Evidence-based radiology: why and how?

Impact and effect of imaging referral guidelines on patients and radiology services: a systematic review

OBJECTIVES

The objective of this systematic review was to offer a comprehensive overview and explore the associated outcomes from imaging referral guidelines on various key stakeholders, such as patients and radiologists.

MATERIALS AND METHODS

An electronic database search was conducted in Medline, Embase and Web of Science to retrieve citations published between 2013 and 2023. The search was constructed using medical subject headings and keywords. Only full-text articles and reviews written in English were included. The quality of the included papers was assessed using the mixed methods appraisal tool. A narrative synthesis was undertaken for the selected articles.

RESULTS

The search yielded 4384 records. Following the abstract, full-text screening, and removal of duplication, 31 studies of varying levels of quality were included in the final analysis. Imaging referral guidelines from the American College of Radiology were most commonly used. Clinical decision support systems were the most evaluated mode of intervention, either integrated or standalone. Interventions showed reduced patient radiation doses and waiting times for imaging. There was a general reduction in radiology workload and utilisation of diagnostic imaging. Low-value imaging utilisation decreased with an increase in the appropriateness of imaging referrals and ratings and cost savings. Clinical effectiveness was maintained during the intervention period without notable adverse consequences.

CONCLUSION

Using evidence-based imaging referral guidelines improves the quality of healthcare and outcomes while reducing healthcare costs. Imaging referral guidelines are one essential component of improving the value of radiology in the healthcare system.

CLINICAL RELEVANCE STATEMENT

There is a need for broader dissemination of imaging referral guidelines to healthcare providers globally in tandem with the harmonisation of the application of these guidelines to improve the overall value of radiology within the healthcare system.

KEY POINTS

The application of imaging referral guidelines has an impact and effect on patients, radiologists, and health policymakers. The adoption of imaging referral guidelines in clinical practice can impact healthcare costs and improve healthcare quality and outcomes. Implementing imaging referral guidelines contributes to the attainment of value-based radiology.

Sample size calculation for data reliability and diagnostic performance: a go-to review

Sample size, namely the number of subjects that should be included in a study to reach the desired endpoint and statistical power, is a fundamental concept of scientific research. Indeed, sample size must be planned a priori, and tailored to the main endpoint of the study, to avoid including too many subjects, thus possibly exposing them to additional risks while also wasting time and resources, or too few subjects, failing to reach the desired purpose. We offer a simple, go-to review of methods for sample size calculation for studies concerning data reliability (repeatability/reproducibility) and diagnostic performance. For studies concerning data reliability, we considered Cohen's κ or intraclass correlation coefficient (ICC) for hypothesis testing, estimation of Cohen's κ or ICC, and Bland-Altman analyses. With regards to diagnostic performance, we considered accuracy or sensitivity/specificity versus reference standards, the comparison of diagnostic performances, and the comparisons of areas under the receiver operating characteristics curve. Finally, we considered the special cases of dropouts or retrospective case exclusions, multiple endpoints, lack of prior data estimates, and the selection of unusual thresholds for α and β errors. For the most frequent cases, we provide example of software freely available on the Internet.Relevance statement Sample size calculation is a fundamental factor influencing the quality of studies on repeatability/reproducibility and diagnostic performance in radiology.Key points• Sample size is a concept related to precision and statistical power.• It has ethical implications, especially when patients are exposed to risks.• Sample size should always be calculated before starting a study.• This review offers simple, go-to methods for sample size calculations.

ESR essentials: MRI of the knee-practice recommendations by ESSR

Many studies and systematic reviews have been published about MRI of the knee and its structures, discussing detailed anatomy, imaging findings, and correlations between imaging and clinical findings. This paper includes evidence-based recommendations for a general radiologist regarding choice of imaging sequences and reporting basic MRI examinations of the knee. We recommend using clinicians' terminology when it is applicable to the imaging findings, for example, when reporting meniscal, ligament and tendon, or cartilage pathology. The intent is to standardise reporting language and to make reports less equivocal. The aim of the paper is to improve the usefulness of the MRI report by understanding the strengths and limitations of the MRI exam with regard to clinical correlation. We hope the implementation of these recommendations into radiological practice will increase diagnostic accuracy and consistency by avoiding pitfalls and reducing overcalling of pathology on MRI of the knee. CLINICAL RELEVANCE STATEMENT: The recommendations presented here are meant to aid general radiologists in planning and assessing studies to evaluate acute and chronic knee findings by advocating the use of unequivocal terminology and discussing the strengths and limitations of MRI examination of the knee. KEY POINTS: • On MRI, the knee should be examined and assessed in three orthogonal imaging planes. • The basic general protocol must yield T2-weighted fluid-sensitive and T1-weighted images. • The radiological assessment should include evaluation of ligamentous structures, cartilage, bony structures and bone marrow, soft tissues, bursae, alignment, and incidental findings.

Evolving paradigms in breast cancer screening: Balancing efficacy, personalization, and equity

Breast cancer remains a significant global health challenge, with projections indicating a troubling increase in incidence. Breast cancer screening programs have long been hailed as life-saving initiatives, yet their true impact on mortality rates is a subject of ongoing debate. Screening poses the risk of false positives and the detection of indolent tumors, potentially leading to overtreatment. Bias factors, including lead time, length time, and selection biases, further complicate the assessment of screening efficacy. Recent studies suggest that AI-driven image analysis may revolutionize breast cancer screening, maintaining diagnostic accuracy while reducing radiologists' workload. However, the generalizability of these findings to diverse populations is a critical consideration. Personalized screening approaches and equitable access to advanced technologies are essential to mitigate disparities. In conclusion, the breast cancer screening landscape is evolving, emphasizing the need for risk stratification, appropriate imaging modalities, and a personalized approach to reduce overdiagnosis and focus on cancers with the potential to impact lives while prioritizing patient-centered care.

Radiology AI Deployment and Assessment Rubric (RADAR) to bring value-based AI into radiological practice

OBJECTIVE

To provide a comprehensive framework for value assessment of artificial intelligence (AI) in radiology.

METHODS

This paper presents the RADAR framework, which has been adapted from Fryback and Thornbury's imaging efficacy framework to facilitate the valuation of radiology AI from conception to local implementation. Local efficacy has been newly introduced to underscore the importance of appraising an AI technology within its local environment. Furthermore, the RADAR framework is illustrated through a myriad of study designs that help assess value.

RESULTS

RADAR presents a seven-level hierarchy, providing radiologists, researchers, and policymakers with a structured approach to the comprehensive assessment of value in radiology AI. RADAR is designed to be dynamic and meet the different valuation needs throughout the AI's lifecycle. Initial phases like technical and diagnostic efficacy (RADAR-1 and RADAR-2) are assessed pre-clinical deployment via in silico clinical trials and cross-sectional studies. Subsequent stages, spanning from diagnostic thinking to patient outcome efficacy (RADAR-3 to RADAR-5), require clinical integration and are explored via randomized controlled trials and cohort studies. Cost-effectiveness efficacy (RADAR-6) takes a societal perspective on financial feasibility, addressed via health-economic evaluations. The final level, RADAR-7, determines how prior valuations translate locally, evaluated through budget impact analysis, multi-criteria decision analyses, and prospective monitoring.

CONCLUSION

The RADAR framework offers a comprehensive framework for valuing radiology AI. Its layered, hierarchical structure, combined with a focus on local relevance, aligns RADAR seamlessly with the principles of value-based radiology.

CRITICAL RELEVANCE STATEMENT

The RADAR framework advances artificial intelligence in radiology by delineating a much-needed framework for comprehensive valuation.

KEYPOINTS

• Radiology artificial intelligence lacks a comprehensive approach to value assessment. • The RADAR framework provides a dynamic, hierarchical method for thorough valuation of radiology AI. • RADAR advances clinical radiology by bridging the artificial intelligence implementation gap.

O-RADS MRI to classify adnexal tumors: from clinical problem to daily use

Eighteen to 35% of adnexal masses remain non-classified following ultrasonography, leading to unnecessary surgeries and inappropriate management. This finding led to the conclusion that ultrasonography was insufficient to accurately assess adnexal masses and that a standardized MRI criteria could improve these patients' management. The aim of this work is to present the different steps from the identification of the clinical issue to the daily use of a score and its inclusion in the latest international guidelines. The different steps were the following: (1) preliminary work to formalize the issue, (2) physiopathological analysis and finding dynamic parameters relevant to increase MRI performances, (3) construction and internal validation of a score to predict the nature of the lesion, (4) external multicentric validation (the EURAD study) of the score named O-RADS MRI, and (5) communication and education work to spread its use and inclusion in guidelines. Future steps will include studies at patients' levels and a cost-efficiency analysis. Critical relevance statement We present translating radiological research into a clinical application based on a step-by-step structured and systematic approach methodology to validate MR imaging for the characterization of adnexal mass with the ultimate step of incorporation in the latest worldwide guidelines of the O-RADS MRI reporting system that allows to distinguish benign from malignant ovarian masses with a sensitivity and specificity higher than 90%. Key points • The initial diagnostic test accuracy studies show the limitation of a preoperative assessment of adnexal masses using solely ultrasonography.• The technical developments (DCE/DWI) were investigated with the value of dynamic MRI to accurately predict the nature of benign or malignant lesions to improve management.• The first developing score named ADNEX MR Score was constructed using multiple easily assessed criteria on MRI to classify indeterminate adnexal lesions following ultrasonography.• The multicentric adnexal study externally validated the score creating the O-RADS MR score and leading to its inclusion for daily use in international guidelines.

Translating radiological research into practice-from discovery to clinical impact

At the European Society of Radiology (ESR), we strive to provide evidence for radiological practices that improve patient outcomes and have a societal impact. Successful translation of radiological research into clinical practice requires multiple factors including tailored methodology, a multidisciplinary approach aiming beyond technical validation, and a focus on unmet clinical needs. Low levels of evidence are a threat to radiology, resulting in low visibility and credibility. Here, we provide the background and rationale for the thematic series Translating radiological research into practice-from discovery to clinical impact, inviting authors to describe their processes of achieving clinically impactful radiological research. We describe the challenges unique to radiological research. Additionally, a survey was sent to non-radiological clinical societies. The majority of respondents (6/11) were in the field of gastrointestinal/abdominal medicine. The implementation of CT/MRI techniques for disease characterisation, detection and staging of cancer, and treatment planning and radiological interventions were mentioned as the most important radiological developments in the past years. The perception was that patients are substantially unaware of the impact of these developments. Unmet clinical needs were mostly early diagnosis and staging of cancer, microstructural/functional assessment of tissues and organs, and implant assessment. All but one respondent considered radiology important for research in their discipline, but five indicated that radiology is currently not involved in their research. Radiology research holds the potential for being transformative to medical practice. It is our responsibility to take the lead in studies including radiology and strive towards the highest levels of evidence.Critical relevance statement For radiological research to make a clinical and societal impact, radiologists should take the lead in radiological studies, go beyond the assessment of technical feasibility and diagnostic accuracy, and-in a multidisciplinary approach-address clinical unmet needs.Key points• Multiple factors are essential for radiological research to make a clinical and societal impact.• Radiological research needs to go beyond diagnostic accuracy and address unmet clinical needs.• Radiologists should take the lead in radiological studies with a multidisciplinary approach.

New Horizons in Vascular Imaging With Photon-Counting Detector CT

ABSTRACT

Noninvasive vascular imaging with computed tomography (CT) has become the clinical mainstay for many indications and body regions. The recent introduction of photon-counting detector (PCD)-CT into clinical routine has further broadened the spectrum of vascular applications. Technical improvements of PCD-CT, such as the decreased noise levels, improved contrast-to-noise ratio, and full spectral multienergy data information from every acquisition, have the potential to further impact on clinical decision making and ultimately on outcome of vascular patients. Early experience with the new PCD-CT technology demonstrates these improvements in various aspects. This review summarizes the main advantages of PCD-CT for vascular imaging a discussion of the PureLumen and PureCalcium algorithms.

Open issues for education in radiological research: data integrity, study reproducibility, peer-review, levels of evidence, and cross-fertilization with data scientists

We are currently facing extraordinary changes. A harder and harder competition in the field of science is open in each country as well as in continents and worldwide. In this context, what should we teach to young students and doctors? There is a need to look backward and return to "fundamentals", i.e. the deep characteristics that must characterize the research in every field, even in radiology. In this article, we focus on data integrity (including the "declarations" given by the authors who submit a manuscript), reproducibility of study results, and the peer-review process. In addition, we highlight the need of raising the level of evidence of radiological research from the estimation of diagnostic performance to that of diagnostic impact, therapeutic impact, patient outcome, and social impact. Finally, on the emerging topic of radiomics and artificial intelligence, the recommendation is to aim for cross-fertilization with data scientists, possibly involving them in the clinical departments.

Framework for health care quality and evidence-based practice in radiology departments: A regional study on radiographer's perceptions

INTRODUCTION

The main goals of this study were to describe, in an integrated and multidimensional way, the conditions related to the quality of care in radiology departments from Algarve (Portugal), to assess the perspective of radiographers on the use of scientific evidence in clinical practice and to validate a model that characterizes the conditions for continuous improvement.

METHODS

A cross sectional study was performed in four radiology departments from public and private healthcare facilities from Algarve region (Portugal). A paper-based survey was sent to all radiographers to assess the quality systems implemented in their radiology departments and their perspective on the use of scientific evidence in clinical practice.

RESULTS

In total, 62 radiographers (61.4%) completed the survey. The quality dimensions that obtained the highest degree of compliance were the existence of quality assurance and improvement activities (43.0%), existence of standards in clinical practice of radiographers (42.7%) and the existence of special provisions (37.6%). The quality dimension related to patient's involvement was the one with the lowest level of compliance. Moreover, from the radiographers perspective, positive responses were obtained related to evidence-based actions (83.0%), sources of evidence (76.0%) and the significance of research activities (74.0%).

CONCLUSION

These findings suggest that a new framework based on four factors (Support for Information; Organizational Capability to Technical Quality of Care; Patient Involvement and Evidence-Based Radiology), should be considered in the establishment of strategic policies that better define the provision of diagnostic procedures and professional practices in radiology departments from Algarve region, based on quality improvement systems and better patient safety.

IMPLICATIONS FOR PRACTICE

There is a need to include patients in the decision-making process, to involve radiographers in quality assurance and improvement activities and to implement quality monitoring mechanisms within radiology departments under study.

Deep Mining Generation of Lung Cancer Malignancy Models from Chest X-ray Images

Lung cancer is the leading cause of cancer death and morbidity worldwide. Many studies have shown machine learning models to be effective in detecting lung nodules from chest X-ray images. However, these techniques have yet to be embraced by the medical community due to several practical, ethical, and regulatory constraints stemming from the “black-box” nature of deep learning models. Additionally, most lung nodules visible on chest X-rays are benign; therefore, the narrow task of computer vision-based lung nodule detection cannot be equated to automated lung cancer detection. Addressing both concerns, this study introduces a novel hybrid deep learning and decision tree-based computer vision model, which presents lung cancer malignancy predictions as interpretable decision trees. The deep learning component of this process is trained using a large publicly available dataset on pathological biomarkers associated with lung cancer. These models are then used to inference biomarker scores for chest X-ray images from two independent data sets, for which malignancy metadata is available. Next, multi-variate predictive models were mined by fitting shallow decision trees to the malignancy stratified datasets and interrogating a range of metrics to determine the best model. The best decision tree model achieved sensitivity and specificity of 86.7% and 80.0%, respectively, with a positive predictive value of 92.9%. Decision trees mined using this method may be considered as a starting point for refinement into clinically useful multi-variate lung cancer malignancy models for implementation as a workflow augmentation tool to improve the efficiency of human radiologists.

Human, All Too Human? An All-Around Appraisal of the "Artificial Intelligence Revolution" in Medical Imaging

Artificial intelligence (AI) has seen dramatic growth over the past decade, evolving from a niche super specialty computer application into a powerful tool which has revolutionized many areas of our professional and daily lives, and the potential of which seems to be still largely untapped. The field of medicine and medical imaging, as one of its various specialties, has gained considerable benefit from AI, including improved diagnostic accuracy and the possibility of predicting individual patient outcomes and options of more personalized treatment. It should be noted that this process can actively support the ongoing development of advanced, highly specific treatment strategies (e.g., target therapies for cancer patients) while enabling faster workflow and more efficient use of healthcare resources. The potential advantages of AI over conventional methods have made it attractive for physicians and other healthcare stakeholders, raising much interest in both the research and the industry communities. However, the fast development of AI has unveiled its potential for disrupting the work of healthcare professionals, spawning concerns among radiologists that, in the future, AI may outperform them, thus damaging their reputations or putting their jobs at risk. Furthermore, this development has raised relevant psychological, ethical, and medico-legal issues which need to be addressed for AI to be considered fully capable of patient management. The aim of this review is to provide a brief, hopefully exhaustive, overview of the state of the art of AI systems regarding medical imaging, with a special focus on how AI and the entire healthcare environment should be prepared to accomplish the goal of a more advanced human-centered world.

Content-Based Medical Image Retrieval and Intelligent Interactive Visual Browser for Medical Education, Research and Care

Medical imaging is essential nowadays throughout medical education, research, and care. Accordingly, international efforts have been made to set large-scale image repositories for these purposes. Yet, to date, browsing of large-scale medical image repositories has been troublesome, time-consuming, and generally limited by text search engines. A paradigm shift, by means of a query-by-example search engine, would alleviate these constraints and beneficially impact several practical demands throughout the medical field. The current project aims to address this gap in medical imaging consumption by developing a content-based image retrieval (CBIR) system, which combines two image processing architectures based on deep learning. Furthermore, a first-of-its-kind intelligent visual browser was designed that interactively displays a set of imaging examinations with similar visual content on a similarity map, making it possible to search for and efficiently navigate through a large-scale medical imaging repository, even if it has been set with incomplete and curated metadata. Users may, likewise, provide text keywords, in which case the system performs a content- and metadata-based search. The system was fashioned with an anonymizer service and designed to be fully interoperable according to international standards, to stimulate its integration within electronic healthcare systems and its adoption for medical education, research and care. Professionals of the healthcare sector, by means of a self-administered questionnaire, underscored that this CBIR system and intelligent interactive visual browser would be highly useful for these purposes. Further studies are warranted to complete a comprehensive assessment of the performance of the system through case description and protocolized evaluations by medical imaging specialists.

To buy or not to buy-evaluating commercial AI solutions in radiology (the ECLAIR guidelines)

Abstract

Artificial intelligence (AI) has made impressive progress over the past few years, including many applications in medical imaging. Numerous commercial solutions based on AI techniques are now available for sale, forcing radiology practices to learn how to properly assess these tools. While several guidelines describing good practices for conducting and reporting AI-based research in medicine and radiology have been published, fewer efforts have focused on recommendations addressing the key questions to consider when critically assessing AI solutions before purchase. Commercial AI solutions are typically complicated software products, for the evaluation of which many factors are to be considered. In this work, authors from academia and industry have joined efforts to propose a practical framework that will help stakeholders evaluate commercial AI solutions in radiology (the ECLAIR guidelines) and reach an informed decision. Topics to consider in the evaluation include the relevance of the solution from the point of view of each stakeholder, issues regarding performance and validation, usability and integration, regulatory and legal aspects, and financial and support services.

Key Points

• Numerous commercial solutions based on artificial intelligence techniques are now available for sale, and radiology practices have to learn how to properly assess these tools.

• We propose a framework focusing on practical points to consider when assessing an AI solution in medical imaging, allowing all stakeholders to conduct relevant discussions with manufacturers and reach an informed decision as to whether to purchase an AI commercial solution for imaging applications.

• Topics to consider in the evaluation include the relevance of the solution from the point of view of each stakeholder, issues regarding performance and validation, usability and integration, regulatory and legal aspects, and financial and support services.

Supplementary Information

The online version contains supplementary material available at 10.1007/s00330-020-07684-x.

Cost-risk-benefit analysis in diagnostic radiology with special reference to the application of referral guidelines

Cost-risk-benefit analysis has been applied to protection of the patient in diagnostic radiology with special reference to the application of referral guidelines. The analysis presented has extended previous work in this field to provide a theoretical framework that encompasses key factors that need to be considered in the optimisation of patient protection from both diagnostic and radiation risks. The fraction of patients whose symptoms do not meet criteria contained in the referral guidelines and, therefore, for whom an X-ray examination is not indicated has been termed the selectivity of the guidelines. Also included are the detriments arising from rejected or repeated examinations as well as the levels of patient dose employed in order to achieve given levels of true and false diagnostic outcomes. A comprehensive framework for optimisation is outlined and its relationship to justification discussed.

How scientific mobility can help current and future radiology research: a radiology trainee's perspective

One of the ways in which modern radiology is manifesting itself in higher education and research is through the increasing importance of scientific mobility. This article seeks to provide an overview and a prospective of radiology fellows in their last year of training about the current trends and policy tools for promoting mobility among young radiologists, especially inside the European Union. Nowadays, the need to promote international cooperation is even greater to ensure that the best evidence-based medical practices become a common background of a next cross-border generation of radiologists. Organisations such as the European Society of Radiology (ESR) and the Radiological Society of North America (RSNA) are called upon to play as guarantors of the training of young radiologists building know-how and world-class excellence. Today, it is not just being certified radiologist that matters, the place where the training was done plays an important role in enhancing chances when applying for a high-level job or fellowship. The article argues that the mobility of radiology trainees is an indispensable prerequisite to face new challenges, including the application of artificial intelligence to medical imaging, which will require a large multicentre collaboration.

Radiological Clinical Practice Guidelines Published in the Last Decade: A Bibliometric Analysis

Objectives:

To provide a comprehensive review of radiological clinical practice guidelines (CPGs) and to establish their characteristics and impact in the field of radiology.

Material and Methods:

A MEDLINE search was performed for CPGs in which at least half of the authors were from the radiology or imaging department. The following information was extracted from each CPG: year of publication, journal, provider, number of authors, number of pages, number of references, collaboration, country of origin, radiological subspecialty, imaging modality used, topic, source of funding, and number and pattern of citations.

Results:

In total, 120 radiological CPGs published between July 2006 and June 2016 were identified. One hundred nine (90.8%) radiological CPGs were published in radiology journals, 96 (80.0%) were provided by the scientific community, 108 (90.0%) were collaborative studies, 64 (53.3%) originated from the United States, 36 (30.0%) were concerned with the field of vascular/interventional radiology, 38 (31.7%) used combined imaging techniques, 52 (43.3%) were focused on interpretation and management, and 118 (98.4%) were not funded. Radiological CPGs included a median of 8 authors, 9 pages, and 49 references. The median number of citations and annual citations were 18 (range, 0–540) and 3.5 (range, 0–75.6), respectively.

Conclusion:

Our study presents several interesting insights into the characteristics and impact of radiological CPGs.

Encouraging MSK imaging research towards clinical impact is a necessity: opinion paper of the European Society of Musculoskeletal Radiology (ESSR)

Radiology has not been spared in recent economic crises with a substantial reduction in the turnover of imaging equipment. These problems are exacerbated by increasing demand for healthcare across Europe. Therefore, using existing radiological services while rigorously following evidence-based guidelines might improve patient care. Thus, diagnostic pathways should be assessed not only for technical and diagnostic performance but also for their impact on medical and social outcome. In this paper, we report the advice of the Research Committee of ESSR on how we may guide musculoskeletal radiological research towards studies that have useful clinical impact. The ESSR Research Committee intends to encourage research with potential to influence treatment, patient outcome, and social impact. Key Points • Research in medical imaging has the potential to improve human health. • High-level studies have the potential to place radiology at the pinnacle of quality in evidence-based practice. • The ESSR Research Committee intends to encourage research with potential to influence treatment, patient outcome, and social impact.

Clinical decision support: the role of ACR Appropriateness Criteria

Clinical decision support is a way to decrease inappropriate imaging exams and promote judicious use of imaging resources. The adoption of clinical decision support will be incentivized by requiring the use of approved mechanisms to qualify for Medicare reimbursement starting in January 2020. Insurance providers base their reimbursement policies on Medicare, so clinical decision support could soon become relevant to pediatric imaging. We present the process behind the American College of Radiology (ACR) Appropriateness Criteria (a set of appropriate use criteria developed by the ACR) that will form the basis for software that can be used to fulfill the criteria for clinical decision support. For most organizations, this software is expected to be the easiest way to implement clinical decision support. Clinical decision support will affect how providers order imaging exams. This article should help readers understand how clinical decision support is expected to change the practice of the ordering providers, how the ACR Appropriateness Criteria are related to clinical decision support and how the ACR Appropriateness Criteria are developed. This will help the interpreting radiologist better communicate with the referring clinician, including informing the latter about how the clinical decision support software is making decisions.

Artificial intelligence in medical imaging: threat or opportunity? Radiologists again at the forefront of innovation in medicine

One of the most promising areas of health innovation is the application of artificial intelligence (AI), primarily in medical imaging. This article provides basic definitions of terms such as “machine/deep learning” and analyses the integration of AI into radiology. Publications on AI have drastically increased from about 100–150 per year in 2007–2008 to 700–800 per year in 2016–2017. Magnetic resonance imaging and computed tomography collectively account for more than 50% of current articles. Neuroradiology appears in about one-third of the papers, followed by musculoskeletal, cardiovascular, breast, urogenital, lung/thorax, and abdomen, each representing 6–9% of articles. With an irreversible increase in the amount of data and the possibility to use AI to identify findings either detectable or not by the human eye, radiology is now moving from a subjective perceptual skill to a more objective science. Radiologists, who were on the forefront of the digital era in medicine, can guide the introduction of AI into healthcare. Yet, they will not be replaced because radiology includes communication of diagnosis, consideration of patient’s values and preferences, medical judgment, quality assurance, education, policy-making, and interventional procedures. The higher efficiency provided by AI will allow radiologists to perform more value-added tasks, becoming more visible to patients and playing a vital role in multidisciplinary clinical teams.

Artificial intelligence as a medical device in radiology: ethical and regulatory issues in Europe and the United States

Abstract

Worldwide interest in artificial intelligence (AI) applications is growing rapidly. In medicine, devices based on machine/deep learning have proliferated, especially for image analysis, presaging new significant challenges for the utility of AI in healthcare. This inevitably raises numerous legal and ethical questions. In this paper we analyse the state of AI regulation in the context of medical device development, and strategies to make AI applications safe and useful in the future. We analyse the legal framework regulating medical devices and data protection in Europe and in the United States, assessing developments that are currently taking place. The European Union (EU) is reforming these fields with new legislation (General Data Protection Regulation [GDPR], Cybersecurity Directive, Medical Devices Regulation, In Vitro Diagnostic Medical Device Regulation). This reform is gradual, but it has now made its first impact, with the GDPR and the Cybersecurity Directive having taken effect in May, 2018. As regards the United States (U.S.), the regulatory scene is predominantly controlled by the Food and Drug Administration. This paper considers issues of accountability, both legal and ethical. The processes of medical device decision-making are largely unpredictable, therefore holding the creators accountable for it clearly raises concerns. There is a lot that can be done in order to regulate AI applications. If this is done properly and timely, the potentiality of AI based technology, in radiology as well as in other fields, will be invaluable.

Teaching Points

• AI applications are medical devices supporting detection/diagnosis, work-flow, cost-effectiveness.

• Regulations for safety, privacy protection, and ethical use of sensitive information are needed.

• EU and U.S. have different approaches for approving and regulating new medical devices.

• EU laws consider cyberattacks, incidents (notification and minimisation), and service continuity.

• U.S. laws ask for opt-in data processing and use as well as for clear consumer consent.

Breast Arterial Calcification on screening mammography can predict significant Coronary Artery Disease in women

INTRODUCTION

Breast Arterial Calcification (BAC) on digital mammography has been associated with an increased risk of Coronary Artery Disease (CAD). We aimed to investigate the association of BAC with findings on Coronary Computed Tomography Angiography (CCTA) within a cohort of women from the national breast screening program.

METHODS

Symptomatic women (chest pain) aged between 50 and 65 who underwent a CCTA and who also had a screening mammography between 2014 and 2015 were recorded. BAC and CAD-RADS™: Coronary Artery Disease-Reporting and Data System were scored by separate blinded specialist radiologists. Cardiac risk factors were recorded. Patients' cardiac follow up (with Exercise Stress Test, Percutaneous Coronary Intervention or echocardiography) and cardio-protective medications were also documented.

RESULTS

219 eligible women underwent a CCTA. Of these, 104 patients also underwent digital mammography. Using standard linear regression BAC was identified as a significant predictor of CAD-RADs ≥3 disease. Using binomial logistic regression, BAC remained associated with CAD-RADs ≥3 (p=0.023). A significantly higher proportion of patients with BAC >1 were on cardio-protective medications (p=0.041) and had medications initiated or changed, or had further cardiac investigation (p=0.037 and p=0.019, respectively) than those with no BAC, after a mean follow-up of 20.6 (range 15-27) months.

CONCLUSION

BAC diagnosed on 2 yearly screening mammography predicts CAD-RADs ≥3 disease in symptomatic patients.

Trends in radiology and experimental research

European Radiology Experimental, the new journal launched by the European Society of Radiology, is placed in the context of three general and seven radiology-specific trends. After describing the impact of population aging, personalized/precision medicine, and information technology development, the article considers the following trends: the tension between subspecialties and the unity of the discipline; attention to patient safety; the challenge of reproducibility for quantitative imaging; standardized and structured reporting; search for higher levels of evidence in radiology (from diagnostic performance to patient outcome); the increasing relevance of interventional radiology; and continuous technological evolution. The new journal will publish not only studies on phantoms, cells, or animal models but also those describing development steps of imaging biomarkers or those exploring secondary end-points of large clinical trials. Moreover, consideration will be given to studies regarding: computer modelling and computer aided detection and diagnosis; contrast materials, tracers, and theranostics; advanced image analysis; optical, molecular, hybrid and fusion imaging; radiomics and radiogenomics; three-dimensional printing, information technology, image reconstruction and post-processing, big data analysis, teleradiology, clinical decision support systems; radiobiology; radioprotection; and physics in radiology. The journal aims to establish a forum for basic science, computer and information technology, radiology, and other medical subspecialties.

Another Time, Another Space: The Evolution of the Virtual Journal Club

Virtual journal clubs (VJCs) provide a standardized, easily accessible forum for evidence-based discussion. The new virtual reality setting in which journal clubs and other online education events now take place offers great advantages and new opportunities for radiologists in academic medicine and private practice. VJCs continue to evolve, largely due to many emerging technologies and platforms. VJCs will continue to play an increasingly important role in medical education, interdisciplinary interaction, and multi-institutional collaboration. In this article, we discuss how to conduct and lead a critical review of medical literature in the setting of a virtual or traditional journal club. We discuss the current applications of VJCs in medical and graduate medical education and continued lifelong learning. We also explain the advantages and disadvantages of VJCs over traditional venues. Finally, the reader will be given the tools to successfully implement and run a VJC.

Should the automatic exposure control system of CT be disabled when scanning patients with endoaortic stents or mechanical heart valves? A phantom study

OBJECTIVES

To estimate the impact of endoaortic stents/mechanical heart valves on the output of an automatic exposure control (AEC) system and CT radiation dose.

METHODS

In this phantom study, seven stents and two valves were scanned with varying tube voltage (80/100/120 kVp), AEC activation (enabled/disabled) and prosthesis (present/absent), for a total of 540 scans. For each prosthesis, the dose-length product (DLP) was compared between scans with the AEC enabled and disabled. Percentage confidence levels for differences due to the prosthesis were calculated.

RESULTS

Differences between results with the AEC enabled and disabled were not statistically significant (p ≥ 0.059). In the comparison with and without the prosthesis, DLP was unchanged at 80 kVp and 100 kVp, while a slight increase was observed at 120 kVp. The radiation dose varied from 1.8 mGy to 2.4 mGy without the prosthesis and from 1.8 mGy to 2.5 mGy with the prosthesis (confidence level 37-100%).

CONCLUSIONS

The effect of the prosthesis on the AEC system was negligible and not clinically relevant. Therefore, disabling the AEC system when scanning these patients is not likely to provide a benefit.

KEY POINTS

• CT-AEC system is not impaired in patients with endoaortic prostheses/heart valves. • Negligible differences may be observed only at 120 kVp. • Disabling the AEC system in these patients is not recommended.

Designing Radiology Outcomes Studies-Essential Principles

Health outcomes research is essential to align radiology with current standards of high-value patient care, through the assessment of end results of diagnostic tests, interventions, or policy on patient health. To bridge studies of diagnostic test accuracy and health outcomes research, key considerations include: (1) how to determine when a diagnostic test merits evaluation of impact on outcomes, (2) when study of intermediate/surrogate outcomes can be useful, (3) how to consider the possible harms as well as potential benefits of a test, and (4) how to integrate evidence of an imaging test's efficacy/effectiveness with clinical data to assess outcomes. Due to challenges in conducting studies of long-term outcomes consequent to imaging use, intermediate health outcomes may capture a test's impact on successful diagnosis and therapy, and can provide readily measurable, incremental insights into the role of imaging in health-care delivery and efficiency. In an era marked by recognition of quality and value of care, outcomes research will provide essential evidence to inform radiologists' guidance of imaging use toward improved patient care, creation of clinical guidelines, and policy decisions.

Advanced neuroimaging in the clinic: critical appraisal of the evidence base

The shortage of high-quality systematic reviews in the field of radiology limits evidence-based integration of imaging methods into clinical practice and may perpetuate misconceptions regarding the efficacy and appropriateness of imaging techniques for specific applications. Diffusion tensor imaging for patients with mild traumatic brain injury (DTI-mTBI) and dynamic susceptibility contrast MRI for patients with glioma (DSC-glioma) are applications of quantitative neuroimaging, which similarly detect manifestations of disease where conventional neuroimaging techniques cannot. We performed a critical appraisal of reviews, based on the current evidence-based medicine methodology, addressing the ability of DTI-mTBI and DSC-glioma to (a) detect brain abnormalities and/or (b) predict clinical outcomes. 23 reviews of DTI-mTBI and 26 reviews of DSC-glioma met criteria for inclusion. All reviews addressed detection of brain abnormalities, whereas 12 DTI-mTBI reviews and 22 DSC-glioma reviews addressed prediction of a clinical outcome. All reviews were assessed using a critical appraisal worksheet consisting of 19 yes/no questions. Reviews were graded according to the total number of positive responses and the 2011 Oxford Centre for evidence-based medicine levels of evidence criteria. Reviews addressing DTI-mTBI detection had moderate quality, while those addressing DSC-glioma were of low quality. Reviews addressing prediction of outcomes for both applications were of low quality. Five DTI-mTBI reviews, but only one review of DSC-glioma met criteria for classification as a meta-analysis/systematic/quantitative review.

Gouty arthritis: the diagnostic and therapeutic impact of dual-energy CT

OBJECTIVES

To determine the diagnostic and therapeutic impact of dual-energy computed tomography (DECT) in gout.

METHODS

Forty-three patients with (n = 20) and without a history of gout (n = 23) showing non-specific soft tissue deposits underwent DECT after unrewarding arthrocentesis. Two blinded, independent readers evaluated DECT for the presence of urate crystals. Clinical diagnosis, clinically suspected urate crystal locations, diagnostic thinking and therapeutic decisions were noted before and after DECT. Clinical 1-month follow-up was obtained.

RESULTS

DECT showed urate in 26/43 patients (60 %). After DECT, clinical diagnosis of gout was withdrawn in 17/43 (40 %) and was maintained in 16/43 patients (37 %). In 10/43 patients (23 %) the diagnosis was maintained, but DECT revealed urate in clinically unsuspected locations. In 23/43 patients (53 %), a treatment-change based on DECT occurred. Changes in diagnostic thinking occurred more frequently in patients without a history of gout (p < 0.001), changes in therapeutic decisions more frequently in patients with a history of gout (p = 0.014). Clinical follow-up indicated beneficial effects of DECT-based diagnoses in 83 % of patients.

CONCLUSIONS

In patients with or without a history of gout and a recent suspicion for gouty arthritis with an unrewarding arthrocentesis, DECT has a marked diagnostic and therapeutic impact when hyperdense soft-tissue deposits are present.

KEY POINTS

• This study evaluates the concept of evidence-based radiology • In patients with suspected gout, DECT can help clinicians make the diagnosis • DECT has a marked impact on therapy • Clinical follow-up after 1 month indicated reliable results of DECT.

Imaging Guidelines for Enhancing Justifications for Radiologic Studies

Justification in the field of radiology refers to the appropriate use of radiologic imaging modalities, and may be achieved by establishing clinical imaging guidelines (CIGs). Recently, CIGs have been shown to be useful in selecting the proper medical imaging modality, resulting in the reduction of inappropriate radiologic examinations, thereby enhancing justifications. However, the development of CIGs is both time-consuming and difficult as the methodology of evidence-based medicine should be adhered to. Thus, although the radiologic societies in developed countries such as the United Kingdom and USA are already developing and implementing CIGs in their clinical practices, CIGs are not yet readily available in many other countries owing to differences in medical circumstances and resources. In this review, we assess the role and limitations of CIGs by examining the current status of CIGs in developed countries, and also describe the specific efforts made to establish CIGs in Korea.

Research education in Europe: an opinion paper by the European Society of Radiology

Research is a major positive driver for radiology. Therefore, research education needs to be a major topic for the radiology leadership, including the research committee of the ESR. Professional (radiological and non-radiological) and scientific publications as well as Research Committee questionnaires provide the basis for this opinion paper. Although radiology is well-positioned to deal with current and future challenges, there are still some gaps, such as the presence of radiology in basic research, radiology-specific research versus research services for other disciplines, need of adaptation to new research topics, general attitude towards research, issues of career planning, lack of incentives for researchers, gender issues with loss of women from the researcher pipeline, limited financing of research education and variability between countries and institutions. There is no easy answer to such challenges. However, all stakeholders, from the ESR to subspecialty societies, university departments, general radiology departments and the individual radiologist must recognise and promote research within their competencies. Many means and structures are already available but need to be used more extensively and systematically. Additional means need to be developed, scientific and professional trends must be actively followed, and minimal standards in research education should be maintained throughout Europe.

Main Messages

• Radiology research includes a broad spectrum, from basic to health services research.

• Research education needs to be widely available and systematically promoted.

• Existing means such as the European Institute for Biomedical Imaging Research (EIBIR) need to be advanced.

• New developments in research topics and professional life must be continuously monitored and evaluated.

[Critical reading of articles about diagnostic tests (part I): Are the results of the study valid?]

In the era of evidence-based medicine, one of the most important skills a radiologist should have is the ability to analyze the diagnostic literature critically. This tutorial aims to present guidelines for determining whether primary diagnostic articles are valid for clinical practice. The following elements should be evaluated: whether the study can be applied to clinical practice, whether the technique was compared to the reference test, whether an appropriate spectrum of patients was included, whether expectation bias and verification bias were limited, the reproducibility of the study, the practical consequences of the study, the confidence intervals for the parameters analyzed, the normal range for continuous variables, and the placement of the test in the context of other diagnostic tests. We use elementary practical examples to illustrate how to select and interpret the literature on diagnostic imaging and specific references to provide more details.

Biostatistics primer for the radiologist

OBJECTIVE

The purpose of this article is to review the most common data analysis methods encountered in radiology-based studies. Initially, description of variable types and their corresponding summary measures are provided; subsequent discussion focuses on comparison of these summary measures between groups, with a particular emphasis on regression analysis.

CONCLUSION

Knowledge of statistical applications is critical for radiologists to accurately evaluate the current literature and to conduct scientifically rigorous studies. Misapplication of statistical methods can lead to inappropriate conclusions and clinical recommendations.

Osteoid osteoma treated by percutaneous thermal ablation: when do we fail? A systematic review and guidelines for future reporting

PURPOSE

Osteoid osteoma (OO) is a painful benign bone tumor of the young that is widely treated by percutaneous thermal ablation (PTA) with success rates close to 100%. Nevertheless, some patients have recurrences. We reviewed the literature to understand whether these are true recurrences or incomplete treatments; to analyze safety and efficacy during long-term follow-up in a extremely large cohort of patients; to detail best-practice suggestions from the largest clinical trials as well as report their complications; and to recommend standards for future reporting.

MATERIALS AND METHODS

This study followed the Cochrane's guidelines for Systematic Reviews of Interventions. Inclusion criteria were as follows: (1) prospective or retrospective cohort study for PTA of OO under computed tomography (CT) guidance; (2) CT or magnetic resonance diagnosis; (3) radiofrequency ablation or interstitial laser ablation technique; (4) English language; (5) population <10 patients; (6) follow-up ≥12 months; and (7) original research. Risk of bias was assessed with a modified Newcastle-Ottawa Scale.

RESULTS

Two hundred fourteen articles were initially found. After applying the criteria mentioned previously, 27 PTA articles concerning 1,772 patients were chosen for inclusion. No exclusions were made due to risk of bias.

CONCLUSION

The investigators proved the long-term efficacy and superiority of PTA for OO compared with other techniques. In 5% of patients, however, the technique failed, and the researchers did not offer detailed exhaustive explanations. Future clinical trials for OO ablation should consider reporting essential procedure details and follow-up findings to allow for a meta-analysis. We provide both recommended standards for future reporting and suggestions for the prevention of recurrence.

Declining impact factor of radiologic journals: a matter for debate

OBJECTIVE

As reported in the 2011 Journal Citation Reports, despite an increase of the impact factor (IF) for 54% of all journals, only 39% of imaging journals experienced an IF increase. Of the 41 imaging journals with an IF of higher than 2.0, only 17 obtained a higher IF value in 2011 than in 2010 and the IF of most key international imaging journals decreased. How to manage and contrast this decline in IF is still a challenge for the radiologic community all over the world.

CONCLUSION

Radiologic journals may try to increase their IF by soliciting important articles, such as meta-analyses, cost-effective analyses, and guidelines, that are frequently published in high-IF clinical journals, and encouraging nonradiologists to submit relevant radiologic-related clinical articles.

Validity and reliability of radiological methods to assess proximal hip geometry in children with cerebral palsy: a systematic review

AIM

The aim of this systematic review was to assess the current validity and reliability of radiological methods used to measure proximal hip geometry in children with cerebral palsy.

METHOD

A search was conducted using relevant keywords and inclusion/exclusion criteria of the MEDLINE, CINALH Plus, Embase, Web of Science, Academic Search Premier, The Cochrane Library, and PsychINFO databases.

RESULTS

The migration percentage using X-rays showed excellent reliability and concurrent validity with three-dimensional (3D) measurements from computed tomography (CT) scans. The acetabular index, measured using X-rays had good reliability but moderate concurrent validity with 3D CT measurements; 3D CT scan indexes had greater reliability. The measurement of the neck shaft angle using X-rays showed excellent concurrent validity with measurements from 3D CT scans and excellent reliability. Regarding femoral anteversion, one study found an excellent correlation between two-dimensional CT and clinical assessment and excellent reliability. Two others showed less evidence for the use of CT ultrasounds.

INTERPRETATION

Most of the X-ray-based measurements showed good to excellent metrological properties. More metrological evidence is needed for the assessment of femoral anteversion. Magnetic resonance imaging and ultrasound-based measurements have great potential although very little metrological evidence is available.