Deep Learning and Medical Image Analysis for COVID-19 Diagnosis and Prediction.

The coronavirus disease 2019 (COVID-19) pandemic has imposed dramatic challenges to health-care organizations worldwide. To combat the global crisis, the use of thoracic imaging has played a major role in the diagnosis, prediction, and management of COVID-19 patients with moderate to severe symptoms or with evidence of worsening respiratory status. In response, the medical image analysis community acted quickly to develop and disseminate deep learning models and tools to meet the urgent need of managing and interpreting large amounts of COVID-19 imaging data. This review aims to not only summarize existing deep learning and medical image analysis methods but also offer in-depth discussions and recommendations for future investigations. We believe that the wide availability of high-quality, curated, and benchmarked COVID-19 imaging data sets offers the great promise of a transformative test bed to develop, validate, and disseminate novel deep learning methods in the frontiers of data science and artificial intelligence.

Implementing Virtual and Augmented Reality Tools for Radiology Education and Training, Communication, and Clinical Care.

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.

Interpreting Radiographs with Concurrently Obtained Patient Photographs.

A technology for automatically obtaining patient photographs along with portable radiographs was implemented clinically at a large academic hospital. This article highlights several cases in which image-related clinical context, provided by the patient photographs, provided quality control information regarding patient identification, laterality, or position and assisted the radiologist with the interpretation. The information in the photographs can easily minimize unnecessary calls to the patient's nursing staff for clarifications and can lead to new methods of physically assessing patients. Published under a CC BY 4.0 license.

CT-Based 3D Printing of the Glenoid Prior to Shoulder Arthroplasty: Bony Morphology and Model Evaluation.

To demonstrate the 3D printed appearance of glenoid morphologies relevant to shoulder replacement surgery and to evaluate the benefits of printed models of the glenoid with regard to surgical planning. A retrospective review of patients referred for shoulder CT was performed, leading to a cohort of nine patients without arthroplasty hardware and exhibiting glenoid changes relevant to shoulder arthroplasty planning. Thin slice CT images were used to create both humerus-subtracted volume renderings of the glenoid, as well as 3D surface models of the glenoid, and 11 printed models were created. Volume renderings, surface models, and printed models were reviewed by a musculoskeletal radiologist for accuracy. Four fellowship-trained orthopaedic surgeons specializing in shoulder surgery reviewed each case individually as follows: First, the source CT images were reviewed, and a score for the clarity of the bony morphologies relevant to shoulder arthroplasty surgery was given. The volume rendering was reviewed, and the clarity was again scored. Finally, the printed model was reviewed, and the clarity again scored. Each printed model was also scored for morphologic complexity, expected usefulness of the printed model, and physical properties of the model. Mann-Whitney-Wilcoxon signed rank tests of the clarity scores were calculated, and the Spearman's ρ correlation coefficient between complexity and usefulness scores was computed. Printed models demonstrated a range of glenoid bony changes including osteophytes, glenoid bone loss, retroversion, and biconcavity. Surgeons rated the glenoid morphology as more clear after review of humerus-subtracted volume rendering, compared with review of the source CT images (p = 0.00903). Clarity was also better with 3D printed models compared to CT (p = 0.00903) and better with 3D printed models compared to humerus-subtracted volume rendering (p = 0. 00879). The expected usefulness of printed models demonstrated a positive correlation with morphologic complexity, with Spearman's ρ 0.73 (p = 0.0108). 3D printing of the glenoid based on pre-operative CT provides a physical representation of patient anatomy. Printed models enabled shoulder surgeons to appreciate glenoid bony morphology more clearly compared to review of CT images or humerus-subtracted volume renderings. These models were more useful as glenoid complexity increased.

CT Prediction Model for Major Arterial Injury after Blunt Pelvic Ring Disruption.

Purpose To develop and test a computed tomography (CT)-based predictive model for major arterial injury after blunt pelvic ring disruptions that incorporates semiautomated pelvic hematoma volume quantification. Materials and Methods A multivariable logistic regression model was developed in patients with blunt pelvic ring disruptions who underwent arterial phase abdominopelvic CT before angiography from 2008 to 2013. Arterial injury at angiography requiring transarterial embolization (TAE) served as the outcome. Areas under the receiver operating characteristic (ROC) curve (AUCs) for the model and for two trauma radiologists were compared in a validation cohort of 36 patients from 2013 to 2015 by using the Hanley-McNeil method. Hematoma volume cutoffs for predicting the need for TAE and probability cutoffs for the secondary outcome of mortality not resulting from closed head injuries were determined by using ROC analysis. Correlation between hematoma volume and transfusion was assessed by using the Pearson coefficient. Results Independent predictor variables included hematoma volume, intravenous contrast material extravasation, atherosclerosis, rotational instability, and obturator ring fracture. In the validation cohort, the model (AUC, 0.78) had similar performance to reviewers (AUC, 0.69-0.72; P = .40-.80). A hematoma volume cutoff of 433 mL had a positive predictive value of 87%-100% for predicting major arterial injury requiring TAE. Hematoma volumes correlated with units of packed red blood cells transfused (r = 0.34-0.57; P = .0002-.0003). Predicted probabilities of 0.64 or less had a negative predictive value of 100% for excluding mortality not resulting from closed head injuries. Conclusion A logistic regression model incorporating semiautomated hematoma volume segmentation produced objective probability estimates of major arterial injury. Hematoma volumes correlated with 48-hour transfusion requirement, and low predicted probabilities excluded mortality from causes other than closed head injury. © RSNA, 2018 Online supplemental material is available for this article.

Machine Meets Biology: a Primer on Artificial Intelligence in Cardiology and Cardiac Imaging.

PURPOSE OF REVIEW: An understanding of the basics concepts of deep learning can be helpful in not only understanding the potential applications of this technique but also in critically reviewing literature in which neural networks are utilized for analysis and modeling. RECENT FINDINGS: The term "deep learning" has been applied to a subset of machine learning that utilizes a "neural network" and is often used interchangeably with "artificial intelligence." It has been increasingly utilized in healthcare for computational "learning", especially for pattern recognition for diagnostic imaging. Another promising application is the potential for these neural networks to improve the accuracy in the identification of patients who are at risk for cardiovascular events and could benefit most from preventive treatment in comparison with more conventional statistical techniques. The importance of such tailored cardiovascular risk assessment and disease management in individual patients is far reaching given that cardiovascular disease is the leading cause of morbidity and mortality in the world. Nearly half of myocardial infarctions and strokes occur in patients who are not predicted to be at risk for cardiovascular events by current guideline-based approaches. Equally important are individuals who are not at risk for cardiovascular events and yet are given expensive and unnecessary preventive treatment with potential untoward side effects. The application of powerful artificial intelligence/deep learning tools in medicine is likely to result in more effective and efficient health care delivery with the potential for significant cost savings by shifting preventative treatment from inappropriate to appropriate patient subgroups.

RSNA Diagnosis Live: A Novel Web-based Audience Response Tool to Promote Evidence-based Learning.

Audience response systems have become more commonplace in radiology residency programs in the last 10 years, as a means to engage learners and promote improved learning and retention. A variety of systems are currently in use. RSNA Diagnosis Live™ provides unique features that are innovative, particularly for radiology resident education. One specific example is the ability to annotate questions with subspecialty tags, which allows resident performance to be tracked over time. In addition, deficiencies in learning can be monitored for each trainee and analytics can be provided, allowing documentation of resident performance improvement. Finally, automated feedback is given not only to the instructor, but also to the trainee. Online supplemental material is available for this article. © RSNA, 2017.

Use of Radiology Procedure Codes in Health Care: The Need for Standardization and Structure.

Radiology procedure codes are a fundamental part of most radiology workflows, such as ordering, scheduling, billing, and image interpretation. Nonstandardized unstructured procedure codes have typically been used in radiology departments. Such codes may be sufficient for specific purposes, but they offer limited support for interoperability. As radiology workflows and the various forms of clinical data exchange have become more sophisticated, the need for more advanced interoperability with use of standardized structured codes has increased. For example, structured codes facilitate the automated identification of relevant prior imaging studies and the collection of data for radiation dose tracking. The authors review the role of imaging procedure codes in radiology departments and across the health care enterprise. Standards for radiology procedure coding are described, and the mechanisms of structured coding systems are reviewed. In particular, the structure of the RadLex™ Playbook coding system and examples of the use of this system are described. Harmonization of the RadLex Playbook system with the Logical Observation Identifiers Names and Codes standard, which is currently in progress, also is described. The benefits and challenges of adopting standardized codes-especially the difficulties in mapping local codes to standardized codes-are reviewed. Tools and strategies for mitigating these challenges, including the use of billing codes as an intermediate step in mapping, also are reviewed. In addition, the authors describe how to use the RadLex Playbook Web service application programming interface for partial automation of code mapping. © RSNA, 2017.

Radiologist Digital Workspace Use and Preference: a Survey-Based Study.

Literature regarding the heterogeneity of and preferences for radiology workstation design-and, in particular, the digital workspace of the radiology workstation-is scant. The purpose of this study was to determine the nature of the digital environments across the specialty and the degree of satisfaction users associated with the particular facets of those environments. A survey was sent to the membership of the Association of University Radiologists in February 2015. The survey comprised 10 questions establishing demographics, current typical workstation setup, perceived satisfaction with that setup, and preferences for potential altered setups. A total of 336 radiologists responded, with a cross-section similar to that described in the 2015 ACR annual workforce survey (1). Although there was a rough split in the number of radiologists using one or two non-diagnostic monitors (46 vs. 51%, respectively), the strong majority (75%) of radiologists use two diagnostic monitors. Users of two non-diagnostic monitors were more likely to keep open the case info (87 vs. 68%) and EMR (84 vs 68%). More senior radiologists tended to find the current setup easy more frequent than younger radiologists, and the latter group was more likely to believe additional monitors would be helpful. Although many radiologists are comfortable with their computing workflows, a significant number indicate dissatisfaction and may be interested in being able to specify the amount of monitor space with which they can work. Additional monitors may promote improved quality in addition to any potential productivity gains.

Computer-Aided Reporting of Chest Radiographs: Efficient and Effective Screening in the Value-Based Imaging Era.

In the post-PACS era, mammography is unique in adopting specialized ergonomic interfaces to improve efficiency in a high volume setting. Chest radiography is also a high volume area of radiology. The authors hypothesize that applying a novel interface for chest radiography interpretation and reporting could create high productivity while maintaining quality. A custom version of the ClearCanvas open source software, EzRad, was created with a workflow re-designed specifically for tuberculosis screening chest radiographs, which utilized standardized computer generated reports. The preliminary reports from 881,792 studies evaluated by radiology residents over a nine-year period were analyzed for productivity as RVU/FTE and compared to the finalized reports from a subspecialty attending chest radiologist for accuracy. Radiology residents were able to produce 7480 RVU/FTE per year in screening chest radiography productivity when using a custom interface at a large academic medical center with a miss rate of 0.1%. Sensitivity was 77% and specificity was 99.9%. An ergonomic user interface allowed high productivity in interpretation of chest radiography for tuberculosis screening while maintaining quality.

Standardization of terminology in dermoscopy/dermatoscopy: Results of the third consensus conference of the International Society of Dermoscopy.

BACKGROUND: Evolving dermoscopic terminology motivated us to initiate a new consensus. OBJECTIVE: We sought to establish a dictionary of standardized terms. METHODS: We reviewed the medical literature, conducted a survey, and convened a discussion among experts. RESULTS: Two competitive terminologies exist, a more metaphoric terminology that includes numerous terms and a descriptive terminology based on 5 basic terms. In a survey among members of the International Society of Dermoscopy (IDS) 23.5% (n = 201) participants preferentially use descriptive terminology, 20.1% (n = 172) use metaphoric terminology, and 484 (56.5%) use both. More participants who had been initially trained by metaphoric terminology prefer using descriptive terminology than vice versa (9.7% vs 2.6%, P < .001). Most new terms that were published since the last consensus conference in 2003 were unknown to the majority of the participants. There was uniform consensus that both terminologies are suitable, that metaphoric terms need definitions, that synonyms should be avoided, and that the creation of new metaphoric terms should be discouraged. The expert panel proposed a dictionary of standardized terms taking account of metaphoric and descriptive terms. LIMITATIONS: A consensus seeks a workable compromise but does not guarantee its implementation. CONCLUSION: The new consensus provides a revised framework of standardized terms to enhance the consistent use of dermoscopic terminology.

Focused Decision Support: a Data Mining Tool to Query the Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial Dataset and Guide Screening Management for the Individual Patient.

The Prostate, Lung, Colorectal, and Ovarian Cancer (PLCO) Screening Trial enrolled ~155,000 participants to determine whether certain screening exams reduced mortality from prostate, lung, colorectal, and ovarian cancer. Repurposing the data provides an unparalleled resource for matching patients with the outcomes of demographically or diagnostically comparable patients. A web-based application was developed to query this subset of patient information against a given patient's demographics and risk factors. Analysis of the matched data yields outcome information which can then be used to guide management decisions and imaging software. Prognostic information is also estimated via the proportion of matched patients that progress to cancer. The US Preventative Services Task Force provides screening recommendations for cancers of the breast, colorectal tract, and lungs. There is wide variability in adherence of clinicians to these guidelines and others published by the Fleischner Society and various cancer organizations. Data mining the PLCO dataset for clinical decision support can optimize the use of limited healthcare resources, focusing screening on patients for whom the benefit to risk ratio is the greatest and most efficacious. A data driven, personalized approach to cancer screening maximizes the economic and clinical efficacy and enables early identification of patients in which the course of disease can be improved. Our dynamic decision support system utilizes a subset of the PLCO dataset as a reference model to determine imaging and testing appropriateness while offering prognostic information for various cancers.

Patient Perceptions of Participating in the RSNA Image Share Project: a Preliminary Study.

The purpose of this study was to gauge patient perceptions of the RSNA Image Share Project (ISP), a pilot program that provides patients access to their imaging studies online via secure Personal Health Record (PHR) accounts. Two separate Institutional Review Board exempted surveys were distributed to patients depending on whether they decided to enroll or opt out of enrollment in the ISP. For patients that enrolled, a survey gauged baseline computer usage, perceptions of online access to images through the ISP, effect of patient access to images on patient-physician relationships, and interest in alternative use of images. The other survey documented the age and reasons for declining participation for those that opted out of enrolling in the ISP. Out of 564 patients, 470 enrolled in the ISP (83 % participation rate) and 456 of these 470 individuals completed the survey for a survey participation rate of 97 %. Patients who enrolled overwhelmingly perceived access to online images as beneficial and felt it bolstered their patient-physician relationship. Out of 564 patients, 94 declined enrollment in the ISP and all 94 individuals completed the survey for a survey participation rate of 100 %. Patients who declined to participate in the ISP cited unreliable access to Internet and existing availability of non-web-based intra-network images to their physicians. Patients who participated in the ISP found having a measure of control over their images to be beneficial and felt that patient-physician relationships could be negatively affected by challenges related to image accessibility.

Evaluation of Low-Contrast Detectability of Iterative Reconstruction across Multiple Institutions, CT Scanner Manufacturers, and Radiation Exposure Levels.

PURPOSE: To compare image resolution from iterative reconstruction with resolution from filtered back projection for low-contrast objects on phantom computed tomographic (CT) images across vendors and exposure levels. MATERIALS AND METHODS: Randomized repeat scans of an American College of Radiology CT accreditation phantom (module 2, low contrast) were performed for multiple radiation exposures, vendors, and vendor iterative reconstruction algorithms. Eleven volunteers were presented with 900 images by using a custom-designed graphical user interface to perform a task created specifically for this reader study. Results were analyzed by using statistical graphics and analysis of variance. RESULTS: Across three vendors (blinded as A, B, and C) and across three exposure levels, the mean correct classification rate was higher for iterative reconstruction than filtered back projection (P < .01): 87.4% iterative reconstruction and 81.3% filtered back projection at 20 mGy, 70.3% iterative reconstruction and 63.9% filtered back projection at 12 mGy, and 61.0% iterative reconstruction and 56.4% filtered back projection at 7.2 mGy. There was a significant difference in mean correct classification rate between vendor B and the other two vendors. Across all exposure levels, images obtained by using vendor B's scanner outperformed the other vendors, with a mean correct classification rate of 74.4%, while the mean correct classification rate for vendors A and C was 68.1% and 68.3%, respectively. Across all readers, the mean correct classification rate for iterative reconstruction (73.0%) was higher compared with the mean correct classification rate for filtered back projection (67.0%). CONCLUSION: The potential exists to reduce radiation dose without compromising low-contrast detectability by using iterative reconstruction instead of filtered back projection. There is substantial variability across vendor reconstruction algorithms.

An Improved Index of Image Quality for Task-based Performance of CT Iterative Reconstruction across Three Commercial Implementations.

PURPOSE: To develop and validate a metric of computed tomographic (CT) image quality that incorporates the noise texture and resolution properties of an image. MATERIALS AND METHODS: Images of the American College of Radiology CT quality assurance phantom were acquired by using three commercial CT systems at seven dose levels with filtered back projection (FBP) and iterative reconstruction (IR). Image quality was characterized by the contrast-to-noise ratio (CNR) and a detectability index (d') that incorporated noise texture and spatial resolution. The measured CNR and d' were compared with a corresponding observer study by using the Spearman rank correlation coefficient to determine how well each metric reflects the ability of an observer to detect subtle lesions. Statistical significance of the correlation between each metric and observer performance was determined by using a Student t distribution; P values less than .05 indicated a significant correlation. Additionally, each metric was used to estimate the dose reduction potential of IR algorithms while maintaining image quality. RESULTS: Across all dose levels, scanner models, and reconstruction algorithms, the d' correlated strongly with observer performance in the corresponding observer study (ρ = 0.95; P < .001), whereas the CNR correlated weakly with observer performance (ρ = 0.31; P = .21). Furthermore, the d' showed that the dose-reduction capabilities differed between clinical implementations (range, 12%-35%) and were less than those predicted from the CNR (range, 50%-54%). CONCLUSION: The strong correlation between the observer performance and the d' indicates that the d' is superior to the CNR for the evaluation of CT image quality. Moreover, the results of this study indicate that the d' improves less than the CNR with the use of IR, which indicates less potential for IR dose reduction than previously thought.

Ontology-based image navigation: exploring 3.0-T MR neurography of the brachial plexus using AIM and RadLex.

Disorders of the peripheral nervous system have traditionally been evaluated using clinical history, physical examination, and electrodiagnostic testing. In selected cases, imaging modalities such as magnetic resonance (MR) neurography may help further localize or characterize abnormalities associated with peripheral neuropathies, and the clinical importance of such techniques is increasing. However, MR image interpretation with respect to peripheral nerve anatomy and disease often presents a diagnostic challenge because the relevant knowledge base remains relatively specialized. Using the radiology knowledge resource RadLex®, a series of RadLex queries, the Annotation and Image Markup standard for image annotation, and a Web services-based software architecture, the authors developed an application that allows ontology-assisted image navigation. The application provides an image browsing interface, allowing users to visually inspect the imaging appearance of anatomic structures. By interacting directly with the images, users can access additional structure-related information that is derived from RadLex (eg, muscle innervation, muscle attachment sites). These data also serve as conceptual links to navigate from one portion of the imaging atlas to another. With 3.0-T MR neurography of the brachial plexus as the initial area of interest, the resulting application provides support to radiologists in the image interpretation process by allowing efficient exploration of the MR imaging appearance of relevant nerve segments, muscles, bone structures, vascular landmarks, anatomic spaces, and entrapment sites, and the investigation of neuromuscular relationships.

Data-driven decision support for radiologists: re-using the National Lung Screening Trial dataset for pulmonary nodule management.

Real-time mining of large research trial datasets enables development of case-based clinical decision support tools. Several applicable research datasets exist including the National Lung Screening Trial (NLST), a dataset unparalleled in size and scope for studying population-based lung cancer screening. Using these data, a clinical decision support tool was developed which matches patient demographics and lung nodule characteristics to a cohort of similar patients. The NLST dataset was converted into Structured Query Language (SQL) tables hosted on a web server, and a web-based JavaScript application was developed which performs real-time queries. JavaScript is used for both the server-side and client-side language, allowing for rapid development of a robust client interface and server-side data layer. Real-time data mining of user-specified patient cohorts achieved a rapid return of cohort cancer statistics and lung nodule distribution information. This system demonstrates the potential of individualized real-time data mining using large high-quality clinical trial datasets to drive evidence-based clinical decision-making.

Using Standardized Lexicons for Report Template Validation with LexMap, a Web-based Application.

An enormous amount of data exists in unstructured diagnostic and interventional radiology reports. Free text or non-standardized terminologies limit the ability to parse, extract, and analyze these report data elements. Medical lexicons and ontologies contain standardized terms for relevant concepts including disease entities, radiographic technique, and findings. The use of standardized terms offers the potential to improve reporting consistency and facilitate computer analysis. The purpose of this project was to implement an interface to aid in the creation of standards-compliant reporting templates for use in interventional radiology. Non-standardized procedure report text was analyzed and referenced to RadLex, SNOMED-CT, and LOINC. Using JavaScript, a web application was developed which determined whether exact terms or synonyms in reports existed within these three reference resources. The NCBO BioPortal Annotator web service was used to map terms, and output from this application was used to create an interactive annotated version of the original report. The application was successfully used to analyze and modify five distinct reports for the Society of Interventional Radiology's standardized reporting project.