Search

While many case studies have described the implementation of self-scheduling tools, which allow patients to schedule visits and imaging studies asynchronously online, none have explored the impact of self-scheduling on equitable access to care.1 Using an electronic health record patient portal, University of California San Francisco deployed a self-scheduling tool that allowed patients to self-schedule diagnostic imaging studies. We analyzed electronic health record data for the imaging modalities with the option to be self-scheduled from January 1, 2021 to September 1, 2021. We used descriptive statistics to compare demographic characteristics and created a multivariable logistic regression model to identify predictors of patient self-scheduling utilization. Among all active patient portal users, Latinx, Black/African American, and non-English speaking patients were less likely to self-schedule studies. Patients with Medi-Cal, California’s Medicaid program, and Medicare insurance were also less likely to self-schedule when compared with commercially insured patients. Efforts to facilitate use of patient portal-based applications are necessary to increase equitability and decrease disparities in access.

BACKGROUND: Lack of standardization in CT protocol choice contributes to radiation dose variation. PURPOSE: To create a framework to assess radiation doses within broad CT categories defined according to body region and clinical imaging indication and to cluster indications according to the dose required for sufficient image quality. MATERIALS AND METHODS: This was a retrospective study using Digital Imaging and Communications in Medicine metadata. CT examinations in adults from January 1, 2016 to December 31, 2019 from the University of California San Francisco International CT Dose Registry were grouped into 19 categories according to body region and required radiation dose levels. Five body regions had a single dose range (ie, extremities, neck, thoracolumbar spine, combined chest and abdomen, and combined thoracolumbar spine). Five additional regions were subdivided according to dose. Head, chest, cardiac, and abdomen each had low, routine, and high dose categories; combined head and neck had routine and high dose categories. For each category, the median and 75th percentile (ie, diagnostic reference level [DRL]) were determined for dose-length product, and the variation in dose within categories versus across categories was calculated and compared using an analysis of variance. Relative median and DRL (95% CI) doses comparing high dose versus low dose categories were calculated. RESULTS: Among 4.5 million examinations, the median and DRL doses varied approximately 10 times between categories compared with between indications within categories. For head, chest, abdomen, and cardiac (3 266 546 examinations [72%]), the relative median doses were higher in examinations assigned to the high dose categories than in examinations assigned to the low dose categories, suggesting the assignment of indications to the broad categories is valid (head, 3.4-fold higher [95% CI: 3.4, 3.5]; chest, 9.6 [95% CI: 9.3, 10.0]; abdomen, 2.4 [95% CI: 2.4, 2.5]; and cardiac, 18.1 [95% CI: 17.7, 18.6]). Results were similar for DRL doses (all P < .001). CONCLUSION: Broad categories based on image quality requirements are a suitable framework for simplifying radiation dose assessment, according to expected variation between and within categories. © RSNA, 2021 See also the editorial by Mahesh in this issue.

Purpose To create an algorithm able to accurately detect IVC filters on radiographs without human assistance, capable of being used to screen radiographs to identify patients needing IVC filter retrieval. Methods A primary dataset of 5225 images, 30% of which included IVC filters, was assembled and annotated. 85% of the data was used to train a Cascade R-CNN (Region Based Convolutional Neural Network) object detection network incorporating a pre-trained ResNet-50 backbone. The remaining 15% of the data, independently annotated by three radiologists, was used as a test set to assess performance. The algorithm was also assessed on an independently constructed 1424-image dataset, drawn from a different institution than the primary dataset. Results On the primary test set, the algorithm achieved a sensitivity of 96.2% (95% CI 92.7–98.1%) and a specificity of 98.9% (95% CI 97.4–99.5%). Results were similar on the external test set: sensitivity 97.9% (95% CI 96.2–98.9%), specificity 99.6 (95% CI 98.9–99.9%). Conclusion Fully automated detection of IVC filters on radiographs with high sensitivity and excellent specificity required for an automated screening system can be achieved using object detection neural networks. Further work will develop a system for identifying patients for IVC filter retrieval based on this algorithm. Graphical abstract

To create an algorithm able to accurately detect IVC filters on radiographs without human assistance, capable of being used to screen radiographs to identify patients needing IVC filter retrieval.A primary dataset of 5225 images, 30% of which included IVC filters, was assembled and annotated. 85% of the data was used to train a Cascade R-CNN (Region Based Convolutional Neural Network) object detection network incorporating a pre-trained ResNet-50 backbone. The remaining 15% of the data, independently annotated by three radiologists, was used as a test set to assess performance. The algorithm was also assessed on an independently constructed 1424-image dataset, drawn from a different institution than the primary dataset.On the primary test set, the algorithm achieved a sensitivity of 96.2% (95% CI 92.7-98.1%) and a specificity of 98.9% (95% CI 97.4-99.5%). Results were similar on the external test set: sensitivity 97.9% (95% CI 96.2-98.9%), specificity 99.6 (95% CI 98.9-99.9%).Fully automated detection of IVC filters on radiographs with high sensitivity and excellent specificity required for an automated screening system can be achieved using object detection neural networks. Further work will develop a system for identifying patients for IVC filter retrieval based on this algorithm.

The ideal radiology report reduces diagnostic uncertainty, while avoiding ambiguity whenever possible. The purpose of this study was to characterize the use of uncertainty terms in radiology reports at a single institution and compare the use of these terms across imaging modalities, anatomic sections, patient characteristics, and radiologist characteristics. We hypothesized that there would be variability among radiologists and between subspecialities within radiology regarding the use of uncertainty terms and that the length of the impression of a report would be a predictor of use of uncertainty terms. Finally, we hypothesized that use of uncertainty terms would often be interpreted by human readers as “hedging.” To test these hypotheses, we applied a natural language processing (NLP) algorithm to assess and count the number of uncertainty terms within radiology reports. An algorithm was created to detect usage of a published set of uncertainty terms. All 642,569 radiology report impressions from 171 reporting radiologists were collected from 2011 through 2015. For validation, two radiologists without knowledge of the software algorithm reviewed report impressions and were asked to determine whether the report was “uncertain” or “hedging.” The relationship between the presence of 1 or more uncertainty terms and the human readers’ assessment was compared. There were significant differences in the proportion of reports containing uncertainty terms across patient admission status and across anatomic imaging subsections. Reports with uncertainty were significantly longer than those without, although report length was not significantly different between subspecialities or modalities. There were no significant differences in rates of uncertainty when comparing the experience of the attending radiologist. When compared with reader 1 as a gold standard, accuracy was 0.91, sensitivity was 0.92, specificity was 0.9, and precision was 0.88, with an F1-score of 0.9. When compared with reader 2, accuracy was 0.84, sensitivity was 0.88, specificity was 0.82, and precision was 0.68, with an F1-score of 0.77. Substantial variability exists among radiologists and subspecialities regarding the use of uncertainty terms, and this variability cannot be explained by years of radiologist experience or differences in proportions of specific modalities. Furthermore, detection of uncertainty terms demonstrates good test characteristics for predicting human readers’ assessment of uncertainty.

Artificial intelligence is a technique that holds promise for helping radiologists improve the care of our patients. At the same time, implementation decisions we make now can have a long-lasting effect on patient outcomes. In the following article, we discuss four areas with unique considerations for implementation of AI: bias, trust, risk, and design. In each section, we highlight applications of AI to abdominal imaging and prostate cancer specifically.

The objective is to determine patients’ utilization rate of radiology image viewing through an online patient portal and to understand its impact on radiologists. IRB approval was waived. In this two-part, multi-institutional study, patients’ image viewing rate was retrospectively assessed, and radiologists were anonymously surveyed for the impact of patient imaging access on their workflow. Patient access to web-based image viewing via electronic patient portals was enabled at 3 institutions (all had open radiology reports) within the past 5 years. The number of exams viewed online was compared against the total number of viewable imaging studies. An anonymized survey was distributed to radiologists at the 3 institutions, and responses were collected over 2 months. Patients viewed 14.2% of available exams – monthly open rate varied from 7.3 to 41.0%. A total of 254 radiologists responded to the survey (response rate 32.8%); 204 were aware that patients could view images. The majority (155/204; 76.0%) felt no impact on their role as radiologists; 11.8% felt negative and 9.3% positive. The majority (63.8%) were never approached by patients. Of the 86 who were contacted, 46.5% were contacted once or twice, 46.5% 3–4 times a year, and 4.7% 3–4 times a month. Free text comments included support for healthcare transparency (71), concern for patient confusion and anxiety (45), and need for attention to radiology reports and image annotations (15). A small proportion of patients viewed their radiology images. Overall, patients’ image viewing had minimal impact on radiologists. Radiologists were seldom contacted by patients. While many radiologists feel supportive, some are concerned about causing patient confusion and suggest minor workflow modifications. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1007/s10278-021-00565-9.

Purpose The objectives of this study was to assess the performance of ultrasound (US) for suspected appendicitis in adult patients and to evaluated the additive value of short-interval (within 1 week) computed tomography (CT) or Magnetic Resonance Imaging (MRI) after performing an initial US. Methods In this IRB-approved, HIPAA-compliant, retrospective study, electronic medical records (EMRs) were queried for "US appendicitis" performed over a 2-year interval. EMR was reviewed for CT or MRI performed within 1 week of this exam, and if any new or additional information was available at subsequent exam. White count, patient disposition, and pathology, if surgery was performed, were also recorded. Results 682 patients underwent US for appendicitis over a 2-year duration, age range from 18 to 92 years (average: 30.1 years, M:F = 141:541). Findings showed 126/682 patients with normal appendix, 75/682 uncomplicated appendicitis, and 4/682 with complicated appendicitis. When performed, no additional findings were seen in these groups on short-interval CT or MRI. 2/682 patients had equivocal findings on US but eventually had normal appendix identified on CT. Four hundred and seventy-three patients had non-visualized appendix, of which only 14/473 (3.1%) eventually had appendicitis. Conclusions Ultrasound is an effective initial modality for evaluating appendicitis even in adult patients. Once a normal appendix, uncomplicated or complicated appendicitis is identified on US, no further imaging is necessary. Very few patients with non-visualization of the appendix eventually have appendicitis. Hence, these patients can be managed with active clinical follow-up rather than immediate CT or MRI. Symptoms and clinical scoring systems can be used for triage of these patients.

Purpose The Radiological Society of North America (RSNA) Pediatric Bone Age Machine Learning Challenge was created to show an application of machine learning (ML) and artificial intelligence (AI) in medical imaging, promote collaboration to catalyze AI model creation, and identify innovators in medical imaging. Materials and Methods The goal of this challenge was to solicit individuals and teams to create an algorithm or model using ML techniques that would accurately determine skeletal age in a curated data set of pediatric hand radiographs. The primary evaluation measure was the mean absolute distance (MAD) in months, which was calculated as the mean of the absolute values of the difference between the model estimates and those of the reference standard, bone age. Results A data set consisting of 14 236 hand radiographs (12 611 training set, 1425 validation set, 200 test set) was made available to registered challenge participants. A total of 260 individuals or teams registered on the Challenge website. A total of 105 submissions were uploaded from 48 unique users during the training, validation, and test phases. Almost all methods used deep neural network techniques based on one or more convolutional neural networks (CNNs). The best five results based on MAD were 4.2, 4.4, 4.4, 4.5, and 4.5 months, respectively. Conclusion The RSNA Pediatric Bone Age Machine Learning Challenge showed how a coordinated approach to solving a medical imaging problem can be successfully conducted. Future ML challenges will catalyze collaboration and development of ML tools and methods that can potentially improve diagnostic accuracy and patient care. © RSNA, 2018 Online supplemental material is available for this article. See also the editorial by Siegel in this issue.

Purpose To subjectively and quantitatively compare the quality of 3 Tesla magnetic resonance imaging of the prostate acquired with a novel flexible surface coil (FSC) and with a conventional endorectal coil (ERC). Methods Six radiologists independently reviewed 200 pairs of axial, high-resolution T2-weighted and diffusion-weighted image data sets, each containing one examination acquired with the FSC and one with the ERC, respectively. Readers selected their preferred examination from each pair and assessed every single examination using six quality criteria on 4-point scales. Signal-to-noise ratios were measured and compared. Results Two readers preferred FSC acquisition (36.5–45%) over ERC acquisition (13.5–15%) for both sequences combined, and four readers preferred ERC acquisition (41–46%). Analysis of pooled responses for both sequences from all readers shows no significant preference for FSC or ERC. Analysis of the individual sequences revealed a pooled preference for the FSC in T2WI (38.7% vs 17.8%) and for the ERC in DWI (50.9% vs 19.6%). Patients’ weight was the only weak predictor of a preference for the ERC acquisition (p = 0.04). SNR and CNR were significantly higher in the ERC acquisitions (pp=0.1). Conclusion Although readers have strong individual preferences, comparable subjective image quality can be obtained for prostate MRI with an ERC and the novel FSC. ERC imaging might be particularly valuable for sequences with inherently lower SNR as DWI and larger patients whereas the FSC is generally preferred in T2WI. FSC imaging generates a lower SNR than with an ERC.

Purpose The coronavirus disease 2019 (COVID-19) pandemic has led to significant disruptions in the healthcare system including surges of infected patients exceeding local capacity, closures of primary care offices, and delays of non-emergent medical care. Government-initiated measures to decrease healthcare utilization (i.e., “flattening the curve”) have included shelter-in-place mandates and social distancing, which have taken effect across most of the USA. We evaluate the immediate impact of the Public Health Messaging and shelter-in-place mandates on Emergency Department (ED) demand for radiology services. Methods We analyzed ED radiology volumes from the five University of California health systems during a 2-week time period following the shelter-in-place mandate and compared those volumes with March 2019 and early April 2019 volumes. Results ED radiology volumes declined from the 2019 baseline by 32 to 40% (p < 0.001) across the five health systems with a total decrease in volumes across all 5 systems by 35% (p < 0.001). Stratifying by subspecialty, the smallest declines were seen in non-trauma thoracic imaging, which decreased 18% (p value < 0.001), while all other non-trauma studies decreased by 48% (p < 0.001). Conclusion Total ED radiology demand may be a marker for public adherence to shelter-in-place mandates, though ED chest radiology demand may increase with an increase in COVID-19 cases. Electronic supplementary material The online version of this article (10.1007/s10140-020-01797-y) contains supplementary material, which is available to authorized users.

Many radiologists are considering investments in artificial intelligence (AI) to improve the quality of care for our patients. This article outlines considerations for the purchasing process beginning with performance evaluation. Practices should decide whether there is a need to independently verify performance or accept vendor-provided data. Successful implementations will consider who will receive AI results, how results will be presented, and the impact on efficiency. The article provides education on infrastructure considerations including the benefits and drawbacks of best-of-breed and platform approaches in addition to highly specialized server requirements like graphical processing unit availability. Finally, the article presents financial and quality and safety considerations, some of which are unique to AI. Examples include whether additional revenue could be obtained, as in the case of mammography, and whether an AI model unintentionally leads to reinforcing healthcare disparities.

Artificial intelligence is a technique that holds promise for helping radiologists improve the care of our patients. At the same time, implementation decisions we make now can have a long-lasting effect on patient outcomes. In the following article, we discuss four areas with unique considerations for implementation of AI: bias, trust, risk, and design. In each section, we highlight applications of AI to abdominal imaging and prostate cancer specifically.

We compared the prevalence of a baseline diagnosis of cancer in patients with and without misty mesentery (MM) and determined its association with the development of a new cancer. This was a retrospective, HIPAA-compliant, IRB-approved case-control study of 148 cases and 4:1 age- and gender-matched controls. Statistical tests included chi-square, t-test, hazard models, and C-statistic. Patients with MM were less likely to have cancer at baseline (RR=0.74, p=0.003), but more likely to develop a new malignancy on follow-up (RR=2.13, p=0.003; survival analysis HR 1.74, p=0.05). MM may confer an increased probability of later developing cancer, particularly genitourinary tumors.

Open-source development can provide a platform for innovation by seeking feedback from community members as well as providing tools and infrastructure to test new standards. Vendors of proprietary systems may delay adoption of new standards until there are sufficient incentives such as legal mandates or financial incentives to encourage/mandate adoption. Moreover, open-source systems in healthcare have been widely adopted in low- and middle-income countries and can be used to bridge gaps that exist in global health radiology. Since 2011, the authors, along with a community of open-source contributors, have worked on developing an open-source radiology information system (RIS) across two communities-OpenMRS and LibreHealth. The main purpose of the RIS is to implement core radiology workflows, on which others can build and test new radiology standards. This work has resulted in three major releases of the system, with current architectural changes driven by changing technology, development of new standards in health and imaging informatics, and changing user needs. At their core, both these communities are focused on building general-purpose EHR systems, but based on user contributions from the fringes, we have been able to create an innovative system that has been used by hospitals and clinics in four different countries. We provide an overview of the history of the LibreHealth RIS, the architecture of the system, overview of standards integration, describe challenges of developing an open-source product, and future directions. Our goal is to attract more participation and involvement to further develop the LibreHealth RIS into an Enterprise Imaging System that can be used in other clinical imaging including pathology and dermatology.

Health Level 7’s (HL7’s) new standard, FHIR (Fast Health Interoperability Resources), is setting healthcare information technology and medical imaging specifically ablaze with excitement. This paper aims to describe the protocol’s advantages in some detail and explore an easy path for those unfamiliar with FHIR to begin learning the standard using free, open-source tools, namely the HL7 application programming interface (HAPI) FHIR server and the SIIM Hackathon Dataset.

Numerous initiatives are in place to support value based care in radiology including decision support using appropriateness criteria, quality metrics like radiation dose monitoring, and efforts to improve the quality of the radiology report for consumption by referring providers. These initiatives are largely data driven. Organizations can choose to purchase proprietary registry systems, pay for software as a service solution, or deploy/build their own registry systems. Traditionally, registries are created for a single purpose like radiation dosage or specific disease tracking like diabetes registry. This results in a fragmented view of the patient, and increases overhead to maintain such single purpose registry system by requiring an alternative data entry workflow and additional infrastructure to host and maintain multiple registries for different clinical needs. This complexity is magnified in the health care enterprise whereby radiology systems usually are run parallel to other clinical systems due to the different clinical workflow for radiologists. In the new era of value based care where data needs are increasing with demand for a shorter turnaround time to provide data that can be used for information and decision making, there is a critical gap to develop registries that are more adapt to the radiology workflow with minimal overhead on resources for maintenance and setup. We share our experience of developing and implementing an open source registry system for quality improvement and research in our academic institution that is driven by our radiology workflow.

OBJECTIVE. The purposes of this article are to describe concepts that radiologists should understand to evaluate machine learning projects, including common algorithms, supervised as opposed to unsupervised techniques, statistical pitfalls, and data considerations for training and evaluation, and to briefly describe ethical dilemmas and legal risk. CONCLUSION. Machine learning includes a broad class of computer programs that improve with experience. The complexity of creating, training, and monitoring machine learning indicates that the success of the algorithms will require radiologist involvement for years to come, leading to engagement rather than replacement.

Respiratory diseases account for a significant proportion of deaths and disabilities across the world. Chest X-ray (CXR) analysis remains a common diagnostic imaging modality for confirming intra-thoracic cardiopulmonary abnormalities. However, there remains an acute shortage of expert radiologists, particularly in under-resourced settings, resulting in severe interpretation delays. These issues can be mitigated by a computer-aided diagnostic (CADx) system to supplement decision-making and improve throughput while preserving and possibly improving the standard-of-care. Systems reported in the literature or popular media use handcrafted features and/or data-driven algorithms like deep learning (DL) to learn underlying data distributions. The remarkable success of convolutional neural networks (CNN) toward image recognition tasks has made them a promising choice for automated medical image analyses. However, CNNs suffer from high variance and may overfit due to their sensitivity to training data fluctuations. Ensemble learning helps to reduce this variance by combining predictions of multiple learning algorithms to construct complex, non-linear functions and improve robustness and generalization. This study aims to construct and assess the performance of an ensemble of machine learning (ML) models applied to the challenge of classifying normal and abnormal CXRs and significantly reducing the diagnostic load of radiologists and primary-care physicians.

The purpose of this report is to describe our experience with the implementation of a practice quality improvement (PQI) project in thoracic imaging as part of the American Board of Radiology Maintenance of Certification process. The goal of this PQI project was to reduce the effective radiation dose of routine chest CT imaging in a busy clinical practice by employing the iDose(4) (Philips Healthcare) iterative reconstruction technique. The dose reduction strategy was implemented in a stepwise process on a single 64-slice CT scanner with a volume of 1141 chest CT scans during the year. In the first annual quarter, a baseline effective dose was established using the standard filtered back projection (FBP) algorithm protocol and standard parameters such as kVp and mAs. The iDose(4) technique was then applied in the second and third annual quarters while keeping all other parameters unchanged. In the fourth quarter, a reduction in kVp was also implemented. Throughout the process, the images were continually evaluated to assure that the image quality was comparable to the standard protocol from multiple other scanners. Utilizing a stepwise approach, the effective radiation dose was reduced by 23.62 and 43.63% in quarters two and four, respectively, compared to our initial standard protocol with no perceived difference in diagnostic quality. This practice quality improvement project demonstrated a significant reduction in the effective radiation dose of thoracic CT scans in a busy clinical practice.

Machine learning has potential to play a key role across a variety of medical imaging applications. This review seeks to elucidate the ways in which machine learning can aid and enhance diagnosis, treatment, and follow-up in neurooncology.Given the rapid pace of development in machine learning over the past several years, a basic proficiency of the key tenets and use cases in the field is critical to assessing potential opportunities and challenges of this exciting new technology.

Artificial intelligence (AI) will reshape radiology over the coming years. The radiology community has a strong history of embracing new technology for positive change, and AI is no exception. As with any new technology, rapid, successful implementation faces several challenges that will require creation and adoption of new integration technology. Use cases important to real-world application of AI are described, including clinical registries, AI research, AI product validation, and computer assistance for radiology reporting. Furthermore, the informatics technologies required for successful implementation of the use cases are described, including open Computer-Assisted Radiologist Decision Support, ACR Assist, ACR Data Science Institute use cases, common data elements (radelement.org), RadLex (radlex.org), LOINC/RSNA RadLex Playbook (loinc.org), and Radiology Report Templates (radreport.org).

With nearly 70% of adults in the United States using at least one social media platform, a social media presence is increasingly important for departments and practices. Patients, prospective faculty and trainees, and referring physicians look to social media to find information about our organizations. The authors present a stepwise process for planning, executing, and evaluating an organizational social media strategy. This process begins with alignment with a strategic plan to set goals, identification of the target audience(s), selection of appropriate social media channels, tracking effectiveness, and resource allocation. The article concludes with a discussion of advantages and disadvantages of social media through a review of current literature. ©RSNA, 2018.

Due to recent advances in artificial intelligence, there is renewed interest in automating interpretation of imaging tests. Chest radiographs are particularly interesting due to many factors: relatively inexpensive equipment, importance to public health, commonly performed throughout the world, and deceptively complex taking years to master. This article presents a brief introduction to artificial intelligence, reviews the progress to date in chest radiograph interpretation, and provides a snapshot of the available datasets and algorithms available to chest radiograph researchers. Finally, the limitations of artificial intelligence with respect to interpretation of imaging studies are discussed.

Chest x-ray (CXR) analysis is a common part of the protocol for confirming active pulmonary Tuberculosis (TB). However, many TB endemic regions are severely resource constrained in radiological services impairing timely detection and treatment. Computer-aided diagnosis (CADx) tools can supplement decision-making while simultaneously addressing the gap in expert radiological interpretation during mobile field screening. These tools use hand-engineered and/or convolutional neural networks (CNN) computed image features. CNN, a class of deep learning (DL) models, has gained research prominence in visual recognition. It has been shown that Ensemble learning has an inherent advantage of constructing non-linear decision making functions and improve visual recognition. We create a stacking of classifiers with hand-engineered and CNN features toward improving TB detection in CXRs. The results obtained are highly promising and superior to the state-of-the-art.

According to the World Health Organization (WHO), tuberculosis (TB) remains the most deadly infectious disease in the world. In a 2015 global annual TB report, 1.5 million TB related deaths were reported. The conditions worsened in 2016 with 1.7 million reported deaths and more than 10 million people infected with the disease. Analysis of frontal chest X-rays (CXR) is one of the most popular methods for initial TB screening, however, the method is impacted by the lack of experts for screening chest radiographs. Computer-aided diagnosis (CADx) tools have gained significance because they reduce the human burden in screening and diagnosis, particularly in countries that lack substantial radiology services. State-of-the-art CADx software typically is based on machine learning (ML) approaches that use hand-engineered features, demanding expertise in analyzing the input variances and accounting for the changes in size, background, angle, and position of the region of interest (ROI) on the underlying medical imagery. More automatic Deep Learning (DL) tools have demonstrated promising results in a wide range of ML applications. Convolutional Neural Networks (CNN), a class of DL models, have gained research prominence in image classification, detection, and localization tasks because they are highly scalable and deliver superior results with end-to-end feature extraction and classification. In this study, we evaluated the performance of CNN based DL models for population screening using frontal CXRs. The results demonstrate that pre-trained CNNs are a promising feature extracting tool for medical imagery including the automated diagnosis of TB from chest radiographs but emphasize the importance of large data sets for the most accurate classification.

In order to support innovation, the Society of Imaging Informatics in Medicine (SIIM) elected to create a collaborative computing experience called a “hackathon.” The SIIM Hackathon has always consisted of two components, the event itself and the infrastructure and resources provided to the participants. In 2014, SIIM provided a collection of servers to participants during the annual meeting. After initial server setup, it was clear that clinical and imaging “test” data were also needed in order to create useful applications. We outline the goals, thought process, and execution behind the creation and maintenance of the clinical and imaging data used to create DICOM and FHIR Hackathon resources.

As part of its ongoing effort to improve healthcare quality, the Center of Medicare and Medicaid Services (CMS) has transitioned from monetary rewards to reimbursement penalties for noncompliance or nonparticipation with its quality measurement initiatives. More specifically, eligible providers who bill for CMS patient care, such as radiologists, will face a 2 % negative payment adjustment, if they fail to report adequate participation and compliance with sufficient CMS quality measures in 2015. Although several methods exist to report participation and compliance, each method requires the gathering of relevant studies and assessing the reports for compliance. To aid in this data gathering and to prevent reduced reimbursements, radiology groups should consider implementing automated processes to monitor compliance with these quality measure standards. This article describes one method of creating an automated report scanner, utilizing an open source interface engine called Mirth Connect, that may facilitate the data gathering and monitoring related to reporting compliance with CMS standard #195 Stenosis measurement in Ultrasound Carotid Imaging Reports. The process described in this article is currently utilized by a large multi-institutional radiology group to assess for report compliance and offers the user near real time surveillance of compliance with the quality measure.

Informatics innovations of the past 30 years have improved radiology quality and efficiency immensely. Radiologists are groundbreaking leaders in clinical information technology (IT), and often radiologists and imaging informaticists created, specified, and implemented these technologies, while also carrying the ongoing burdens of training, maintenance, support, and operation of these IT solutions. Being pioneers of clinical IT had advantages of local radiology control and radiology-centric products and services. As health care businesses become more clinically IT savvy, however, they are standardizing IT products and procedures across the enterprise, resulting in the loss of radiologists' local control and flexibility. Although this inevitable consequence may provide new opportunities in the long run, several questions arise.What will happen to the informatics expertise within the radiology domain? Will radiology's current and future concerns be heard and their needs addressed? What should radiologists do to understand, obtain, and use informatics products to maximize efficiency and provide the most value and quality for patients and the greater health care community? This article will propose some insights and considerations as we rethink radiology informatics.

At the first annual Conference on Machine Intelligence in Medical Imaging (C-MIMI), held in September 2016, a conference session on medical image data and datasets for machine learning identified multiple issues. The common theme from attendees was that everyone participating in medical image evaluation with machine learning is data starved. There is an urgent need to find better ways to collect, annotate, and reuse medical imaging data. Unique domain issues with medical image datasets require further study, development, and dissemination of best practices and standards, and a coordinated effort among medical imaging domain experts, medical imaging informaticists, government and industry data scientists, and interested commercial, academic, and government entities. High-level attributes of reusable medical image datasets suitable to train, test, validate, verify, and regulate ML products should be better described. NIH and other government agencies should promote and, where applicable, enforce, access to medical image datasets. We should improve communication among medical imaging domain experts, medical imaging informaticists, academic clinical and basic science researchers, government and industry data scientists, and interested commercial entities.

Background By the traditional definition of unintended weight loss, cachexia develops in ~80% of patients with pancreatic ductal adenocarcinoma (PDAC). Here, we measure the longitudinal body composition changes in patients with advanced PDAC undergoing 5‐fluorouracil, leucovorin, irinotecan, and oxaliplatin therapy. Methods We performed a retrospective review of 53 patients with advanced PDAC on 5‐fluorouracil, leucovorin, irinotecan, and oxaliplatin as first line therapy at Indiana University Hospital from July 2010 to August 2015. Demographic, clinical, and survival data were collected. Body composition measurement by computed tomography (CT), trend, univariate, and multivariate analysis were performed. Results Among all patients, three cachexia phenotypes were identified. The majority of patients, 64%, had Muscle and Fat Wasting (MFW), while 17% had Fat‐Only Wasting (FW) and 19% had No Wasting (NW). NW had significantly improved overall median survival (OMS) of 22.6 months vs. 13.0 months for FW and 12.2 months for MFW (P = 0.02). FW (HR = 5.2; 95% confidence interval = 1.5–17.3) and MFW (HR = 1.8; 95% confidence interval = 1.1–2.9) were associated with an increased risk of mortality compared with NW. OMS and risk of mortality did not differ between FW and MFW. Progression of disease, sarcopenic obesity at diagnosis, and primary tail tumours were also associated with decreased OMS. On multivariate analysis, cachexia phenotype and chemotherapy response were independently associated with survival. Notably, CT‐based body composition analysis detected tissue loss of >5% in 81% of patients, while the traditional definition of >5% body weight loss identified 56.6%. Conclusions Distinct cachexia phenotypes were observed in this homogeneous population of patients with equivalent stage, diagnosis, and first‐line treatment. This suggests cellular, molecular, or genetic heterogeneity of host or tumour. Survival among patients with FW was as poor as for MFW, indicating adipose tissue plays a crucial role in cachexia and PDAC mortality. Adipose tissue should be studied for its mechanistic contributions to cachexia.

Tuberculosis (TB) is a severe comorbidity of HIV and chest x-ray (CXR) analysis is a necessary step in screening for the infective disease. Automatic analysis of digital CXR images for detecting pulmonary abnormalities is critical for population screening, especially in medical resource constrained developing regions. In this article, we describe steps that improve previously reported performance of NLM’s CXR screening algorithms and help advance the state of the art in the field. We propose a local-global classifier fusion method where two complementary classification systems are combined. The local classifier focuses on subtle and partial presentation of the disease leveraging information in radiology reports that roughly indicates locations of the abnormalities. In addition, the global classifier models the dominant spatial structure in the gestalt image using GIST descriptor for the semantic differentiation. Finally, the two complementary classifiers are combined using linear fusion, where the weight of each decision is calculated by the confidence probabilities from the two classifiers. We evaluated our method on three datasets in terms of the area under the Receiver Operating Characteristic (ROC) curve, sensitivity, specificity and accuracy. The evaluation demonstrates the superiority of our proposed local-global fusion method over any single classifier.

Technology standards form the basis for clinical workflow in radiology. This article reviews 3 types of standards relevant for radiology: the DICOM standard for handling images; the Health Level 7 standard for communicating with the health care enterprise; and standards in coding and terminology such as International Classification of Diseases, Current Procedural Terminology, and RadLex. This third category has an impact on radiology reporting and practice management. Familiarity with all these standards can help radiologists optimize operations and plan for the future.

Over the past 20 years, imaging informatics has been driven by the widespread adoption of radiology information and picture archiving and communication and speech recognition systems. These three clinical information systems are commonplace and are intuitive to most radiologists as they replicate familiar paper and film workflow. So what is next? There is a surge of innovation in imaging informatics around advanced workflow, search, electronic medical record aggregation, dashboarding, and analytics tools for quality measures (Nance et al., AJR Am J Roentgenol 200:1064–1070, 2013). The challenge lies in not having to rebuild the technological wheel for each of these new applications but instead attempt to share common components through open standards and modern development techniques. The next generation of applications will be built with moving parts that work together to satisfy advanced use cases without replicating databases and without requiring fragile, intense synchronization from clinical systems. The purpose of this paper is to identify building blocks that can position a practice to be able to quickly innovate when addressing clinical, educational, and research-related problems. This paper is the result of identifying common components in the construction of over two dozen clinical informatics projects developed at the University of Maryland Radiology Informatics Research Laboratory. The systems outlined are intended as a mere foundation rather than an exhaustive list of possible extensions.

The residency review committee of the Accreditation Council of Graduate Medical Education (ACGME) collects data on resident exam volume and sets minimum requirements. However, this data is not made readily available, and the ACGME does not share their tools or methodology. It is therefore difficult to assess the integrity of the data and determine if it truly reflects relevant aspects of the resident experience. This manuscript describes our experience creating a multi-institutional case log, incorporating data from three American diagnostic radiology residency programs. Each of the three sites independently established automated query pipelines from the various radiology information systems in their respective hospital groups, thereby creating a resident-specific database. Then, the three institutional resident case log databases were aggregated into a single centralized database schema. Three hundred thirty residents and 2,905,923 radiologic examinations over a 4-year span were catalogued using 11 ACGME categories. Our experience highlights big data challenges including internal data heterogeneity and external data discrepancies faced by informatics researchers.

Determining practice parameters for interventional procedures is challenging due to many factors including unreliable laboratory tests to measure bleeding risk, variable usage of standardized terminology for adverse events, poorly defined standards for administration of blood products, and the growing numbers of anticoagulant and antiplatelet medications. We aim to address these and other issues faced by radiologists performing invasive procedures through a review of available literature, and experiential guidance from three academic medical centers. We discuss the significant limitations with respect to using prothrombin-time and international normalized ratio to measure bleeding risk, especially in patients with synthetic defects due to liver function. Factors affecting platelet function including the impact of uremia; recent advances in laboratory testing, including platelet function testing; and thromboelastography are also discussed. A review of the existing literature of fresh-frozen plasma replacement therapy is included. The literature regarding comorbidities affecting coagulation including malignancy, liver failure, and uremia are also reviewed. Finally, the authors present a set of recommendations for laboratory thresholds, corrective transfusions, as well as withholding and restarting medications.

This study aims to determine whether a modified four-view hand and wrist study performs comparably to the traditional seven views in the evaluation of acute hand and wrist fractures. This retrospective study was approved by the institutional review board with waiver of informed consent. Two hundred forty patients (50 % male; ages 18–92 years) with unilateral three-view hand (posteroanterior, oblique, and lateral) and four-view wrist (posteroanterior, oblique, lateral, and ulnar deviation) radiographs obtained concurrently following trauma were included in this study. Four emergency radiologists interpreted the original seven images, with two radiologists independently evaluating each study. The patients’ radiographs were then recombined into four-view series using the three hand images and the ulnar deviated wrist image. These were interpreted by the same radiologists following an 8-week delay. Kappa statistics were generated to measure inter-observer and inter-method agreement. Generalized linear mixed model analysis was performed between the seven- and four-view methods. Of the 480 reports generated in each of the seven- and four-view image sets, 142 (29.6 %) of the seven-view and 126 (26.2 %) of the four-view reports conveyed certain or suspected acute osseous findings. Average inter-observer kappa coefficients were 0.7845 and 0.8261 for the seven- and four-view protocols, respectively. The average inter-method kappa was 0.823. The odds ratio of diagnosing injury using the four-view compared to the seven-view algorithm was 0.69 (CI 0.45–1.06, P = 0.0873). The modified four-view hand and wrist radiographic series produces diagnostic results comparable to the traditional seven views for acute fracture evaluation.

Determine factors that increase the risk of bleeding after liver biopsy. Retrospective review of radiology and clinical databases from Jan 2008 to Jun 2014 revealed 847 patients with liver biopsy. Of these, 154 (group I) had targeted biopsy of focal lesion and 142 (group 2) had random core biopsy for diffuse liver disease. The rest of the patients were excluded due to insufficient post-biopsy data. Data including pre-biopsy laboratory results, history of transfusion, and biopsy complications were recorded in the study cohort. After review of initial results, a “Risk Score” for bleeding was created using platelet count, INR, estimated glomerular filtration rate (eGFR), and suspicion of malignancy. Zero point was given for normal laboratory results or absence of malignancy. One point was given for mildly abnormal laboratory values or presence of malignancy. Severe biochemical abnormalities, e.g., INR > 2.0, eGFR

In recent years, there has been enormous interest in applying artificial intelligence (AI) to radiology. Although some of this interest may have been driven by exaggerated expectations that the technology can outperform radiologists in some tasks, there is a growing body of evidence that illustrates its limitations in medical imaging. The true potential of the technique probably lies somewhere in the middle, and AI will ultimately play a key role in medical imaging in the future. The limitless power of computers makes AI an ideal candidate to provide the standardization, consistency, and dependability needed to support radiologists in their mission to provide excellent patient care. However, important roadblocks currently limit the expansion of this field in medical imaging. This article reviews some of the challenges and potential solutions to advance the field forward, with focus on the experience gained by hosting image-based competitions.

The clinical treatment of patients with anorectal and pelvic floor dysfunction is often difficult. Dynamic cystocolpoproctography (DCP) has evolved from a method of evaluating the anorectum for functional disorders to its current status as a functional method of evaluating the global pelvic floor for defecatory disorders and pelvic organ prolapse. It has both high observer accuracy and a high yield of positive diagnoses. Clinicians find it a useful diagnostic tool that can alter management decisions from surgical to medical and vice versa in many cases. Functional radiography provides the maximum stress to the pelvic floor, resulting in levator ani relaxation accompanied by rectal emptying-which is needed to diagnose defecatory disorders. It also provides organ-specific quantificative information about female pelvic organ prolapse-information that usually can only be inferred by means of physical examination. The application of functional radiography to the assessment of defecatory disorders and pelvic organ prolapse has highlighted the limitations of physical examination. It has become clear that pelvic floor disorders rarely occur in isolation and that global pelvic floor assessment is necessary. Despite the advances in other imaging methods, DCP has remained a practical, cost-effective procedure for the evaluation of anorectal and pelvic floor dysfunction. In this article, the authors describe the technique they use when performing DCP, define the radiographic criteria used for diagnosis, and discuss the limitations and clinical utility of DCP.

Access to basic imaging and laboratory services remains a major challenge in rural, resource-limited settings in sub-Saharan Africa. In 2016, the Academic Model Providing Access to Healthcare programme in western Kenya implemented a mobile diagnostic unit (MDU) outfitted with a generator-powered X-ray machine and basic laboratory tests to address the lack of these services at rural, low-resource, public health facilities. The objective of this paper is to describe the design, implementation, preliminary impact and operational challenges of the MDU in western Kenya. Since implementing the MDU at seven rural health facilities serving a catchment of over half a million people, over 4500 chest radiographs have been performed, with one or more abnormalities detected in approximately 30% of radiographs. We observed favorable feedback and uptake of MDU services by healthcare workers and patients. However, various operational challenges in the design and construction of the MDU and the transmission and reporting of radiographs in remote areas were encountered. Our experience supports the feasibility of deploying an MDU to increase access to basic radiology and laboratory services in rural, resource-limited settings.

Training as a radiology resident is a complex task. Residents frequently encounter multiple hospital systems, each with unique workflow patterns and heterogenous information systems. We identified an opportunity to ease some of the resulting anxiety and frustration by centralizing high-quality resources using a wiki. In this manuscript, we describe our choice of wiki software, give basic information about hardware requirements, detail steps for configuration, outline information included on the wiki, and present the results of a resident acceptance survey.

The timing of surgical intervention as well as the optimal method of radiologic investigation for patients with incomplete, open loop small bowel obstruction has changed over the past two decades. This review focuses on the role of computed tomographic enteroclysis in the evaluation of patients with suspected small bowel obstruction. The technique of examination is described and an overview of its clinical applications and imaging controversy are presented.

Objective Clinical documents made available for secondary use play an increasingly important role in discovery of clinical knowledge, development of research methods, and education. An important step in facilitating secondary use of clinical document collections is easy access to descriptions and samples that represent the content of the collections. This paper presents an approach to developing a collection of radiology examinations, including both the images and radiologist narrative reports, and making them publicly available in a searchable database. Materials and Methods The authors collected 3996 radiology reports from the Indiana Network for Patient Care and 8121 associated images from the hospitals’ picture archiving systems. The images and reports were de-identified automatically and then the automatic de-identification was manually verified. The authors coded the key findings of the reports and empirically assessed the benefits of manual coding on retrieval. Results The automatic de-identification of the narrative was aggressive and achieved 100% precision at the cost of rendering a few findings uninterpretable. Automatic de-identification of images was not quite as perfect. Images for two of 3996 patients (0.05%) showed protected health information. Manual encoding of findings improved retrieval precision. Conclusion Stringent de-identification methods can remove all identifiers from text radiology reports. DICOM de-identification of images does not remove all identifying information and needs special attention to images scanned from film. Adding manual coding to the radiologist narrative reports significantly improved relevancy of the retrieved clinical documents. The de-identified Indiana chest X-ray collection is available for searching and downloading from the National Library of Medicine ( http://openi.nlm.nih.gov/ ).