Perceptual and Interpretive Error in Diagnostic Radiology-Causes and Potential Solutions

The application of deep learning in abdominal trauma diagnosis by CT imaging

BACKGROUND

Abdominal computed tomography (CT) scan is a crucial imaging modality for creating cross-sectional images of the abdominal area, particularly in cases of abdominal trauma, which is commonly encountered in traumatic injuries. However, interpreting CT images is a challenge, especially in emergency. Therefore, we developed a novel deep learning algorithm-based detection method for the initial screening of abdominal internal organ injuries.

METHODS

We utilized a dataset provided by the Kaggle competition, comprising 3,147 patients, of which 855 were diagnosed with abdominal trauma, accounting for 27.16% of the total patient population. Following image data pre-processing, we employed a 2D semantic segmentation model to segment the images and constructed a 2.5D classification model to assess the probability of injury for each organ. Subsequently, we evaluated the algorithm's performance using 5k-fold cross-validation.

RESULTS

With particularly noteworthy performance in detecting renal injury on abdominal CT scans, we achieved an acceptable accuracy of 0.932 (with a positive predictive value (PPV) of 0.888, negative predictive value (NPV) of 0.943, sensitivity of 0.887, and specificity of 0.944). Furthermore, the accuracy for liver injury detection was 0.873 (with PPV of 0.789, NPV of 0.895, sensitivity of 0.789, and specificity of 0.895), while for spleen injury, it was 0.771 (with PPV of 0.630, NPV of 0.814, sensitivity of 0.626, and specificity of 0.816).

CONCLUSIONS

The deep learning model demonstrated the capability to identify multiple organ injuries simultaneously on CT scans and holds potential for application in preliminary screening and adjunctive diagnosis of trauma cases beyond abdominal injuries.

Diagnostic Accuracy of Artificial Intelligence-Based Algorithms in Automated Detection of Neck of Femur Fracture on a Plain Radiograph: A Systematic Review and Meta-analysis

Objectives

To evaluate the diagnostic accuracy of artificial intelligence-based algorithms in identifying neck of femur fracture on a plain radiograph.

Design

Systematic review and meta-analysis.

Data sources

PubMed, Web of science, Scopus, IEEE, and the Science direct databases were searched from inception to 30 July 2023.

Eligibility criteria for study selection

Eligible article types were descriptive, analytical, or trial studies published in the English language providing data on the utility of artificial intelligence (AI) based algorithms in the detection of the neck of the femur (NOF) fracture on plain X-ray.

Main outcome measures

The prespecified primary outcome was to calculate the sensitivity, specificity, accuracy, Youden index, and positive and negative likelihood ratios. Two teams of reviewers (each consisting of two members) extracted the data from available information in each study. The risk of bias was assessed using a mix of the CLAIM (the Checklist for AI in Medical Imaging) and QUADAS-2 (A Revised Tool for the Quality Assessment of Diagnostic Accuracy Studies) criteria.

Results

Of the 437 articles retrieved, five were eligible for inclusion, and the pooled sensitivity of AIs in diagnosing the fracture NOF was 85%, with a specificity of 87%. For all studies, the pooled Youden index (YI) was 0.73. The average positive likelihood ratio (PLR) was 19.88, whereas the negative likelihood ratio (NLR) was 0.17. The random effects model showed an overall odds of 1.16 (0.84-1.61) in the forest plot, comparing the AI system with those of human diagnosis. The overall heterogeneity of the studies was marginal (I2 = 51%). The CLAIM criteria for risk of bias assessment had an overall >70% score.

Conclusion

Artificial intelligence (AI)-based algorithms can be used as a diagnostic adjunct, benefiting clinicians by taking less time and effort in neck of the femur (NOF) fracture diagnosis.

Study registration

PROSPERO CRD42022375449.

Supplementary Information

The online version contains supplementary material available at 10.1007/s43465-024-01130-6.

Capturing artificial intelligence applications' value proposition in healthcare - a qualitative research study

Artificial intelligence (AI) applications pave the way for innovations in the healthcare (HC) industry. However, their adoption in HC organizations is still nascent as organizations often face a fragmented and incomplete picture of how they can capture the value of AI applications on a managerial level. To overcome adoption hurdles, HC organizations would benefit from understanding how they can capture AI applications' potential.We conduct a comprehensive systematic literature review and 11 semi-structured expert interviews to identify, systematize, and describe 15 business objectives that translate into six value propositions of AI applications in HC.Our results demonstrate that AI applications can have several business objectives converging into risk-reduced patient care, advanced patient care, self-management, process acceleration, resource optimization, and knowledge discovery.We contribute to the literature by extending research on value creation mechanisms of AI to the HC context and guiding HC organizations in evaluating their AI applications or those of the competition on a managerial level, to assess AI investment decisions, and to align their AI application portfolio towards an overarching strategy.

Preliminary Evidence of the Use of Generative AI in Health Care Clinical Services: Systematic Narrative Review

BACKGROUND

Generative artificial intelligence tools and applications (GenAI) are being increasingly used in health care. Physicians, specialists, and other providers have started primarily using GenAI as an aid or tool to gather knowledge, provide information, train, or generate suggestive dialogue between physicians and patients or between physicians and patients' families or friends. However, unless the use of GenAI is oriented to be helpful in clinical service encounters that can improve the accuracy of diagnosis, treatment, and patient outcomes, the expected potential will not be achieved. As adoption continues, it is essential to validate the effectiveness of the infusion of GenAI as an intelligent technology in service encounters to understand the gap in actual clinical service use of GenAI.

OBJECTIVE

This study synthesizes preliminary evidence on how GenAI assists, guides, and automates clinical service rendering and encounters in health care The review scope was limited to articles published in peer-reviewed medical journals.

METHODS

We screened and selected 0.38% (161/42,459) of articles published between January 1, 2020, and May 31, 2023, identified from PubMed. We followed the protocols outlined in the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines to select highly relevant studies with at least 1 element on clinical use, evaluation, and validation to provide evidence of GenAI use in clinical services. The articles were classified based on their relevance to clinical service functions or activities using the descriptive and analytical information presented in the articles.

RESULTS

Of 161 articles, 141 (87.6%) reported using GenAI to assist services through knowledge access, collation, and filtering. GenAI was used for disease detection (19/161, 11.8%), diagnosis (14/161, 8.7%), and screening processes (12/161, 7.5%) in the areas of radiology (17/161, 10.6%), cardiology (12/161, 7.5%), gastrointestinal medicine (4/161, 2.5%), and diabetes (6/161, 3.7%). The literature synthesis in this study suggests that GenAI is mainly used for diagnostic processes, improvement of diagnosis accuracy, and screening and diagnostic purposes using knowledge access. Although this solves the problem of knowledge access and may improve diagnostic accuracy, it is oriented toward higher value creation in health care.

CONCLUSIONS

GenAI informs rather than assisting or automating clinical service functions in health care. There is potential in clinical service, but it has yet to be actualized for GenAI. More clinical service-level evidence that GenAI is used to streamline some functions or provides more automated help than only information retrieval is needed. To transform health care as purported, more studies related to GenAI applications must automate and guide human-performed services and keep up with the optimism that forward-thinking health care organizations will take advantage of GenAI.

Streamlining Radiology Workflows Through the Development and Deployment of Automated Microservices

Microservices are a software development approach where an application is structured as a collection of loosely coupled, independently deployable services, each focusing on executing a specific purpose. The development of microservices could have a significant impact on radiology workflows, allowing routine tasks to be automated and improving the efficiency and accuracy of radiologic tasks. This technical report describes the development of several microservices that have been successfully deployed in a tertiary cancer center, resulting in substantial time savings for radiologists and other staff involved in radiology workflows. These microservices include the automatic generation of shift emails, notifying administrative staff and faculty about fellows on rotation, notifying referring physicians about outside examinations, and populating report templates with information from PACS and RIS. The report outlines the common thought process behind developing these microservices, including identifying a problem, connecting various APIs, collecting data in a database, writing a prototype and deploying it, gathering feedback and refining the service, putting it in production, and identifying staff who are in charge of maintaining the service. The report concludes by discussing the benefits and challenges of microservices in radiology workflows, highlighting the importance of multidisciplinary collaboration, interoperability, security, and privacy.

Why do errors arise in artificial intelligence diagnostic tools in histopathology and how can we minimize them?

Artificial intelligence (AI)-based diagnostic tools can offer numerous benefits to the field of histopathology, including improved diagnostic accuracy, efficiency and productivity. As a result, such tools are likely to have an increasing role in routine practice. However, all AI tools are prone to errors, and these AI-associated errors have been identified as a major risk in the introduction of AI into healthcare. The errors made by AI tools are different, in terms of both cause and nature, to the errors made by human pathologists. As highlighted by the National Institute for Health and Care Excellence, it is imperative that practising pathologists understand the potential limitations of AI tools, including the errors made. Pathologists are in a unique position to be gatekeepers of AI tool use, maximizing patient benefit while minimizing harm. Furthermore, their pathological knowledge is essential to understanding when, and why, errors have occurred and so to developing safer future algorithms. This paper summarises the literature on errors made by AI diagnostic tools in histopathology. These include erroneous errors, data concerns (data bias, hidden stratification, data imbalances, distributional shift, and lack of generalisability), reinforcement of outdated practices, unsafe failure mode, automation bias, and insensitivity to impact. Methods to reduce errors in both tool design and clinical use are discussed, and the practical roles for pathologists in error minimisation are highlighted. This aims to inform and empower pathologists to move safely through this seismic change in practice and help ensure that novel AI tools are adopted safely.

Medical Malpractice and Diagnostic Radiology: Challenges and Opportunities

Medicolegal challenges in radiology are broad and impact both radiologists and patients. Radiologists may be affected directly by malpractice litigation or indirectly due to defensive imaging ordering practices. Patients also could be harmed physically, emotionally, or financially by unnecessary tests or procedures. As technology advances, the incorporation of artificial intelligence into medicine will bring with it new medicolegal challenges and opportunities. This article reviews the current and emerging direct and indirect effects of medical malpractice on radiologists and summarizes evidence-based solutions.

Design and Perceived Value of a Novel Solution for Asynchronous Communication in Radiology

RATIONALE AND OBJECTIVES

Communication with and within the Radiology Department is typically initiated over phone, face-to-face or general-purpose chat, causing frequent interruptions, additional mental workload, workflow inefficiencies and diagnostic errors. We developed and evaluated a new communication solution that aims to reduce avoidable interruptions caused by technologist-radiologist communication.

MATERIALS AND METHODS

Following an iterative design process with future end users, a scalable web-based software solution, RadConnect, was developed enabling a chat-based communication workflow between a technologist and a radiologist. As a first experimental implementation, technologists can send categorized tickets to a radiology section account. Radiologists receive the tickets in a worklist that is prioritized by urgency. Consented radiologists and technologists performed scripted tasks in 2 hr sessions and completed a structured questionnaire on perceived value and comparison to standard communication modes.

RESULTS

Of 17 participants from three academic European institutes, 65% (11/17) believed they would use RadConnect frequently; 53% (9/17) believed that it reduces phone calls >80%; and 88% (15/17) believed it adds value compared to general-purpose enterprise chat applications.

DISCUSSION

Participants recognized the value of RadConnect especially its categorized tickets, prioritized worklist and role-based interaction model. Inter-institute differences in perceived value of RadConnect may have been caused by technologist-radiologist proximity and communication alternatives in the institutions.

CONCLUSION

Chat-based role-based communication might be a viable mode of communication between technologists and radiologists to reduce avoidable interruptions. Tailoring the chat solution to the needs of and tightly integrated with the radiology workflow is valued by future end users after exposure to the tool in a simulated environment.

Diagnostic error and bias in the department of radiology: a pictorial essay

Diagnostic imaging is an essential and indispensable part of medical diagnosis and treatment, and diagnostic errors or biases are also common in the department of radiology, sometimes even having a severe impact on the diagnosis and treatment of patients. There are various reasons for diagnostic errors or biases in imaging. In this review, we analyze and summarize the causes of diagnostic imaging errors and biases based on practical cases. We propose solutions for dealing with diagnostic imaging errors and reducing their probability, thereby helping radiologists in their clinical practice.Critical relevance statement Diagnostic errors or bias contribute to most medical errors in the radiology department. Solutions for dealing with diagnostic imaging errors are pivotal for patients.Key points• Diagnostic errors or bias contribute to most medical errors in radiology department.• Solutions for dealing with diagnostic imaging errors are pivotal for patients.• This review summarizes the causes of diagnostic errors and offers solutions to them.

Discrimination tasks in simulated low-dose CT noise

BACKGROUND

This study reports the results of a set of discrimination experiments using simulated images that represent the appearance of subtle lesions in low-dose computed tomography (CT) of the lungs. Noise in these images has a characteristic ramp-spectrum before apodization by noise control filters. We consider three specific diagnostic features that determine whether a lesion is considered malignant or benign, two system-resolution levels, and four apodization levels for a total of 24 experimental conditions.

PURPOSE

The goal of the investigation is to better understand how well human observers perform subtle discrimination tasks like these, and the mechanisms of that performance. We use a forced-choice psychophysical paradigm to estimate observer efficiency and classification images. These measures quantify how effectively subjects can read the images, and how they use images to perform discrimination tasks across the different imaging conditions.

MATERIALS AND METHODS

The simulated CT images used as stimuli in the psychophysical experiments are generated from high-resolution objects passed through a modulation transfer function (MTF) before down-sampling to the image-pixel grid. Acquisition noise is then added with a ramp noise-power spectrum (NPS), with subsequent smoothing through apodization filters. The features considered are lesion size, indistinct lesion boundary, and a nonuniform lesion interior. System resolution is implemented by an MTF with resolution (10% max.) of 0.47 or 0.58 cyc/mm. Apodization is implemented by a Shepp-Logan filter (Sinc profile) with various cutoffs. Six medically naïve subjects participated in the psychophysical studies, entailing training and testing components for each condition. Training consisted of staircase procedures to find the 80% correct threshold for each subject, and testing involved 2000 psychophysical trials at the threshold value for each subject. Human-observer performance is compared to the Ideal Observer to generate estimates of task efficiency. The significance of imaging factors is assessed using ANOVA. Classification images are used to estimate the linear template weights used by subjects to perform these tasks. Classification-image spectra are used to analyze subject weights in the spatial-frequency domain.

RESULTS

Overall, average observer efficiency is relatively low in these experiments (10%-40%) relative to detection and localization studies reported previously. We find significant effects for feature type and apodization level on observer efficiency. Somewhat surprisingly, system resolution is not a significant factor. Efficiency effects of the different features appear to be well explained by the profile of the linear templates in the classification images. Increasingly strong apodization is found to both increase the classification-image weights and to increase the mean-frequency of the classification-image spectra. A secondary analysis of "Unapodized" classification images shows that this is largely due to observers undoing (inverting) the effects of apodization filters.

CONCLUSIONS

These studies demonstrate that human observers can be relatively inefficient at feature-discrimination tasks in ramp-spectrum noise. Observers appear to be adapting to frequency suppression implemented in apodization filters, but there are residual effects that are not explained by spatial weighting patterns. The studies also suggest that the mechanisms for improving performance through the application of noise-control filters may require further investigation.

Reducing Errors Resulting From Commonly Missed Chest Radiography Findings

Chest radiography (CXR), the most frequently performed imaging examination, is vulnerable to interpretation errors resulting from commonly missed findings. Methods to reduce these errors are presented. A practical approach using a systematic and comprehensive visual search strategy is described. The use of a checklist for quality control in the interpretation of CXR images is proposed to avoid overlooking commonly missed findings of clinical importance. Artificial intelligence is among the emerging and promising methods to enhance detection of CXR abnormalities. Despite their potential adverse consequences, errors offer opportunities for continued education and quality improvements in patient care, if managed within a just, supportive culture.

Eye-tracking differences between free text and template radiology reports: a pilot study

Purpose

One possible limitation of structured template radiology reports is that radiologists look back and forth between viewing and dictation monitors, thereby impacting the length of time spent reviewing images and generating a report. We hypothesize that the total time spent viewing case images is diminished and/or the total time spent creating a report is prolonged when the report is generated using a structured template compared with free text format.

Approach

Three neuroradiologists and three senior residents viewed five brain magnetic resonance imaging cases with unique findings while eye position was recorded. Participants generated reports for each case utilizing both structured templates and free text dictation. The time spent viewing images was compared with the time spent looking at the dictation screen.

Results

The two main hypotheses were confirmed: the total time viewing images diminished with templates versus free text dictation and the total time to create a report was prolonged with templates. The mean time (s) spent on the "image" region of interest approached statistical significance as a function of the report type [free: attendings = 236.79 (154.43), residents = 223.55 (77.79); template: attendings = 163.40 (73.42), residents = 182.48 (77.47)] and was overall lower with the template reporting for both attendings and residents ( F = 3.77 , p = 0.0623 ), but it did not differ as a function of seniority ( F = 0.017 , p = 0.8977 ).

Conclusions

Template-based radiology reports have significant potential to alter the way radiologists view images and report on them, spending more time viewing the report monitor rather than diagnostic images compared with free text dictation. Many radiologists prefer templates for reporting as the structured format may aid in conducting a more systematic or thorough search for findings, although prior work on this assumption is mixed. Future eye-tracking studies could further elucidate whether and how templates and free reports impact the detection and classification of radiographic findings.

Factors affecting interpretation of dental radiographs

OBJECTIVES

To identify the factors influencing errors in the interpretation of dental radiographs.

METHODS

A protocol was registered on Prospero. All studies published until May 2022 were included in this review. The search of the electronic databases spanned Ovid Medline, PubMed, EMBASE, Web of Science and Scopus. The quality of the studies was assessed using the MMAT tool. Due to the heterogeneity of the included studies, a meta-analysis was not conducted.

RESULTS

The search yielded 858 articles, of which eight papers met the inclusion and exclusion criteria and were included in the systematic review. These studies assessed the factors influencing the accuracy of the interpretation of dental radiographs. Six factors were identified as being significant that affected the occurrence of interpretation errors. These include clinical experience, clinical knowledge, and technical ability, case complexity, time pressure, location and duration of dental education and training and cognitive load.

CONCLUSIONS

The occurrence of interpretation errors has not been widely investigated in dentistry. The factors identified in this review are interlinked. Further studies are needed to better understand the extent of the occurrence of interpretive errors and their impact on the practice of dentistry.

Influence of Prior Imaging Information on Diagnostic Accuracy for Focal Skeletal Processes—A Retrospective Analysis of the Consistency between Biopsy-Verified Imaging Diagnoses

Introduction: Comparing imaging examinations with those previously obtained is considered mandatory in imaging guidelines. To our knowledge, no studies are available on neither the influence, nor the sequence, of prior imaging and reports on diagnostic accuracy using biopsy as the reference standard. Such data are important to minimize diagnostic errors and to improve the preparation of diagnostic imaging guidelines. The aim of our study was to provide such data. Materials and methods: A retrospective cohort of 216 consecutive skeletal biopsies from patients with at least 2 different imaging modalities (X-ray, CT and MRI) performed within 6 months of biopsy was identified. The diagnostic accuracy of the individual imaging modality was assessed. Finally, the possible influence of the sequence of imaging modalities was investigated. Results: No significant difference in the accuracy of the imaging modalities was shown, being preceded by another imaging modality or not. However, the sequence analyses indicate sequential biases, particularly if MRI was the first imaging modality. Conclusion: The sequence of the imaging modalities seems to influence the diagnostic accuracy against a pathology reference standard. Further studies are needed to establish evidence-based guidelines for the strategy of using previous imaging and reports to improve diagnostic accuracy.

Angiographically Occult Subarachnoid Hemorrhage: Yield of Repeat Angiography, Influence of Initial CT Bleed Pattern, and Sources of Diagnostic Error in 242 Consecutive Patients

BACKGROUND AND PURPOSE

Nearly 20% of patients with spontaneous SAH have no definitive source on initial DSA. The purpose of this study was to investigate the timing and yield of repeat DSA, to clarify the influence of initial CT bleed pattern, and to characterize sources of diagnostic error in this scenario.

MATERIALS AND METHODS

We evaluated the yield of repeat DSA and clinical outcomes stratified by hemorrhage pattern on CT in consecutive patients with nontraumatic SAH with negative initial DSA findings at a referral center. Cases in which the culprit lesion was subsequently diagnosed were classified as physiologically occult (ie, undetectable) on the initial DSA, despite adequate technique and interpretation or misdiagnosed due to operator-dependent error.

RESULTS

Two hundred forty-two of 1163 (20.8%) patients with spontaneous SAH had negative initial DSA findings between 2009 and 2018. The SAH CT pattern was nonperimesencephalic (41%), perimesencephalic (36%), sulcal (18%), and CT-negative (5%). Repeat DSA in 135/242 patients (55.8%) revealed a source in 10 patients (7.4%): 4 saccular aneurysms, 4 atypical aneurysms, and 2 arteriovenous shunts. The overall yield of repeat DSA was 11.3% with nonperimesencephalic and 2.2% for perimesencephalic patterns. The yield of the second and third DSAs with a nonperimesencephalic pattern was 7.7% and 12%, respectively. Physiologically occult lesions accounted for 6/242 (2.5%) and operator-dependent errors accounted for 7/242 (2.9%) of all angiographically occult lesions on the first DSA.

CONCLUSIONS

Atypical aneurysms and small arteriovenous shunts are important causes of SAH negative on angiography. Improving DSAs technique can modestly reduce the need for repeat DSA; however, a small fraction of SAH sources remain occult despite adequate technique. These findings support the practice of repeating DSA in patients with a nonperimesencephalic SAH pattern.

The medical algorithmic audit

Artificial intelligence systems for health care, like any other medical device, have the potential to fail. However, specific qualities of artificial intelligence systems, such as the tendency to learn spurious correlates in training data, poor generalisability to new deployment settings, and a paucity of reliable explainability mechanisms, mean they can yield unpredictable errors that might be entirely missed without proactive investigation. We propose a medical algorithmic audit framework that guides the auditor through a process of considering potential algorithmic errors in the context of a clinical task, mapping the components that might contribute to the occurrence of errors, and anticipating their potential consequences. We suggest several approaches for testing algorithmic errors, including exploratory error analysis, subgroup testing, and adversarial testing, and provide examples from our own work and previous studies. The medical algorithmic audit is a tool that can be used to better understand the weaknesses of an artificial intelligence system and put in place mechanisms to mitigate their impact. We propose that safety monitoring and medical algorithmic auditing should be a joint responsibility between users and developers, and encourage the use of feedback mechanisms between these groups to promote learning and maintain safe deployment of artificial intelligence systems.

Artificial intelligence system for identification of false-negative interpretations in chest radiographs

OBJECTIVES

To investigate the efficacy of an artificial intelligence (AI) system for the identification of false negatives in chest radiographs that were interpreted as normal by radiologists.

METHODS

We consecutively collected chest radiographs that were read as normal during 1 month (March 2020) in a single institution. A commercialized AI system was retrospectively applied to these radiographs. Radiographs with abnormal AI results were then re-interpreted by the radiologist who initially read the radiograph ("AI as the advisor" scenario). The reference standards for the true presence of relevant abnormalities in radiographs were defined by majority voting of three thoracic radiologists. The efficacy of the AI system was evaluated by detection yield (proportion of true-positive identification among the entire examination) and false-referral rate (FRR, proportion of false-positive identification among all examinations). Decision curve analyses were performed to evaluate the net benefits of applying the AI system.

RESULTS

A total of 4208 radiographs from 3778 patients (M:F = 1542:2236; median age, 56 years) were included. The AI system identified initially overlooked relevant abnormalities with a detection yield and an FRR of 2.4% and 14.0%, respectively. In the "AI as the advisor" scenario, radiologists detected initially overlooked relevant abnormalities with a detection yield and FRR of 1.2% and 0.97%, respectively. In a decision curve analysis, AI as an advisor scenario exhibited a positive net benefit when the cost-to-benefit ratio was below 1:0.8.

CONCLUSION

An AI system could identify relevant abnormalities overlooked by radiologists and could enable radiologists to correct their false-negative interpretations by providing feedback to radiologists.

KEY POINTS

• In consecutive chest radiographs with normal interpretations, an artificial intelligence system could identify relevant abnormalities that were initially overlooked by radiologists. • The artificial intelligence system could enable radiologists to correct their initial false-negative interpretations by providing feedback to radiologists when overlooked abnormalities were present.

Enhanced lung image segmentation using deep learning

With the advances in technology, assistive medical systems are emerging with rapid growth and helping healthcare professionals. The proactive diagnosis of diseases with artificial intelligence (AI) and its aligned technologies has been an exciting research area in the last decade. Doctors usually detect tuberculosis (TB) by checking the lungs' X-rays. Classification using deep learning algorithms is successfully able to achieve accuracy almost similar to a doctor in detecting TB. It is found that the probability of detecting TB increases if classification algorithms are implemented on segmented lungs instead of the whole X-ray. The paper's novelty lies in detailed analysis and discussion of U-Net +  + results and implementation of U-Net +  + in lung segmentation using X-ray. A thorough comparison of U-Net +  + with three other benchmark segmentation architectures and segmentation in diagnosing TB or other pulmonary lung diseases is also made in this paper. To the best of our knowledge, no prior research tried to implement U-Net +  + for lung segmentation. Most of the papers did not even use segmentation before classification, which causes data leakage. Very few used segmentations before classification, but they only used U-Net, which U-Net +  + can easily replace because accuracy and mean_iou of U-Net +  + are greater than U-Net accuracy and mean_iou , discussed in results, which can minimize data leakage. The authors achieved more than 98% lung segmentation accuracy and mean_iou 0.95 using U-Net +  + , and the efficacy of such comparative analysis is validated.

Social Asymmetry, Artificial Intelligence and the Medical Imaging Landscape

Social and health care equity and justice should be prioritized by the mantra of medicine, first do no harm. Despite highly motivated national and global health strategies, there remains significant health care inequity. Intrinsic and extrinsic factors, including a number of biases, are key drivers of ongoing health inequity including equity of access and opportunity for nuclear medicine and radiology services. There is a substantial gap in the global practice of nuclear medicine in particular, but also radiology, between developed health economies and those considered developing or undeveloped. At a local level, even in developed health economies, there can be a significant disparity between health services, including medical imaging, between communities based on socioeconomic, cultural or geographic differences. Artificial intelligence (AI) has the potential to either widen the health inequity divide or substantially reduce it. Distributed generally, AI technology could be used to overcome geographic boundaries to health care, thus bringing general and specialist care into underserved communities. However, should AI technology be limited to localities already enjoying ample healthcare access and direct access to health infrastructure, like radiology and nuclear medicine, it could then accentuate the gap. There are a number of challenges across the AI pipeline that need careful attention to ensure beneficence over maleficence. Fully realized, AI augmented health care could be crafted as an integral part of the broader strategy convergence on local, national and global health equity. The applications of AI in nuclear medicine and radiology could emerge as a powerful tool in social and health equity.

Role of Ergonomic Improvements in Decreasing Repetitive Stress Injuries and Promoting Well-Being in a Radiology Department

RATIONALE AND OBJECTIVES

To determine if ergonomic improvements in a radiology department can decrease repetitive stress injuries (RSIs), advance ergonomics knowledge, and improve well-being.

MATERIALS AND METHODS

Radiologists in an academic institution were surveyed regarding physician wellness, workstations, RSIs, and ergonomics knowledge before and after interventions over 1 year. Interventions included committee formation, education, wrist pads and wireless mice, broken table and chair replacement, and cord organization. Mann-Whitney U test was used for analysis.

RESULTS

Survey response was 40% preinterventions (59/147), and 42% (66/157) postinterventions. Preinterventions, of radiologists with RSI history, 17/40 (42%) reported the RSI caused symptoms which can lead to burnout, and 15/40 (37%) responded their RSI made them think about leaving their job. Twenty-three of 59 (39%) radiologists had an active RSI preinterventions. Postinterventions, 9/25 (36%) RSI resolved, 13/25 (52%) RSI improved, and 3/25 (12%) RSI did not improve. RSI improvements were attributed to ergonomic interventions in 19/25 (76%) and therapy in 2/25 (8%). Radiologists who thought their workstation was designed with well-being in mind increased from 9/59 (15%) to 52/64 (81%). The percentage of radiologists knowing little or nothing about ergonomics decreased from 15/59 (25%) to 5/64 (8%). After ergonomics interventions, more radiologists thought the administration cared about safety and ergonomics, equipment was distributed fairly, and radiologists had the ability to ask for equipment (p < .01). Fifty-three of 64 (83%) of radiologists after interventions said improving workstation ergonomic design contributed to well-being.

CONCLUSION

Ergonomic improvements in radiology can decrease RSIs, advance ergonomics knowledge, and improve well-being.

Shortened total spine MRI protocol in the detection of spinal cord compression and pathology for emergent settings: a noninferiority study

BACKGROUND AND PURPOSE

Spinal cord compression (SCC) requires rapid diagnosis in the emergent setting; however, current MRI protocols may be cumbersome for patients and clinicians. We sought to validate an abbreviated total spine MRI (TS-MRI) protocol using standard non-contrast sequences in the detection of SCC and other clinically significant findings (OCSF).

METHODS

Two hundred six TS-MRI scans obtained over a 30-month period for SCC were included. Sagittal T2 (T2_sag), sagittal T1 (T1_sag), and sagittal STIR (IR_sag), as well as axial T2 (T2_ax) images, were individually assessed independently by 2 reviewers for SCC, cauda equina compression (CEC), and OCSF. A protocol consisting of all the sequences was considered the gold standard. Sensitivity and specificity of single and combined MRI sequences for SCC/CEC and OCSF were determined and were tested for noninferiority relative to standard non-contrast sequences using a 5% noninferiority margin.

RESULTS

An abbreviated protocol of IR_sag + T2_ax provided the best performance with sensitivity and specificity of 100% (95%CI, 96.0-100.0) and 98.6% (95%CI, 95.6-99.7) for SCC/CEC and 100.0% (95%CI, 96.7-100.0), and 99.3% (95%CI, 96.6-99.9) for OCSF. The mean difference of sensitivity and specificity between IR_sag + T2_ax and standard protocol was 0.0% (95%CI, 0.0-4.0) and - 2.1% (95%CI, - 5.4 to - 0.6) for SCC/CEC and 0.0% (95%CI, 0.0-3.3) and - 1.5% (95%CI, - 4.8 to - 0.3) for OCSF, all within the noninferiority margin of 5%.

CONCLUSIONS

An abbreviated TS-MRI protocol of IR_sag + T2_ax is noninferior to the standard non-contrast protocol, potentially allowing for faster emergent imaging diagnosis and triage.

[Peer Review of Teleradiology at a Teleradiology Clinic: Comparison of Unacceptable Diagnosis and Clinically Significant Discrepancy between Radiology Sections and Imaging Modalities]

Purpose

The purpose of this study was to evaluate the rates of unacceptable diagnosis and clinically significant diagnostic discrepancy in radiology sections and imaging modalities through a peer review of teleradiology.

Materials and Methods

Teleradiology peer reviews in a Korean teleradiology clinic in 2018 and 2019 were included. The peer review scores were classified as acceptable and unacceptable diagnoses and clinically insignificant and significant diagnostic discrepancy. The diagnostic discrepancy rates and clinical significance were compared among radiology sections and imaging modalities using the chi-square test.

Results

Of 1312 peer reviews, 117 (8.9%) cases had unacceptable diagnoses. Of 462 diagnostic discrepancies, the clinically significant discrepancy was observed in 104 (21.6%) cases. In radiology sections, the unacceptable diagnosis was highest in the musculoskeletal section (21.4%) (p < 0.05), followed by the abdominal section (7.3%) and neuro section (1.3%) (p < 0.05). The proportion of significant discrepancy was higher in the chest section (32.7%) than in the musculoskeletal (19.5%) and abdominal sections (17.1%) (p < 0.05). Regarding modalities, the number of unacceptable diagnoses was higher with MRI (16.2%) than plain radiology (7.8%) (p < 0.05). There was no significant difference in significant discrepancy.

Conclusion

Peer review provides the rates of unacceptable diagnosis and clinically significant discrepancy in teleradiology. These rates also differ with subspecialty and modality.

Sabre-sheath trachea: An underused diagnostic weapon in the thoracic armoury

INTRODUCTION

A sabre-sheath trachea is a highly specific sign for chronic obstructive pulmonary disease. It also correlates well with the degree of disease. We hypothesized that the term is vastly under-utilized in radiologic reporting, despite its high diagnostic value.

METHODS

We interrogated our multisite metropolitan-wide radiology information system to find the number of CT reports containing the phrase 'sabre-sheath trachea' and conceivable variants thereof, over the 10 years to present. We compared this with the entire number of CT chest reports in the same time period in order to estimate the utilization of the sign.

RESULTS

The results confirmed our hypothesis that the sign is rarely invoked, likely around 1 in 41 times relative to opportunity.

CONCLUSIONS

This highly specific sign of chronic obstructive small airway disease should be reinforced in training and utilized by radiologists.

Novel Chest Radiographic Biomarkers for COVID-19 Using Radiomic Features Associated with Diagnostics and Outcomes

COVID-19 is a highly contagious disease that can cause severe pneumonia. Patients with pneumonia undergo chest X-rays (XR) to assess infiltrates that identify the infection. However, the radiographic characteristics of COVID-19 are similar to the other acute respiratory syndromes, hindering the imaging diagnosis. In this work, we proposed identifying quantitative/radiomic biomarkers for COVID-19 to support XR assessment of acute respiratory diseases. This retrospective study used different cohorts of 227 patients diagnosed with pneumonia; 49 of them had COVID-19. Automatically segmented images were characterized by 558 quantitative features, including gray-level histogram and matrices of co-occurrence, run-length, size zone, dependence, and neighboring gray-tone difference. Higher-order features were also calculated after applying square and wavelet transforms. Mann-Whitney U test assessed the diagnostic performance of the features, and the log-rank test assessed the prognostic value to predict Kaplan-Meier curves of overall and deterioration-free survival. Statistical analysis identified 51 independently validated radiomic features associated with COVID-19. Most of them were wavelet-transformed features; the highest performance was the small dependence matrix feature of "low gray-level emphasis" (area under the curve of 0.87, sensitivity of 0.85, [Formula: see text]). Six features presented short-term prognostic value to predict overall and deterioration-free survival. The features of histogram "mean absolute deviation" and size zone matrix "non-uniformity" yielded the highest differences on Kaplan-Meier curves with a hazard ratio of 3.20 ([Formula: see text]). The radiomic markers showed potential as quantitative measures correlated with the etiologic agent of acute infectious diseases and to stratify short-term risk of COVID-19 patients.

Artificial Intelligence to Assist in Exclusion of Coronary Atherosclerosis During CCTA Evaluation of Chest Pain in the Emergency Department: Preparing an Application for Real-world Use

Coronary computed tomography angiography (CCTA) evaluation of chest pain patients in an emergency department (ED) is considered appropriate. While a "negative" CCTA interpretation supports direct patient discharge from an ED, labor-intensive analyses are required, with accuracy in jeopardy from distractions. We describe the development of an artificial intelligence (AI) algorithm and workflow for assisting qualified interpreting physicians in CCTA screening for total absence of coronary atherosclerosis. The two-phase approach consisted of (1) phase 1-development and preliminary testing of an algorithm for vessel-centerline extraction classification in a balanced study population (n = 500 with 50% disease prevalence) derived by retrospective random case selection, and (2) phase 2-simulated clinical Trialing of developed algorithm on a per-case (entire coronary artery tree) basis in a more "real-world" study population (n = 100 with 28% disease prevalence) from an ED chest pain series. This allowed pre-deployment evaluation of the AI-based CCTA screening application which provides vessel-by-vessel graphic display of algorithm inference results integrated into a clinically capable viewer. Algorithm performance evaluation used area under the receiver operating characteristic curve (AUC-ROC); confusion matrices reflected ground truth vs AI determinations. The vessel-based algorithm demonstrated strong performance with AUC-ROC = 0.96. In both phase 1 and phase 2, independent of disease prevalence differences, negative predictive values at the case level were very high at 95%. The rate of completion of the algorithm workflow process (96% with inference results in 55-80 s) in phase 2 depended on adequate image quality. There is potential for this AI application to assist in CCTA interpretation to help extricate atherosclerosis from chest pain presentations.

The Effects of Perinodular Features on Solid Lung Nodule Classification

Lung cancer is the most lethal malignant neoplasm worldwide, with an annual estimated rate of 1.8 million deaths. Computed tomography has been widely used to diagnose and detect lung cancer, but its diagnosis remains an intricate and challenging work, even for experienced radiologists. Computer-aided diagnosis tools and radiomics tools have provided support to the radiologist's decision, acting as a second opinion. The main focus of these tools has been to analyze the intranodular zone; nevertheless, recent works indicate that the interaction between the nodule and its surroundings (perinodular zone) could be relevant to the diagnosis process. However, only a few works have investigated the importance of specific attributes of the perinodular zone and have shown how important they are in the classification of lung nodules. In this context, the purpose of this work is to evaluate the impact of using the perinodular zone on the characterization of lung lesions. Motivated by reproducible research, we used a large public dataset of solid lung nodule images and extracted fine-tuned radiomic attributes from the perinodular and intranodular zones. Our best-evaluated model obtained an average AUC of 0.916, an accuracy of 84.26%, a sensitivity of 84.45%, and specificity of 83.84%. The combination of attributes from the perinodular and intranodular zones in the image characterization resulted in an improvement in all the metrics analyzed when compared to intranodular-only characterization. Therefore, our results highlighted the importance of using the perinodular zone in the solid pulmonary nodules classification process.

Not all biases are bad: equitable and inequitable biases in machine learning and radiology

The application of machine learning (ML) technologies in medicine generally but also in radiology more specifically is hoped to improve clinical processes and the provision of healthcare. A central motivation in this regard is to advance patient treatment by reducing human error and increasing the accuracy of prognosis, diagnosis and therapy decisions. There is, however, also increasing awareness about bias in ML technologies and its potentially harmful consequences. Biases refer to systematic distortions of datasets, algorithms, or human decision making. These systematic distortions are understood to have negative effects on the quality of an outcome in terms of accuracy, fairness, or transparency. But biases are not only a technical problem that requires a technical solution. Because they often also have a social dimension, the 'distorted' outcomes they yield often have implications for equity. This paper assesses different types of biases that can emerge within applications of ML in radiology, and discusses in what cases such biases are problematic. Drawing upon theories of equity in healthcare, we argue that while some biases are harmful and should be acted upon, others might be unproblematic and even desirable-exactly because they can contribute to overcome inequities.

Radiologists and Clinical Trials: Part 1 The Truth About Reader Disagreements

The debate over human visual perception and how medical images should be interpreted have persisted since X-rays were the only imaging technique available. Concerns over rates of disagreement between expert image readers are associated with much of the clinical research and at times driven by the belief that any image endpoint variability is problematic. The deeper understanding of the reasons, value, and risk of disagreement are somewhat siloed, leading, at times, to costly and risky approaches, especially in clinical trials. Although artificial intelligence promises some relief from mistakes, its routine application for assessing tumors within cancer trials is still an aspiration. Our consortium of international experts in medical imaging for drug development research, the Pharma Imaging Network for Therapeutics and Diagnostics (PINTAD), tapped the collective knowledge of its members to ground expectations, summarize common reasons for reader discordance, identify what factors can be controlled and which actions are likely to be effective in reducing discordance. Reinforced by an exhaustive literature review, our work defines the forces that shape reader variability. This review article aims to produce a singular authoritative resource outlining reader performance's practical realities within cancer trials, whether they occur within a clinical or an independent central review.

Towards better metainterpretation: improving the clinician's interpretation of the radiology report

How the clinician interprets the radiology report has a major impact on the patient's care. It is a crucial cognitive task, and can also be a significant source of error. Because the clinician must secondarily interpret the radiologist's interpretation of the images, this step can be referred to as a "metainterpretation". Some considerations for that task are offered from the perspective of a radiologist. A revival of the tradition of discussing cases with the radiologist is encouraged.

Communicating traumatic pathology to ensure shared understanding: is there a recipe for the perfect preliminary image evaluation?

Medical imaging and emergency departments work collaboratively to interpret trauma radiographs. In addition to accurate radiographic interpretation, clear communication is crucial to ensure appropriate and timely management of musculoskeletal injuries. This two-step 'how to guide' provides the reviewer with a recipe for effectively evaluating trauma radiographs for traumatic pathology and succinctly documenting the findings. Step 1 is a systematic search of the radiograph: soft tissues, bones, alignment of joints and satisfaction of search (SBASS). Utilising SBASS increases reviewer confidence in identifying traumatic pathology of the appendicular and axial skeleton. Step 2 is a streamlined communication model for the documentation of pathological findings. The WWW acronym (What is it? Where is it? What is it doing?) can be adapted to describe simple or complex traumatic pathology.

Sources of Error in Interventional Radiology: How, Why, and When

This article aims to provide an overview of the sources for error in interventional radiology (IR). Being both a procedure and an imaging-based specialty, IR has unique considerations as to how error can occur. However, compared to the surgical and medical literature, data on error in IR are lacking. The available IR literature is reviewed but supplemented with lessons from other specialties and the World Health Organization. Individual risks such as cognitive bias as well as system-level factors are also considered in order to generate a taxonomy for error in IR that includes the operator, patient, team, and environment.

A Review of Perceptual Expertise in Radiology-How it develops, How we can test it, and Why humans still matter in the era of Artificial Intelligence

As the first step in image interpretation is detection, an error in perception can prematurely end the diagnostic process leading to missed diagnoses. Because perceptual errors of this sort-"failure to detect"-are the most common interpretive error (and cause of litigation) in radiology, understanding the nature of perceptual expertise is essential in decreasing radiology's long-standing error rates. In this article, we review what constitutes a perceptual error, the existing models of radiologic image perception, the development of perceptual expertise and how it can be tested, perceptual learning methods in training radiologists, and why understanding perceptual expertise is still relevant in the era of artificial intelligence. Adding targeted interventions, such as perceptual learning, to existing teaching practices, has the potential to enhance expertise and reduce medical error.

Review of learning opportunity rates: correlation with radiologist assignment, patient type and exam priority

BACKGROUND

Common cause analysis of learning opportunities identified in a peer collaborative improvement process can gauge the potential risk to patients and opportunities to improve.

OBJECTIVE

To study rates of learning opportunities based on radiologist assignment, patient type and exam priority at an academic children's hospital with 24/7 in-house attending coverage.

MATERIALS AND METHODS

Actively submitted peer collaborative improvement learning opportunities from July 2, 2016, to July 31, 2018, were identified. Learning opportunity rates (number of learning opportunities divided by number of exams in each category) were calculated based on the following variables: radiologist assignment at the time of dictation (daytime weekday, daytime weekend and holiday, evening, and night) patient type (inpatient, outpatient or emergency center) and exam priority (stat, urgent or routine). A statistical analysis of rate differences was performed using a chi-square test. Pairwise comparisons were made with Bonferroni method adjusted P-values.

RESULTS

There were 1,370 learning opportunities submitted on 559,584 studies (overall rate: 0.25%). The differences in rates by assignment were statistically significant (P<0.0001), with the highest rates on exams dictated in the evenings (0.31%) and lowest on those on nights (0.19%). Weekend and holiday daytime (0.26%) and weekday daytime (0.24%) rates fell in between. There were significantly higher rates on inpatients (0.33%) than on outpatients (0.22%, P<0.0001) or emergency center patients (0.16%, P<0.0001). There were no significant differences based on exam priority (stat 0.24%, urgent 0.26% and routine 0.24%, P=0.55).

CONCLUSION

In this study, the highest learning opportunity rates occurred on the evening rotation and inpatient studies, which could indicate an increased risk for patient harm and potential opportunities for improvement.

Analysis of Perceptual Expertise in Radiology - Current Knowledge and a New Perspective

Radiologists rely principally on visual inspection to detect, describe, and classify findings in medical images. As most interpretive errors in radiology are perceptual in nature, understanding the path to radiologic expertise during image analysis is essential to educate future generations of radiologists. We review the perceptual tasks and challenges in radiologic diagnosis, discuss models of radiologic image perception, consider the application of perceptual learning methods in medical training, and suggest a new approach to understanding perceptional expertise. Specific principled enhancements to educational practices in radiology promise to deepen perceptual expertise among radiologists with the goal of improving training and reducing medical error.

Physician centred imaging interpretation is dying out - why should I be a nuclear medicine physician?

Radiomics, machine learning, and, more generally, artificial intelligence (AI) provide unique tools to improve the performances of nuclear medicine in all aspects. They may help rationalise the operational organisation of imaging departments, optimise resource allocations, and improve image quality while decreasing radiation exposure and maintaining qualitative accuracy. There is already convincing data that show AI detection, and interpretation algorithms can perform with equal or higher diagnostic accuracy in various specific indications than experts in the field. Preliminary data strongly suggest that AI will be able to process imaging data and information well beyond what is visible to the human eye, and it will be able to integrate features to provide signatures that may further drive personalised medicine. As exciting as these prospects are, they currently remain essentially projects with a long way to go before full validation and routine clinical implementation. AI uses a language that is totally unfamiliar to nuclear medicine physicians, who have not been trained to manage the highly complex concepts that rely primarily on mathematics, computer sciences, and engineering. Nuclear medicine physicians are mostly familiar with biology, pharmacology, and physics, yet, considering the disruptive nature of AI in medicine, we need to start acquiring the knowledge that will keep us in the position of being actors and not merely witnesses of the wonders developed by other stakeholders in front of our incredulous eyes. This will allow us to remain a useful and valid interface between the image, the data, and the patients and free us to pursue other, one might say nobler tasks, such as treating, caring and communicating with our patients or conducting research and development.