Workload of diagnostic radiologists in the foreseeable future based on recent scientific advances: growth expectations and role of artificial intelligence

Nuclear medicine technologists practice impacted by AI denoising applications in PET/CT images

PURPOSE

Artificial intelligence (AI) in positron emission tomography/computed tomography (PET/CT) can be used to improve image quality when it is useful to reduce the injected activity or the acquisition time. Particular attention must be paid to ensure that users adopt this technological innovation when outcomes can be improved by its use. The aim of this study was to identify the aspects that need to be analysed and discussed to implement an AI denoising PET/CT algorithm in clinical practice, based on the representations of Nuclear Medicine Technologists (NMT) from Western-Switzerland, highlighting the barriers and facilitators associated.

METHODS

Two focus groups were organised in June and September 2023, involving ten voluntary participants recruited from all types of medical imaging departments, forming a diverse sample of NMT. The interview guide followed the first stage of the revised model of Ottawa of Research Use. A content analysis was performed following the three-stage approach described by Wanlin. Ethics cleared the study.

RESULTS

Clinical practice, workload, knowledge and resources were de 4 themes identified as necessary to be thought before implementing an AI denoising PET/CT algorithm by ten NMT participants (aged 31-60), not familiar with this AI tool. The main barriers to implement this algorithm included workflow challenges, resistance from professionals and lack of education; while the main facilitators were explanations and the availability of support to ask questions such as a "local champion".

CONCLUSION

To implement a denoising algorithm in PET/CT, several aspects of clinical practice need to be thought to reduce the barriers to its implementation such as the procedures, the workload and the available resources. Participants emphasised also the importance of clear explanations, education, and support for successful implementation.

IMPLICATIONS FOR PRACTICE

To facilitate the implementation of AI tools in clinical practice, it is important to identify the barriers and propose strategies that can mitigate it.

Substantial variability in what is considered important in the radiological report for anterior shoulder instability: a Delphi study with Dutch musculoskeletal radiologists and orthopedic surgeons

Background

Standardized consensus-based radiological reports for shoulder instability may improve clinical quality, reduce heterogeneity, and reduce workload. Therefore, the aim of this study was to determine important elements for the x-ray, magnetic resonance imaging (MRI) arthrography (MRA), and computed tomography (CT) report, the extent of variability, and important MRI views and settings.

Methods

An expert panel of musculoskeletal radiologists and orthopedic surgeons was recruited in a three-round Delphi design. Important elements were identified for the x-ray, MRA, and CT report and important MRI views and setting. These were rated on a 0-9 Likert scale. High variability was defined as at least one score between 1-3 and 7-9. Consensus was reached when ≥80% scored an element 1-3 or 7-9.

Results

The expert panel consisted of 21 musculoskeletal radiologists and 15 orthopedic surgeons. The number of elements identified in the first round was seventeen for the x-ray report, 52 for MRA, 21 for CT, and 23 for the MRI protocol. The number of elements that reached consensus was five for x-ray, twenty for MRA, nine for CT, and two for the MRI protocol. High variability was observed in 76.5% (n = 13) x-ray elements, 85.0% (n = 45) MRA, 76.2% (n = 16) CT, and 85.7% (n = 18) MRI protocol.

Conclusion

Substantial variability was observed in the scoring of important elements in the radiological for the evaluation of anterior shoulder instability, regardless of modality. Consensus was reached for five elements in the x-ray report, twenty in the MRA report, and nine in the CT report. Finally, consensus was reached on two elements regarding MRA views and settings.

Artificial intelligence for MRI stroke detection: a systematic review and meta-analysis

OBJECTIVES

This systematic review and meta-analysis aimed to assess the stroke detection performance of artificial intelligence (AI) in magnetic resonance imaging (MRI), and additionally to identify reporting insufficiencies.

METHODS

PRISMA guidelines were followed. MEDLINE, Embase, Cochrane Central, and IEEE Xplore were searched for studies utilising MRI and AI for stroke detection. The protocol was prospectively registered with PROSPERO (CRD42021289748). Sensitivity, specificity, accuracy, and area under the receiver operating characteristic (ROC) curve were the primary outcomes. Only studies using MRI in adults were included. The intervention was AI for stroke detection with ischaemic and haemorrhagic stroke in separate categories. Any manual labelling was used as a comparator. A modified QUADAS-2 tool was used for bias assessment. The minimum information about clinical artificial intelligence modelling (MI-CLAIM) checklist was used to assess reporting insufficiencies. Meta-analyses were performed for sensitivity, specificity, and hierarchical summary ROC (HSROC) on low risk of bias studies.

RESULTS

Thirty-three studies were eligible for inclusion. Fifteen studies had a low risk of bias. Low-risk studies were better for reporting MI-CLAIM items. Only one study examined a CE-approved AI algorithm. Forest plots revealed detection sensitivity and specificity of 93% and 93% with identical performance in the HSROC analysis and positive and negative likelihood ratios of 12.6 and 0.079.

CONCLUSION

Current AI technology can detect ischaemic stroke in MRI. There is a need for further validation of haemorrhagic detection. The clinical usability of AI stroke detection in MRI is yet to be investigated.

CRITICAL RELEVANCE STATEMENT

This first meta-analysis concludes that AI, utilising diffusion-weighted MRI sequences, can accurately aid the detection of ischaemic brain lesions and its clinical utility is ready to be uncovered in clinical trials.

KEY POINTS

There is a growing interest in AI solutions for detection aid. The performance is unknown for MRI stroke assessment. AI detection sensitivity and specificity were 93% and 93% for ischaemic lesions. There is limited evidence for the detection of patients with haemorrhagic lesions. AI can accurately detect patients with ischaemic stroke in MRI.

Contrast Reactions and Approaches to Staffing the Contrast Reaction Management Team

RATIONALE AND OBJECTIVES

Managing contrast reactions is critical as contrast reactions can be life-threatening and unpredictable. Institutions need an effective system to handle these events. Currently, there is no standard practice for assigning trainees, radiologists, non-radiologist physicians, or other non-physician providers for management of contrast reaction.

MATERIALS AND METHODS

The Association of Academic Radiologists (AAR) created a task force to address this gap. The AAR task force reviewed existing practices, studied available literature, and consulted experts related to contrast reaction management. The Society of Chairs of Academic Radiology Departments (SCARD) members were surveyed using a questionnaire focused on staffing strategies for contrast reaction management.

RESULTS

The task force found disparities in contrast reactions management across institutions and healthcare providers. There is a lack of standardized protocols for assigning personnel for contrast reaction management.

CONCLUSION

The AAR task force suggests developing standardized protocols for contrast reaction management. The protocols should outline clear roles for different healthcare providers involved in these events.

Modern imaging of acute pulmonary embolism

The first-choice imaging test for visualization of thromboemboli in the pulmonary vasculature in patients with suspected acute pulmonary embolism (PE) is multidetector computed tomography pulmonary angiography (CTPA) - a readily available and widely used imaging technique. Through technological advancements over the past years, alternative imaging techniques for the diagnosis of PE have become available, whilst others are still under investigation. In particular, the evolution of artificial intelligence (AI) is expected to enable further innovation in diagnostic management of PE. In this narrative review, current CTPA techniques and the emerging technology photon-counting CT (PCCT), as well as other modern imaging techniques of acute PE are discussed, including CTPA with iodine maps based on subtraction or dual-energy acquisition, single-photon emission CT (SPECT), magnetic resonance angiography (MRA), and magnetic resonance direct thrombus imaging (MRDTI). Furthermore, potential applications of AI are discussed.

Multiparametric MRI for characterization of the tumour microenvironment

Our understanding of tumour biology has evolved over the past decades and cancer is now viewed as a complex ecosystem with interactions between various cellular and non-cellular components within the tumour microenvironment (TME) at multiple scales. However, morphological imaging remains the mainstay of tumour staging and assessment of response to therapy, and the characterization of the TME with non-invasive imaging has not yet entered routine clinical practice. By combining multiple MRI sequences, each providing different but complementary information about the TME, multiparametric MRI (mpMRI) enables non-invasive assessment of molecular and cellular features within the TME, including their spatial and temporal heterogeneity. With an increasing number of advanced MRI techniques bridging the gap between preclinical and clinical applications, mpMRI could ultimately guide the selection of treatment approaches, precisely tailored to each individual patient, tumour and therapeutic modality. In this Review, we describe the evolving role of mpMRI in the non-invasive characterization of the TME, outline its applications for cancer detection, staging and assessment of response to therapy, and discuss considerations and challenges for its use in future medical applications, including personalized integrated diagnostics.

Large Language Models for Automated Synoptic Reports and Resectability Categorization in Pancreatic Cancer

Background Structured radiology reports for pancreatic ductal adenocarcinoma (PDAC) improve surgical decision-making over free-text reports, but radiologist adoption is variable. Resectability criteria are applied inconsistently. Purpose To evaluate the performance of large language models (LLMs) in automatically creating PDAC synoptic reports from original reports and to explore performance in categorizing tumor resectability. Materials and Methods In this institutional review board-approved retrospective study, 180 consecutive PDAC staging CT reports on patients referred to the authors' European Society for Medical Oncology-designated cancer center from January to December 2018 were included. Reports were reviewed by two radiologists to establish the reference standard for 14 key findings and National Comprehensive Cancer Network (NCCN) resectability category. GPT-3.5 and GPT-4 (accessed September 18-29, 2023) were prompted to create synoptic reports from original reports with the same 14 features, and their performance was evaluated (recall, precision, F1 score). To categorize resectability, three prompting strategies (default knowledge, in-context knowledge, chain-of-thought) were used for both LLMs. Hepatopancreaticobiliary surgeons reviewed original and artificial intelligence (AI)-generated reports to determine resectability, with accuracy and review time compared. The McNemar test, t test, Wilcoxon signed-rank test, and mixed effects logistic regression models were used where appropriate. Results GPT-4 outperformed GPT-3.5 in the creation of synoptic reports (F1 score: 0.997 vs 0.967, respectively). Compared with GPT-3.5, GPT-4 achieved equal or higher F1 scores for all 14 extracted features. GPT-4 had higher precision than GPT-3.5 for extracting superior mesenteric artery involvement (100% vs 88.8%, respectively). For categorizing resectability, GPT-4 outperformed GPT-3.5 for each prompting strategy. For GPT-4, chain-of-thought prompting was most accurate, outperforming in-context knowledge prompting (92% vs 83%, respectively; P = .002), which outperformed the default knowledge strategy (83% vs 67%, P < .001). Surgeons were more accurate in categorizing resectability using AI-generated reports than original reports (83% vs 76%, respectively; P = .03), while spending less time on each report (58%; 95% CI: 0.53, 0.62). Conclusion GPT-4 created near-perfect PDAC synoptic reports from original reports. GPT-4 with chain-of-thought achieved high accuracy in categorizing resectability. Surgeons were more accurate and efficient using AI-generated reports. © RSNA, 2024 Supplemental material is available for this article. See also the editorial by Chang in this issue.

Performance of AI to exclude normal chest radiographs to reduce radiologists' workload

INTRODUCTION

This study investigates the performance of a commercially available artificial intelligence (AI) system to identify normal chest radiographs and its potential to reduce radiologist workload.

METHODS

Retrospective analysis included consecutive chest radiographs from two medical centers between Oct 1, 2016 and Oct 14, 2016. Exclusions comprised follow-up exams within the inclusion period, bedside radiographs, incomplete images, imported radiographs, and pediatric radiographs. Three chest radiologists categorized findings into normal, clinically irrelevant, clinically relevant, urgent, and critical. A commercial AI system processed all radiographs, scoring 10 chest abnormalities on a 0-100 confidence scale. AI system performance was evaluated using the area under the ROC curve (AUC), assessing the detection of normal radiographs. Sensitivity was calculated for the default and a conservative operating point. the detection of negative predictive value (NPV) for urgent and critical findings, as well as the potential workload reduction, was calculated.

RESULTS

A total of 2603 radiographs were acquired in 2141 unique patients. Post-exclusion, 1670 radiographs were analyzed. Categories included 479 normal, 332 clinically irrelevant, 339 clinically relevant, 501 urgent, and 19 critical findings. The AI system achieved an AUC of 0.92. Sensitivity for normal radiographs was 92% at default and 53% at the conservative operating point. At the conservative operating point, NPV was 98% for urgent and critical findings, and could result in a 15% workload reduction.

CONCLUSION

A commercially available AI system effectively identifies normal chest radiographs and holds the potential to lessen radiologists' workload by omitting half of the normal exams from reporting.

CLINICAL RELEVANCE STATEMENT

The AI system is able to detect half of all normal chest radiographs at a clinically acceptable operating point, thereby potentially reducing the workload for the radiologists by 15%.

KEY POINTS

The AI system reached an AUC of 0.92 for the detection of normal chest radiographs. Fifty-three percent of normal chest radiographs were identified with a NPV of 98% for urgent findings. AI can reduce the workload of chest radiography reporting by 15%.

Multiple Kernel Synthesis of Head CT Using a Task-Based Loss Function

In CT imaging of the head, multiple image series are routinely reconstructed with different kernels and slice thicknesses. Reviewing the redundant information is an inefficient process for radiologists. We address this issue with a convolutional neural network (CNN)-based technique, synthesiZed Improved Resolution and Concurrent nOise reductioN (ZIRCON), that creates a single, thin, low-noise series that combines the favorable features from smooth and sharp head kernels. ZIRCON uses a CNN model with an autoencoder U-Net architecture that accepts two input channels (smooth- and sharp-kernel CT images) and combines their salient features to produce a single CT image. Image quality requirements are built into a task-based loss function with a smooth and sharp loss terms specific to anatomical regions. The model is trained using supervised learning with paired routine-dose clinical non-contrast head CT images as training targets and simulated low-dose (25%) images as training inputs. One hundred unique de-identified clinical exams were used for training, ten for validation, and ten for testing. Visual comparisons and contrast measurements of ZIRCON revealed that thinner slices and the smooth-kernel loss function improved gray-white matter contrast. Combined with lower noise, this increased visibility of small soft-tissue features that would be otherwise impaired by partial volume averaging or noise. Line profile analysis showed that ZIRCON images largely retained sharpness compared to the sharp-kernel input images. ZIRCON combined desirable image quality properties of both smooth and sharp input kernels into a single, thin, low-noise series suitable for both brain and skull imaging.

Imagine there is no paperwork… it's easy if you try

Artificial Intelligence (AI) applied to radiology is so vast that it provides applications ranging from becoming a complete replacement for radiologists (a potential threat) to an efficient paperwork-saving time assistant (an evident strength). Nowadays, there are AI applications developed to facilitate the diagnostic process of radiologists without directly influencing (or replacing) the proper diagnostic decision step. These tools may help to reduce administrative workload, in different scenarios ranging from assisting in scheduling, study prioritization, or report communication, to helping with patient follow-up, including recommending additional exams. These are just a few of the highly time-consuming tasks that radiologists have to deal with every day in their routine workflow. These tasks hinder the time that radiologists should spend evaluating images and caring for patients, which will have a direct and negative impact on the quality of reports and patient attention, increasing the delay and waiting list of studies pending to be performed and reported. These types of AI applications should help to partially face this worldwide shortage of radiologists.

What is the added value of specialist radiology review of multidisciplinary team meeting cases in a tertiary care center?

PURPOSE

Multidisciplinary team meetings (MDTMs) are an important component of the workload of radiologists. This study investigated how often subspecialized radiologists change patient management in MDTMs at a tertiary care institution.

MATERIALS AND METHODS

Over 2 years, six subspecialty radiologists documented their contributions to MDTMs at a tertiary care center. Both in-house and external imaging examinations were discussed at the MDTMs. All imaging examinations (whether primary or second opinion) were interpreted and reported by subspecialty radiologist prior to the MDTMs. The management change ratio (MCratio) of the radiologist was defined as the number of cases in which the radiologist's input in the MDTM changed patient management beyond the information that was already provided by the in-house (primary or second opinion) radiology report, as a proportion of the total number of cases whose imaging examinations were prepared for demonstration in the MDTM.

RESULTS

Sixty-eight MDTMs were included. The time required for preparing and attending all MDTMs (excluding imaging examinations that had not been reported yet) was 11,000 min, with a median of 172 min (IQR 113-200 min) per MDTM, and a median of 9 min (IQR 8-13 min) per patient. The radiologists' input changed patient management in 113 out of 1138 cases, corresponding to an MCratio of 8.4%. The median MCratio per MDTM was 6% (IQR 0-17%).

CONCLUSION

Radiologists' time investment in MDTMs is considerable relative to the small proportion of cases in which they influence patient management in the MDTM. The use of radiologists for MDTMs should therefore be improved.

CLINICAL RELEVANCE STATEMENT

The use of radiologists for MDTMs (multidisciplinary team meetings) should be improved, because their time investment in MDTMs is considerable relative to the small proportion of cases in which they influence patient management in the MDTM.

KEY POINTS

• Multidisciplinary team meetings (MDTMs) are an important component of the workload of radiologists. • In a tertiary care center in which all imaging examinations have already been interpreted and reported by subspecialized radiologists before the MDTM takes place, the median time investment of a radiologist for preparing and demonstrating one MDTM patient is 9 min. • In this setting, the radiologist changes patient management in only a minority of cases in the MDTM.

Artificial intelligence-assisted double reading of chest radiographs to detect clinically relevant missed findings: a two-centre evaluation

OBJECTIVES

To evaluate an artificial intelligence (AI)-assisted double reading system for detecting clinically relevant missed findings on routinely reported chest radiographs.

METHODS

A retrospective study was performed in two institutions, a secondary care hospital and tertiary referral oncology centre. Commercially available AI software performed a comparative analysis of chest radiographs and radiologists' authorised reports using a deep learning and natural language processing algorithm, respectively. The AI-detected discrepant findings between images and reports were assessed for clinical relevance by an external radiologist, as part of the commercial service provided by the AI vendor. The selected missed findings were subsequently returned to the institution's radiologist for final review.

RESULTS

In total, 25,104 chest radiographs of 21,039 patients (mean age 61.1 years ± 16.2 [SD]; 10,436 men) were included. The AI software detected discrepancies between imaging and reports in 21.1% (5289 of 25,104). After review by the external radiologist, 0.9% (47 of 5289) of cases were deemed to contain clinically relevant missed findings. The institution's radiologists confirmed 35 of 47 missed findings (74.5%) as clinically relevant (0.1% of all cases). Missed findings consisted of lung nodules (71.4%, 25 of 35), pneumothoraces (17.1%, 6 of 35) and consolidations (11.4%, 4 of 35).

CONCLUSION

The AI-assisted double reading system was able to identify missed findings on chest radiographs after report authorisation. The approach required an external radiologist to review the AI-detected discrepancies. The number of clinically relevant missed findings by radiologists was very low.

CLINICAL RELEVANCE STATEMENT

The AI-assisted double reader workflow was shown to detect diagnostic errors and could be applied as a quality assurance tool. Although clinically relevant missed findings were rare, there is potential impact given the common use of chest radiography.

KEY POINTS

• A commercially available double reading system supported by artificial intelligence was evaluated to detect reporting errors in chest radiographs (n=25,104) from two institutions. • Clinically relevant missed findings were found in 0.1% of chest radiographs and consisted of unreported lung nodules, pneumothoraces and consolidations. • Applying AI software as a secondary reader after report authorisation can assist in reducing diagnostic errors without interrupting the radiologist's reading workflow. However, the number of AI-detected discrepancies was considerable and required review by a radiologist to assess their relevance.

Is work overload associated with diagnostic errors on 18F-FDG-PET/CT?

PURPOSE

To determine the association between workload and diagnostic errors on 18F-FDG-PET/CT.

MATERIALS AND METHODS

This study included 103 18F-FDG-PET/CT scans with a diagnostic error that was corrected with an addendum between March 2018 and July 2023. All scans were performed at a tertiary care center. The workload of each nuclear medicine physician or radiologist who authorized the 18F-FDG-PET/CT report was determined on the day the diagnostic error was made and normalized for his or her own average daily production (workloadnormalized). A workloadnormalized of more than 100% indicates that the nuclear medicine physician or radiologist had a relative work overload, while a value of less than 100% indicates a relative work underload on the day the diagnostic error was made. The time of the day the diagnostic error was made was also recorded. Workloadnormalized was compared to 100% using a signed rank sum test, with the hypothesis that it would significantly exceed 100%. A Mann-Kendall test was performed to test the hypothesis that diagnostic errors would increase over the course of the day.

RESULTS

Workloadnormalized (median of 121%, interquartile range: 71 to 146%) on the days the diagnostic errors were made was significantly higher than 100% (P = 0.014). There was no significant upward trend in the frequency of diagnostic errors over the course of the day (Mann-Kendall tau = 0.05, P = 0.7294).

CONCLUSION

Work overload seems to be associated with diagnostic errors on 18F-FDG-PET/CT. Diagnostic errors were encountered throughout the entire working day, without any upward trend towards the end of the day.

Rogue AI: Cautionary Cases in Neuroradiology and What We Can Learn From Them

Introduction In recent years, artificial intelligence (AI) in medical imaging has undergone unprecedented innovation and advancement, sparking a revolutionary transformation in healthcare. The field of radiology is particularly implicated, as clinical radiologists are expected to interpret an ever-increasing number of complex cases in record time. Machine learning software purchased by our institution is expected to help our radiologists come to a more prompt diagnosis by delivering point-of-care quantitative analysis of suspicious findings and streamlining clinical workflow. This paper explores AI's impact on neuroradiology, an area accounting for a substantial portion of recent radiology studies. We present a case series evaluating an AI software's performance in detecting neurovascular findings, highlighting five cases where AI interpretations differed from radiologists' assessments. Our study underscores common pitfalls of AI in the context of CT head angiograms, aiming to guide future AI algorithms.  Methods We conducted a retrospective case series study at Stony Brook University Hospital, a large medical center in Stony Brook, New York, spanning from October 1, 2021 to December 31, 2021, analyzing 140 randomly sampled CT angiograms using AI software. This software assessed various neurovascular parameters, and AI findings were compared with neuroradiologists' interpretations. Five cases with divergent interpretations were selected for detailed analysis. Results Five representative cases in which AI findings were discordant with radiologists' interpretations are presented with diagnoses including diffuse anoxic ischemic injury, cortical laminar necrosis, colloid cyst, right superficial temporal artery-to-middle cerebral artery (STA-MCA) bypass, and subacute bilateral subdural hematomas. Discussion The errors identified in our case series expose AI's limitations in radiology. Our case series reveals that AI's incorrect interpretations can stem from complexities in pathology, challenges in distinguishing densities, inability to identify artifacts, identifying post-surgical changes in normal anatomy, sensitivity limitations, and insufficient pattern recognition. AI's potential for improvement lies in refining its algorithms to effectively recognize and differentiate pathologies. Incorporating more diverse training datasets, multimodal data, deep-reinforcement learning, clinical context, and real-time learning capabilities are some ways to improve AI's performance in the field of radiology. Conclusion Overall, it is apparent that AI applications in radiology have much room for improvement before becoming more widely integrated into clinical workflows. While AI demonstrates remarkable potential to aid in diagnosis and streamline workflows, our case series highlights common pitfalls that underscore the need for continuous improvement. By refining algorithms, incorporating diverse datasets, embracing multimodal information, and leveraging innovative machine learning strategies, AI's diagnostic accuracy can be significantly improved.

Using artificial intelligence to promote equitable care for inpatients with language barriers and complex medical needs: clinical stakeholder perspectives

OBJECTIVES

Inpatients with language barriers and complex medical needs suffer disparities in quality of care, safety, and health outcomes. Although in-person interpreters are particularly beneficial for these patients, they are underused. We plan to use machine learning predictive analytics to reliably identify patients with language barriers and complex medical needs to prioritize them for in-person interpreters.

MATERIALS AND METHODS

This qualitative study used stakeholder engagement through semi-structured interviews to understand the perceived risks and benefits of artificial intelligence (AI) in this domain. Stakeholders included clinicians, interpreters, and personnel involved in caring for these patients or for organizing interpreters. Data were coded and analyzed using NVIVO software.

RESULTS

We completed 49 interviews. Key perceived risks included concerns about transparency, accuracy, redundancy, privacy, perceived stigmatization among patients, alert fatigue, and supply-demand issues. Key perceived benefits included increased awareness of in-person interpreters, improved standard of care and prioritization for interpreter utilization; a streamlined process for accessing interpreters, empowered clinicians, and potential to overcome clinician bias.

DISCUSSION

This is the first study that elicits stakeholder perspectives on the use of AI with the goal of improved clinical care for patients with language barriers. Perceived benefits and risks related to the use of AI in this domain, overlapped with known hazards and values of AI but some benefits were unique for addressing challenges with providing interpreter services to patients with language barriers.

CONCLUSION

Artificial intelligence to identify and prioritize patients for interpreter services has the potential to improve standard of care and address healthcare disparities among patients with language barriers.

Value Creation Through Artificial Intelligence and Cardiovascular Imaging: A Scientific Statement From the American Heart Association

Multiple applications for machine learning and artificial intelligence (AI) in cardiovascular imaging are being proposed and developed. However, the processes involved in implementing AI in cardiovascular imaging are highly diverse, varying by imaging modality, patient subtype, features to be extracted and analyzed, and clinical application. This article establishes a framework that defines value from an organizational perspective, followed by value chain analysis to identify the activities in which AI might produce the greatest incremental value creation. The various perspectives that should be considered are highlighted, including clinicians, imagers, hospitals, patients, and payers. Integrating the perspectives of all health care stakeholders is critical for creating value and ensuring the successful deployment of AI tools in a real-world setting. Different AI tools are summarized, along with the unique aspects of AI applications to various cardiac imaging modalities, including cardiac computed tomography, magnetic resonance imaging, and positron emission tomography. AI is applicable and has the potential to add value to cardiovascular imaging at every step along the patient journey, from selecting the more appropriate test to optimizing image acquisition and analysis, interpreting the results for classification and diagnosis, and predicting the risk for major adverse cardiac events.

The radiology job market in the Netherlands: which subspecialties and other skills are in demand?

OBJECTIVES

To evaluate the current job market for medical specialists in radiology and nuclear medicine (NM) in the Netherlands.

METHODS

Vacancies posted for radiologists and nuclear medicine physicians in the Netherlands between December 2020 and February 2022 were collected and analyzed.

RESULTS

A total of 157 vacancies (146 for radiologist and 11 for nuclear medicine physicians) were included. The most sought-after subspecialties were all-round (22%), abdominal (19%), and interventional radiology (14%), and 30% of vacancies preferred applicants with additional non-clinical skills (research, teaching, management, information and communications technology (ICT)/artificial intelligence (AI)). Non-academic hospitals significantly more frequently requested all-round radiologists (n = 31) than academic hospitals (n = 1) (p = 0.001), while the distribution of other requested subspecialties was not significantly different between non-academic and academic vacancies. Non-academic hospitals also significantly more frequently requested additional research tasks in their vacancies (n = 35) compared to academic hospitals (n = 4) (p = 0.011). There were non-significant trends for non-academic hospitals more frequently requesting teaching tasks in their vacancies (n =18) than academic hospitals (n = 1) (p = 0.051), and for non-academic hospitals more frequently asking for management skills (n = 11) than academic hospitals (n = 0) (p = 0.075).

CONCLUSION

All-round, abdominal, and interventional radiologists are most in demand on the job market in the Netherlands. All-round radiologists are particularly sought after by non-academic hospitals, whereas nuclear radiologists who completed the Dutch integrated NM and radiology residency seem to be welcomed by hospitals searching for a nuclear medicine specialist. Finally, non-clinical skills (research, teaching, management, ICT/AI) are commonly requested. These data can be useful for residents and developers of training curricula.

CLINICAL RELEVANCE STATEMENT

An overview of the radiology job market and the requested skills is important for residents, for those who seek work as a radiologist, and for those who are involved in the design and revision of residency programs.

KEY POINTS

Review of job vacancies over an extended period of time provides valuable information to residents and feedback to potentially improve radiology and nuclear medicine (NM) residency programs. All-round radiologists are wanted in non-academic hospitals and nuclear radiologists (those who have completed an integrated NM-radiology curriculum) are welcomed by hospitals searching for nuclear medicine specialists in the Netherlands. There is a need to train residents in important non-clinical skills, such as research and teaching, but also management and communications technology/artificial intelligence.

How to prepare for a bright future of radiology in Europe

Because artificial intelligence (AI)-powered algorithms allow automated image analysis in a growing number of diagnostic scenarios, some healthcare stakeholders have raised doubts about the future of the entire radiologic profession. Their view disregards not only the role of radiologists in the diagnostic service chain beyond reporting, but also the many multidisciplinary and patient-related consulting tasks for which radiologists are solicited. The time commitment for these non-reporting tasks is considerable but difficult to quantify and often impossible to fulfil considering the current mismatch between workload and workforce in many countries. Nonetheless, multidisciplinary, and patient-centred consulting activities could move up on radiologists' agendas as soon as AI-based tools can save time in daily routine. Although there are many reasons why AI will assist and not replace radiologists as imaging experts in the future, it is important to position the next generation of European radiologists in view of this expected trend. To ensure radiologists' personal professional recognition and fulfilment in multidisciplinary environments, the focus of training should go beyond diagnostic reporting, concentrating on clinical backgrounds, specific communication skills with referrers and patients, and integration of imaging findings with those of other disciplines. Close collaboration between the European Society of Radiology (ESR) and European national radiologic societies can help to achieve these goals. Although each adequate treatment begins with a correct diagnosis, many health politicians see radiologic procedures mainly as a cost factor. Radiologic research should, therefore, increasingly investigate the imaging impact on treatment and outcome rather than focusing mainly on technical improvements and diagnostic accuracy alone.Critical relevance statement Strategies are presented to prepare for a successful future of the radiologic profession in Europe, if AI-powered tools can alleviate the current reporting overload: engaging in multidisciplinary activities (clinical and integrative diagnostics), enhancing the value and recognition of radiologists' role through clinical expertise, focusing radiological research on the impact on diagnosis and outcome, and promoting patient-centred radiology by enhancing communication skills.Key points • AI-powered tools will not replace radiologists but hold promise to reduce the current reporting burden, enabling them to reinvest liberated time in multidisciplinary clinical and patient-related tasks.• The skills and resources for these tasks should be considered when recruiting and teaching the next generation of radiologists, when organising departments and planning staffing.• Communication skills will play an increasing role in both multidisciplinary activities and patient-centred radiology.• The value and importance of a correct and integrative diagnosis and the cost of an incorrect imaging diagnosis should be emphasised when discussing with non-medical stakeholders in healthcare.• The radiologic community in Europe should start now to prepare for a bright future of the profession for the benefit of patients and medical colleagues alike.

Work overload and diagnostic errors in radiology

PURPOSE

To determine the association between workload and diagnostic errors on clinical CT scans.

METHOD

This retrospective study was performed at a tertiary care center and covered the period from January 2020 to March 2023. All clinical CT scans that contained an addendum describing a perceptual error (i.e. failure to detect an important abnormality) in the original report that was issued on office days between 7.30 a.m. and 18.00 p.m., were included. The workload of the involved radiologist on the day of the diagnostic error was calculated in terms of relative value units, and normalized for the known average daily production of each individual radiologist (workloadnormalized). A workloadnormalized of less than 100% indicates relative work underload, while a workloadnormalized of > 100% indicates relative work overload in terms of reported examinations on an individual radiologist's basis.

RESULTS

A total of 49 diagnostic errors were included. Top-five locations of diagnostic errors were lung (n = 8), bone (n = 8), lymph nodes (n = 5), peritoneum (n = 5), and liver (n = 4). Workloadnormalized on the days the diagnostic errors were made was on average 121% (95% confidence interval: 106% to 136%), which was significantly higher than 100% (P = 0.008). There was no significant upward monotonic trend in diagnostic errors over the course of the day (Mann-Kendall tau of 0.005, P = 1.000), and there were no other notable temporal trends either.

CONCLUSIONS

Radiologists appear to have a relative work overload when they make a diagnostic error on CT. Diagnostic errors occurred throughout the entire day, without any increase towards the end of the day.

Artificial intelligence for breast cancer detection in screening mammography in Sweden: a prospective, population-based, paired-reader, non-inferiority study

BACKGROUND

Artificial intelligence (AI) as an independent reader of screening mammograms has shown promise, but there are few prospective studies. Our aim was to conduct a prospective clinical trial to examine how AI affects cancer detection and false positive findings in a real-world setting.

METHODS

ScreenTrustCAD was a prospective, population-based, paired-reader, non-inferiority study done at the Capio Sankt Göran Hospital in Stockholm, Sweden. Consecutive women without breast implants aged 40-74 years participating in population-based screening in the geographical uptake area of the study hospital were included. The primary outcome was screen-detected breast cancer within 3 months of mammography, and the primary analysis was to assess non-inferiority (non-inferiority margin of 0·15 relative reduction in breast cancer diagnoses) of double reading by one radiologist plus AI compared with standard-of-care double reading by two radiologists. We also assessed single reading by AI alone and triple reading by two radiologists plus AI compared with standard-of-care double reading by two radiologists. This study is registered with ClinicalTrials.gov, NCT04778670.

FINDINGS

From April 1, 2021, to June 9, 2022, 58 344 women aged 40-74 years underwent regular mammography screening, of whom 55 581 were included in the study. 269 (0·5%) women were diagnosed with screen-detected breast cancer based on an initial positive read: double reading by one radiologist plus AI was non-inferior for cancer detection compared with double reading by two radiologists (261 [0·5%] vs 250 [0·4%] detected cases; relative proportion 1·04 [95% CI 1·00-1·09]). Single reading by AI (246 [0·4%] vs 250 [0·4%] detected cases; relative proportion 0·98 [0·93-1·04]) and triple reading by two radiologists plus AI (269 [0·5%] vs 250 [0·4%] detected cases; relative proportion 1·08 [1·04-1·11]) were also non-inferior to double reading by two radiologists.

INTERPRETATION

Replacing one radiologist with AI for independent reading of screening mammograms resulted in a 4% higher non-inferior cancer detection rate compared with radiologist double reading. Our study suggests that AI in the study setting has potential for controlled implementation, which would include risk management and real-world follow-up of performance.

FUNDING

Swedish Research Council, Swedish Cancer Society, Region Stockholm, and Lunit.

Australian sonographers' perceptions of patient safety in ultrasound imaging: Part two - translation into practice

Introduction

A lack of patient safety research hampers capacity to improve safety in healthcare.Ultrasound is often considered 'safe' as it does not use ionising radiation, a simplistic view of patient safety. Understanding sonographers' actions towards patient safety is crucial; however, self-reported measures cannot always predict behaviour. This study is part of a PhD exploring patient safety in medical diagnostic ultrasound. The aim of this paper is to explore sonographers' responses to the patient safety concerns identified in Part one of this study. The ultimate aim of the study is to inform the final phase of the doctoral study which will consider the next steps in improving the quality and safety of healthcare experienced by patients.

Methods

A qualitative study using semi-structured, one-on-one interviews. The Theory of Planned Behaviour (TPB) explained how sonographers respond to perceived patient safety risks in practice.

Results

Thirty-one sonographers were interviewed. Based on the seven themes identified in Part one of the study, results showed that incongruences exist between identifying patient safety risks and the actions taken in practice to manage these risks.

Conclusion

The TPB showed that behavioural, normative and control beliefs impact sonographers' responses to perceived patient safety risks in practice and can lead to risk avoidance. Lack of regulation in ultrasound creates a challenge in dealing with Fitness to Practice issues. Collective actions are required to support sonographers in taking appropriate actions to enhance patient safety from multiple stakeholders including accreditation bodies, regulatory authorities, educational institutions and employers.

Retrospective batch analysis to evaluate the diagnostic accuracy of a clinically deployed AI algorithm for the detection of acute pulmonary embolism on CTPA

PURPOSE

To generate and extend the evidence on the clinical validity of an artificial intelligence (AI) algorithm to detect acute pulmonary embolism (PE) on CT pulmonary angiography (CTPA) of patients suspected of PE and to evaluate the possibility of reducing the risk of missed findings in clinical practice with AI-assisted reporting.

METHODS

Consecutive CTPA scan data of 3316 patients referred because of suspected PE between 24-2-2018 and 31-12-2020 were retrospectively analysed by a CE-certified and FDA-approved AI algorithm. The output of the AI was compared with the attending radiologists' report. To define the reference standard, discordant findings were independently evaluated by two readers. In case of disagreement, an experienced cardiothoracic radiologist adjudicated.

RESULTS

According to the reference standard, PE was present in 717 patients (21.6%). PE was missed by the AI in 23 patients, while the attending radiologist missed 60 PE. The AI detected 2 false positives and the attending radiologist 9. The sensitivity for the detection of PE by the AI algorithm was significantly higher compared to the radiology report (96.8% vs. 91.6%, p < 0.001). Specificity of the AI was also significantly higher (99.9% vs. 99.7%, p = 0.035). NPV and PPV of the AI were also significantly higher than the radiology report.

CONCLUSION

The AI algorithm showed a significantly higher diagnostic accuracy for the detection of PE on CTPA compared to the report of the attending radiologist. This finding indicates that missed positive findings could be prevented with the implementation of AI-assisted reporting in daily clinical practice.

CRITICAL RELEVANCE STATEMENT

Missed positive findings on CTPA of patients suspected of pulmonary embolism can be prevented with the implementation of AI-assisted care.

KEY POINTS

The AI algorithm showed excellent diagnostic accuracy detecting PE on CTPA. Accuracy of the AI was significantly higher compared to the attending radiologist. Highest diagnostic accuracy can likely be achieved by radiologists supported by AI. Our results indicate that implementation of AI-assisted reporting could reduce the number of missed positive findings.

Australian sonographers' perceptions of patient safety in ultrasound imaging: Part 1 - identifying the main safety concerns, a qualitative study

Introduction

Patient safety has been an undervalued component of quality healthcare but is a challenging area of research.Ultrasound is the most common imaging modality in the world. Research on patient safety in ultrasound is generally focused on bioeffects and safe operation of ultrasound equipment. However, other safety issues exist in practice that warrant consideration.This paper forms the first part of a PhD study exploring patient safety in medical diagnostic ultrasound, beyond the notion of bioeffects.The ultimate aim of the study is to inform the final phase of the research study which will consider the next steps in improving the quality and safety of healthcare experienced by patients.

Methods

A qualitative study using semi-structured, one-on-one interviews. A thematic analysis categorised data into codes and generated final themes.

Results

A heterogeneous mix of 31 sonographers, who reflected the profile of the profession in Australia, were interviewed between September 2019 and January 2020. Seven themes emerged from the analysis. These were bioeffects, physical safety, workload, reporting, professionalism, intimate examinations and infection control.

Conclusion

This study presents a comprehensive analysis of sonographers' perceptions of patient safety in ultrasound imaging, not previously available in the literature. Consistent with the literature, patient safety in ultrasound tends to be viewed in technical terms through the potential for bioeffects of tissue damage or physical harm to the patient. However, other patient safety issues have emerged, and while not as well recognised, have the potential to negatively impact on patient safety.

Artificial Intelligence Tool for Detection and Worklist Prioritization Reduces Time to Diagnosis of Incidental Pulmonary Embolism at CT

Purpose

To evaluate the diagnostic efficacy of artificial intelligence (AI) software in detecting incidental pulmonary embolism (IPE) at CT and shorten the time to diagnosis with use of radiologist reading worklist prioritization.

Materials and Methods

In this study with historical controls and prospective evaluation, regulatory-cleared AI software was evaluated to prioritize IPE on routine chest CT scans with intravenous contrast agent in adult oncology patients. Diagnostic accuracy metrics were calculated, and temporal end points, including detection and notification times (DNTs), were assessed during three time periods (April 2019 to September 2020): routine workflow without AI, human triage without AI, and worklist prioritization with AI.

Results

In total, 11 736 CT scans in 6447 oncology patients (mean age, 63 years ± 12 [SD]; 3367 men) were included. Prevalence of IPE was 1.3% (51 of 3837 scans), 1.4% (54 of 3920 scans), and 1.0% (38 of 3979 scans) for the respective time periods. The AI software detected 131 true-positive, 12 false-negative, 31 false-positive, and 11 559 true-negative results, achieving 91.6% sensitivity, 99.7% specificity, 99.9% negative predictive value, and 80.9% positive predictive value. During prospective evaluation, AI-based worklist prioritization reduced the median DNT for IPE-positive examinations to 87 minutes (vs routine workflow of 7714 minutes and human triage of 4973 minutes). Radiologists' missed rate of IPE was significantly reduced from 44.8% (47 of 105 scans) without AI to 2.6% (one of 38 scans) when assisted by the AI tool (P < .001).

Conclusion

AI-assisted workflow prioritization of IPE on routine CT scans in oncology patients showed high diagnostic accuracy and significantly shortened the time to diagnosis in a setting with a backlog of examinations.Keywords: CT, Computer Applications, Detection, Diagnosis, Embolism, Thorax, ThrombosisSupplemental material is available for this article.© RSNA, 2023See also the commentary by Elicker in this issue.

Assessing the Impact of Image Resolution on Deep Learning for TB Lesion Segmentation on Frontal Chest X-rays

Deep learning (DL) models are state-of-the-art in segmenting anatomical and disease regions of interest (ROIs) in medical images. Particularly, a large number of DL-based techniques have been reported using chest X-rays (CXRs). However, these models are reportedly trained on reduced image resolutions for reasons related to the lack of computational resources. Literature is sparse in discussing the optimal image resolution to train these models for segmenting the tuberculosis (TB)-consistent lesions in CXRs. In this study, we investigated the performance variations with an Inception-V3 UNet model using various image resolutions with/without lung ROI cropping and aspect ratio adjustments and identified the optimal image resolution through extensive empirical evaluations to improve TB-consistent lesion segmentation performance. We used the Shenzhen CXR dataset for the study, which includes 326 normal patients and 336 TB patients. We proposed a combinatorial approach consisting of storing model snapshots, optimizing segmentation threshold and test-time augmentation (TTA), and averaging the snapshot predictions, to further improve performance with the optimal resolution. Our experimental results demonstrate that higher image resolutions are not always necessary; however, identifying the optimal image resolution is critical to achieving superior performance.

Developing a machine learning model to predict patient need for computed tomography imaging in the emergency department

Overcrowding is a well-known problem in hospitals and emergency departments (ED) that can negatively impact patients and staff. This study aims to present a machine learning model to detect a patient's need for a Computed Tomography (CT) exam in the emergency department at the earliest possible time. The data for this work was collected from ED at Thunder Bay Regional Health Sciences Centre over one year (05/2016-05/2017) and contained administrative triage information. The target outcome was whether or not a patient required a CT exam. Multiple combinations of text embedding methods, machine learning algorithms, and data resampling methods were experimented with to find the optimal model for this task. The final model was trained with 81, 118 visits and tested on a hold-out test set with a size of 9, 013 visits. The best model achieved a ROC AUC score of 0.86 and had a sensitivity of 87.3% and specificity of 70.9%. The most important factors that led to a CT scan order were found to be chief complaint, treatment area, and triage acuity. The proposed model was able to successfully identify patients needing a CT using administrative triage data that is available at the initial stage of a patient's arrival. By determining that a CT scan is needed early in the patient's visit, the ED can allocate resources to ensure these investigations are completed quickly and patient flow is maintained to reduce overcrowding.

Perspectives of radiographers on the emergence of artificial intelligence in diagnostic imaging in Saudi Arabia

Objectives

This study aimed to gain insight into radiographers’ views on the application of artificial intelligence (AI) in Saudi Arabia by conducting a qualitative investigation designed to provide recommendations to assist radiographic workforce improvement.

Materials and methods

We conducted an online cross-sectional online survey of Saudi radiographers regarding perspectives on AI implementation, job security, workforce development, and ethics.

Results

In total, 562 valid responses were received. Most respondents (90.6%) believed that AI was the direction of diagnostic imaging. Among the respondents, 88.5% stated that AI would improve the accuracy of diagnosis. Some challenges in implementing AI in Saudi Arabia include the high cost of equipment, inadequate knowledge, radiologists’ fear of losing employment, and concerns related to potential medical errors and cyber threats.

Conclusion

Radiographers were generally positive about introducing AI to radiology departments. To integrate AI successfully into radiology departments, radiographers need training programs, transparent policies, and motivation.

Learning rate of students detecting and annotating pediatric wrist fractures in supervised artificial intelligence dataset preparations

The use of artificial intelligence (AI) in image analysis is an intensively debated topic in the radiology community these days. AI computer vision algorithms typically rely on large-scale image databases, annotated by specialists. Developing and maintaining them is time-consuming, thus, the involvement of non-experts into the workflow of annotation should be considered. We assessed the learning rate of inexperienced evaluators regarding correct labeling of pediatric wrist fractures on digital radiographs. Students with and without a medical background labeled wrist fractures with bounding boxes in 7,000 radiographs over ten days. Pediatric radiologists regularly discussed their mistakes. We found F1 scores-as a measure for detection rate-to increase substantially under specialist feedback (mean 0.61±0.19 at day 1 to 0.97±0.02 at day 10, p<0.001), but not the Intersection over Union as a parameter for labeling precision (mean 0.27±0.29 at day 1 to 0.53±0.25 at day 10, p<0.001). The times needed to correct the students decreased significantly (mean 22.7±6.3 seconds per image at day 1 to 8.9±1.2 seconds at day 10, p<0.001) and were substantially lower as annotated by the radiologists alone. In conclusion our data showed, that the involvement of undergraduated students into annotation of pediatric wrist radiographs enables a substantial time saving for specialists, therefore, it should be considered.

Performance of a deep learning tool to detect missed aortic dilatation in a large chest CT cohort

Purpose

Thoracic aortic (TA) dilatation (TAD) is a risk factor for acute aortic syndrome and must therefore be reported in every CT report. However, the complex anatomy of the thoracic aorta impedes TAD detection. We investigated the performance of a deep learning (DL) prototype as a secondary reading tool built to measure TA diameters in a large-scale cohort.

Material and methods

Consecutive contrast-enhanced (CE) and non-CE chest CT exams with “normal” TA diameters according to their radiology reports were included. The DL-prototype (AIRad, Siemens Healthineers, Germany) measured the TA at nine locations according to AHA guidelines. Dilatation was defined as &gt;45 mm at aortic sinus, sinotubular junction (STJ), ascending aorta (AA) and proximal arch and &gt;40 mm from mid arch to abdominal aorta. A cardiovascular radiologist reviewed all cases with TAD according to AIRad. Multivariable logistic regression (MLR) was used to identify factors (demographics and scan parameters) associated with TAD classification by AIRad.

Results

18,243 CT scans (45.7% female) were successfully analyzed by AIRad. Mean age was 62.3 ± 15.9 years and 12,092 (66.3%) were CE scans. AIRad confirmed normal diameters in 17,239 exams (94.5%) and reported TAD in 1,004/18,243 exams (5.5%). Review confirmed TAD classification in 452/1,004 exams (45.0%, 2.5% total), 552 cases were false-positive but identification was easily possible using visual outputs by AIRad. MLR revealed that the following factors were significantly associated with correct TAD classification by AIRad: TAD reported at AA [odds ratio (OR): 1.12, p &lt; 0.001] and STJ (OR: 1.09, p = 0.002), TAD found at &gt;1 location (OR: 1.42, p = 0.008), in CE exams (OR: 2.1–3.1, p &lt; 0.05), men (OR: 2.4, p = 0.003) and patients presenting with higher BMI (OR: 1.05, p = 0.01). Overall, 17,691/18,243 (97.0%) exams were correctly classified.

Conclusions

AIRad correctly assessed the presence or absence of TAD in 17,691 exams (97%), including 452 cases with previously missed TAD independent from contrast protocol. These findings suggest its usefulness as a secondary reading tool by improving report quality and efficiency.

Complication Management and Prevention in Vascular and non-vascular Interventions

PURPOSE

 This overview summarizes key points of complication management in vascular and non-vascular interventions, particularly focusing on complication prevention and practiced safety culture. Flowcharts for intervention planning and implementation are outlined, and recording systems and conferences are explained in the context of failure analysis. In addition, troubleshooting by interventionalists on patient cases is presented.

MATERIAL AND METHODS

 The patient cases presented are derived from our institute. Literature was researched on PubMed.

RESULTS

 Checklists, structured intervention planning, standard operating procedures, and opportunities for error and complication discussion are important elements of complication management and essential for a practiced safety culture.

CONCLUSION

 A systematic troubleshooting and a practiced safety culture contribute significantly to patient safety. Primarily, a rational and thorough error analysis is important for quality improvement.

KEY POINTS

· Establishing a safety culture is essential for high-quality interventions with few complications.. · A rational and careful troubleshooting is essential to increase quality of interventions.. · Checklists and SOPs can structure and optimize the procedure of interventions..

CITATION FORMAT

· Weiss D, Wilms LM, Ivan VL et al. Complication Management and Prevention in Vascular and non-vascular Interventions. Fortschr Röntgenstr 2022; DOI: 10.1055/a-1829-6055.

Impact of a content-based image retrieval system on the interpretation of chest CTs of patients with diffuse parenchymal lung disease

OBJECTIVES

Content-based image retrieval systems (CBIRS) are a new and potentially impactful tool for radiological reporting, but their clinical evaluation is largely missing. This study aimed at assessing the effect of CBIRS on the interpretation of chest CT scans from patients with suspected diffuse parenchymal lung disease (DPLD).

MATERIALS AND METHODS

A total of 108 retrospectively included chest CT scans with 22 unique, clinically and/or histopathologically verified diagnoses were read by eight radiologists (four residents, four attending, median years reading chest CT scans 2.1± 0.7 and 12 ± 1.8, respectively). The radiologists read and provided the suspected diagnosis at a certified radiological workstation to simulate clinical routine. Half of the readings were done without CBIRS and half with the additional support of the CBIRS. The CBIRS retrieved the most likely of 19 lung-specific patterns from a large database of 6542 thin-section CT scans and provided relevant information (e.g., a list of potential differential diagnoses).

RESULTS

Reading time decreased by 31.3% (p < 0.001) despite the radiologists searching for additional information more frequently when the CBIRS was available (154 [72%] vs. 95 [43%], p < 0.001). There was a trend towards higher overall diagnostic accuracy (42.2% vs 34.7%, p = 0.083) when the CBIRS was available.

CONCLUSION

The use of the CBIRS had a beneficial impact on the reading time of chest CT scans in cases with DPLD. In addition, both resident and attending radiologists were more likely to consult informational resources if they had access to the CBIRS. Further studies are needed to confirm the observed trend towards increased diagnostic accuracy with the use of a CBIRS in practice.

KEY POINTS

• A content-based image retrieval system for supporting the diagnostic process of reading chest CT scans can decrease reading time by 31.3% (p < 0.001). • The decrease in reading time was present despite frequent usage of the content-based image retrieval system. • Additionally, a trend towards higher diagnostic accuracy was observed when using the content-based image retrieval system (42.2% vs 34.7%, p = 0.083).

Current practical experience with artificial intelligence in clinical radiology: a survey of the European Society of Radiology

A survey among the members of European Society of Radiology (ESR) was conducted regarding the current practical clinical experience of radiologists with Artificial Intelligence (AI)-powered tools. 690 radiologists completed the survey. Among these were 276 radiologists from 229 institutions in 32 countries who had practical clinical experience with an AI-based algorithm and formed the basis of this study. The respondents with clinical AI experience included 143 radiologists (52%) from academic institutions, 102 radiologists (37%) from regional hospitals, and 31 radiologists (11%) from private practice. The use case scenarios of the AI algorithm were mainly related to diagnostic interpretation, image post-processing, and prioritisation of workflow. Technical difficulties with integration of AI-based tools into the workflow were experienced by only 49 respondents (17.8%). Of 185 radiologists who used AI-based algorithms for diagnostic purposes, 140 (75.7%) considered the results of the algorithms generally reliable. The use of a diagnostic algorithm was mentioned in the report by 64 respondents (34.6%) and disclosed to patients by 32 (17.3%). Only 42 (22.7%) experienced a significant reduction of their workload, whereas 129 (69.8%) found that there was no such effect. Of 111 respondents who used AI-based algorithms for clinical workflow prioritisation, 26 (23.4%) considered algorithms to be very helpful for reducing the workload of the medical staff whereas the others found them only moderately helpful (62.2%) or not helpful at all (14.4%). Only 92 (13.3%) of the total 690 respondents indicated that they had intentions to acquire AI tools. In summary, although the assistance of AI algorithms was found to be reliable for different use case scenarios, the majority of radiologists experienced no reduction of practical clinical workload.

On-call abdominal ultrasonography: the rate of negative examinations and incidentalomas in a European tertiary care center

Objectives

To determine the proportions of abdominal US examinations during on-call hours that are negative and that contain an incidentaloma, and to explore temporal changes and determinants.

Methods

This study included 1615 US examinations that were done during on-call hours at a tertiary care center between 2005 and 2017.

Results

The total proportion of negative US examinations was 49.2% (795/1615). The total proportion of US examinations with an incidentaloma was 8.0% (130/1615). There were no significant temporal changes in either one of these proportions. The likelihood of a negative US examination was significantly higher when requested by anesthesiology [odds ratio (OR) 2.609, P = 0.011], or when the indication for US was focused on gallbladder and biliary ducts (OR 1.556, P = 0.007), transplant (OR 2.371, P = 0.005), trauma (OR 3.274, P < 0.001), or urolithiasis/postrenal obstruction (OR 3.366, P < 0.001). In contrast, US examinations were significantly less likely to be negative when requested by urology (OR 0.423, P = 0.014), or when the indication for US was acute oncology (OR 0.207, P = 0.045) or appendicitis (OR 0.260, P < 0.001). The likelihood of an incidentaloma on US was significantly higher in older patients (OR 1.020 per year of age increase, P < 0.001) or when the liver was evaluated with US (OR 3.522, P < 0.001).

Discussion

Nearly 50% of abdominal US examinations during on-call hours are negative, and 8% reveal an incidentaloma. Requesting specialty and indication for US affect the likelihood of a negative examination, and higher patient age and liver evaluations increase the chance of detecting an incidentaloma in this setting. These data may potentially be used to improve clinical reasoning and restrain overutilization of imaging.

Supplementary Information

The online version contains supplementary material available at 10.1007/s00261-022-03525-1.