The 50 most-cited articles in Radiology, 2018

Photon-counting CT is an emerging technology with the potential to dramatically change clinical CT. Photon-counting CT uses new energy-resolving x-ray detectors, with mechanisms that differ substantially from those of conventional energy-integrating detectors. Photon-counting CT detectors count the number of incoming photons and measure photon energy. This technique results in higher contrast-to-noise ratio, improved spatial resolution, and optimized spectral imaging. Photon-counting CT can reduce radiation exposure, reconstruct images at a higher resolution, correct beam-hardening artifacts, optimize the use of contrast agents, and create opportunities for quantitative imaging relative to current CT technology. In this review, the authors will explain the technical principles of photon-counting CT in nonmathematical terms for radiologists and clinicians. Following a general overview of the current status of photon-counting CT, they will explain potential clinical applications of this technology.

The Liver Imaging Reporting and Data System (LI-RADS) is composed of four individual algorithms intended to standardize the lexicon, as well as reporting and care, in patients with or at risk for hepatocellular carcinoma in the context of surveillance with US; diagnosis with CT, MRI, or contrast material-enhanced US; and assessment of treatment response with CT or MRI. This report provides a broad overview of LI-RADS, including its historic development, relationship to other imaging guidelines, composition, aims, and future directions. In addition, readers will understand the motivation for and key components of the 2018 update.

The use of artificial intelligence in medicine is currently an issue of great interest, especially with regard to the diagnostic or predictive analysis of medical images. Adoption of an artificial intelligence tool in clinical practice requires careful confirmation of its clinical utility. Herein, the authors explain key methodology points involved in a clinical evaluation of artificial intelligence technology for use in medicine, especially high-dimensional or overparameterized diagnostic or predictive models in which artificial deep neural networks are used, mainly from the standpoints of clinical epidemiology and biostatistics. First, statistical methods for assessing the discrimination and calibration performances of a diagnostic or predictive model are summarized. Next, the effects of disease manifestation spectrum and disease prevalence on the performance results are explained, followed by a discussion of the difference between evaluating the performance with use of internal and external datasets, the importance of using an adequate external dataset obtained from a…

Recent advances and future perspectives of machine learning techniques offer promising applications in medical imaging. Machine learning has the potential to improve different steps of the radiology workflow including order scheduling and triage, clinical decision support systems, detection and interpretation of findings, postprocessing and dose estimation, examination quality control, and radiology reporting. In this article, the authors review examples of current applications of machine learning and artificial intelligence techniques in diagnostic radiology. In addition, the future impact and natural extension of these techniques in radiology practice are discussed.

Purpose To identify the reproducible and nonredundant radiomics features (RFs) for computed tomography (CT). Materials and Methods Two phantoms were used to test RF reproducibility by using test-retest analysis, by changing the CT acquisition parameters (hereafter, intra-CT analysis), and by comparing five different scanners with the same CT parameters (hereafter, inter-CT analysis). Reproducible RFs were selected by using the concordance correlation coefficient (as a measure of the agreement between variables) and the coefficient of variation (defined as the ratio of the standard deviation to the mean). Redundant features were grouped by using hierarchical cluster analysis. Results A total of 177 RFs including intensity, shape, and texture features were evaluated. The test-retest analysis showed that 91% (161 of 177) of the RFs were reproducible according to concordance correlation coefficient. Reproducibility of intra-CT RFs, based on coefficient of variation, ranged from 89.3% (151 of 177) to 43.1% (76 of 177) where the pitch factor and the reconstruction kernel were modified,…

Purpose To compare breast cancer detection performance of radiologists reading mammographic examinations unaided versus supported by an artificial intelligence (AI) system. Materials and Methods An enriched retrospective, fully crossed, multireader, multicase, HIPAA-compliant study was performed. Screening digital mammographic examinations from 240 women (median age, 62 years; range, 39-89 years) performed between 2013 and 2017 were included. The 240 examinations (100 showing cancers, 40 leading to false-positive recalls, 100 normal) were interpreted by 14 Mammography Quality Standards Act-qualified radiologists, once with and once without AI support. The readers provided a Breast Imaging Reporting and Data System score and probability of malignancy. AI support provided radiologists with interactive decision support (clicking on a breast region yields a local cancer likelihood score), traditional lesion markers for computer-detected abnormalities, and an examination-based cancer likelihood score. The area under the receiver operating characteristic curve (AUC), specificity and…

Purpose To develop and validate a deep learning algorithm that predicts the final diagnosis of Alzheimer disease (AD), mild cognitive impairment, or neither at fluorine 18 (18F) fluorodeoxyglucose (FDG) PET of the brain and compare its performance to that of radiologic readers. Materials and Methods Prospective 18F-FDG PET brain images from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) (2109 imaging studies from 2005 to 2017, 1002 patients) and retrospective independent test set (40 imaging studies from 2006 to 2016, 40 patients) were collected. Final clinical diagnosis at follow-up was recorded. Convolutional neural network of InceptionV3 architecture was trained on 90% of ADNI data set and tested on the remaining 10%, as well as the independent test set, with performance compared to radiologic readers. Model was analyzed with sensitivity, specificity, receiver operating characteristic (ROC), saliency map, and t-distributed stochastic neighbor embedding. Results The algorithm achieved area under the ROC curve of 0.98 (95% confidence interval: 0.94, 1.00) when evaluated on…

Purpose To develop and validate a deep learning-based automatic detection algorithm (DLAD) for malignant pulmonary nodules on chest radiographs and to compare its performance with physicians including thoracic radiologists. Materials and Methods For this retrospective study, DLAD was developed by using 43 292 chest radiographs (normal radiograph-to-nodule radiograph ratio, 34 067:9225) in 34 676 patients (healthy-to-nodule ratio, 30 784:3892; 19 230 men [mean age, 52.8 years; age range, 18-99 years]; 15 446 women [mean age, 52.3 years; age range, 18-98 years]) obtained between 2010 and 2015, which were labeled and partially annotated by 13 board-certified radiologists, in a convolutional neural network. Radiograph classification and nodule detection performances of DLAD were validated by using one internal and four external data sets from three South Korean hospitals and one U.S. hospital. For internal and external validation, radiograph classification and nodule detection performances of DLAD were evaluated by using the area under the receiver operating characteristic curve (AUROC)…

Purpose The Radiological Society of North America (RSNA) Pediatric Bone Age Machine Learning Challenge was created to show an application of machine learning (ML) and artificial intelligence (AI) in medical imaging, promote collaboration to catalyze AI model creation, and identify innovators in medical imaging. Materials and Methods The goal of this challenge was to solicit individuals and teams to create an algorithm or model using ML techniques that would accurately determine skeletal age in a curated data set of pediatric hand radiographs. The primary evaluation measure was the mean absolute distance (MAD) in months, which was calculated as the mean of the absolute values of the difference between the model estimates and those of the reference standard, bone age. Results A data set consisting of 14 236 hand radiographs (12 611 training set, 1425 validation set, 200 test set) was made available to registered challenge participants. A total of 260 individuals or teams registered on the Challenge website. A total of 105 submissions were uploaded from 48 unique users during the…

Purpose To analyze how automatic segmentation translates in accuracy and precision to morphology and relaxometry compared with manual segmentation and increases the speed and accuracy of the work flow that uses quantitative magnetic resonance (MR) imaging to study knee degenerative diseases such as osteoarthritis (OA). Materials and Methods This retrospective study involved the analysis of 638 MR imaging volumes from two data cohorts acquired at 3.0 T: (a) spoiled gradient-recalled acquisition in the steady state T1ρ-weighted images and (b) three-dimensional (3D) double-echo steady-state (DESS) images. A deep learning model based on the U-Net convolutional network architecture was developed to perform automatic segmentation. Cartilage and meniscus compartments were manually segmented by skilled technicians and radiologists for comparison. Performance of the automatic segmentation was evaluated on Dice coefficient overlap with the manual segmentation, as well as by the automatic segmentations’ ability to quantify, in a longitudinally repeatable way, relaxometry and morphology.…

Purpose To investigate the value of T2-weighted–based radiomics compared with qualitative assessment at T2-weighted imaging and diffusion-weighted (DW) imaging for diagnosis of clinical complete response in patients with rectal cancer after neoadjuvant chemotherapy–radiation therapy (CRT). Materials and Methods This retrospective study included 114 patients with rectal cancer who underwent magnetic resonance (MR) imaging after CRT between March 2012 and February 2016. Median age among women (47 of 114, 41%) was 55.9 years (interquartile range, 45.4–66.7 years) and median age among men (67 of 114, 59%) was 55 years (interquartile range, 48–67 years). Surgical histopathologic analysis was the reference standard for pathologic complete response (pCR). For qualitative assessment, two radiologists reached a consensus. For radiomics, one radiologist segmented the volume of interest on high-spatial-resolution T2-weighted images. A random forest classifier was trained to separate the patients by their outcomes after balancing the number of patients in each response category by using the…

Purpose To develop and evaluate a fully automated algorithm for segmenting the abdomen from CT to quantify body composition. Materials and Methods For this retrospective study, a convolutional neural network based on the U-Net architecture was trained to perform abdominal segmentation on a data set of 2430 two-dimensional CT examinations and was tested on 270 CT examinations. It was further tested on a separate data set of 2369 patients with hepatocellular carcinoma (HCC). CT examinations were performed between 1997 and 2015. The mean age of patients was 67 years; for male patients, it was 67 years (range, 29-94 years), and for female patients, it was 66 years (range, 31-97 years). Differences in segmentation performance were assessed by using two-way analysis of variance with Bonferroni correction. Results Compared with reference segmentation, the model for this study achieved Dice scores (mean ± standard deviation) of 0.98 ± 0.03, 0.96 ± 0.02, and 0.97 ± 0.01 in the test set, and 0.94 ± 0.05, 0.92 ± 0.04, and 0.98 ± 0.02 in the HCC data set, for the subcutaneous, muscle, and…

Computed tomography and magnetic resonance enterography have become routine small bowel imaging tests to evaluate patients with established or suspected Crohn’s disease, but the interpretation and use of these imaging modalities can vary widely. A shared understanding of imaging findings, nomenclature, and utilization will improve the utility of these imaging techniques to guide treatment options, as well as assess for treatment response and complications. Representatives from the Society of Abdominal Radiology Crohn’s Disease-Focused Panel, the Society of Pediatric Radiology, the American Gastroenterological Association, and other experts, systematically evaluated evidence for imaging findings associated with small bowel Crohn’s disease enteric inflammation and established recommendations for the evaluation, interpretation, and use of computed tomography and magnetic resonance enterography in small bowel Crohn’s disease. This work makes recommendations for imaging findings that indicate small bowel Crohn’s disease, how inflammatory small bowel Crohn’s disease and its complications…

Purpose To evaluate ability of radiomic (computer-extracted imaging) features to distinguish non-small cell lung cancer adenocarcinomas from granulomas at noncontrast CT. Materials and Methods For this retrospective study, screening or standard diagnostic noncontrast CT images were collected for 290 patients (mean age, 68 years; range, 18-92 years; 125 men [mean age, 67 years; range, 18-90 years] and 165 women [mean age, 68 years; range, 33-92 years]) from two institutions between 2007 and 2013. Histopathologic analysis was available for one nodule per patient. Corresponding nodule of interest was identified on axial CT images by a radiologist with manual annotation. Nodule shape, wavelet (Gabor), and texture-based (Haralick and Laws energy) features were extracted from intra- and perinodular regions. Features were pruned to train machine learning classifiers with 145 patients. In a test set of 145 patients, classifier results were compared against a convolutional neural network (CNN) and diagnostic readings of two radiologists. Results Support vector machine classifier with…

Cerebral microbleeds (CMBs), also referred to as microhemorrhages, appear on magnetic resonance (MR) images as hypointense foci notably at T2*-weighted or susceptibility-weighted (SW) imaging. CMBs are detected with increasing frequency because of the more widespread use of high magnetic field strength and of newer dedicated MR imaging techniques such as three-dimensional gradient-echo T2*-weighted and SW imaging. The imaging appearance of CMBs is mainly because of changes in local magnetic susceptibility and reflects the pathologic iron accumulation, most often in perivascular macrophages, because of vasculopathy. CMBs are depicted with a true-positive rate of 48%–89% at 1.5 T or 3.0 T and T2*-weighted or SW imaging across a wide range of diseases. False-positive “mimics” of CMBs occur at a rate of 11%–24% and include microdissections, microaneurysms, and microcalcifications; the latter can be differentiated by using phase images. Compared with postmortem histopathologic analysis, at least half of CMBs are missed with premortem clinical MR imaging. In general, CMB detection rate…

In 2017, the Thyroid Imaging Reporting and Data System (TI-RADS) Committee of the American College of Radiology (ACR) published a white paper that presented a new risk-stratification system for classifying thyroid nodules on the basis of their appearance at ultrasonography (US). In ACR TI-RADS, points in five feature categories are summed to determine a risk level from TR1 to TR5. Recommendations for biopsy or US follow-up are based on the nodule’s ACR TI-RADS level and its maximum diameter. The purpose of this article is to offer practical guidance on how to implement and apply ACR TI-RADS based on the authors’ experience with the system. © RSNA, 2018 An earlier incorrect version of this article appeared online and in print. This article was corrected on April 2, 2018.

Purpose To determine the feasibility of using a deep learning approach to detect cartilage lesions (including cartilage softening, fibrillation, fissuring, focal defects, diffuse thinning due to cartilage degeneration, and acute cartilage injury) within the knee joint on MR images. Materials and Methods A fully automated deep learning-based cartilage lesion detection system was developed by using segmentation and classification convolutional neural networks (CNNs). Fat-suppressed T2-weighted fast spin-echo MRI data sets of the knee of 175 patients with knee pain were retrospectively analyzed by using the deep learning method. The reference standard for training the CNN classification was the interpretation provided by a fellowship-trained musculoskeletal radiologist of the presence or absence of a cartilage lesion within 17 395 small image patches placed on the articular surfaces of the femur and tibia. Receiver operating curve (ROC) analysis and the κ statistic were used to assess diagnostic performance and intraobserver agreement for detecting cartilage lesions for two individual…

Purpose To develop a deep learning (DL) algorithm to assess mammographic breast density. Materials and Methods In this retrospective study, a deep convolutional neural network was trained to assess Breast Imaging Reporting and Data System (BI-RADS) breast density based on the original interpretation by an experienced radiologist of 41 479 digital screening mammograms obtained in 27 684 women from January 2009 to May 2011. The resulting algorithm was tested on a held-out test set of 8677 mammograms in 5741 women. In addition, five radiologists performed a reader study on 500 mammograms randomly selected from the test set. Finally, the algorithm was implemented in routine clinical practice, where eight radiologists reviewed 10 763 consecutive mammograms assessed with the model. Agreement on BI-RADS category for the DL model and for three sets of readings-(a) radiologists in the test set, (b) radiologists working in consensus in the reader study set, and (c) radiologists in the clinical implementation set-were estimated with linear-weighted κ statistics and were compared across 5000…

Purpose To determine if the change in tumor apparent diffusion coefficient (ADC) at diffusion-weighted (DW) MRI is predictive of pathologic complete response (pCR) to neoadjuvant chemotherapy for breast cancer. Materials and Methods In this prospective multicenter study, 272 consecutive women with breast cancer were enrolled at 10 institutions (from August 2012 to January 2015) and were randomized to treatment with 12 weekly doses of paclitaxel (with or without an experimental agent), followed by 12 weeks of treatment with four cycles of anthracycline. Each woman underwent breast DW MRI before treatment, at early treatment (3 weeks), at midtreatment (12 weeks), and after treatment. Percentage change in tumor ADC from that before treatment (ΔADC) was measured at each time point. Performance for predicting pCR was assessed by using the area under the receiver operating characteristic curve (AUC) for the overall cohort and according to tumor hormone receptor (HR)/human epidermal growth factor receptor 2 (HER2) disease subtype. Results The final analysis included 242 patients with…

Gadolinium-based contrast agents (GBCAs) have revolutionized MRI, enabling physicians to obtain crucial life-saving medical information that often cannot be obtained with other imaging modalities. Since initial approval in 1988, over 450 million intravenous GBCA doses have been administered worldwide, with an extremely favorable pharmacologic safety profile; however, recent information has raised new concerns over the safety of GBCAs. Mounting evidence has shown there is long-term retention of gadolinium in human tissues. Further, a small subset of patients have attributed a constellation of symptoms to GBCA exposure, although the association of these symptoms with GBCA administration or gadolinium retention has not been proven by scientific investigation. Despite evidence that macrocyclic GBCAs show less gadolinium retention than linear GBCAs, the safety implications of gadolinium retention are unknown. The mechanism and chemical forms of gadolinium retention, as well as the biologic activity and clinical importance of these retained gadolinium species, remain poorly understood and…

Precision medicine is medicine optimized to the genotypic and phenotypic characteristics of an individual and, when present, his or her disease. It has a host of targets, including genes and their transcripts, proteins, and metabolites. Studying precision medicine involves a systems biology approach that integrates mathematical modeling and biology genomics, transcriptomics, proteomics, and metabolomics. Moreover, precision medicine must consider not only the relatively static genetic codes of individuals, but also the dynamic and heterogeneous genetic codes of cancers. Thus, precision medicine relies not only on discovering identifiable targets for treatment and surveillance modification, but also on reliable, noninvasive methods of identifying changes in these targets over time. Imaging via radiomics and radiogenomics is poised for a central role. Radiomics, which extracts large volumes of quantitative data from digital images and amalgamates these together with clinical and patient data into searchable shared databases, potentiates radiogenomics, which is the combination of…

Purpose To compare the diagnostic performance of radiomic analysis (RA) and a convolutional neural network (CNN) to radiologists for classification of contrast agent-enhancing lesions as benign or malignant at multiparametric breast MRI. Materials and Methods Between August 2011 and August 2015, 447 patients with 1294 enhancing lesions (787 malignant, 507 benign; median size, 15 mm ± 20) were evaluated. Lesions were manually segmented by one breast radiologist. RA was performed by using L1 regularization and principal component analysis. CNN used a deep residual neural network with 34 layers. All algorithms were also retrained on half the number of lesions (n = 647). Machine interpretations were compared with prospective interpretations by three breast radiologists. Standard of reference was histologic analysis or follow-up. Areas under the receiver operating curve (AUCs) were used to compare diagnostic performance. Results CNN trained on the full cohort was superior to training on the half-size cohort (AUC, 0.88 vs 0.83, respectively; P = .01), but there was no difference for RA…

Chronic liver diseases often result in the development of liver fibrosis and ultimately, cirrhosis. Treatment strategies and prognosis differ greatly depending on the severity of liver fibrosis, thus liver fibrosis staging is clinically relevant. Traditionally, liver biopsy has been the method of choice for fibrosis evaluation. Because of liver biopsy limitations, noninvasive methods have become a key research interest in the field. Elastography enables the noninvasive measurement of tissue mechanical properties through observation of shear-wave propagation in the tissue of interest. Increasing fibrosis stage is associated with increased liver stiffness, providing a discriminatory feature that can be exploited by elastographic methods. Ultrasonographic (US) and magnetic resonance (MR) imaging elastographic methods are commercially available, each with their respective strengths and limitations. Here, the authors review the technical basis, acquisition techniques, and results and limitations of US- and MR-based elastography techniques. Diagnostic performance in the most common…

Purpose To report long-term outcomes of radiation segmentectomy (RS) for early hepatocellular carcinoma (HCC). The authors hypothesized that outcomes are comparable to curative treatments for patients with solitary HCC less than or equal to 5 cm and preserved liver function. Materials and Methods This retrospective study included 70 patients (median age, 71 years; range, 22–96 years) with solitary HCC less than or equal to 5 cm not amenable to percutaneous ablation who underwent RS (dose of >190 Gy) between 2003 and 2016. Patients who underwent subsequent curative liver transplantation were excluded to eliminate this confounding variable affecting survival. Radiologic response of time to progression and median overall survival were estimated by using the Kaplan-Meier method per the guidelines of the European Association for the Study of the Liver (EASL) and the World Health Organization (WHO). Results Seventy patients were treated with RS over 14 years. Sixty-three patients (90%) showed response by using EASL criteria, of which 41 (59%) showed complete response. Fifty patients (71%)…

Purpose To investigate whether radiomic features at MRI improve survival prediction in patients with glioblastoma multiforme (GBM) when they are integrated with clinical and genetic profiles. Materials and Methods Data in patients with a diagnosis of GBM between December 2009 and January 2017 (217 patients) were retrospectively reviewed up to May 2017 and allocated to training and test sets (3:1 ratio). Radiomic features (n = 796) were extracted from multiparametric MRI. A random survival forest (RSF) model was trained with the radiomic features along with clinical and genetic profiles (O-6-methylguanine-DNA-methyltransferase promoter methylation and isocitrate dehydrogenase 1 mutation statuses) to predict overall survival (OS) and progression-free survival (PFS). The RSF models were validated on the test set. The incremental values of radiomic features were evaluated by using the integrated area under the receiver operating characteristic curve (iAUC). Results The 217 patients had a mean age of 57.9 years, and there were 87 female patients (age range, 22-81 years) and 130 male…

Purpose To reduce radiotracer requirements for amyloid PET/MRI without sacrificing diagnostic quality by using deep learning methods. Materials and Methods Forty data sets from 39 patients (mean age ± standard deviation [SD], 67 years ± 8), including 16 male patients and 23 female patients (mean age, 66 years ± 6 and 68 years ± 9, respectively), who underwent simultaneous amyloid (fluorine 18 [18F]–florbetaben) PET/MRI examinations were acquired from March 2016 through October 2017 and retrospectively analyzed. One hundredth of the raw list-mode PET data were randomly chosen to simulate a low-dose (1%) acquisition. Convolutional neural networks were implemented with low-dose PET and multiple MR images (PET-plus-MR model) or with low-dose PET alone (PET-only) as inputs to predict full-dose PET images. Quality of the synthesized images was evaluated while Bland-Altman plots assessed the agreement of regional standard uptake value ratios (SUVRs) between image types. Two readers scored image quality on a five-point scale (5 = excellent) and determined amyloid status (positive or…

Purpose To compare two technical approaches for determination of coronary computed tomography (CT) angiography–derived fractional flow reserve (FFR)—FFR derived from coronary CT angiography based on computational fluid dynamics (hereafter, FFRCFD) and FFR derived from coronary CT angiography based on machine learning algorithm (hereafter, FFRML)—against coronary CT angiography and quantitative coronary angiography (QCA). Materials and Methods A total of 85 patients (mean age, 62 years ± 11 [standard deviation]; 62% men) who had undergone coronary CT angiography followed by invasive FFR were included in this single-center retrospective study. FFR values were derived on-site from coronary CT angiography data sets by using both FFRCFD and FFRML. The performance of both techniques for detecting lesion-specific ischemia was compared against visual stenosis grading at coronary CT angiography, QCA, and invasive FFR as the reference standard. Results On a per-lesion and per-patient level, FFRML showed a sensitivity of 79% and 90% and a specificity of 94% and 95%, respectively, for detecting…

Purpose To investigate the feasibility of tumor type prediction with MRI radiomic image features of different brain metastases in a multiclass machine learning approach for patients with unknown primary lesion at the time of diagnosis. Materials and methods This single-center retrospective analysis included radiomic features of 658 brain metastases from T1-weighted contrast material-enhanced, T1-weighted nonenhanced, and fluid-attenuated inversion recovery (FLAIR) images in 189 patients (101 women, 88 men; mean age, 61 years; age range, 32-85 years). Images were acquired over a 9-year period (from September 2007 through December 2016) with different MRI units, reflecting heterogeneous image data. Included metastases originated from breast cancer (n = 143), small cell lung cancer (n = 151), non-small cell lung cancer (n = 225), gastrointestinal cancer (n = 50), and melanoma (n = 89). A total of 1423 quantitative image features and basic clinical data were evaluated by using random forest machine learning algorithms. Validation was performed with model-external fivefold cross…

Purpose To develop and evaluate a supportive algorithm using deep learning for detecting cerebral aneurysms at time-of-flight MR angiography to provide a second assessment of images already interpreted by radiologists. Materials and Methods MR images reported by radiologists to contain aneurysms were extracted from four institutions for the period from November 2006 through October 2017. The images were divided into three data sets: training data set, internal test data set, and external test data set. The algorithm was constructed by deep learning with the training data set, and its sensitivity to detect aneurysms in the test data sets was evaluated. To find aneurysms that had been overlooked in the initial reports, two radiologists independently performed a blinded interpretation of aneurysm candidates detected by the algorithm. When there was disagreement, the final diagnosis was made in consensus. The number of newly detected aneurysms was also evaluated. Results The training data set, which provided training and validation data, included 748 aneurysms (mean size, 3.1 mm ± 2.0…

Purpose To determine whether visually assessed patterns of emphysema at CT might provide a simple assessment of mortality risk among cigarette smokers. Materials and Methods Of the first 4000 cigarette smokers consecutively enrolled between 2007 and 2011 in this COPDGene study, 3171 had data available for both visual emphysema CT scores and survival. Each CT scan was retrospectively visually scored by two analysts using the Fleischner Society classification system. Severity of emphysema was also evaluated quantitatively by using percentage lung volume occupied by low-attenuation areas (voxels with attenuation of −950 HU or less) (LAA-950). Median duration of follow-up was 7.4 years. Regression analysis for the relationship between imaging patterns and survival was based on the Cox proportional hazards model, with adjustment for age, race, sex, height, weight, pack-years of cigarette smoking, current smoking status, educational level, LAA-950, and (in a second model) forced expiratory volume in 1 second (FEV1). Results Observer agreement in visual scoring was good (weighted κ values,…

Purpose To characterize intratumoral spatial heterogeneity at perfusion magnetic resonance (MR) imaging and investigate intratumoral heterogeneity as a predictor of recurrence-free survival (RFS) in breast cancer. Materials and Methods In this retrospective study, a discovery cohort (n = 60) and a multicenter validation cohort (n = 186) were analyzed. Each tumor was divided into multiple spatially segregated, phenotypically consistent subregions on the basis of perfusion MR imaging parameters. The authors first defined a multiregional spatial interaction (MSI) matrix and then, based on this matrix, calculated 22 image features. A network strategy was used to integrate all image features and classify patients into different risk groups. The prognostic value of imaging-based stratification was evaluated in relation to clinical-pathologic factors with multivariable Cox regression. Results Three intratumoral subregions with high, intermediate, and low MR perfusion were identified and showed high consistency between the two cohorts. Patients in both cohorts were stratified according to…

Purpose To develop and validate a deep learning system (DLS) for staging liver fibrosis by using CT images in the liver. Materials and Methods DLS for CT-based staging of liver fibrosis was created by using a development data set that included portal venous phase CT images in 7461 patients with pathologically confirmed liver fibrosis. The diagnostic performance of the DLS was evaluated in separate test data sets for 891 patients. The influence of patient characteristics and CT techniques on the staging accuracy of the DLS was evaluated by logistic regression analysis. In a subset of 421 patients, the diagnostic performance of the DLS was compared with that of the radiologist's assessment, aminotransferase-to-platelet ratio index (APRI), and fibrosis-4 index by using the area under the receiver operating characteristic curve (AUROC) and Obuchowski index. Results In the test data sets, the DLS had a staging accuracy of 79.4% (707 of 891) and an AUROC of 0.96, 0.97, and 0.95 for diagnosing significant fibrosis (F2-4), advanced fibrosis (F3-4), and cirrhosis (F4), respectively. At…

Cholangiocarcinoma, the second most common hepatobiliary malignancy after hepatocellular carcinoma, is a heterogeneous disease entity with widely varying radiologic features, clinical behavior, and treatment approaches. In the detection, characterization, staging, and resectability assessment of cholangiocarcinoma, imaging studies are indispensable. Herein, an overview of the state-of-the-art imaging techniques is presented for the evaluation of intrahepatic and perihilar cholangiocarcinoma, as well as complementary multimodality and multiparametric imaging approaches for a more comprehensive evaluation. In addition, classification systems, new pathologic concepts in cholangiocarcinogenesis and premalignant lesions, and current trends in treatment approaches, which are vital to the imaging interpretation of cholangiocarcinoma, will be discussed. © RSNA, 2018

Cancer immunotherapy using immune-checkpoint inhibitors has emerged as an effective treatment option for a variety of advanced cancers in the past decade. Because of the distinct mechanisms of immunotherapy that activate the host immunity to treat cancers, unconventional immune-related phenomena are encountered in terms of tumor response and progression, as well as drug toxicity. Imaging plays an important role in objectively characterizing immune-related tumor responses and progression and in detecting and monitoring immune-related adverse events. Moreover, emerging data suggest a promise for molecular imaging that can visualize the specific target molecules involved in immune-checkpoint pathways. In this article, the background and current status of cancer immunotherapy are summarized, and the current methods for imaging evaluations of immune-related responses and toxicities are reviewed along with their limitations and pitfalls. Emerging approaches with molecular imaging are also discussed as a future direction to address unmet needs.

Purpose To develop a deep learning–based method for fully automated quantification of left ventricular (LV) function from short-axis cine MR images and to evaluate its performance in a multivendor and multicenter setting. Materials and Methods This retrospective study included cine MRI data sets obtained from three major MRI vendors in four medical centers from 2008 to 2016. Three convolutional neural networks (CNNs) with the U-NET architecture were trained on data sets of increasing variability: (a) a single-vendor, single-center, homogeneous cohort of 100 patients (CNN1); (b) a single-vendor, multicenter, heterogeneous cohort of 200 patients (CNN2); and (c) a multivendor, multicenter, heterogeneous cohort of 400 patients (CNN3). All CNNs were tested on an independent multivendor, multicenter data set of 196 patients. CNN performance was evaluated with respect to the manual annotations from three experienced observers in terms of (a) LV detection accuracy, (b) LV segmentation accuracy, and (c) LV functional parameter accuracy. Automatic and manual results were compared with the…

Purpose To compare the diagnostic performance of ultrasonography (US)-based fine-needle aspiration biopsy (FNAB) criteria from seven international societies in the detection of thyroid malignancy. Materials and Methods This study included a total of 2000 consecutive thyroid nodules (≥1 cm) in 1802 patients with final diagnoses from January 2010 to May 2011. US features of the thyroid nodules were retrospectively reviewed and were classified according to the categories defined by the seven international society guidelines. The diagnostic performance of US-based FNAB criteria in the detection of thyroid malignancy and unnecessary FNAB rates were calculated and compared by using a generalized estimating equation method. Results Of the 2000 thyroid nodules, 1546 (78.3%) were benign and 454 (22.7%) were malignant, with papillary carcinoma comprising 85.5% of all malignancies. The Korean Thyroid Association/Korean Society of Thyroid Radiology (KTA/KSThR) (94.5%), National Comprehensive Cancer Network (NCCN) (92.5%), and American Thyroid Association (ATA) (89.6%) guidelines were more…

Purpose To assess the ability of convolutional neural networks (CNNs) to enable high-performance automated binary classification of chest radiographs. Materials and Methods In a retrospective study, 216 431 frontal chest radiographs obtained between 1998 and 2012 were procured, along with associated text reports and a prospective label from the attending radiologist. This data set was used to train CNNs to classify chest radiographs as normal or abnormal before evaluation on a held-out set of 533 images hand-labeled by expert radiologists. The effects of development set size, training set size, initialization strategy, and network architecture on end performance were assessed by using standard binary classification metrics; detailed error analysis, including visualization of CNN activations, was also performed. Results Average area under the receiver operating characteristic curve (AUC) was 0.96 for a CNN trained with 200 000 images. This AUC value was greater than that observed when the same model was trained with 2000 images (AUC = 0.84, P < .005) but was not significantly…

Purpose To compare biparametric contrast-free radiomic machine learning (RML), mean apparent diffusion coefficient (ADC), and radiologist assessment for characterization of prostate lesions detected during prospective MRI interpretation. Materials and Methods This single-institution study included 316 men (mean age ± standard deviation, 64.0 years ± 7.8) with an indication for MRI–transrectal US fusion biopsy between May 2015 and September 2016 (training cohort, 183 patients; test cohort, 133 patients). Lesions identified by prospective clinical readings were manually segmented for mean ADC and radiomics analysis. Global and zone-specific random forest RML and mean ADC models for classification of clinically significant prostate cancer (Gleason grade group ≥ 2) were developed on the training set and the fixed models tested on an independent test set. Clinical readings, mean ADC, and radiomics were compared by using the McNemar test and receiver operating characteristic (ROC) analysis. Results In the test set, radiologist interpretation had a per-lesion sensitivity of 88% (53 of 60)…

Purpose To evaluate a radiomics model for predicting lymph node (LN) metastasis in biliary tract cancers (BTCs) and to determine its prognostic value for disease-specific and recurrence-free survival. Materials and Methods For this retrospective study, a radiomics model was developed on the basis of a primary cohort of 177 patients with BTC who underwent resection and LN dissection between June 2010 and December 2016. Radiomic features were extracted from portal venous CT scans. A radiomics signature was built on the basis of reproducible features by using the least absolute shrinkage and selection operator method. Multivariable logistic regression model was adopted to establish a radiomics nomogram. Nomogram performance was determined by its discrimination, calibration, and clinical usefulness. The model was internally validated in 70 consecutive patients with BTC between January 2017 and February 2018. Results The radiomics signature, composed of three LN-status-related features, was associated with LN metastasis in primary and validation cohorts (P < .001). The radiomics nomogram…

Purpose To compare the biopsy rate and diagnostic accuracy before and after applying the American College of Radiology (ACR) Thyroid Imaging Reporting and Data System (TI-RADS) criteria for thyroid nodule evaluation. Materials and Methods In this retrospective study, eight radiologists with 3–32 years experience in thyroid ultrasonography (US) reviewed US features of 100 thyroid nodules that were cytologically proven, pathologically proven, or both in December 2016. The radiologists evaluated nodule features in five US categories and provided biopsy recommendations based on their own practice patterns without knowledge of ACR TI-RADS criteria. Another three expert radiologists served as the reference standard readers for the imaging findings. ACR TI-RADS criteria were retrospectively applied to the features assigned by the eight radiologists to produce biopsy recommendations. Comparison was made for biopsy rate, sensitivity, specificity, and accuracy. Results Fifteen of the 100 nodules (15%) were malignant. The mean number of nodules recommended for biopsy by the eight radiologists…

Pancreatic ductal adenocarcinoma (PDA) remains among the most challenging malignancies to treat. At diagnosis, the tumor often already extends beyond the confines of the pancreas, spreading to an extent such that primary surgery with curative intent is very rarely feasible. Considerable momentum is now being given to a treatment strategy involving neoadjuvant chemotherapy or chemotherapy and radiation therapy in patients with nonmetastatic PDA. The main advantage of this strategy is better selection of patients likely to benefit from curative-intent surgery through the achievement of negative resection margins. Patients with rapidly progressive disease are identified and are spared ineffective surgery with its attendant morbidity. Neoadjuvant therapy can downstage tumors classified as locally advanced at initial imaging studies to resectable tumors. However, the imaging study evaluation of the response to neoadjuvant therapy is extremely complex. Thus, the diagnostic performance of imaging studies is not sufficient to ensure the accurate selection of patients in whom negative-margin…

Purpose To (a) evaluate the postsurgical prognostic implication of the Liver Imaging Reporting and Data System (LI-RADS) categories of primary liver cancers and (b) determine the performance of LI-RADS version 2017 in differentiating hepatocellular carcinoma (HCC) from intrahepatic cholangiocarcinoma (IHCC) and combined hepatocellular-cholangiocarcinoma (cHCC-CC) at gadoxetic acid-enhanced MRI. Materials and Methods In this retrospective study, 194 patients with cirrhosis and surgically proven single primary liver cancer (53 with cHCC-CC, 44 with IHCC, and 97 with HCC) were evaluated with gadoxetic acid-enhanced MRI between 2009 and 2014. The mean patient age was 57 years (age range, 30-83 years). There were 155 men with a mean age of 56 years (range, 30-81 years) and 39 women with a mean age of 58 years (range, 38-83 years). Two independent readers assigned an LI-RADS category for each nodule. Overall survival (OS), recurrence-free survival (RFS), and their associated factors were evaluated by using the Kaplan-Meier method, log-rank test, and Cox proportional hazard model. Results…

Chronic pancreatitis is an inflammatory condition of the pancreas with clinical manifestations ranging from abdominal pain, acute pancreatitis, exocrine and/or endocrine dysfunction, and pancreatic cancer. There is a need for longitudinal studies in well-phenotyped patients to ascertain the utility of cross-sectional imaging findings of chronic pancreatitis for diagnosis and assessment of disease severity. CT and MR cholangiopancreatography are the most common cross-sectional imaging studies performed for the evaluation of chronic pancreatitis. Currently, there are no universal reporting standards for chronic pancreatitis. Several features of chronic pancreatitis are applied clinically, such as calcifications, parenchymal T1 signal changes, focal or diffuse gland atrophy, or irregular contour of the gland. Such findings have not been incorporated into standardized diagnostic criteria. There is also lack of consensus on quantification of disease severity in chronic pancreatitis, other than by using ductal features alone as described in the Cambridge classification. The Consortium for…

Purpose To determine the prognostic value of peritumoral edema identified at preoperative breast magnetic resonance (MR) imaging for disease recurrence in patients with invasive breast cancer. Materials and Methods Between January 2011 and December 2012, 353 women (median age, 49 years; range, 27–77 years) with invasive breast cancer who had undergone preoperative MR imaging and mastectomy or breast-conserving surgery were identified. Two radiologists independently reviewed peritumoral edema on the basis of the degree of the signal intensity surrounding the tumor on T2-weighted images. The association of disease recurrence with peritumoral edema and clinical-pathologic features was assessed by using the multivariate Cox proportional hazards model and the integrated discrimination improvement (IDI) and continuous net reclassification improvement (NRI) indexes. Results Twenty-four patients (6.8%) had disease recurrence after 27.2 months of median follow-up. At multivariate analysis, higher N stage (hazard ratio = 4.84, P = .002) and the presence of lymphovascular invasion (hazard…

Purpose To compare the diagnostic accuracy of gallium 68 (68Ga)–labeled prostate-specific membrane antigen (PSMA)–11 PET/MRI with that of multiparametric MRI in the detection of prostate cancer. Materials and Methods The authors performed a retrospective study of men with biopsy-proven prostate cancer who underwent simultaneous 68Ga-PSMA-11 PET/MRI before radical prostatectomy between December 2015 and June 2017. The reference standard was whole-mount pathologic examination. Readers were blinded to radiologic and pathologic findings. Tumor localization was based on 30 anatomic regions. Region-specific sensitivity and specificity were calculated for PET/MRI and multiparametric MRI by using raw stringent and alternative neighboring approaches. Maximum standardized uptake value (SUVmax) in the tumor and Prostate Imaging Reporting and Data System (PI-RADS) version 2 grade were compared with tumor Gleason score. Generalized estimating equations were used to estimate population-averaged sensitivity and specificity and to determine the association between tumor characteristics and SUVmax…

Purpose To determine if postmortem computed tomography (CT) and postmortem CT angiography help to detect more lesions than autopsy in postmortem examinations, to evaluate the strengths and weaknesses of each method, and to define their indications. Materials and Methods Postmortem CT angiography was performed on 500 human corpses and followed by conventional autopsy. Nine centers were involved. All CT images were read by an experienced team including one forensic pathologist and one radiologist, blinded to the autopsy results. All findings were recorded for each method and categorized by anatomic structure (bone, organ parenchyma, soft tissue, and vascular) and relative importance in the forensic case (essential, useful, and unimportant). Results Among 18 654 findings, autopsies helped to identify 61.3% (11 433 of 18 654), postmortem CT helped to identify 76.0% (14 179 of 18 654), and postmortem CT angiography helped to identify 89.9% (16 780 of 18 654; P < .001). Postmortem CT angiography was superior to autopsy, especially at helping to identify essential skeletal lesions (96.1%…

Purpose To evaluate the performance of major features, ancillary features, and categories of Liver Imaging Reporting and Data System (LI-RADS) version 2014 at magnetic resonance (MR) imaging for the diagnosis of hepatocellular carcinoma (HCC). Materials and Methods This retrospective institutional review board-approved study included patients with liver MR imaging and at least one pathologically proved lesion. Between 2004 and 2016, 102 patients (275 observations including 113 HCCs) met inclusion criteria. Two radiologists independently assessed major and ancillary imaging features for each liver observation and assigned a LI-RADS category. Per-lesion estimates of diagnostic performance of major features, ancillary features, and LI-RADS categories were assessed by using generalized estimating equation models. Results Major features (arterial phase hyperenhancement, washout, capsule, and threshold growth) had a sensitivity of 88.5%, 60.6%, 32.9%, and 41.6%, and a specificity of 18.6%, 84.8%, 98.8%, and 83.2% for HCC, respectively. Ancillary features (mild-moderate T2 hyperintensity,…

Purpose To identify wall enhancement patterns on vessel wall MRI that discriminate between stable and unstable unruptured intracranial aneurysm (UIA). Materials and Methods Patients were included from November 2012 through January 2016. Vessel wall MR images were acquired at 3 T in patients with stable (incidental and nonchanging over 6 months) or unstable (symptomatic or changing over 6 months) UIA. Each aneurysm was evaluated by using a four-grade classification of enhancement: 0, none; 1, focal; 2, thin circumferential; and 3, thick (>1 mm) circumferential. Inter- and intrareader agreement for the presence and the grade of enhancement were assessed by using κ statistics and 95% confidence interval (CI). The sensitivity, specificity, and negative and positive predictive values of each enhancement grade for differentiating stable from unstable aneurysms was compared. Results The study included 263 patients with 333 aneurysms. Inter- and intrareader agreement was excellent for both the presence of enhancement (κ values, 0.82 [95% CI: 0.67, 0.99] and 0.87 [95% CI: 0.7, 1.0],…

Purpose To evaluate MRI T1 and T2 mapping with calculation of extracellular volume (ECV) for diagnosis and grading of liver fibrosis. Materials and Methods Different grades of fibrosis were induced in 60 male Sprague-Dawley rats by bile duct ligation (BDL) and carbon-tetrachloride (CCl4) intoxication. Portal pressure was measured invasively, whereas hepatic fibrosis was quantified by hydroxyproline content, Sirius red staining, and α smooth muscle actin staining. T1 values, T2 values, and ECV were assessed by using quantitative MRI mapping techniques. Results T1 values in animals 4 weeks after BDL were greater than in control animals (718 msec ± 74 vs 578 msec ± 33, respectively; P < .001). T2 values at 4 weeks were also greater in animals that underwent BDL than in control animals (46 msec ± 6 vs 29 msec ± 2, respectively; P < .001). Similar T1 and T2 findings were observed after CCl4 intoxication. ECV was greater in animals 4 weeks after BDL compared with control animals (31.3% ± 1.3 vs 18.2% ± 3.5, respectively; P < .001), with similar results after CCl4 intoxication. High…

Purpose To compare different methods for generating features from radiology reports and to develop a method to automatically identify findings in these reports. Materials and Methods In this study, 96 303 head computed tomography (CT) reports were obtained. The linguistic complexity of these reports was compared with that of alternative corpora. Head CT reports were preprocessed, and machine-analyzable features were constructed by using bag-of-words (BOW), word embedding, and Latent Dirichlet allocation–based approaches. Ultimately, 1004 head CT reports were manually labeled for findings of interest by physicians, and a subset of these were deemed critical findings. Lasso logistic regression was used to train models for physician-assigned labels on 602 of 1004 head CT reports (60%) using the constructed features, and the performance of these models was validated on a held-out 402 of 1004 reports (40%). Models were scored by area under the receiver operating characteristic curve (AUC), and aggregate AUC statistics were reported for (a) all labels, (b) critical labels, and (c) the…