Opportunistic Screening: Radiology Scientific Expert Panel.

Radiologic tests often contain rich imaging data not relevant to the clinical indication. Opportunistic screening refers to the practice of systematically leveraging these incidental imaging findings. Although opportunistic screening can apply to imaging modalities such as conventional radiography, US, and MRI, most attention to date has focused on body CT by using artificial intelligence (AI)-assisted methods. Body CT represents an ideal high-volume modality whereby a quantitative assessment of tissue composition (eg, bone, muscle, fat, and vascular calcium) can provide valuable risk stratification and help detect unsuspected presymptomatic disease. The emergence of "explainable" AI algorithms that fully automate these measurements could eventually lead to their routine clinical use. Potential barriers to widespread implementation of opportunistic CT screening include the need for buy-in from radiologists, referring providers, and patients. Standardization of acquiring and reporting measures is needed, in addition to expanded normative data according to age, sex, and race and ethnicity. Regulatory and reimbursement hurdles are not insurmountable but pose substantial challenges to commercialization and clinical use. Through demonstration of improved population health outcomes and cost-effectiveness, these opportunistic CT-based measures should be attractive to both payers and health care systems as value-based reimbursement models mature. If highly successful, opportunistic screening could eventually justify a practice of standalone "intended" CT screening.

Determination of lower radiation dose limit for automatic measurement of adipose tissue.

The purpose of this study was to determine the lower limit of radiation dose required to measure visceral adipose tissue (VAT) and subcutaneous adipose tissue (SAT) volumes when a fat quantification and noise reduction techniques (NRTs) are combined. For this purpose, we utilized CT colonography (CTC) images taken at low doses and manually segmented VAT and SAT fat volumes as ground truth. In order to derive the acceptable precision of the measurements needed to estimate the lower limit of radiation dose, we estimated the effect of different positioning during CT scanning on fat measurements using manually segmented VAT and SAT against normal dose. As a result, the acceptable accuracy of SAT and VAT was found to be 94.5% and 85.2%, respectively. Using these thresholds, the lower radiation dose limit required to accurately measure SAT using 5.25-mm slice-thick images was 1.5 mGy of size-specific dose estimates (SSDE), while the lower radiation dose limit required to accurately measure VAT was 0.4 mGy of SSDE. The lower dose limit for SAT and VAT combined was 1.5 mGy, which was equivalent to an estimated effective dose of 0.38 mSv. Alternatively, without noise reduction, SAT could not achieve acceptable accuracy even for images with a slice thickness of 5.25 mm, while VAT required noise reduction for images with a slice thickness of 1.25 mm, but could achieve acceptable accuracy without noise reduction for images with a slice thickness of 5.25 mm.

Association of sarcopenia with survival in advanced NSCLC patients receiving concurrent immunotherapy and chemotherapy.

Background: Frailty, sarcopenia and malnutrition are powerful predictors of clinical outcomes that are not routinely measured in patients with non-small cell lung cancer (NSCLC). The primary aim of this study was to investigate the association of sarcopenia, determined by the psoas muscle index (PMI) with overall survival (OS) in patients with advanced NSCLC treated with concurrent immune checkpoint inhibitor (ICI) and chemotherapy (CTX). Methods: We retrospectively reviewed data from a cohort of patients with locally advanced or metastatic NSCLC who were treated between 2015 and 2021 at the University of Virginia Medical Center. The cross-sectional area of the psoas muscle was assessed on CT or PET/CT imaging prior to treatment initiation. Multivariate analysis was performed using Cox proportional hazards regression models. Results: A total of 92 patients (median age: 64 years, range 36-89 years), 48 (52.2%) men and 44 (47.8%) women, were included in the study. The median follow-up was 29.6 months. The median OS was 17.8 months. Sarcopenia, defined by a PMI below the 25th percentile, was associated with significantly lower OS (9.1 months in sarcopenic patients vs. 22.3 months in non-sarcopenic patients, P = 0.002). Multivariate analysis revealed that sarcopenia (HR 2.12, P = 0.0209), ECOG ≥ 2 (HR 2.88, P = 0.0027), prognostic nutritional index (HR 3.02, P = 0.0034) and the absence of immune related adverse events (HR 2.04, P = 0.0185) were independently associated with inferior OS. Conclusions: Sarcopenia is independently associated with poor OS in patients with advanced NSCLC undergoing concurrent ICI and CTX.

Moving Toward Seamless Interinstitutional Electronic Image Transfer.

The fact that medical images are still predominately exchanged between institutions via physical media is unacceptable in the era of value-driven health care. Although better solutions are technically possible, problems of coordination and market dynamics may be inhibiting progress more than technical factors. We provide a macrosystem analysis of the problem of interinstitutional medical image exchange and propose a strategy for nudging the market toward a patient-friendly solution. The system can be viewed as a network, with autonomous nodes interconnected by links through which information is exchanged. A variety of potential network configurations include those that depend on individual carriers, peer-to-peer links, one or multiple hubs, or a hybrid of models. We find the linked multihub model, in which individual institutions are connected to other institutions via image exchange companies, to be the configuration most likely to create a patient-friendly electronic image exchange system. To achieve this configuration, image exchange companies, which operate in a competitive marketplace, must exchange images with each other. We call on these vendors to immediately commit to coordinating in this manner. We call on all other stakeholders, including local care provider institutions, medical societies, payers, and regulators, to actively encourage and facilitate this behavior. Specifically, we call on institutions to create appropriate market incentives by only contracting with image exchange vendors who are committed to begin vendor-to-vendor image exchange by no later than 2024.

Radiologists' Increasing Role in Population Health Management: AJR Expert Panel Narrative Review.

Population health management (PHM) is the holistic process of improving health outcomes of groups of individuals through the support of appropriate financial and care models. Radiologists' presence at the intersection of many aspects of health care, including screening, diagnostic imaging, and image-guided therapies, provides the opportunity for increased radiologist engagement in PHM. Furthermore, innovations in artificial intelligence and imaging informatics will serve as critical tools to improve value in health care through evidence-based and equitable approaches. Given radiologists' limited engagement in PHM to date, it is imperative to define the PHM priorities of the specialty so that radiologists' full value in improving population health is realized. The purpose of this expert review is to explore programs and future directions for radiologists in PHM.

Patient and Provider Feedback for Radiology Reports: Implementation of a Quality Improvement Project in a Multi-Institutional Setting.

BACKGROUND: Radiology does not routinely solicit feedback on radiology reports. The aim of the study is to report the feasibility and initial results of a multi-institutional quality improvement project implementing patient and provider feedback for radiology reports. METHODS: A HIPAA-compliant, institutional review board-waived quality improvement effort at two institutions obtaining patient and provider feedback for radiology reports was implemented from January 2018 to May 2020. INTERVENTION: A two-question survey (quantitative review and open text box feedback) was embedded into the electronic health records for patients and providers. Text-based feedback was evaluated, and patterns of feedback were categorized: thoroughness of reports, error in reports, timeliness of reports, access to reports, desire for patient summary, and desire for key images. We performed the χ2 test for categorical variables. P < .05 was considered significant. RESULTS: Of 367 responses, patients provided 219 of 367 (60%), and providers provided 148 of 367 (40%) of the feedback. A higher proportion of patients reported satisfaction with reports (76% versus 65%, P = .023) and provided more feedback compared with providers (71% versus 50%, P < .0001). Both patients and providers commented on the thoroughness of reports (12% of patients versus 9% of providers) and errors in reports (8% of patients and 9% of providers). Patients disproportionately commented on timeliness of reports (11%) and access to the reports (6%) compared with providers (3% each). In addition, 7% of patients expressed a desire for patient summaries. CONCLUSION: Report-specific patient and provider feedback demonstrate the feasibility of embedding surveys into electronic medical records. Up to 9% of the feedback addressed an error in reports.

Patient-centered Reporting in Radiology: A Single-site Survey Study of Lung Cancer Screening Results.

PURPOSE: This study aimed to assess whether patients preferred traditional or patient-friendly radiology reports and, secondarily, whether one reporting style led to a subjective improvement in patients' understanding of their imaging results and next steps in their clinical care. MATERIALS AND METHODS: This randomized study included patients who had previously enrolled in an institutional comprehensive lung cancer screening program. Three hundred patients were randomly selected from the program database to receive both traditional and patient-centered radiology reports. Randomization also occurred at both the risk level of the fictitious test results (low, intermediate, or high) and the order in which the reports were read by each participant. Participants completed a survey providing demographic information and indicating which report style was preferred and which report style led to a better understanding of screening results and future options. In addition, each report style was rated (from 1 to 5) for clarity, understandability, attractiveness, and helpfulness. RESULTS: A total of 46 responses for report preference data and 41 responses for attribute rating data were obtained. Overall, participants demonstrate a preference for patient-friendly reports (65.2%) over traditional reports (21.7%). On a 5-point scale, average ratings for patient-friendly reports were higher than traditional reports by 1.2 (P<0.001) for clarity, 1.5 (P<0.001) for understandability, 1.5 (P<0.001) for attractiveness, and 1.0 (P<0.001) for helpfulness. CONCLUSION: Data suggest that patients prefer patient-friendly reports over traditional reports and find them to be clearer, more comprehensible, more attractive, and more helpful.

Deep Learning-based Quantification of Abdominal Subcutaneous and Visceral Fat Volume on CT Images.

RATIONALE AND OBJECTIVES: Develop a deep learning-based algorithm using the U-Net architecture to measure abdominal fat on computed tomography (CT) images. MATERIALS AND METHODS: Sequential CT images spanning the abdominal region of seven subjects were manually segmented to calculate subcutaneous fat (SAT) and visceral fat (VAT). The resulting segmentation maps of SAT and VAT were augmented using a template-based data augmentation approach to create a large dataset for neural network training. Neural network performance was evaluated on both sequential CT slices from three subjects and randomly selected CT images from the upper, central, and lower abdominal regions of 100 subjects. RESULTS: Both subcutaneous and abdominal cavity segmentation images created by the two methods were highly comparable with an overall Dice similarity coefficient of 0.94. Pearson's correlation coefficients between the subcutaneous and visceral fat volumes quantified using the two methods were 0.99 and 0.99 and the overall percent residual squared error were 5.5% and 8.5%. Manual segmentation of SAT and VAT on the 555 CT slices used for testing took approximately 46 hours while automated segmentation took approximately 1 minute. CONCLUSION: Our data demonstrates that deep learning methods utilizing a template-based data augmentation strategy can be employed to accurately and rapidly quantify total abdominal SAT and VAT with a small number of training images.

Patient-centered Radiology Reporting for Lung Cancer Screening.

Medicine is slowly transitioning toward a more patient-centered approach, with patients taking a more central role in their own care. A key part of this movement has involved giving patients increased access to their medical record and imaging results via electronic health portals. However, most patients lack the knowledge to fully understand medical documents, which are generally written above their comprehension level. Radiology reports, in particular, utilize complex terminology due to radiologists' historic function as consultants to other physicians, with little direct communication to patients. As a result, typical radiology reports lack standardized formatting, and they are often inscrutable to patients. Numerous studies examining patient preference also point to a trend for more accessible radiology reports geared toward patients. Reports designed with an infographic format, combining simple pictures and standardized text, may be an ideal format that radiologists can pursue to provide patient-centered care. Our team, through feedback from patient advisory groups, developed a patient-friendly low-dose computed tomography lung cancer screening report with an infographic format that is both visually attractive and comprehensible to the average patient. The report is designed with sections including a description of low-dose computed tomography, a section on individualized patient results, the meaning of the results, and a list of the next steps in their care. We believe that this form of the report has the potential to serve as a bridge between radiologists and patients, allowing for a better patient understanding of their health and empowering patients to participate in their health and health care.

Utility of 3-Week Follow-Up of Upper Gastrointestinal Study in Asymptomatic Patients Following Laparoscopic Sleeve Gastrectomy.

RATIONALE AND OBJECTIVES: One day following laparoscopic sleeve gastrectomy (LSG), routine practice has historically dictated that an upper gastrointestinal (UGI) study be performed to assess for staple line leak, clinically significant stenosis, or other complications requiring surgical revision. Recent literature has cast doubt on the utility of performing an UGI immediately following surgery due to its poor sensitivity in detecting leaks and hence referrals for this post-operative study have decreased. However, routine practice at our institution is to perform an UGI study at three weeks following LSG to assess for late complications despite a similar lack of evidence supporting the yield of this exam. The purpose of our study is to assess the utility and cost effectiveness of UGI exams in asymptomatic patients three weeks following LSG. METHODS AND MATERIALS: A retrospective chart review of patients who underwent LSG for obesity performed at our institution between January 2014 and October 2018 and subsequently had an UGI within two-four weeks following the surgery was conducted. RESULTS: A total of seventy three asymptomatic patients underwent an UGI study, of which no clinically significant stricture, leak or other complications were identified. Of the fifteen patients who were symptomatic between two-four weeks after surgery, twelve (80%) were found to have complications ranging from staple line leak or gastric narrowing. CONCLUSIONS: The utility of UGI following LSG in the absence of symptoms is doubtful. Additionally, the added cost and radiation does not add value to the patient's care. We recommend UGI study utilization when there is a clinical suspicion for a complication.

Simulating Dual-Energy X-Ray Absorptiometry in CT Using Deep-Learning Segmentation Cascade.

PURPOSE: Osteoporosis is an underdiagnosed condition despite effective screening modalities. Dual-energy x-ray absorptiometry (DEXA) screening, although recommended in clinical guidelines, remains markedly underutilized. In contrast to DEXA, CT utilization is high and presents a valuable data source for opportunistic osteoporosis screening. The purpose of this study was to describe a method to simulate lumbar DEXA scores from routinely acquired CT studies using a machine-learning algorithm. METHODS: Between January 2010 and September 2014, 610 CT studies of the abdomen and pelvis were used to develop spinal column and L1 to L4 multiclass segmentation. DEXA simulation training and validation used 1,843 pairs of CT studies accompanied by DEXA results obtained within a 6-month interval from the same individual. Machine learning-based regression was used to determine correlation between calculated grade (on the basis of vertebrae L1-L4) and DEXA t score. RESULTS: Analysis of the t score equivalent, generated by the algorithm, revealed true positives in 1,144 patients, false positives in 92 patients, true negatives in 245 patients, and false negatives in 212 patients, resulting in an accuracy of 82%. Sensitivity for the detection of osteoporosis or osteopenia was 84.4% (95% confidence interval, 82.3%-86.2%), and specificity was 72.7% (95% confidence interval, 67.7%-77.2%). CONCLUSIONS: The presented algorithm can identify osteoporosis and osteopenia with a high degree of accuracy (82%) and a small proportion of false positives. Efforts to cull greater information using machine-learning algorithms from pre-existing data have the potential to have a marked impact on population health efforts such as bone mineral density screening for osteoporosis, in which gaps in screening currently exist.

Navigating Generational Differences in Radiology.

Generations are cohorts of individuals born in a particular time period who share similar values or value systems owing to historic events that occurred at crucial times during their development. Generations are defined to study how views and values change over time and to assess the differential impact that formative experiences have on groups. Understanding and navigating generational differences will be a critical skill for radiology leaders in the coming decade, as four distinct generations are working side by side for the first time in history. The four generations currently in the workforce are categorized as traditionalists, baby boomers, Generation Xers, and millennials. Beginning in 2016, millennials became the largest generation in the U.S. workforce, surpassing the number of Generation Xers. This major demographic shift will have a profound impact on workplace culture, recruitment efforts, and trainee education. While each generation has similar basic needs, meeting those needs and motivating individuals of different generations are best accomplished using different approaches. Radiology leaders must encourage and support these varied generations to work harmoniously to foster high-performance organizations. ©RSNA, 2018.

Developing an Interactive Data Visualization Tool to Assess the Impact of Decision Support on Clinical Operations.

Due to mandates from recent legislation, clinical decision support (CDS) software is being adopted by radiology practices across the country. This software provides imaging study decision support for referring providers at the point of order entry. CDS systems produce a large volume of data, providing opportunities for research and quality improvement. In order to better visualize and analyze trends in this data, an interactive data visualization dashboard was created using a commercially available data visualization platform. Following the integration of a commercially available clinical decision support product into the electronic health record, a dashboard was created using a commercially available data visualization platform (Tableau, Seattle, WA). Data generated by the CDS were exported from the data warehouse, where they were stored, into the platform. This allowed for real-time visualization of the data generated by the decision support software. The creation of the dashboard allowed the output from the CDS platform to be more easily analyzed and facilitated hypothesis generation. Integrating data visualization tools into clinical decision support tools allows for easier data analysis and can streamline research and quality improvement efforts.