Natural Language Processing Model for Identifying Critical Findings-A Multi-Institutional Study.

Improving detection and follow-up of recommendations made in radiology reports is a critical unmet need. The long and unstructured nature of radiology reports limits the ability of clinicians to assimilate the full report and identify all the pertinent information for prioritizing the critical cases. We developed an automated NLP pipeline using a transformer-based ClinicalBERT++ model which was fine-tuned on 3 M radiology reports and compared against the traditional BERT model. We validated the models on both internal hold-out ED cases from EUH as well as external cases from Mayo Clinic. We also evaluated the model by combining different sections of the radiology reports. On the internal test set of 3819 reports, the ClinicalBERT++ model achieved 0.96 f1-score while the BERT also achieved the same performance using the reason for exam and impression sections. However, ClinicalBERT++ outperformed BERT on the external test dataset of 2039 reports and achieved the highest performance for classifying critical finding reports (0.81 precision and 0.54 recall). The ClinicalBERT++ model has been successfully applied to large-scale radiology reports from 5 different sites. Automated NLP system that can analyze free-text radiology reports, along with the reason for the exam, to identify critical radiology findings and recommendations could enable automated alert notifications to clinicians about the need for clinical follow-up. The clinical significance of our proposed model is that it could be used as an additional layer of safeguard to clinical practice and reduce the chance of important findings reported in a radiology report is not overlooked by clinicians as well as provide a way to retrospectively track large hospital databases for evaluating the documentation of the critical findings.

Imaging Informatics Fellowship Curriculum: Building Consensus on the Most Critical Topics and the Future of the Informatics Fellowship.

The existing fellowship imaging informatics curriculum, established in 2004, has not undergone formal revision since its inception and inaccurately reflects present-day radiology infrastructure. It insufficiently equips trainees for today's informatics challenges as current practices require an understanding of advanced informatics processes and more complex system integration. We sought to address this issue by surveying imaging informatics fellowship program directors across the country to determine the components and cutline for essential topics in a standardized imaging informatics curriculum, the consensus on essential versus supplementary knowledge, and the factors individual programs may use to determine if a newly developed topic is an essential topic. We further identified typical program structural elements and sought fellowship director consensus on offering official graduate trainee certification to imaging informatics fellows. Here, we aim to provide an imaging informatics fellowship director consensus on topics considered essential while still providing a framework for informatics fellowship programs to customize their individual curricula.

Shared Decision Making: Radiology's Role and Opportunities.

OBJECTIVE. The purpose of this article is to present a targeted literature review describing the current state of radiology initiatives in support of shared decision making and gaps that offer opportunities for innovation and improvement. CONCLUSION. Breaking down the shared decision-making process into its four major components (access to information, comprehension of the information, appraisal of the information, application of knowledge in care decisions) reveals the role of radiologists in the decision-making process and opportunities for expanding this role.

A Survey of Imaging Informatics Fellowships and Their Curricula: Current State Assessment.

In a 2016 survey of imaging informatics ("II") fellowship graduates, the surveyed fellowship graduates expressed the "opinion that II fellowships needed further formalization and standardization" Liao et al. (J Digit Imaging, 2016). This, coupled with the fact that the original published "standardized" curriculum is about 15 years out of date in our rapidly changing systems, suggests an opportunity for curriculum improvement. Before agreeing on improved structural and content suggestions for fellowships, we completed a current-state assessment of how each fellowship organizes its education and what requirements each have for fellowship completion. In this work, we aimed to collect existing information about imaging informatics fellowship curricula by contacting institutions across the country. A survey was completed by phone with the fellowship directors of existing imaging informatics fellowships across the country. Additionally, we collected existing documentation that outlines the curricula currently in use at institutions. We reviewed both the interview responses and documentation to assess overlapping trends and institutional differences in curriculum structure and content. All fellowships had suggested reading lists, didactic lectures, and a required project for each fellow. There were required practicum activities or teaching experience each in two fellowships, and one fellowship had a mandatory certification requirement for graduation. Curriculum topics in Technical Informatics or Business and Management were covered by a majority of institutions, while Quality and Safety and Research topics had inconsistent coverage across fellowships. Our plan is to reengage II fellowship directors to develop a core curriculum, which is part of the Society of Imaging Informatics in Medicine strategic plan.

Imaging of Children With Nontraumatic Headaches.

OBJECTIVE: Headache in children is a common symptom and often is worrisome for clinicians and parents because of the breadth of possible underlying significant abnormalities, including meningitis, brain neoplasms, and intracranial hemorrhage. For this reason, many children with headaches undergo neuroimaging. Most neuroimaging studies performed of children with headaches have normal findings but may lead to significant downstream effects, including unnecessary exposure to ionizing radiation or sedation, as well as unnecessary cost to the health care system. In this article, we review the current evidence and discuss the role of neuroimaging in the diagnosis and management of pediatric headaches, with a special focus on tools that may aid in increasing the rate of positive findings, such as classification systems, algorithms, and red flag criteria. CONCLUSION: Many tools exist that can help in improving the appropriateness of neuroimaging in pediatric headache. The main issues that remain to be addressed include scientific proof of safety and validity of these tools and clarity regarding the risks, benefits, and cost-effectiveness of CT versus MRI in various clinical settings and scenarios.

Confidence-Aware Severity Assessment of Lung Disease from Chest X-Rays Using Deep Neural Network on a Multi-Reader Dataset.

In this study, we present a method based on Monte Carlo Dropout (MCD) as Bayesian neural network (BNN) approximation for confidence-aware severity classification of lung diseases in COVID-19 patients using chest X-rays (CXRs). Trained and tested on 1208 CXRs from Hospital 1 in the USA, the model categorizes severity into four levels (i.e., normal, mild, moderate, and severe) based on lung consolidation and opacity. Severity labels, determined by the median consensus of five radiologists, serve as the reference standard. The model's performance is internally validated against evaluations from an additional radiologist and two residents that were excluded from the median. The performance of the model is further evaluated on additional internal and external datasets comprising 2200 CXRs from the same hospital and 1300 CXRs from Hospital 2 in South Korea. The model achieves an average area under the curve (AUC) of 0.94 ± 0.01 across all classes in the primary dataset, surpassing human readers in each severity class and achieves a higher Kendall correlation coefficient (KCC) of 0.80 ± 0.03. The performance of the model is consistent across varied datasets, highlighting its generalization. A key aspect of the model is its predictive uncertainty (PU), which is inversely related to the level of agreement among radiologists, particularly in mild and moderate cases. The study concludes that the model outperforms human readers in severity assessment and maintains consistent accuracy across diverse datasets. Its ability to provide confidence measures in predictions is pivotal for potential clinical use, underscoring the BNN's role in enhancing diagnostic precision in lung disease analysis through CXR.

The EMory BrEast imaging Dataset (EMBED): A Racially Diverse, Granular Dataset of 3.4 Million Screening and Diagnostic Mammographic Images.

Supplemental material is available for this article. Keywords: Mammography, Breast, Machine Learning © RSNA, 2023.

Intellectual property: A primer for radiologists.

Typically the creative product of the mind, intellectual property often forms the basis of a new product, service line, or company. Intellectual property law is complicated and nuanced, and poorly understood by many physicians, innovators, and entrepreneurs. Successfully navigating the process of intellectual property protection is critical in facilitating the translation of innovation into clinical practice. We define intellectual property and common terms used in intellectual property law and offer justification for the importance of intellectual property protections. We additionally highlight resources to assist radiologists with intellectual property protection and outline basic guidelines to successfully initiate discussions around intellectual property with third party vendors and consultants. SUMMARY: Proactive intellectual property protection is critically important for radiologist innovators seeking to bring new ideas to the marketplace.

COVID-19 pneumonia chest radiographic severity score: variability assessment among experienced and in-training radiologists and creation of a multireader composite score database for artificial intelligence algorithm development.

OBJECTIVE: The purpose was to evaluate reader variability between experienced and in-training radiologists of COVID-19 pneumonia severity on chest radiograph (CXR), and to create a multireader database suitable for AI development. METHODS: In this study, CXRs from polymerase chain reaction positive COVID-19 patients were reviewed. Six experienced cardiothoracic radiologists and two residents classified each CXR according to severity. One radiologist performed the classification twice to assess intraobserver variability. Severity classification was assessed using a 4-class system: normal (0), mild (1), moderate (2), and severe (3). A median severity score (Rad Med) for each CXR was determined for the six radiologists for development of a multireader database (XCOMS). Kendal Tau correlation and percentage of disagreement were calculated to assess variability. RESULTS: A total of 397 patients (1208 CXRs) were included (mean age, 60 years SD ± 1), 189 men). Interobserver variability between the radiologists ranges between 0.67 and 0.78. Compared to the Rad Med score, the radiologists show good correlation between 0.79-0.88. Residents show slightly lower interobserver agreement of 0.66 with each other and between 0.69 and 0.71 with experienced radiologists. Intraobserver agreement was high with a correlation coefficient of 0.77. In 220 (18%), 707 (59%), 259 (21%) and 22 (2%) CXRs there was a 0, 1, 2 or 3 class-difference. In 594 (50%) CXRs the median scores of the residents and the radiologists were similar, in 578 (48%) and 36 (3%) CXRs there was a 1 and 2 class-difference. CONCLUSION: Experienced and in-training radiologists demonstrate good inter- and intraobserver agreement in COVID-19 pneumonia severity classification. A higher percentage of disagreement was observed in moderate cases, which may affect training of AI algorithms. ADVANCES IN KNOWLEDGE: Most AI algorithms are trained on data labeled by a single expert. This study shows that for COVID-19 X-ray severity classification there is significant variability and disagreement between radiologist and between residents.

Balancing the Scales: An Analysis of Social Determinants of Health, Radiology Report Acuity, and Radiology Staffing Models in an Academic Health System.

PURPOSE: Social determinants of health, including race and insurance status, contribute to patient outcomes. In academic health systems, care is provided by a mix of trainees and faculty members. The optimal staffing ratio of trainees to faculty members (T/F) in radiology is unknown but may be related to the complexity of patients requiring care. Hospital characteristics, patient demographics, and radiology report findings may serve as markers of risk for poor outcomes because of patient complexity. METHODS: Descriptive characteristics of each hospital in an urban five-hospital academic health system, including payer distribution and race, were collected. Radiology department T/F ratios were calculated. A natural language processing model was used to classify multimodal report findings into nonacute, acute, and critical, with report acuity calculated as the fraction of acute and critical findings. Patient race, payer type, T/F ratio, and report acuity score for hospital 1, a safety net hospital, were compared with these factors for hospitals 2 to 5. RESULTS: The fraction of patients at hospital 1 who are Black (79%) and have Medicaid insurance (28%) is significantly higher than at hospitals 2 to 5 (P < .0001), with the exception of hospital 3 (80.1% black). The T/F ratio of 1.37 at hospital 1 as well as its report acuity (28.9%) were significantly higher (P < .0001 for both). CONCLUSIONS: T/F ratio and report acuity are highest at hospital 1, which serves the most at-risk patient population. This suggests a potential overreliance on trainees at a site whose patients may require the greatest expertise to optimize care.