The 50 most-cited articles in Radiology, 2017

RSNA, 2017 Online supplemental material is available for this article. An earlier incorrect version of this article appeared online. This article was corrected on March 13, 2017.

Purpose To evaluate the efficacy of deep convolutional neural networks (DCNNs) for detecting tuberculosis (TB) on chest radiographs. Materials and Methods Four deidentified HIPAA-compliant datasets were used in this study that were exempted from review by the institutional review board, which consisted of 1007 posteroanterior chest radiographs. The datasets were split into training (68.0%), validation (17.1%), and test (14.9%). Two different DCNNs, AlexNet and GoogLeNet, were used to classify the images as having manifestations of pulmonary TB or as healthy. Both untrained and pretrained networks on ImageNet were used, and augmentation with multiple preprocessing techniques. Ensembles were performed on the best-performing algorithms. For cases where the classifiers were in disagreement, an independent board-certified cardiothoracic radiologist blindly interpreted the images to evaluate a potential radiologist-augmented workflow. Receiver operating characteristic curves and areas under the curve (AUCs) were used to assess model performance by using the DeLong method for statistical…

Purpose To investigate diagnostic performance by using a deep learning method with a convolutional neural network (CNN) for the differentiation of liver masses at dynamic contrast agent–enhanced computed tomography (CT). Materials and Methods This clinical retrospective study used CT image sets of liver masses over three phases (noncontrast-agent enhanced, arterial, and delayed). Masses were diagnosed according to five categories (category A, classic hepatocellular carcinomas [HCCs]; category B, malignant liver tumors other than classic and early HCCs; category C, indeterminate masses or mass-like lesions [including early HCCs and dysplastic nodules] and rare benign liver masses other than hemangiomas and cysts; category D, hemangiomas; and category E, cysts). Supervised training was performed by using 55 536 image sets obtained in 2013 (from 460 patients, 1068 sets were obtained and they were augmented by a factor of 52 [rotated, parallel-shifted, strongly enlarged, and noise-added images were generated from the original images]). The CNN was composed of six convolutional, three…

Purpose To compare the performance of a deep-learning bone age assessment model based on hand radiographs with that of expert radiologists and that of existing automated models. Materials and Methods The institutional review board approved the study. A total of 14 036 clinical hand radiographs and corresponding reports were obtained from two children’s hospitals to train and validate the model. For the first test set, composed of 200 examinations, the mean of bone age estimates from the clinical report and three additional human reviewers was used as the reference standard. Overall model performance was assessed by comparing the root mean square (RMS) and mean absolute difference (MAD) between the model estimates and the reference standard bone ages. Ninety-five percent limits of agreement were calculated in a pairwise fashion for all reviewers and the model. The RMS of a second test set composed of 913 examinations from the publicly available Digital Hand Atlas was compared with published reports of an existing automated model. Results The mean difference between bone age estimates…

Purpose To evaluate the effect of middle meningeal artery (MMA) embolization on chronic subdural hematoma (CSDH) and compare the treatment outcomes of MMA embolization and conventional treatment. Materials and Methods All consecutive patients 20 years or older with CSDH were assessed for eligibility. CSDHs with a focal location, a thickness of 10 mm or less, no mass effect, or underlying conditions were excluded. Seventy-two prospectively enrolled patients with CSDH underwent MMA embolization (embolization group; as the sole treatment in 27 [37.5%] asymptomatic patients and with additional hematoma removal for symptom relief in 45 [62.5%] symptomatic patients). For comparison, 469 patients who underwent conventional treatment were included as a historical control group (conventional treatment group; close, nonsurgical follow-up in 67 [14.3%] and hematoma removal in 402 [85.7%] patients). Primary outcome was treatment failure defined as a composite of incomplete hematoma resolution (remaining or reaccumulated hematoma with thickness > 10 mm) or surgical rescue (hematoma removal for…

Purpose To develop and evaluate the feasibility of deep learning approaches for magnetic resonance (MR) imaging–based attenuation correction (AC) (termed deep MRAC) in brain positron emission tomography (PET)/MR imaging. Materials and Methods A PET/MR imaging AC pipeline was built by using a deep learning approach to generate pseudo computed tomographic (CT) scans from MR images. A deep convolutional auto-encoder network was trained to identify air, bone, and soft tissue in volumetric head MR images coregistered to CT data for training. A set of 30 retrospective three-dimensional T1-weighted head images was used to train the model, which was then evaluated in 10 patients by comparing the generated pseudo CT scan to an acquired CT scan. A prospective study was carried out for utilizing simultaneous PET/MR imaging for five subjects by using the proposed approach. Analysis of covariance and paired-sample t tests were used for statistical analysis to compare PET reconstruction error with deep MRAC and two existing MR imaging–based AC approaches with CT-based AC. Results Deep MRAC…

Purpose To develop diagnostic reference levels (DRLs) and achievable doses (ADs) for the 10 most common adult computed tomographic (CT) examinations in the United States as a function of patient size by using the CT Dose Index Registry. Materials and Methods Data from the 10 most commonly performed adult CT head, neck, and body examinations from 583 facilities were analyzed. For head examinations, the lateral thickness was used as an indicator of patient size; for neck and body examinations, water-equivalent diameter was used. Data from 1 310 727 examinations (analyzed by using SAS 9.3) provided median values, as well as means and 25th and 75th (DRL) percentiles for volume CT dose index (CTDIvol), dose-length product (DLP), and size-specific dose estimate (SSDE). Applicable results were compared with DRLs from eight countries. Results More than 46% of the facilities were community hospitals; 13% were academic facilities. More than 48% were in metropolitan areas, 39% were suburban, and 13% were rural. More than 50% of the facilities performed fewer than 500 examinations per month.…

These recommendations for measuring pulmonary nodules at computed tomography (CT) are a statement from the Fleischner Society and, as such, incorporate the opinions of a multidisciplinary international group of thoracic radiologists, pulmonologists, surgeons, pathologists, and other specialists. The recommendations address nodule size measurements at CT, which is a topic of importance, given that all available guidelines for nodule management are essentially based on nodule size or changes thereof. The recommendations are organized according to practical questions that commonly arise when nodules are measured in routine clinical practice and are, together with their answers, summarized in a table. The recommendations include technical requirements for accurate nodule measurement, directions on how to accurately measure the size of nodules at the workstation, and directions on how to report nodule size and changes in size. The recommendations are designed to provide practical advice based on the available evidence from the literature; however, areas of uncertainty are also discussed,…

Purpose To determine the linearity, bias, and precision of hepatic proton density fat fraction (PDFF) measurements by using magnetic resonance (MR) imaging across different field strengths, imager manufacturers, and reconstruction methods. Materials and Methods This meta-analysis was performed in accordance with Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. A systematic literature search identified studies that evaluated the linearity and/or bias of hepatic PDFF measurements by using MR imaging (hereafter, MR imaging–PDFF) against PDFF measurements by using colocalized MR spectroscopy (hereafter, MR spectroscopy-PDFF) or the precision of MR imaging–PDFF. The quality of each study was evaluated by using the Quality Assessment of Studies of Diagnostic Accuracy 2 tool. De-identified original data sets from the selected studies were pooled. Linearity was evaluated by using linear regression between MR imaging–PDFF and MR spectroscopy-PDFF measurements. Bias, defined as the mean difference between MR imaging–PDFF and MR spectroscopy-PDFF measurements, was…

Purpose To determine whether gadolinium deposits in neural tissues of patients with intracranial abnormalities following intravenous gadolinium-based contrast agent (GBCA) exposure might be related to blood-brain barrier integrity by studying adult patients with normal brain pathologic characteristics. Materials and Methods After obtaining antemortem consent and institutional review board approval, the authors compared postmortem neuronal tissue samples from five patients who had undergone four to 18 gadolinium-enhanced magnetic resonance (MR) examinations between 2005 and 2014 (contrast group) with samples from 10 gadolinium-naive patients who had undergone at least one MR examination during their lifetime (control group). All patients in the contrast group had received gadodiamide. Neuronal tissues from the dentate nuclei, pons, globus pallidus, and thalamus were harvested and analyzed with inductively coupled plasma mass spectrometry (ICP-MS), transmission electron microscopy with energy-dispersive x-ray spectroscopy, and light microscopy to quantify, localize, and assess the…

The Liver Imaging Reporting and Data System (LI-RADS) standardizes the interpretation, reporting, and data collection for imaging examinations in patients at risk for hepatocellular carcinoma (HCC). It assigns category codes reflecting relative probability of HCC to imaging-detected liver observations based on major and ancillary imaging features. LI-RADS also includes imaging features suggesting malignancy other than HCC. Supported and endorsed by the American College of Radiology (ACR), the system has been developed by a committee of radiologists, hepatologists, pathologists, surgeons, lexicon experts, and ACR staff, with input from the American Association for the Study of Liver Diseases and the Organ Procurement Transplantation Network/United Network for Organ Sharing. Development of LI-RADS has been based on literature review, expert opinion, rounds of testing and iteration, and feedback from users. This article summarizes and assesses the quality of evidence supporting each LI-RADS major feature for diagnosis of HCC, as well as of the LI-RADS imaging features suggesting…

Purpose To characterize clinically important prostate cancers missed at multiparametric (MP) magnetic resonance (MR) imaging. Materials and Methods The local institutional review board approved this HIPAA-compliant retrospective single-center study, which included 100 consecutive patients who had undergone MP MR imaging and subsequent radical prostatectomy. A genitourinary pathologist blinded to MP MR findings outlined prostate cancers on whole-mount pathology slices. Two readers correlated mapped lesions with reports of prospectively read MP MR images. Readers were blinded to histopathology results during prospective reading. At histopathologic examination, 80 clinically unimportant lesions (<5 mm; Gleason score, 3+3) were excluded. The same two readers, who were not blinded to histopathologic findings, retrospectively reviewed cancers missed at MP MR imaging and assigned a Prostate Imaging Reporting and Data System (PI-RADS) version 2 score to better understand false-negative lesion characteristics. Descriptive statistics were used to define patient characteristics, including age,…

Purpose To determine the diagnostic accuracy for clinically significant prostate cancer achieved with abbreviated biparametric prostate magnetic resonance (MR) imaging in comparison with full multiparametric contrast material–enhanced prostate MR imaging in men with elevated prostate-specific antigen (PSA) and negative transrectal ultrasonography (US)–guided biopsy findings; to determine the significant cancer detection rate of biparametric versus full multiparametric contrast-enhanced MR imaging and between-reader agreement for interpretation of biparametric MR imaging. Materials and Methods In this institutional review board–approved retrospective review of prospectively acquired data, men with PSA greater than or equal to 3 ng/mL after negative transrectal US-guided biopsy findings underwent state-of-the-art, full multiparametric contrast-enhanced MR imaging at 3.0-T including high-spatial-resolution structural imaging in several planes, diffusion-weighted imaging at 0, 800, 1000, and 1400 mm2/sec, and dynamic contrast-enhanced MR imaging, obtained without endorectal coil within…

Purpose To evaluate the performance of an artificial intelligence (AI) tool using a deep learning algorithm for detecting hemorrhage, mass effect, or hydrocephalus (HMH) at non—contrast material–enhanced head computed tomographic (CT) examinations and to determine algorithm performance for detection of suspected acute infarct (SAI). Materials and Methods This HIPAA-compliant retrospective study was completed after institutional review board approval. A training and validation dataset of noncontrast-enhanced head CT examinations that comprised 100 examinations of HMH, 22 of SAI, and 124 of noncritical findings was obtained resulting in 2583 representative images. Examinations were processed by using a convolutional neural network (deep learning) using two different window and level configurations (brain window and stroke window). AI algorithm performance was tested on a separate dataset containing 50 examinations with HMH findings, 15 with SAI findings, and 35 with noncritical findings. Results Final algorithm performance for HMH showed 90% (45 of 50) sensitivity (95% confidence…

The use of neoadjuvant systemic therapy in the treatment of breast cancer patients is increasing beyond the scope of locally advanced disease. Imaging provides important information in assessing response to therapy as a complement to conventional tumor measurements via physical examination. The purpose of this article is to discuss the advantages and limitations of current assessment methods, as well as review functional and molecular imaging approaches being investigated as emerging techniques for evaluating neoadjuvant therapy response for patients with primary breast cancer. © RSNA, 2017

Purpose To determine if patient survival and mechanisms of right ventricular failure in pulmonary hypertension could be predicted by using supervised machine learning of three-dimensional patterns of systolic cardiac motion. Materials and Methods The study was approved by a research ethics committee, and participants gave written informed consent. Two hundred fifty-six patients (143 women; mean age ± standard deviation, 63 years ± 17) with newly diagnosed pulmonary hypertension underwent cardiac magnetic resonance (MR) imaging, right-sided heart catheterization, and 6-minute walk testing with a median follow-up of 4.0 years. Semiautomated segmentation of short-axis cine images was used to create a three-dimensional model of right ventricular motion. Supervised principal components analysis was used to identify patterns of systolic motion that were most strongly predictive of survival. Survival prediction was assessed by using difference in median survival time and area under the curve with time-dependent receiver operating characteristic analysis for 1-year survival. Results At the…

Intracranial vessel wall magnetic resonance (MR) imaging has gained much attention in the past decade and has become part of state-of-the-art MR imaging protocols to assist in diagnosing the cause of ischemic stroke. With intracranial vessel wall imaging, vessel wall characteristics have tentatively been described for atherosclerosis, vasculitis, dissections, Moyamoya disease, and aneurysms. With the increasing demand and subsequently increased use of intracranial vessel wall imaging in clinical practice, radiologists should be aware of the choices in imaging parameters and how they affect image quality, the clinical indications, methods of assessment, and limitations in the interpretation of these images. In this How I do It article, the authors will discuss the technical requirements and considerations for vessel wall image acquisition in general, describe their own vessel wall imaging protocol at 3 T and 7 T, show a step-by-step basic assessment of intracranial vessel wall imaging as performed at their institution-including commonly encountered artifacts and pitfalls-and…

Purpose To compare the benefits and harms of radiofrequency ablation (RFA) and hepatic resection (HR) and to test the consistency of currently available evidence. Materials and Methods PubMed, Embase, and the Cochrane Library were systematically searched for randomized controlled trials (RCTs) that compared the effects of HR and RFA for Barcelona Clinic Liver Cancer very early or early stage hepatocellular carcinoma (HCC). The primary outcome was overall survival, and secondary outcomes were recurrence rate, complication rate, and hospitalization duration. A random- or fixed-effects model according to the level of heterogeneity was applied. The meta-analysis was performed by using software, and trial sequential analysis (TSA) was performed. Results Five trials examining 742 patients were included in this study (sizes of trials: 161, 230, 168, 120, and 63 patients). The meta-analysis showed that RFA and HR had similar overall survival at 1 year (relative risk [RR], 1.39; 95% confidence interval [CI]: 0.36, 5.33; P = .63) and 3 years (RR, 1.40; 95% CI: 0.75, 2.62; P = .29), whereas…

Purpose To evaluate the performance of a deep learning convolutional neural network (CNN) model compared with a traditional natural language processing (NLP) model in extracting pulmonary embolism (PE) findings from thoracic computed tomography (CT) reports from two institutions. Materials and Methods Contrast material–enhanced CT examinations of the chest performed between January 1, 1998, and January 1, 2016, were selected. Annotations by two human radiologists were made for three categories: the presence, chronicity, and location of PE. Classification of performance of a CNN model with an unsupervised learning algorithm for obtaining vector representations of words was compared with the open-source application PeFinder. Sensitivity, specificity, accuracy, and F1 scores for both the CNN model and PeFinder in the internal and external validation sets were determined. Results The CNN model demonstrated an accuracy of 99% and an area under the curve value of 0.97. For internal validation report data, the CNN model had a statistically significant larger F1 score (0.938) than did…

Focused assessment with sonography in trauma (FAST) has been extensively utilized and studied in blunt and penetrating trauma for the past 3 decades. Prior to FAST, invasive procedures such as diagnostic peritoneal lavage and exploratory laparotomy were commonly utilized to diagnose intraabdominal injury. Today the FAST examination has evolved into a more comprehensive study of the abdomen, heart, chest, and inferior vena cava, and many variations in technique, protocols, and interpretation exist. Trauma management strategies such as laparotomy, laparoscopy, endoscopy, computed tomographic angiography, angiographic intervention, serial imaging, and clinical observation have also changed over the years. This state of the art review will discuss the evolution of the FAST examination to its current state in 2017 and evaluate its evolving role in the acute management of the trauma patient. The authors also report on the utility of FAST in special patient populations, such as pediatric and pregnant trauma patients, and the potential for future research, applications, and portions of this…

Purpose To compare the diagnostic utility of an investigational optoacoustic imaging device that fuses laser optical imaging (OA) with grayscale ultrasonography (US) to grayscale US alone in differentiating benign and malignant breast masses. Materials and Methods This prospective, 16-site study of 2105 women (study period: 12/21/2012 to 9/9/2015) compared Breast Imaging Reporting and Data System (BI-RADS) categories assigned by seven blinded independent readers to benign and malignant breast masses using OA/US versus US alone. BI-RADS 3, 4, or 5 masses assessed at diagnostic US with biopsy-proven histologic findings and BI-RADS 3 masses stable at 12 months were eligible. Independent readers reviewed US images obtained with the OA/US device, assigned a probability of malignancy (POM) and BI-RADS category, and locked results. The same independent readers then reviewed OA/US images, scored OA features, and assigned OA/US POM and a BI-RADS category. Specificity and sensitivity were calculated for US and OA/US. Benign and malignant mass upgrade and downgrade rates, positive and negative…

Purpose To determine the accuracy of dual-energy computed tomographic (CT) quantitation in a phantom system comparing fast kilovolt peak–switching, dual-source, split-filter, sequential-scanning, and dual-layer detector systems. Materials and Methods A large elliptical phantom containing iodine (2, 5, and 15 mg/mL), simulated contrast material–enhanced blood, and soft-tissue inserts with known elemental compositions was scanned three to five times with seven dual-energy CT systems and a total of 10 kilovolt peak settings. Monochromatic images (50, 70, and 140 keV) and iodine concentration images were created. Mean iodine concentration and monochromatic attenuation for each insert and reconstruction energy level were recorded. Measurement bias was assessed by using the sum of the mean signed errors measured across relevant inserts for each monochromatic energy level and iodine concentration. Iodine and monochromatic errors were assessed by using the root sum of the squared error of all measurements. Results At least one acquisition paradigm per scanner had iodine biases (range, −2.6…

Invasive coronary angiography (ICA) with measurement of fractional flow reserve (FFR) by means of a pressure wire technique is the established reference standard for the functional assessment of coronary artery disease (CAD) (1,2). Coronary computed tomographic (CT) angiography has emerged as a noninvasive method for direct assessment of CAD and plaque characterization with high diagnostic accuracy compared with ICA (3,4). However, the solely anatomic assessment provided with both coronary CT angiography and ICA has poor discriminatory power for ischemia-inducing lesions. FFR derived from standard coronary CT angiography (FFRCT) data sets by using any of several advanced computational analytic approaches enables combined anatomic and hemodynamic assessment of a coronary lesion by a single noninvasive test. Current technical approaches to the calculation of FFRCT include algorithms based on full- and reduced-order computational fluid dynamic modeling, as well as artificial intelligence deep machine learning (5,6). A growing body of evidence has validated the diagnostic accuracy of…

Purpose To investigate the performance of a deep convolutional neural network (DCNN) model in the staging of liver fibrosis using gadoxetic acid–enhanced hepatobiliary phase magnetic resonance (MR) imaging. Materials and Methods This retrospective study included patients for whom input data (hepatobiliary phase MR images, static magnetic field of the imaging unit, and hepatitis B and C virus testing results available, either positive or negative) and reference standard data (liver fibrosis stage evaluated from biopsy or surgical specimens obtained within 6 months of the MR examinations) were available were assigned to the training (534 patients) or the test (100 patients) group. For the training group (54, 53, 81, 113, and 233 patients with fibrosis stages F0, F1, F2, F3, and F4, respectively; mean patient age, 67.4 ± 9.7 years; 388 men and 146 women), MR images with three different section levels were augmented 90-fold (rotated, parallel-shifted, brightness-changed and contrast-changed images were generated; a total of 144 180 images). Supervised training was performed by using the…

Purpose To compare gadolinium tissue concentrations of multiple linear and macrocyclic chelates in a rat model to better understand the scope and extent of tissue deposition following multiple intravenous doses of gadolinium-based contrast agent (GBCA). Materials and Methods In this Institutional Animal Care and Use Committee-approved study, healthy rats received 20 intravenous injections of 2.5 mmol gadolinium per kilogram (gadolinium-exposed group) or saline (control group) over a 26-day period. Unenhanced T1 signal intensities of the dentate nucleus were measured from magnetic resonance (MR) images obtained prior to GBCA injection and 3 days after final injection. Rat brain and renal, hepatic, and splenic tissues were harvested 7 days after final injection and subjected to inductively coupled plasma mass spectrometry and transmission electron microscopy for quantification and characterization of gadolinium deposits. Results Gadolinium deposition in brain tissue significantly varied with GBCA type (F = 31.2; P < .0001), with median concentrations of 0 μg gadolinium per gram of…

Purpose To test whether texture analysis (TA) allows for the diagnosis of subacute and chronic myocardial infarction (MI) on noncontrast material–enhanced cine cardiac magnetic resonance (MR) images. Materials and Methods In this retrospective, institutional review board–approved study, 120 patients who underwent cardiac MR imaging and showed large transmural (volume of enhancement on late gadolinium enhancement [LGE] images >20%, n = 72) or small (enhanced volume ≤20%, n = 48) subacute or chronic ischemic scars were included. Sixty patients with normal cardiac MR imaging findings served as control subjects. Regions of interest for TA encompassing the left ventricle were drawn by two blinded, independent readers on cine images in end systole by using a freely available software package. Stepwise dimension reduction and texture feature selection based on reproducibility, machine learning, and correlation analyses were performed for selecting features, enabling the diagnosis of MI on nonenhanced cine MR images by using LGE imaging as the standard of reference. Results Five independent…

Purpose To create and validate a computer system with which to detect, localize, and classify compression fractures and measure bone density of thoracic and lumbar vertebral bodies on computed tomographic (CT) images. Materials and Methods Institutional review board approval was obtained, and informed consent was waived in this HIPAA-compliant retrospective study. A CT study set of 150 patients (mean age, 73 years; age range, 55–96 years; 92 women, 58 men) with (n = 75) and without (n = 75) compression fractures was assembled. All case patients were age and sex matched with control subjects. A total of 210 thoracic and lumbar vertebrae showed compression fractures and were electronically marked and classified by a radiologist. Prototype fully automated spinal segmentation and fracture detection software were then used to analyze the study set. System performance was evaluated with free-response receiver operating characteristic analysis. Results Sensitivity for detection or localization of compression fractures was 95.7% (201 of 210; 95% confidence interval [CI]: 87.0%, 98.9%), with…

In children, hematogenous osteomyelitis is an infection that primarily affects the most vascularized regions of the growing skeleton. The disease has increased in frequency, virulence, and degree of soft-tissue involvement. The change in clinical manifestations and management over the past 2 decades should be reflected in the current imaging approach to the disease. Imaging of infection must depict the location of a single focus or of multiple foci of involvement and the presence of drainable collections. This review provides an overview of the imaging implications directed by the changing epidemiology, the newer insights of anatomy and pathophysiology, the imaging characteristics with emphasis on specific locations and disease complications, and the differential diagnosis considerations. In addition, basic imaging guidelines for appropriate extent of area to image based on patient age are provided. © RSNA, 2017

Purpose To create a radiogenomic map linking computed tomographic (CT) image features and gene expression profiles generated by RNA sequencing for patients with non-small cell lung cancer (NSCLC). Materials and Methods A cohort of 113 patients with NSCLC diagnosed between April 2008 and September 2014 who had preoperative CT data and tumor tissue available was studied. For each tumor, a thoracic radiologist recorded 87 semantic image features, selected to reflect radiologic characteristics of nodule shape, margin, texture, tumor environment, and overall lung characteristics. Next, total RNA was extracted from the tissue and analyzed with RNA sequencing technology. Ten highly coexpressed gene clusters, termed metagenes, were identified, validated in publicly available gene-expression cohorts, and correlated with prognosis. Next, a radiogenomics map was built that linked semantic image features to metagenes by using the t statistic and the Spearman correlation metric with multiple testing correction. Results RNA sequencing analysis resulted in 10 metagenes that capture a variety of…

Purpose To evaluate associations between pulmonary function and both quantitative analysis and visual assessment of thin-section computed tomography (CT) images at baseline and at 15-month follow-up in subjects with idiopathic pulmonary fibrosis (IPF). Materials and Methods This retrospective analysis of preexisting anonymized data, collected prospectively between 2007 and 2013 in a HIPAA-compliant study, was exempt from additional institutional review board approval. The extent of lung fibrosis at baseline inspiratory chest CT in 280 subjects enrolled in the IPF Network was evaluated. Visual analysis was performed by using a semiquantitative scoring system. Computer-based quantitative analysis included CT histogram-based measurements and a data-driven textural analysis (DTA). Follow-up CT images in 72 of these subjects were also analyzed. Univariate comparisons were performed by using Spearman rank correlation. Multivariate and longitudinal analyses were performed by using a linear mixed model approach, in which models were compared by using asymptotic χ2 tests. Results At…

Purpose To investigate whether photon-counting detector (PCD) technology can improve dose-reduced chest computed tomography (CT) image quality compared with that attained with conventional energy-integrating detector (EID) technology in vivo. Materials and Methods This was a HIPAA-compliant institutional review board-approved study, with informed consent from patients. Dose-reduced spiral unenhanced lung EID and PCD CT examinations were performed in 30 asymptomatic volunteers in accordance with manufacturer-recommended guidelines for CT lung cancer screening (120-kVp tube voltage, 20-mAs reference tube current–time product for both detectors). Quantitative analysis of images included measurement of mean attenuation, noise power spectrum (NPS), and lung nodule contrast-to-noise ratio (CNR). Images were qualitatively analyzed by three radiologists blinded to detector type. Reproducibility was assessed with the intraclass correlation coefficient (ICC). McNemar, paired t, and Wilcoxon signed-rank tests were used to compare image quality. Results Thirty study subjects were evaluated…

Purpose To determine the intermodality agreement of morphologic grading and clinically relevant quantitative measurements between computed tomography (CT) and zero echo time (ZTE) magnetic resonance (MR) imaging of the shoulder. The primary objective was to demonstrate the clinical applicability of ZTE in osseous shoulder imaging. Materials and Methods Thirty-four patients undergoing standard-of-care (SOC) MR imaging with concomitant CT were enrolled in this institutional review board–approved study. ZTE images were acquired after SOC MR imaging. Glenoid morphology (version, vault depth, erosion), injury or disease (osteoarthritis, Bankart and Hill-Sachs lesions, subchondral cysts), and evidence of prior surgery were graded or measured. κ Values, intraclass correlation coefficients (ICCs), and Bland-Altman limits of agreement were used to establish agreement. Qualitative comparison of osseous findings was performed between ZTE and SOC MR imaging. Results Binary classification and nominal/ordinal grades showed substantial or better agreement between raters and modalities (κ or ICC >…

Purpose To perform a systematic review and meta-analysis to determine if there are differences in rates of immediate allergic events between classes of gadolinium-based contrast agents (GBCAs). Materials and Methods PubMed and Google Scholar databases were searched for studies in which rates of immediate adverse events to GBCAs were reported. The American College of Radiology classification system was used to characterize allergic-like events as mild, moderate, or severe, and the total number of administrations of each GBCA was recorded. Where necessary, authors of studies were contacted to clarify data and eliminate physiologic reactions. Relative risks of GBCA types were estimated by using the Mantel-Haenszel type method. Results Nine studies in which immediate reactions to GBCA were recorded from a total of 716 978 administrations of GBCA met the criteria for inclusion and exclusion. The overall and severe rates of GBCA allergic-like adverse events were 9.2 and 0.52 per 10 000 administrations, respectively: 81% (539 of 662) were mild, 13% (86 of 662) were moderate, and 6% (37 of…

Artificial intelligence (AI), machine learning, and deep learning are terms now seen frequently, all of which refer to computer algorithms that change as they are exposed to more data. Many of these algorithms are surprisingly good at recognizing objects in images. The combination of large amounts of machine-consumable digital data, increased and cheaper computing power, and increasingly sophisticated statistical models combine to enable machines to find patterns in data in ways that are not only cost-effective but also potentially beyond humans’ abilities. Building an AI algorithm can be surprisingly easy. Understanding the associated data structures and statistics, on the other hand, is often difficult and obscure. Converting the algorithm into a sophisticated product that works consistently in broad, general clinical use is complex and incompletely understood. To show how these AI products reduce costs and improve outcomes will require clinical translation and industrial-grade integration into routine workflow. Radiology has the chance to leverage AI to become a center of…

Purpose To develop a machine learning model that allows high-risk breast lesions (HRLs) diagnosed with image-guided needle biopsy that require surgical excision to be distinguished from HRLs that are at low risk for upgrade to cancer at surgery and thus could be surveilled. Materials and Methods Consecutive patients with biopsy-proven HRLs who underwent surgery or at least 2 years of imaging follow-up from June 2006 to April 2015 were identified. A random forest machine learning model was developed to identify HRLs at low risk for upgrade to cancer. Traditional features such as age and HRL histologic results were used in the model, as were text features from the biopsy pathologic report. Results One thousand six HRLs were identified, with a cancer upgrade rate of 11.4% (115 of 1006). A machine learning random forest model was developed with 671 HRLs and tested with an independent set of 335 HRLs. Among the most important traditional features were age and HRL histologic results (eg, atypical ductal hyperplasia). An important text feature from the pathologic reports was “severely…

Primary brain tumors, most commonly gliomas, are histopathologically typed and graded as World Health Organization (WHO) grades I–IV according to increasing degrees of malignancy. These grades provide prognostic information and guidance on treatment such as radiation therapy and chemotherapy after surgery. Despite the confirmed value of the WHO grading system, results of a multitude of studies and prospective interventional trials now indicate that tumors with identical morphologic criteria can have highly different outcomes. Molecular markers can allow subtypes of tumors of the same morphologic type and WHO grade to be distinguished and are, therefore, of great interest in personalization of brain tumor treatment. Recent genomic-wide studies have resulted in a far more comprehensive understanding of the genomic alterations in gliomas and provide suggestions for a new molecularly based classification. Magnetic resonance (MR) imaging phenotypes can serve as noninvasive surrogates for tumor genotypes and can provide important information for diagnosis, prognosis, and, eventually,…

Amid mounting concerns about nephrogenic sclerosis and gadolinium deposition in the brain, physicians and patients alike are starting to question the use of gadolinium chelates for clinical magnetic resonance (MR) imaging. The search for safer alternatives is currently underway. In North America, the iron supplement ferumoxytol has gained considerable interest as an MR contrast agent. In Europe, ferumoxtran-10 is entering phase III clinical trials. As these agents are starting to be used by a new generation of radiologists, important clinical questions have re-emerged, including those that have been answered in the past. This article offers 10 important insights for the use of iron oxide nanoparticles in clinical MR imaging.

In sports-related muscle injuries, the main goal of the sports medicine physician is to return the athlete to competition—balanced against the need to prevent the injury from worsening or recurring. Prognosis based on the available clinical and imaging information is crucial. Imaging is crucial to confirm and assess the extent of sports-related muscle injuries and may help to guide management, which directly affects the prognosis. This is especially important when the diagnosis or grade of injury is unclear, when recovery is taking longer than expected, and when interventional or surgical management may be necessary. Several imaging techniques are widely available, with ultrasonography and magnetic resonance imaging currently the most frequently applied in sports medicine. This state of the art review will discuss the main imaging modalities for the assessment of sports-related muscle injuries, including advanced imaging techniques, with the focus on the clinical relevance of imaging features of muscle injuries. © RSNA, 2017 Online supplemental material is available for this…

Purpose To test whether computer-aided diagnosis (CAD) approaches can increase the positive predictive value (PPV) and reduce the false-positive rate in lung cancer screening for small nodules compared with human reading by thoracic radiologists. Materials and Methods A matched case-control sample of low-dose computed tomography (CT) studies in 186 participants with 4–20-mm noncalcified lung nodules who underwent biopsy in the National Lung Screening Trial (NLST) was selected. Variables used for matching were age, sex, smoking status, chronic obstructive pulmonary disease status, body mass index, study year of the positive screening test, and screening results. Studies before lung biopsy were randomly split into a training set (70 cancers plus 70 benign controls) and a validation set (20 cancers plus 26 benign controls). Image features from within and outside dominant nodules were extracted. A CAD algorithm developed from the training set and a random forest classifier were applied to the validation set to predict biopsy outcomes. Receiver operating characteristic analysis was used…

Purpose To evaluate the associations among mathematical modeling with the use of magnetic resonance (MR) imaging–based texture features and deep myometrial invasion (DMI), lymphovascular space invasion (LVSI), and histologic high-grade endometrial carcinoma. Materials and Methods Institutional review board approval was obtained for this retrospective study. This study included 137 women with endometrial carcinomas measuring greater than 1 cm in maximal diameter who underwent 1.5-T MR imaging before hysterectomy between January 2011 and December 2015. Texture analysis was performed with commercial research software with manual delineation of a region of interest around the tumor on MR images (T2-weighted, diffusion-weighted, and dynamic contrast material–enhanced images and apparent diffusion coefficient maps). Areas under the receiver operating characteristic curve and diagnostic performance of random forest models determined by using a subset of the most relevant texture features were estimated and compared with those of independent and blinded visual assessments by three…

Purpose To determine whether quantitative tissue characterization with T1 and T2 mapping supports recognition of myocardial involvement in patients with systemic sarcoidosis. Materials and Methods Fifty-three consecutive patients with a biopsy-proven extracardiac diagnosis of systemic sarcoidosis (21 men; median age, 45 years; interquartile range, 22 years) and 36 normotensive previously healthy control subjects (14 men; median age, 43 years; interquartile range, 18 years) underwent cardiovascular magnetic resonance imaging, which was performed to assess cardiac function and late gadolinium enhancement, and T1 and T2 mapping. A follow-up substudy was performed in 40 patients (mean follow-up interval, 144 days ± 35 [standard deviation]); of these 40 patients, 18 underwent anti-inflammatory treatment for systemic symptoms. Binary logistic regression and receiver operating characteristic curve analyses were used to assess discrimination between health and disease; Wilcoxon signed rank test was used to assess the effect of treatment. Results When compared with control subjects, patients…

Purpose To develop and evaluate an examination consisting of magnetic resonance (MR) fingerprinting–based T1, T2, and standard apparent diffusion coefficient (ADC) mapping for multiparametric characterization of prostate disease. Materials and Methods This institutional review board–approved, HIPAA-compliant retrospective study of prospectively collected data included 140 patients suspected of having prostate cancer. T1 and T2 mapping was performed with fast imaging with steady-state precession–based MR fingerprinting with ADC mapping. Regions of interest were drawn by two independent readers in peripheral zone lesions and normal-appearing peripheral zone (NPZ) tissue identified on clinical images. T1, T2, and ADC were recorded for each region. Histopathologic correlation was based on systematic transrectal biopsy or cognitively targeted biopsy results, if available. Generalized estimating equations logistic regression was used to assess T1, T2, and ADC in the differentiation of (a) cancer versus NPZ, (b) cancer versus prostatitis, (c) prostatitis versus NPZ, and (d) high- or…

Purpose To assess the determinants of technical failure of magnetic resonance (MR) elastography of the liver in a large single-center study. Materials and Methods This retrospective study was approved by the institutional review board. Seven hundred eighty-one MR elastography examinations performed in 691 consecutive patients (mean age, 58 years; male patients, 434 [62.8%]) in a single center between June 2013 and August 2014 were retrospectively evaluated. MR elastography was performed at 3.0 T (n = 443) or 1.5 T (n = 338) by using a gradient-recalled-echo pulse sequence. MR elastography and anatomic image analysis were performed by two observers. Additional observers measured liver T2* and fat fraction. Technical failure was defined as no pixel value with a confidence index higher than 95% and/or no apparent shear waves imaged. Logistic regression analysis was performed to assess potential predictive factors of technical failure of MR elastography. Results The technical failure rate of MR elastography at 1.5 T was 3.5% (12 of 338), while it was higher, 15.3% (68 of 443), at 3.0 T.…

Purpose To assess the safety and efficacy of irreversible electroporation (IRE) in the treatment of patients with inoperable hepatocellular carcinoma (HCC) who are ineligible for thermal ablative techniques. Materials and Methods This retrospective study was approved by an ethics review board, and the requirement to obtain informed written consent was waived. From March 2012 to June 2015, 58 patients (median age, 65.4 years; range 41.6–90 years) with cirrhosis received IRE for the treatment of 75 HCC tumors. The median tumor diameter was 24 mm (range, 6–90 mm). IRE was selected because of tumor location (48 patients) or the patient’s poor general condition (10 patients). Treatment response was assessed with magnetic resonance (MR) imaging 1 month after treatment and every 3 months thereafter. Overall local tumor progression-free survival (PFS) per nodule (including initial treatment failures) was assessed by using the Kaplan-Meier method. The marginal Cox proportional hazards model was used to assess the factors associated with overall local tumor PFS. Complications were recorded…

Purpose To evaluate the potential role of diffusion kurtosis imaging and conventional magnetic resonance (MR) imaging findings including standard monoexponential model of diffusion-weighted imaging and morphologic features for preoperative prediction of microvascular invasion (MVI) of hepatocellular carcinoma (HCC). Materials and Methods Institutional review board approval and written informed consent were obtained. Between September 2015 and November 2016, 84 patients (median age, 54 years; range, 29–79 years) with 92 histopathologically confirmed HCCs (40 MVI-positive lesions and 52 MVI-negative lesions) were analyzed. Preoperative MR imaging examinations including diffusion kurtosis imaging (b values: 0, 200, 500, 1000, 1500, and 2000 sec/mm2) were performed and kurtosis, diffusivity, and apparent diffusion coefficient maps were calculated. Morphologic features of conventional MR images were also evaluated. Univariate and multivariate logistic regression analyses were used to evaluate the relative value of these parameters as potential predictors of MVI. Results Features…

Purpose To identify cerebral radiomic features related to diagnosis and subtyping of attention deficit hyperactivity disorder (ADHD) and to build and evaluate classification models for ADHD diagnosis and subtyping on the basis of the identified features. Materials and Methods A consecutive cohort of 83 age- and sex-matched children with newly diagnosed and never-treated ADHD (mean age 10.83 years ± 2.30; range, 7–14 years; 71 boys, 40 with ADHD-inattentive [ADHD-I] and 43 with ADHD-combined [ADHD-C, or inattentive and hyperactive]) and 87 healthy control subjects (mean age, 11.21 years ± 2.51; range, 7–15 years; 72 boys) underwent anatomic and diffusion-tensor magnetic resonance (MR) imaging. Features representing the shape properties of gray matter and diffusion properties of white matter were extracted for each participant. The initial feature set was input into an all-relevant feature selection procedure within cross-validation loops to identify features with significant discriminative power for diagnosis and subtyping. Random forest classifiers were constructed and evaluated on…

Purpose To validate the dominant pulse sequence paradigm and limited role of dynamic contrast material–enhanced magnetic resonance (MR) imaging in the Prostate Imaging Reporting and Data System (PI-RADS) version 2 for prostate multiparametric MR imaging by using data from a multireader study. Materials and Methods This HIPAA-compliant retrospective interpretation of prospectively acquired data was approved by the local ethics committee. Patients were treatment-naïve with endorectal coil 3-T multiparametric MR imaging. A total of 163 patients were evaluated, 110 with prostatectomy after multiparametric MR imaging and 53 with negative multiparametric MR imaging and systematic biopsy findings. Nine radiologists participated in this study and interpreted images in 58 patients, on average (range, 56–60 patients). Lesions were detected with PI-RADS version 2 and were compared with whole-mount prostatectomy findings. Probability of cancer detection for overall, T2-weighted, and diffusion-weighted (DW) imaging PI-RADS scores was calculated in the peripheral zone (PZ) and transition zone…

Purpose To compare the diagnostic performances of contrast material–enhanced spectral mammography and breast magnetic resonance (MR) imaging in the detection of index and secondary cancers in women with newly diagnosed breast cancer by using histologic or imaging follow-up as the standard of reference. Materials and Methods This institutional review board–approved, HIPAA-compliant, retrospective study included 52 women who underwent breast MR imaging and contrast-enhanced spectral mammography for newly diagnosed unilateral breast cancer between March 2014 and October 2015. Of those 52 patients, 46 were referred for contrast-enhanced spectral mammography and targeted ultrasonography because they had additional suspicious lesions at MR imaging. In six of the 52 patients, breast cancer had been diagnosed at an outside institution. These patients were referred for contrast-enhanced spectral mammography and targeted US as part of diagnostic imaging. Images from contrast-enhanced spectral mammography were analyzed by two fellowship-trained breast imagers with 2.5 years of experience with…

Purpose To investigate the utility of magnetic resonance (MR) elastography–derived mechanical properties in the discrimination of hepatic inflammation and fibrosis in the early stages of chronic liver diseases. Materials and Methods All studies were approved by the institutional animal care and use committee. A total of 187 animals were studied, including 182 mice and five pigs. These animals represented five different liver diseases with a varying combination and extent of hepatic inflammation, fibrosis, congestion, and portal hypertension. Multifrequency three-dimensional MR elastography was performed, and shear stiffness, storage modulus, shear loss modulus, and damping ratio were calculated for all animals. Necroinflammation, fibrosis, and portal pressure were either histologically scored or biochemically and physically quantified in all animals. Two-sided Welch t tests were used to evaluate mean differences between disease and control groups. Spearman correlation analyses were used to evaluate the relationships between mechanical parameters and quantitative fibrosis extent…

A consistent feature of many lung nodule management guidelines is the recommendation to evaluate nodule size by using diameter measurements and electronic calipers. Traditionally, the use of nodule volumetry applications has primarily been reserved for certain lung cancer screening trials rather than clinical practice. However, even before the first nodule management guidelines were published more than a decade ago, research has been ongoing into the use of nodule volumetry as a means of measuring nodule size, and this research has accelerated in recent years. This article aims to provide radiologists with an up-to-date review of the most recent literature on volumetry and volume doubling times in lung nodule management, outlining their benefits and drawbacks. A brief technical review of typical volumetry applications is also provided. © RSNA, 2017