4D flow MRI has potential for quantitatively grading the severity of inlet valve regurgitation in adult patients.PurposeTo evaluate the accuracy and reproducibility of volumetric 4D flow MRI for quantification of inlet valvular regurgitation compared to conventional multiplanar MRI, which may simplify and improve accessibility of cardiac MRI.Study typeThis retrospective, HIPAA-compliant imaging-based comparison study was conducted at a single institution.SubjectsTwenty-one patients who underwent concurrent multiplanar and 4D flow cardiac MRI between April 2015 and January 2017.Field strength/sequences3T; steady-state free-precession (SSFP), 2D phase contrast (2D-PC), and postcontrast 4D flow.AssessmentWe evaluated the intertechnique (4D flow vs. 2D-PC), intermethod (direct vs. indirect measurement), interobserver and intraobserver reproducibility of measurements of regurgitant flow volume (RFV), fraction (RF), and volume (RVol).Statistical testsStatistical analysis included Pearson correlation, Bland-Altman statistics, and intraclass correlation coefficients.ResultsThere was high concordance between 4D flow and multiplanar MRI, whether using direct or indirect methods of quantifying regurgitation (r = 0.813-0.985). Direct interrogation of the regurgitant jet with 4D flow showed high intraobserver consistency (r = 0.976-0.999) and interobserver consistency (r = 0.861-0.992), and correlated well with traditional indirect measurements obtained as the difference between stroke volume and forward outlet valve flow.Data conclusion4D flow MRI provides highly reproducible measurements of mitral and tricuspid regurgitant volume, and may be used in place of conventional multiplanar MRI.Level of evidence4 Technical Efficacy: Stage 2 J. Magn. Reson. Imaging 2018;48:1147-1158.

PurposeTo develop a rapid segmentation-free method to visualize and compute wall shear stress (WSS) throughout the aorta using 4D Flow MRI data. WSS is the drag force-per-area the vessel endothelium exerts on luminal blood; abnormal levels of WSS are associated with cardiovascular pathologies. Previous methods for computing WSS are bottlenecked by labor-intensive manual segmentation of vessel boundaries. A rapid automated segmentation-free method for computing WSS is presented.Theory and methodsShear stress is the dot-product of the viscous stress tensor and the inward normal vector. The inward normal vectors are approximated as the gradient of fluid speed at every voxel. Subsequently, a 4D map of shear stress is computed as the partial derivatives of velocity with respect to the inward normal vectors. We highlight the shear stress near the wall by fusing visualization with edge-emphasized anatomical data.ResultsAs a proof-of-concept, four cases with aortic pathologies are presented. Visualization allows for rapid localization of pathologic WSS. Subsequent analysis of these pathological regions enables quantification of WSS. Average WSS during peak systole measures approximately 50-60 cPa in nonpathological regions of the aorta and is elevated in regions of stenosis, coarctation, and dissection. WSS is reduced in regions of aneurysm.ConclusionA volumetric technique for calculation and visualization of WSS from 4D Flow MRI data is presented. Traditional labor-intensive methods for WSS rely on explicit manual segmentation of vessel boundaries before visualization. This automated volumetric strategy for visualization and quantification of WSS may facilitate its clinical translation.

OBJECTIVE:We report here an initial experience using 4D flow MRI in pelvic imaging-specifically, in imaging uterine fibroids. We hypothesized that blood flow might correlate with fibroid volume and that quantifying blood flow might help to predict the amount of embolic required to achieve stasis at subsequent uterine fibroid embolization (UFE). MATERIALS AND METHODS:Thirty-three patients with uterine fibroids and seven control subjects underwent pelvic MRI with 4D flow imaging. Of the patients with fibroids, 10 underwent 4D flow imaging before UFE and seven after UFE; in the remaining 16 patients with fibroids, UFE had yet to be performed. Four-dimensional flow measurements were performed using Arterys CV Flow. The flow fraction of the internal iliac artery was expressed as the ratio of internal iliac artery flow to external iliac artery flow and was compared between groups. The flow ratios between the internal iliac arteries on each side were calculated. Fibroid volume versus internal iliac flow fraction, embolic volume versus internal iliac flow fraction, and embolic volume ratio between sides versus the ratio of internal iliac artery flows between sides were compared. RESULTS:The mean internal iliac flow fraction was significantly higher in the 26 patients who underwent imaging before UFE (mean ± standard error, 0.78 ± 0.06) than in the seven patients who underwent imaging after UFE (0.48 ± 0.07, p < 0.01) and in the seven control patients without fibroids (0.48 ± 0.08, p < 0.0001). The internal iliac flow fraction correlated well with fibroid volumes before UFE (r = 0.7754, p < 0.0001) and did not correlate with fibroid volumes after UFE (r = -0.3051, p = 0.51). The ratio of embolic required to achieve stasis between sides showed a modest correlation with the ratio of internal iliac flow (r = 0.6776, p = 0.03). CONCLUSION:Internal iliac flow measured by 4D flow MRI correlates with fibroid volume and is predictive of the ratio of embolic required to achieve stasis on each side at subsequent UFE and may be useful for preprocedural evaluation of patients with uterine fibroids.