The 50 most-cited articles in Radiology, 2019

MRI of the breast has the highest sensitivity for breast cancer detection among current clinical imaging modalities and is indispensable for breast imaging practice. While the basis of breast MRI consists of T1-weighted contrast-enhanced imaging, T2-weighted, ultrafast, and diffusion-weighted imaging may be used to improve lesion characterization. Such multiparametric assessment of breast lesions allows for excellent discrimination between benign and malignant breast lesions. Indications for breast MRI are expanding. In preoperative staging, multiple studies confirm the superiority of MRI to other imaging modalities for tumor size estimation and detection of additional tumor foci in the ipsilateral and contralateral breast. Ongoing studies show that in experienced hands this can be used to improve breast cancer surgery, although there is no evidence of improved long-term outcomes. Screening indications are likewise growing as evidence is accumulating that OncologicRI depicts cancers at an earlier stage than mammography in all women. To manage the associated costs for screening, the…

Background Mammographic density improves the accuracy of breast cancer risk models. However, the use of breast density is limited by subjective assessment, variation across radiologists, and restricted data. A mammography-based deep learning (DL) model may provide more accurate risk prediction. Purpose To develop a mammography-based DL breast cancer risk model that is more accurate than established clinical breast cancer risk models. Materials and Methods This retrospective study included 88 994 consecutive screening mammograms in 39 571 women between January 1, 2009, and December 31, 2012. For each patient, all examinations were assigned to either training, validation, or test sets, resulting in 71 689, 8554, and 8751 examinations, respectively. Cancer outcomes were obtained through linkage to a regional tumor registry. By using risk factor information from patient questionnaires and electronic medical records review, three models were developed to assess breast cancer risk within 5 years: a risk-factor-based logistic regression model (RF-LR) that used traditional risk factors, a…

Cystic renal cell carcinoma (RCC) is almost certainly overdiagnosed and overtreated. Efforts to diagnose and treat RCC at a curable stage result in many benign neoplasms and indolent cancers being resected without clear benefit. This is especially true for cystic masses, which compared with solid masses are more likely to be benign and, when malignant, less aggressive. For more than 30 years, the Bosniak classification has been used to stratify the risk of malignancy in cystic renal masses. Although it is widely used and still effective, the classification does not formally incorporate masses identified at MRI or US or masses that are incompletely characterized but are highly likely to be benign, and it is affected by interreader variability and variable reported malignancy rates. The Bosniak classification system cannot fully differentiate aggressive from indolent cancers and results in many benign masses being resected. This proposed update to the Bosniak classification addresses some of these shortcomings. The primary modifications incorporate MRI, establish definitions for…

Deep learning has rapidly advanced in various fields within the past few years and has recently gained particular attention in the radiology community. This article provides an introduction to deep learning technology and presents the stages that are entailed in the design process of deep learning radiology research. In addition, the article details the results of a survey of the application of deep learning-specifically, the application of convolutional neural networks-to radiologic imaging that was focused on the following five major system organs: chest, breast, brain, musculoskeletal system, and abdomen and pelvis. The survey of the studies is followed by a discussion about current challenges and future trends and their potential implications for radiology. This article may be used as a guide for radiologists planning research in the field of radiologic image analysis using convolutional neural networks.

The Ovarian-Adnexal Reporting and Data System (O-RADS) US risk stratification and management system is designed to provide consistent interpretations, to decrease or eliminate ambiguity in US reports resulting in a higher probability of accuracy in assigning risk of malignancy to ovarian and other adnexal masses, and to provide a management recommendation for each risk category. It was developed by an international multidisciplinary committee sponsored by the American College of Radiology and applies the standardized reporting tool for US based on the 2018 published lexicon of the O-RADS US working group. For risk stratification, the O-RADS US system recommends six categories (O-RADS 0-5), incorporating the range of normal to high risk of malignancy. This unique system represents a collaboration between the pattern-based approach commonly used in North America and the widely used, European-based, algorithmic-style International Ovarian Tumor Analysis (IOTA) Assessment of Different Neoplasias in the Adnexa model system, a risk prediction model that has undergone successful…

Background Radiomics extracts features from medical images more precisely and more accurately than visual assessment. However, radiomics features are affected by CT scanner parameters such as reconstruction kernel or section thickness, thus obscuring underlying biologically important texture features. Purpose To investigate whether a compensation method could correct for the variations of radiomic feature values caused by using different CT protocols. Materials and Methods Phantom data involving 10 texture patterns and 74 patients in cohorts 1 (19 men; 42 patients; mean age, 60.4 years; September-October 2013) and 2 (16 men; 32 patients; mean age, 62.1 years; January-September 2007) scanned by using different CT protocols were retrospectively included. For any radiomic feature, the compensation approach identified a protocol-specific transformation to express all data in a common space that were devoid of protocol effects. The differences in statistical distributions between protocols were assessed by using Friedman tests before and after compensation. Principal component analyses…

Background Commercial low-field-strength MRI systems are generally not equipped with state-of-the-art MRI hardware, and are not suitable for demanding imaging techniques. An MRI system was developed that combines low field strength (0.55 T) with high-performance imaging technology. Purpose To evaluate applications of a high-performance low-field-strength MRI system, specifically MRI-guided cardiovascular catheterizations with metallic devices, diagnostic imaging in high-susceptibility regions, and efficient image acquisition strategies. Materials and Methods A commercial 1.5-T MRI system was modified to operate at 0.55 T while maintaining high-performance hardware, shielded gradients (45 mT/m; 200 T/m/sec), and advanced imaging methods. MRI was performed between January 2018 and April 2019. T1, T2, and T2* were measured at 0.55 T; relaxivity of exogenous contrast agents was measured; and clinical applications advantageous at low field were evaluated. Results There were 83 0.55-T MRI examinations performed in study participants (45 women; mean age, 34 years ± 13). On average, T1 was…

This is a condensed summary of an international multisociety statement on ethics of artificial intelligence (AI) in radiology produced by the ACR, European Society of Radiology, RSNA, Society for Imaging Informatics in Medicine, European Society of Medical Imaging Informatics, Canadian Association of Radiologists, and American Association of Physicists in Medicine. AI has great potential to increase efficiency and accuracy throughout radiology, but it also carries inherent pitfalls and biases. Widespread use of AI-based intelligent and autonomous systems in radiology can increase the risk of systemic errors with high consequence and highlights complex ethical and societal issues. Currently, there is little experience using AI for patient care in diverse clinical settings. Extensive research is needed to understand how to best deploy AI in clinical practice. This statement highlights our consensus that ethical use of AI in radiology should promote well-being, minimize harm, and ensure that the benefits and harms are distributed among stakeholders in a just manner. We believe AI…

Background Deep learning (DL) algorithms are gaining extensive attention for their excellent performance in image recognition tasks. DL models can automatically make a quantitative assessment of complex medical image characteristics and achieve increased accuracy in diagnosis with higher efficiency. Purpose To determine the feasibility of using a DL approach to predict clinically negative axillary lymph node metastasis from US images in patients with primary breast cancer. Materials and Methods A data set of US images in patients with primary breast cancer with clinically negative axillary lymph nodes from Tongji Hospital (974 imaging studies from 2016 to 2018, 756 patients) and an independent test set from Hubei Cancer Hospital (81 imaging studies from 2018 to 2019, 78 patients) were collected. Axillary lymph node status was confirmed with pathologic examination. Three different convolutional neural networks (CNNs) of Inception V3, Inception-ResNet V2, and ResNet-101 architectures were trained on 90% of the Tongji Hospital data set and tested on the remaining 10%, as well as on the…

Imaging research laboratories are rapidly creating machine learning systems that achieve expert human performance using open-source methods and tools. These artificial intelligence systems are being developed to improve medical image reconstruction, noise reduction, quality assurance, triage, segmentation, computer-aided detection, computer-aided classification, and radiogenomics. In August 2018, a meeting was held in Bethesda, Maryland, at the National Institutes of Health to discuss the current state of the art and knowledge gaps and to develop a roadmap for future research initiatives. Key research priorities include: 1, new image reconstruction methods that efficiently produce images suitable for human interpretation from source data; 2, automated image labeling and annotation methods, including information extraction from the imaging report, electronic phenotyping, and prospective structured image reporting; 3, new machine learning methods for clinical imaging data, such as tailored, pretrained model architectures, and federated machine learning methods; 4, machine learning…

Background Abdominal and thoracic CT provide a valuable opportunity for osteoporosis screening regardless of the clinical indication for imaging. Purpose To establish reference normative ranges for first lumbar vertebra (L1) trabecular attenuation values across all adult ages to measure bone mineral density (BMD) at routine CT. Materials and Methods Reference data were constructed from 20 374 abdominal and/or thoracic CT examinations performed at 120 kV. Data were derived from adults (mean age, 60 years ± 12 [standard deviation]; 56.1% [11 428 of 20 374] women). CT examinations were performed with (n = 4263) or without (n = 16 111) intravenous contrast agent administration for a variety of unrelated clinical indications between 2000 and 2018. L1 Hounsfield unit measurement was obtained either with a customized automated tool (n = 11 270) or manually by individual readers (n = 9104). The effects of patient age, sex, contrast agent, and manual region-of-interest versus fully automated L1 Hounsfield unit measurement were assessed using multivariable logistic regression analysis.…

Background Nasopharyngeal carcinoma (NPC) may be cured with radiation therapy. Tumor proximity to critical structures demands accuracy in tumor delineation to avoid toxicities from radiation therapy; however, tumor target contouring for head and neck radiation therapy is labor intensive and highly variable among radiation oncologists. Purpose To construct and validate an artificial intelligence (AI) contouring tool to automate primary gross tumor volume (GTV) contouring in patients with NPC. Materials and Methods In this retrospective study, MRI data sets covering the nasopharynx from 1021 patients (median age, 47 years; 751 male, 270 female) with NPC between September 2016 and September 2017 were collected and divided into training, validation, and testing cohorts of 715, 103, and 203 patients, respectively. GTV contours were delineated for 1021 patients and were defined by consensus of two experts. A three-dimensional convolutional neural network was applied to 818 training and validation MRI data sets to construct the AI tool, which was tested in 203 independent MRI data sets.…

Hyperpolarized (HP) carbon 13 (13C) MRI is an emerging molecular imaging method that allows rapid, noninvasive, and pathway-specific investigation of dynamic metabolic and physiologic processes that were previously inaccessible to imaging. This technique has enabled real-time in vivo investigations of metabolism that are central to a variety of diseases, including cancer, cardiovascular disease, and metabolic diseases of the liver and kidney. This review provides an overview of the methods of hyperpolarization and 13C probes investigated to date in preclinical models of disease. The article then discusses the progress that has been made in translating this technology for clinical investigation. In particular, the potential roles and emerging clinical applications of HP [1-13C]pyruvate MRI will be highlighted. The future directions to enable the adoption of this technology to advance the basic understanding of metabolism, to improve disease diagnosis, and to accelerate treatment assessment are also detailed. ©RSNA, 2019 Online supplemental material is available for this article.

Acknowledging the increasingly important role of whole-body MRI for directing patient care in myeloma, a multidisciplinary, international, and expert panel of radiologists, medical physicists, and hematologists with specific expertise in whole-body MRI in myeloma convened to discuss the technical performance standards, merits, and limitations of currently available imaging methods. Following guidance from the International Myeloma Working Group and the National Institute for Clinical Excellence in the United Kingdom, the Myeloma Response Assessment and Diagnosis System (or MY-RADS) imaging recommendations are designed to promote standardization and diminish variations in the acquisition, interpretation, and reporting of whole-body MRI in myeloma and allow response assessment. This consensus proposes a core clinical protocol for whole-body MRI and an extended protocol for advanced assessments. Published under a CC BY 4.0 license. Online supplemental material is available for this article.

Background Men suspected of having clinically significant prostate cancer (sPC) increasingly undergo prostate MRI. The potential of deep learning to provide diagnostic support for human interpretation requires further evaluation. Purpose To compare the performance of clinical assessment to a deep learning system optimized for segmentation trained with T2-weighted and diffusion MRI in the task of detection and segmentation of lesions suspicious for sPC. Materials and Methods In this retrospective study, T2-weighted and diffusion prostate MRI sequences from consecutive men examined with a single 3.0-T MRI system between 2015 and 2016 were manually segmented. Ground truth was provided by combined targeted and extended systematic MRI–transrectal US fusion biopsy, with sPC defined as International Society of Urological Pathology Gleason grade group greater than or equal to 2. By using split-sample validation, U-Net was internally validated on the training set (80% of the data) through cross validation and subsequently externally validated on the test set (20% of the data). U-Net–derived…

Osteoarthritis (OA) of the hip and knee is among the most common joint disorders. Intra-articular corticosteroid (IACS) injections are frequently performed to treat OA and other joint-related pain syndromes; however, there is conflicting evidence on their potential benefit. There is a lack of prospective and large retrospective studies evaluating potential joint findings, including increased risk for accelerated OA progression or adverse joint events, after treatment with IACS injection. Four main adverse joint findings have been structurally observed in patients after IACS injections: accelerated OA progression, subchondral insufficiency fracture, complications of osteonecrosis, and rapid joint destruction, including bone loss. Physicians, including radiologists, should be familiar with imaging findings and patient characteristics that may help them identify potential joints at risk for such events. The purpose of this report is to review the existing literature, describe observed adverse joint events after IACS injections, and provide an outlook on how this may affect clinical…

Purpose To develop and test an artificial intelligence (AI) system, based on deep convolutional neural networks (CNNs), for automated real-time triaging of adult chest radiographs on the basis of the urgency of imaging appearances. Materials and Methods An AI system was developed by using 470 388 fully anonymized institutional adult chest radiographs acquired from 2007 to 2017. The free-text radiology reports were preprocessed by using an in-house natural language processing (NLP) system modeling radiologic language. The NLP system analyzed the free-text report to prioritize each radiograph as critical, urgent, nonurgent, or normal. An AI system for computer vision using an ensemble of two deep CNNs was then trained by using labeled radiographs to predict the clinical priority from radiologic appearances only. The system's performance in radiograph prioritization was tested in a simulation by using an independent set of 15 887 radiographs. Prediction performance was assessed with the area under the receiver operating characteristic curve; sensitivity, specificity, positive…

Although computer-aided diagnosis (CAD) is widely used in mammography, conventional CAD programs that use prompts to indicate potential cancers on the mammograms have not led to an improvement in diagnostic accuracy. Because of the advances in machine learning, especially with use of deep (multilayered) convolutional neural networks, artificial intelligence has undergone a transformation that has improved the quality of the predictions of the models. Recently, such deep learning algorithms have been applied to mammography and digital breast tomosynthesis (DBT). In this review, the authors explain how deep learning works in the context of mammography and DBT and define the important technical challenges. Subsequently, they discuss the current status and future perspectives of artificial intelligence-based clinical applications for mammography, DBT, and radiomics. Available algorithms are advanced and approach the performance of radiologists-especially for cancer detection and risk prediction at mammography. However, clinical validation is largely lacking, and it is not clear how the…

Background Intratumor heterogeneity in lung cancer may influence outcomes. CT radiomics seeks to assess tumor features to provide detailed imaging features. However, CT radiomic features vary according to the reconstruction kernel used for image generation. Purpose To investigate the effect of different reconstruction kernels on radiomic features and assess whether image conversion using a convolutional neural network (CNN) could improve reproducibility of radiomic features between different kernels. Materials and Methods In this retrospective analysis, patients underwent non–contrast material–enhanced and contrast material–enhanced axial chest CT with soft kernel (B30f) and sharp kernel (B50f) reconstruction using a single CT scanner from April to June 2017. To convert different kernels without sinogram, the CNN model was developed using residual learning and an end-to-end way. Kernel-converted images were generated, from B30f to B50f and from B50f to B30f. Pulmonary nodules or masses were semiautomatically segmented and 702 radiomic features (tumor intensity, texture, and wavelet…

Background Deep learning has the potential to augment the use of chest radiography in clinical radiology, but challenges include poor generalizability, spectrum bias, and difficulty comparing across studies. Purpose To develop and evaluate deep learning models for chest radiograph interpretation by using radiologist-adjudicated reference standards. Materials and Methods Deep learning models were developed to detect four findings (pneumothorax, opacity, nodule or mass, and fracture) on frontal chest radiographs. This retrospective study used two data sets. Data set 1 (DS1) consisted of 759 611 images from a multicity hospital network and ChestX-ray14 is a publicly available data set with 112 120 images. Natural language processing and expert review of a subset of images provided labels for 657 954 training images. Test sets consisted of 1818 and 1962 images from DS1 and ChestX-ray14, respectively. Reference standards were defined by radiologist-adjudicated image review. Performance was evaluated by area under the receiver operating characteristic curve analysis, sensitivity,…

Background Results of recent phantom studies show that variation in CT acquisition parameters and reconstruction techniques may make radiomic features largely nonreproduceable and of limited use for prognostic clinical studies. Purpose To investigate the effect of CT radiation dose and reconstruction settings on the reproducibility of radiomic features, as well as to identify correction factors for mitigating these sources of variability. Materials and Methods This was a secondary analysis of a prospective study of metastatic liver lesions in patients who underwent staging with single-energy dual-source contrast material–enhanced staging CT between September 2011 and April 2012. Technique parameters were altered, resulting in 28 CT data sets per patient that included different dose levels, section thicknesses, kernels, and reconstruction algorithm settings. By using a training data set (n = 76), reproducible intensity, shape, and texture radiomic features (reproducibility threshold, R2 ≥ 0.95) were selected and correction factors were calculated by using a linear model to convert…

High-quality evidence shows that MRI in biopsy-naive men can reduce the number of men who need prostate biopsy and can reduce the number of diagnoses of clinically insignificant cancers that are unlikely to cause harm. In men with prior negative biopsy results who remain under persistent suspicion, MRI improves the detection and localization of life-threatening prostate cancer with greater clinical utility than the current standard of care, systematic transrectal US-guided biopsy. Systematic analyses show that MRI-directed biopsy increases the effectiveness of the prostate cancer diagnosis pathway. The incorporation of MRI-directed pathways into clinical care guidelines in prostate cancer detection has begun. The widespread adoption of the Prostate Imaging Reporting and Data System (PI-RADS) for multiparametric MRI data acquisition, interpretation, and reporting has promoted these changes in practice. The PI-RADS MRI-directed biopsy pathway enables the delivery of key diagnostic benefits to men suspected of having cancer based on clinical suspicion. Herein, the PI-RADS Steering…

Background Multicenter studies are required to validate the added benefit of using deep convolutional neural network (DCNN) software for detecting malignant pulmonary nodules on chest radiographs. Purpose To compare the performance of radiologists in detecting malignant pulmonary nodules on chest radiographs when assisted by deep learning–based DCNN software with that of radiologists or DCNN software alone in a multicenter setting. Materials and Methods Investigators at four medical centers retrospectively identified 600 lung cancer–containing chest radiographs and 200 normal chest radiographs. Each radiograph with a lung cancer had at least one malignant nodule confirmed by CT and pathologic examination. Twelve radiologists from the four centers independently analyzed the chest radiographs and marked regions of interest. Commercially available deep learning–based computer-aided detection software separately trained, tested, and validated with 19 330 radiographs was used to find suspicious nodules. The radiologists then reviewed the images with the assistance of DCNN software. The…

Digital breast tomosynthesis (DBT) is emerging as the standard of care for breast imaging based on improvements in both screening and diagnostic imaging outcomes. The additional information obtained from the tomosynthesis acquisition decreases the confounding effect of overlapping tissue, allowing for improved lesion detection, characterization, and localization. In addition, the quasi three-dimensional information obtained from the reconstructed DBT data set allows a more efficient imaging work-up than imaging with two-dimensional full-field digital mammography alone. Herein, the authors review the benefits of DBT imaging in screening and diagnostic breast imaging.

In this article, the authors discuss the technical background and summarize the current body of literature regarding virtual monoenergetic (VM) images derived from dual-energy CT data, which can be reconstructed between 40 and 200 keV. Substantially improved iodine attenuation at lower kiloelectron volt levels and reduced beam-hardening artifacts at higher kiloelectron volt levels have been demonstrated from all major manufacturers of dual-energy CT units. Improved contrast attenuation with VM imaging at lower kiloelectron volt levels enables better delineation and diagnostic accuracy in the detection of various vascular or oncologic abnormalities. Low-kiloelectron-volt VM imaging may be useful for salvaging CT studies with suboptimal contrast material delivery or providing additional information on the arterial vasculature obtained from venous phase acquisitions. For patients with renal impairment, substantial reductions in the use of iodinated contrast material can be achieved by using lower-energy VM imaging. The authors recommend routine reconstruction of VM images at 50 keV…

Due to the exponential growth of computational algorithms, artificial intelligence (AI) methods are poised to improve the precision of diagnostic and therapeutic methods in medicine. The field of radiomics in neuro-oncology has been and will likely continue to be at the forefront of this revolution. A variety of AI methods applied to conventional and advanced neuro-oncology MRI data can already delineate infiltrating margins of diffuse gliomas, differentiate pseudoprogression from true progression, and predict recurrence and survival better than methods used in daily clinical practice. Radiogenomics will also advance our understanding of cancer biology, allowing noninvasive sampling of the molecular environment with high spatial resolution and providing a systems-level understanding of underlying heterogeneous cellular and molecular processes. By providing in vivo markers of spatial and molecular heterogeneity, these AI-based radiomic and radiogenomic tools have the potential to stratify patients into more precise initial diagnostic and therapeutic pathways and enable better dynamic…

Purpose To evaluate MRI features associated with pathologically defined extraprostatic extension (EPE) of prostate cancer and to propose an MRI grading system for pathologic EPE. Materials and Methods In this prospective study, consecutive male study participants underwent preoperative 3.0-T MRI from June 2007 to March 2017 followed by robotic-assisted laparoscopic radical prostatectomy. An MRI-based EPE grading system was defined as follows: curvilinear contact length of 1.5 cm or capsular bulge and irregularity were grade 1, both features were grade 2, and frank capsular breach were grade 3. Multivariable logistic regression and decision curve analyses were performed to compare the MRI grade model and clinical parameters (prostate-specific antigen, Gleason score) for pathologic EPE prediction by using the area under the receiver operating characteristic curve (AUC) value. Results Among 553 study participants, the mean age was 60 years ± 8 (standard deviation); the median prostate-specific antigen value was 6.3 ng/mL. A total of 125 of 553 (22%) participants had pathologic EPE at…

Background Renal impairment is common in patients with coronary artery disease and, if severe, late gadolinium enhancement (LGE) imaging for myocardial infarction (MI) evaluation cannot be performed. Purpose To develop a fully automatic framework for chronic MI delineation via deep learning on non–contrast material–enhanced cardiac cine MRI. Materials and Methods In this retrospective single-center study, a deep learning model was developed to extract motion features from the left ventricle and delineate MI regions on nonenhanced cardiac cine MRI collected between October 2015 and March 2017. Patients with chronic MI, as well as healthy control patients, had both nonenhanced cardiac cine (25 phases per cardiac cycle) and LGE MRI examinations. Eighty percent of MRI examinations were used for the training data set and 20% for the independent testing data set. Chronic MI regions on LGE MRI were defined as ground truth. Diagnostic performance was assessed by analysis of the area under the receiver operating characteristic curve (AUC). MI area and MI area percentage from nonenhanced…

The development and widespread adoption of iterative reconstruction (IR) algorithms for CT have greatly facilitated the contemporary practice of radiation dose reduction during abdominal CT examinations. IR mitigates the increased image noise typically associated with reduced radiation dose levels, thereby maintaining subjective image quality and diagnostic confidence for a variety of clinical tasks. Mounting evidence, however, points to important limitations of this method involving radiologists' ability to perform low-contrast diagnostic tasks, such as the detection of liver metastases or pancreatic masses. Radiologists need to be aware that use of IR can result in a decline of spatial resolution for low-contrast structures and degradation of low-contrast detectability when radiation dose reductions exceed approximately 25%. This article will review the principles of IR algorithm technology, describe the various commercial implementations of IR in CT, and review published studies that have evaluated the ability of IR to preserve diagnostic performance for low-contrast diagnostic…

Background Fluorine 18 (18F)−fluorodeoxyglucose (FDG) PET/CT is a routine tool for staging patients with lymphoma and lung cancer. Purpose To evaluate configurations of deep convolutional neural networks (CNNs) to localize and classify uptake patterns of whole-body 18F-FDG PET/CT images in patients with lung cancer and lymphoma. Materials and Methods This was a retrospective analysis of consecutive patients with lung cancer or lymphoma referred to a single center from August 2011 to August 2013. Two nuclear medicine experts manually delineated foci with increased 18F-FDG uptake, specified the anatomic location, and classified these findings as suspicious for tumor or metastasis or nonsuspicious. By using these expert readings as the reference standard, a CNN was developed to detect foci positive for 18F-FDG uptake, predict the anatomic location, and determine the expert classification. Examinations were divided into independent training (60%), validation (20%), and test (20%) subsets. Results This study included 629 patients (mean age, 52.2 years ± 20.4 [standard deviation]; 394…

Background The Vesical Imaging-Reporting and Data System (VI-RADS) scoring system was created in 2018 to standardize imaging and reporting of bladder cancer staging with multiparametric MRI. The system provides a five-point VI-RADS score, which suggests the likelihood of detrusor muscle invasion. Muscle-invasive disease carries a worse prognosis and requires radical surgery. Purpose To determine the performance of the VI-RADS score in detecting muscle-invasive bladder cancer in a cohort of patients undergoing multiparametric MRI before surgery. Materials and Methods In this retrospective study, a total of 340 patients with bladder cancer were identified from a database of consecutive patients undergoing multiparametric MRI from November 2011 to August 2018. The tumor with the largest burden was selected in those patients with multifocal tumors. Bladder tumors were retrospectively categorized according to the VI-RADS five-point scoring system by two readers, independently and in consensus, who were blinded to histologic findings. The VI-RADS score was compared with postoperative…

Background Recent deep learning (DL) approaches have shown promise in improving sensitivity but have not addressed limitations in radiologist specificity or efficiency. Purpose To develop a DL model to triage a portion of mammograms as cancer free, improving performance and workflow efficiency. Materials and Methods In this retrospective study, 223 109 consecutive screening mammograms performed in 66 661 women from January 2009 to December 2016 were collected with cancer outcomes obtained through linkage to a regional tumor registry. This cohort was split by patient into 212 272, 25 999, and 26 540 mammograms from 56 831, 7021, and 7176 patients for training, validation, and testing, respectively. A DL model was developed to triage mammograms as cancer free and evaluated on the test set. A DL-triage workflow was simulated in which radiologists skipped mammograms triaged as cancer free (interpreting them as negative for cancer) and read mammograms not triaged as cancer free by using the original interpreting radiologists’ assessments. Sensitivities, specificities, and percentage of…

Background Management of thyroid nodules may be inconsistent between different observers and time consuming for radiologists. An artificial intelligence system that uses deep learning may improve radiology workflow for management of thyroid nodules. Purpose To develop a deep learning algorithm that uses thyroid US images to decide whether a thyroid nodule should undergo a biopsy and to compare the performance of the algorithm with the performance of radiologists who adhere to American College of Radiology (ACR) Thyroid Imaging Reporting and Data System (TI-RADS). Materials and Methods In this retrospective analysis, studies in patients referred for US with subsequent fine-needle aspiration or with surgical histologic analysis used as the standard were evaluated. The study period was from August 2006 to May 2010. A multitask deep convolutional neural network was trained to provide biopsy recommendations for thyroid nodules on the basis of two orthogonal US images as the input. In the training phase, the deep learning algorithm was first evaluated by using 10-fold cross-validation.…

Background Obesity has been associated with increased risk of accelerated cognitive decline and dementia, which suggests underlying neurobiological changes. Purpose To investigate the associations between obesity and brain structure (overall and regional brain volumes, and white matter microstructure) assessed at MRI in a sample of the general population. Materials and Methods Between March 2014 and January 2018, 12 087 participants (52.8% women [6381 of 12 087]; mean age, 62 years; age range, 45–76 years) in the prospective observational UK Biobank study underwent 3.0-T multiparametric (ie, three-dimensional T1-weighted diffusion tensor imaging [DTI]) brain imaging. Percentage of total body fat (TBF) was assessed by body impedance. Volumetric measures included brain volume, gray matter volume, white matter volume, volumes of subcortical gray matter structures, and regional cortical volumes. Global and tract-specific microstructure was assessed by fractional anisotropy (FA) and mean diffusivity (MD) by using DTI. Linear regression was performed by using TBF as determinant and brain…

Background Contrast agent–enhanced digital mammography (CEDM) has been shown to be more sensitive and specific than two-dimensional full-field digital mammography in the diagnostic setting. Few studies have reported on its performance in the screening setting. Purpose To evaluate the performance of CEDM for breast cancer screening. Materials and Methods This retrospective study included women who underwent dual-energy CEDM for breast cancer screening from December 2012 through April 2016. Medical records were reviewed for age, risk factors, short-interval follow-up and biopsies recommended, and cancers detected. Sensitivity, specificity, positive predictive value of abnormal findings at screening (PPV1), positive predictive value of biopsy performed (PPV3), and negative predictive value were determined. Results In the study period 904 baseline CEDMs were performed. Mean age was 51.8 years ± 9.4 (standard deviation). Of 904 patients, 700 (77.4%) had dense breasts, 247 (27.3%) had a family history of breast cancer in a first-degree relative age 50 years or younger, and 363 (40.2%) a…

Lung cancer remains the leading cause of cancer mortality in the United States. Lung cancer screening (LCS) with low-dose CT reduces mortality among high-risk current and former smokers and has been covered by public and private insurers without cost sharing since 2015. Patients and referring providers confront numerous barriers to participation in screening. To best serve in multidisciplinary efforts to expand LCS nationwide, radiologists must be knowledgeable of these challenges. A better understanding of the difficulties confronted by other stakeholders will help radiologists continue to collaboratively guide the growth of LCS programs in their communities. This article reviews barriers to participation in LCS for patients and referring providers, as well as possible solutions and interventions currently underway.

Background Computational models on the basis of deep neural networks are increasingly used to analyze health care data. However, the efficacy of traditional computational models in radiology is a matter of debate. Purpose To evaluate the accuracy and efficiency of a combined machine and deep learning approach for early breast cancer detection applied to a linked set of digital mammography images and electronic health records. Materials and Methods In this retrospective study, 52 936 images were collected in 13 234 women who underwent at least one mammogram between 2013 and 2017, and who had health records for at least 1 year before undergoing mammography. The algorithm was trained on 9611 mammograms and health records of women to make two breast cancer predictions: to predict biopsy malignancy and to differentiate normal from abnormal screening examinations. The study estimated the association of features with outcomes by using t test and Fisher exact test. The model comparisons were performed with a 95% confidence interval (CI) or by using the DeLong test. Results The resulting…

Advances in virtual immersive and augmented reality technology, commercially available for the entertainment and gaming industry, hold potential for education and clinical use in medicine and the field of medical imaging. Radiology departments have begun exploring the use of these technologies to help with radiology education and clinical care. The purpose of this review article is to summarize how three institutions have explored using virtual and augmented reality for radiology.

Background Managing soft-tissue sarcoma (STS) relies on histologic grade, which is the strongest prognostic factor and a routine assessment at biopsy. However, underestimation of histologic grade may occur because of tumor heterogeneity. Purpose To identify MRI features that are associated with high-grade STS (grade III) and to determine the relationship between MRI features and patient survival. Materials and Methods In this retrospective single-center study, patients (age ≥16 years) were included if they presented with STS diagnosed between 2008 and 2015, had a baseline contrast material–enhanced MRI study, had a pathologic grade assessed on the whole surgical specimen, and had no history of neoadjuvant treatment. Visceral sarcomas, well-differentiated liposarcomas, and angiosarcomas were excluded. Images were evaluated for size, heterogeneity, architecture, margins, and surrounding tissue at T2-weighted, T1-weighted precontrast, and T1-weighted postcontrast MRI. χ2 tests, Fisher tests, and multivariable binary logistic regression were performed to identify features associated…

Background Multicenter studies may be required for establishing guidelines for safe usage of iodinated contrast media (ICM). Purpose To identify the prevalence, patterns, risk factors, and preventive measures for ICM-related hypersensitivity reactions (HSRs). Materials and Methods Between March 2017 and October 2017, a total of 196 081 patients who underwent ICM administration were enrolled from seven participating institutions. The occurrence of HSRs and baseline patient information were recorded. χ2 and Student t test were performed, and logistic regression analyses were used to identify risk factors that predict occurrence and recurrence of HSR. Results Among 196 081 patients (mean age ± standard deviation, 59.1 years ± 16.0; 105 014 men and 91 067 women) who underwent ICM administration, the overall prevalence of HSRs was 0.73% (1433 of 196 081), and severe reactions occurred in 0.01% (17 of 196 081). Conditional logistic regression for patients with HSR (n = 1433) and a control group (1:1 matched group for age, sex, ICM product, and institution) demonstrated that a patient’s…

Background Digital breast tomosynthesis (DBT) is replacing digital mammography (DM) in the clinical workflow. Currently, there are limited prospective studies comparing the diagnostic accuracy of both examinations and the role of synthetic mammography (SM) and computer-aided detection (CAD). Purpose To compare the accuracy of DM versus DM + DBT in population-based breast cancer screening. Materials and Methods This prospective study, performed from November 2010 to December 2012, included 24 301 women (mean age, 59.1 years ± 5.7 [standard deviation]) with 281 cancers, of which 51 were interval cancers. Each examination was independently interpreted with four reading modes: DM, DM + CAD, DM + DBT, and SM + DBT. Sensitivity and specificity were compared for DM versus DM + DBT, DM versus DM + CAD, DM + DBT versus SM + DBT, and DM versus DM + DBT at double reading. Reader-adjusted performance characteristics of reading modes were evaluated on the basis of pre-arbitration (initial interpretation) scores. Statistical analysis was based on cluster bootstrap analysis using 10 000 random…

Background Ferumoxytol is approved for use in the treatment of iron deficiency anemia, but it can serve as an alternative to gadolinium-based contrast agents. On the basis of postmarketing surveillance data, the Food and Drug Administration issued a black box warning regarding the risks of rare but serious acute hypersensitivity reactions during fast high-dose injection (510 mg iron in 17 seconds) for therapeutic use. Whereas single-center safety data for diagnostic use have been positive, multicenter data are lacking. Purpose To report multicenter safety data for off-label diagnostic ferumoxytol use. Materials and Methods The multicenter ferumoxytol MRI registry was established as an open-label nonrandomized surveillance databank without industry involvement. Each center monitored all ferumoxytol administrations, classified adverse events (AEs) using the National Cancer Institute Common Terminology Criteria for Adverse Events (grade 1–5), and assessed the relationship of AEs to ferumoxytol administration. AEs related to or possibly related to ferumoxytol injection were considered…

Background Most risk prediction models for breast cancer are based on questionnaires and mammographic density assessments. By training a deep neural network, further information in the mammographic images can be considered. Purpose To develop a risk score that is associated with future breast cancer and compare it with density-based models. Materials and Methods In this retrospective study, all women aged 40–74 years within the Karolinska University Hospital uptake area in whom breast cancer was diagnosed in 2013–2014 were included along with healthy control subjects. Network development was based on cases diagnosed from 2008 to 2012. The deep learning (DL) risk score, dense area, and percentage density were calculated for the earliest available digital mammographic examination for each woman. Logistic regression models were fitted to determine the association with subsequent breast cancer. False-negative rates were obtained for the DL risk score, age-adjusted dense area, and age-adjusted percentage density. Results A total of 2283 women, 278 of whom were later diagnosed with breast…

Background The performance of a deep learning (DL) algorithm should be validated in actual clinical situations, before its clinical implementation. Purpose To evaluate the performance of a DL algorithm for identifying chest radiographs with clinically relevant abnormalities in the emergency department (ED) setting. Materials and Methods This single-center retrospective study included consecutive patients who visited the ED and underwent initial chest radiography between January 1 and March 31, 2017. Chest radiographs were analyzed with a commercially available DL algorithm. The performance of the algorithm was evaluated by determining the area under the receiver operating characteristic curve (AUC), sensitivity, and specificity at predefined operating cutoffs (high-sensitivity and high-specificity cutoffs). The sensitivities and specificities of the algorithm were compared with those of the on-call radiology residents who interpreted the chest radiographs in the actual practice by using McNemar tests. If there were discordant findings between the algorithm and resident, the…

Imaging plays a central role in the 2018 International Federation of Gynecology and Obstetrics staging system for uterine cervical cancer. The revision calls for a more precise measurement of primary tumor size, best assessed with imaging. Evaluation for abdominopelvic retroperitoneal lymphadenopathy, either with imaging alone or with pathologic analysis, is now also part of staging. Choice of modality depends on the technology available within the practice setting. In high-resource settings, pelvic MRI (to assess tumor size and central pelvic spread) and torso fluorodeoxyglucose PET/CT (to assess lymphadenopathy and distant metastases) are used to assign stage and to plan therapy. In lower-resource settings, analogous modalities are pelvic US and chest radiography. Although imaging is already a part of pretreatment planning in some high-resource settings, its incorporation into assigning stage is a new development.

Background Pattern of emphysema at chest CT, scored visually by using the Fleischner Society system, is associated with physiologic impairment and mortality risk. Purpose To determine whether participant-level emphysema pattern could predict impairment and mortality when classified by using a deep learning method. Materials and Methods This retrospective analysis of Genetic Epidemiology of COPD (COPDGene) study participants enrolled between 2007 and 2011 included those with baseline CT, visual emphysema scores, and survival data through 2018. Participants were partitioned into nonoverlapping sets of 2407 for algorithm training, 100 for validation and parameter tuning, and 7143 for testing. A deep learning algorithm using convolutional neural network and long short-term memory architectures was trained to classify pattern of emphysema according to Fleischner criteria. Deep learning scores were compared with visual scores and clinical parameters including pulmonary function tests. Cox proportional hazard models were used to evaluate relationships between emphysema scores and survival.…

Background Patients with locally advanced pancreatic cancer have a dismal prognosis, with a median overall survival (OS) of 12–14 months with systemic therapies. Irreversible electroporation (IRE), a nonthermal ablative technique, may prolong survival of patients with locally advanced pancreatic cancer. Purpose To investigate the safety and efficacy of percutaneous IRE for locally advanced pancreatic cancer and locally recurring pancreatic cancer in a prospective phase II trial. Materials and Methods Between December 2012 and September 2017, participants with locally advanced pancreatic cancer or postresection local recurrence were prospectively treated with percutaneous CT-guided IRE (ClinicalTrials.gov identifier: NCT01939665). The primary end point was median OS from diagnosis. The target median OS was 11.6 months for participants receiving no induction chemotherapy or gemcitabine-based induction chemotherapy and 14.9 months for those receiving induction 5-fluorouracil, leucovorin, irinotecan, and oxaliplatin (FOLFIRINOX). Results Fifty participants (25 men and 25 women; median…

Diffusion-weighted (DW) MRI is a rapid technique that measures the mobility of water molecules within tissue, reflecting the cellular microenvironment. At DW MRI, breast cancers typically exhibit reduced diffusivity and appear hyperintense to surrounding tissues. On the basis of this characteristic, DW MRI may offer an unenhanced method to detect breast cancer without the costs and safety concerns associated with dynamic contrast material-enhanced MRI, the current reference standard in the setting of high-risk screening. This application of DW MRI has not been widely explored but is particularly timely given the growing health concerns related to the long-term use of gadolinium-based contrast material. Moreover, increasing breast density notification legislation across the United States is raising awareness of the limitations of mammography in women with dense breasts, emphasizing the need for additional cost-effective supplemental screening examinations. Preliminary studies suggest unenhanced MRI with DW MRI may provide higher sensitivity than screening mammography for the…

Background Nonalcoholic fatty liver disease and its consequences are a growing public health concern requiring cross-sectional imaging for noninvasive diagnosis and quantification of liver fat. Purpose To investigate a deep learning–based automated liver fat quantification tool at nonenhanced CT for establishing the prevalence of steatosis in a large screening cohort. Materials and Methods In this retrospective study, a fully automated liver segmentation algorithm was applied to noncontrast abdominal CT examinations from consecutive asymptomatic adults by using three-dimensional convolutional neural networks, including a subcohort with follow-up scans. Automated volume-based liver attenuation was analyzed, including conversion to CT fat fraction, and compared with manual measurement in a large subset of scans. Results A total of 11 669 CT scans in 9552 adults (mean age ± standard deviation, 57.2 years ± 7.9; 5314 women and 4238 men; median body mass index [BMI], 27.8 kg/m2) were evaluated, including 2117 follow-up scans in 1862 adults (mean age, 59.2 years; 971 women and 891 men;…

Background Risk stratification systems for thyroid nodules are often complicated and affected by low specificity. Continual improvement of these systems is necessary to reduce the number of unnecessary thyroid biopsies. Purpose To use artificial intelligence (AI) to optimize the American College of Radiology (ACR) Thyroid Imaging Reporting and Data System (TI-RADS). Materials and Methods A total of 1425 biopsy-proven thyroid nodules from 1264 consecutive patients (1026 women; mean age, 52.9 years [range, 18–93 years]) were evaluated retrospectively. Expert readers assigned points based on five ACR TI-RADS categories (composition, echogenicity, shape, margin, echogenic foci), and a genetic AI algorithm was applied to a training set (1325 nodules). Point and pathologic data were used to create an optimized scoring system (hereafter, AI TI-RADS). Performance of the systems was compared by using a test set of the final 100 nodules with interpretations from the expert reader, eight nonexpert readers, and an expert panel. Initial performance of AI TI-RADS was calculated by using a test…