Boosting automatic COVID-19 detection performance with self-supervised learning and batch knowledge ensembling.

PROBLEM: Detecting COVID-19 from chest X-ray (CXR) images has become one of the fastest and easiest methods for detecting COVID-19. However, the existing methods usually use supervised transfer learning from natural images as a pretraining process. These methods do not consider the unique features of COVID-19 and the similar features between COVID-19 and other pneumonia. AIM: In this paper, we want to design a novel high-accuracy COVID-19 detection method that uses CXR images, which can consider the unique features of COVID-19 and the similar features between COVID-19 and other pneumonia. METHODS: Our method consists of two phases. One is self-supervised learning-based pertaining; the other is batch knowledge ensembling-based fine-tuning. Self-supervised learning-based pretraining can learn distinguished representations from CXR images without manually annotated labels. On the other hand, batch knowledge ensembling-based fine-tuning can utilize category knowledge of images in a batch according to their visual feature similarities to improve detection performance. Unlike our previous implementation, we introduce batch knowledge ensembling into the fine-tuning phase, reducing the memory used in self-supervised learning and improving COVID-19 detection accuracy. RESULTS: On two public COVID-19 CXR datasets, namely, a large dataset and an unbalanced dataset, our method exhibited promising COVID-19 detection performance. Our method maintains high detection accuracy even when annotated CXR training images are reduced significantly (e.g., using only 10% of the original dataset). In addition, our method is insensitive to changes in hyperparameters. CONCLUSION: The proposed method outperforms other state-of-the-art COVID-19 detection methods in different settings. Our method can reduce the workloads of healthcare providers and radiologists.

Exploring Radiologists' Burnout in the COVID-19 Era: A Narrative Review.

Since its beginning in March 2020, the COVID-19 pandemic has claimed an exceptionally high number of victims and brought significant disruption to the personal and professional lives of millions of people worldwide. Among medical specialists, radiologists have found themselves at the forefront of the crisis due to the pivotal role of imaging in the diagnostic and interventional management of COVID-19 pneumonia and its complications. Because of the disruptive changes related to the COVID-19 outbreak, a proportion of radiologists have faced burnout to several degrees, resulting in detrimental effects on their working activities and overall wellbeing. This paper aims to provide an overview of the literature exploring the issue of radiologists' burnout in the COVID-19 era.

Trends in Academic Productivity Among Radiologists During the COVID-19 Pandemic.

PURPOSE: There is a scarcity of literature examining changes in radiologist research productivity during the COVID-19 pandemic. The current study aimed to investigate changes in academic productivity as measured by publication volume before and during the COVID-19 pandemic. METHODS: This single-center, retrospective cohort study included the publication data of 216 researchers consisting of associate professors, assistant professors, and professors of radiology. Wilcoxon's signed-rank test was used to identify changes in publication volume between the 1-year-long defined prepandemic period (publications between May 1, 2019, and April 30, 2020) and COVID-19 pandemic period (May 1, 2020, to April 30, 2021). RESULTS: There was a significantly increased mean annual volume of publications in the pandemic period (5.98, SD = 7.28) compared with the prepandemic period (4.98, SD = 5.53) (z = -2.819, P = .005). Subset analysis demonstrated a similar (17.4%) increase in publication volume for male researchers when comparing the mean annual prepandemic publications (5.10, SD = 5.79) compared with the pandemic period (5.99, SD = 7.60) (z = -2.369, P = .018). No statistically significant changes were found in similar analyses with the female subset. DISCUSSION: Significant increases in radiologist publication volume were found during the COVID-19 pandemic compared with the year before. Changes may reflect an overall increase in academic productivity in response to clinical and imaging volume ramp down.

The role of teleradiology during COVID-19 outbreak: Saudi radiologists' perspectives.

OBJECTIVES: To evaluate the role of teleradiology during the COVID-19 pandemic from Saudi radiologists' perspectives to improve the radiology quality service. METHODS: A cross-sectional study was carried out in Saudi Arabia among radiologists working at local hospitals from October to November 2021. It contains 21 questions involved demographic information; general information on teleradiology services; and the impact of teleradiology during COVID-19. One-way ANOVA was used to compare demographic groups. Chi-square test was used to compare demographic groups regarding their distribution of responses. All tests were carried out <0.05 level of significance. RESULTS: A total of 102 radiologists participated in this study (56% males, 44% females), 58.8% of them were sub-specialized in chest radiology. Regarding the general status of teleradiology, 69.6% of participants believed that teleradiology is a helpful tool for imaging interpretation. However, 44% of them were uncertain on the impact of teleradiology on patients' confidentiality. Approximately 87% of participants agreed that there is a positive contribution of teleradiology during COVID-19, which enables decreasing risk of infection and workload. There was a significant difference between professional degrees and overall participant responses (p<0.05). Academicians agreed that it enhances radiology departments' work (mean=17.78, SD=1.86). CONCLUSION: Concerns raised on complicated cases that require physical presence of patients, cannot be performed by teleradiology. Additionally, it might provide insufficient communication with other professionals to discuss images.

An Integrated Radiologic-Pathologic Understanding of COVID-19 Pneumonia.

This article reviews the radiologic and pathologic findings of the epithelial and endothelial injuries in COVID-19 pneumonia to help radiologists understand the fundamental nature of the disease. The radiologic and pathologic manifestations of COVID-19 pneumonia result from epithelial and endothelial injuries based on viral toxicity and immunopathologic effects. The pathologic features of mild and reversible COVID-19 pneumonia involve nonspecific pneumonia or an organizing pneumonia pattern, while the pathologic features of potentially fatal and irreversible COVID-19 pneumonia are characterized by diffuse alveolar damage followed by fibrosis or acute fibrinous organizing pneumonia. These pathologic responses of epithelial injuries observed in COVID-19 pneumonia are not specific to SARS-CoV-2 but rather constitute universal responses to viral pneumonia. Endothelial injury in COVID-19 pneumonia is a prominent feature compared with other types of viral pneumonia and encompasses various vascular abnormalities at different levels, including pulmonary thromboembolism, vascular engorgement, peripheral vascular reduction, a vascular tree-in-bud pattern, and lung perfusion abnormality. Chest CT with different imaging techniques (eg, CT quantification, dual-energy CT perfusion) can fully capture the various manifestations of epithelial and endothelial injuries. CT can thus aid in establishing prognosis and identifying patients at risk for deterioration.

Application of deep learning-based diagnostic systems in screening asymptomatic COVID-19 patients among oversea returnees.

INTRODUCTION: Our study aimed to investigate the performance of deep learning (DL)-based diagnostic systems in alerting against COVID-19, especially among asymptomatic individuals coming from overseas, and to analyze the features of identified asymptomatic patients in detail. METHODOLOGY: DL diagnostic systems were deployed to assist in the screening of COVID-19, including the pneumonia system and pulmonary nodules system. 1,917 overseas returnees who underwent CT examination and rRT-PCR tests were enrolled. DL pneumonia system promptly alerted clinicians to suspected COVID-19 after CT examinations while the performance was evaluated with rRT-PCR results as the reference. The radiological features of asymptomatic COVID-19 cases were described according to the Nomenclature of the Fleischner Society. RESULTS: Fifty-three cases were confirmed as COVID-19 patients by rRT-PCR tests, including 5 asymptomatic cases. DL pneumonia system correctly alerted 50 cases as suspected COVID-19 with a sensitivity of 0.9434 and specificity of 0.9592 (within 2 minutes per case); while the pulmonary nodules system alerted 2 of the 3 missed asymptomatic cases. Additionally, five asymptomatic patients presented different characteristics such as elevated creatine kinase level and prolonged prothrombin time, as well as atypical radiological features. CONCLUSIONS: DL diagnostic systems are promising complementary approaches for prompt screening of imported COVID-19 patients, even the imported asymptomatic cases. Unique clinical and radiological characteristics of asymptomatic cases might be of great value in screening as well. ADVANCES IN KNOWLEDGE: DL-based systems are practical, efficient, and reliable to assist radiologists in screening COVID-19 patients. Differential features of asymptomatic patients might be useful to clinicians in the frontline to differentiate asymptomatic cases.

Two in one: Overlapping CT findings of COVID-19 and underlying lung diseases.

Coronavirus disease 2019 (COVID-19) is associated with pneumonia and has various pulmonary manifestations on computed tomography (CT). Although COVID-19 pneumonia is usually seen as bilateral predominantly peripheral ground-glass opacities with or without consolidation, it can present with atypical radiological findings and resemble the imaging findings of other lung diseases. Diagnosis of COVID-19 pneumonia is much more challenging for both clinicians and radiologists in the presence of pre-existing lung disease. The imaging features of COVID-19 and underlying lung disease can overlap and obscure the findings of each other. Knowledge of the radiological findings of both diseases and possible complications, correct diagnosis, and multidisciplinary consensus play key roles in the appropriate management of diseases. In this pictorial review, the chest CT findings are presented of patients with underlying lung diseases and overlapping COVID-19 pneumonia and the various reasons for radiological lung abnormalities in these patients are discussed.

Spectrum of opportunistic fungal lung co-infections in COVID-19: What the radiologist needs to know.

Fungal lung co-infections associated with COVID-19 may occur in severely ill patients or those with underlying co-morbidities, and immunosuppression. The most common invasive fungal infections are caused by aspergillosis, mucormycosis, pneumocystis, cryptococcus, and candida. Radiologists integrate the clinical disease features with the CT pattern-based approach and play a crucial role in identifying these co-infections in COVID-19 to assist clinicians to make a confident diagnosis, initiate treatment and prevent complications.

Axillary Lymphadenopathy in the COVID-19 Era: What the Radiologist Needs to Know.

Axillary lymphadenopathy caused by the high immunogenicity of messenger RNA (mRNA) COVID-19 vaccines presents radiologists with new diagnostic dilemmas in differentiating vaccine-related benign reactive lymphadenopathy from that due to malignant causes. Understanding axillary anatomy and lymphatic drainage is key to radiologic evaluation of the axilla. US plays a critical role in evaluation and classification of axillary lymph nodes on the basis of their cortical and hilar morphology, which allows prediction of metastatic disease. Guidelines for evaluation and management of axillary lymphadenopathy continue to evolve as radiologists gain more experience with axillary lymphadenopathy related to COVID-19 vaccines. General guidelines recommend documenting vaccination dates and laterality and administering all vaccine doses contralateral to the site of primary malignancy whenever applicable. Guidelines also recommend against postponing imaging for urgent clinical indications or for treatment planning in patients with newly diagnosed breast cancer. Although conservative management approaches to axillary lymphadenopathy initially recommended universal short-interval imaging follow-up, updates to those approaches as well as risk-stratified approaches recommend interpreting lymphadenopathy in the context of both vaccination timing and the patient's overall risk of metastatic disease. Patients with active breast cancer in the pretreatment or peritreatment phase should be evaluated with standard imaging protocols regardless of vaccination status. Tissue sampling and multidisciplinary discussion remain useful in management of complex cases, including increasing lymphadenopathy at follow-up imaging, MRI evaluation of extent of disease, response to neoadjuvant treatment, and potentially confounding cases. An invited commentary by Weinstein is available online. ©RSNA, 2022.

Multi-population generalizability of a deep learning-based chest radiograph severity score for COVID-19.

To tune and test the generalizability of a deep learning-based model for assessment of COVID-19 lung disease severity on chest radiographs (CXRs) from different patient populations. A published convolutional Siamese neural network-based model previously trained on hospitalized patients with COVID-19 was tuned using 250 outpatient CXRs. This model produces a quantitative measure of COVID-19 lung disease severity (pulmonary x-ray severity (PXS) score). The model was evaluated on CXRs from 4 test sets, including 3 from the United States (patients hospitalized at an academic medical center (N = 154), patients hospitalized at a community hospital (N = 113), and outpatients (N = 108)) and 1 from Brazil (patients at an academic medical center emergency department (N = 303)). Radiologists from both countries independently assigned reference standard CXR severity scores, which were correlated with the PXS scores as a measure of model performance (Pearson R). The Uniform Manifold Approximation and Projection (UMAP) technique was used to visualize the neural network results. Tuning the deep learning model with outpatient data showed high model performance in 2 United States hospitalized patient datasets (R = 0.88 and R = 0.90, compared to baseline R = 0.86). Model performance was similar, though slightly lower, when tested on the United States outpatient and Brazil emergency department datasets (R = 0.86 and R = 0.85, respectively). UMAP showed that the model learned disease severity information that generalized across test sets. A deep learning model that extracts a COVID-19 severity score on CXRs showed generalizable performance across multiple populations from 2 continents, including outpatients and hospitalized patients.

Integrating patient symptoms, clinical readings, and radiologist feedback with computer-aided diagnosis system for detection of infectious pulmonary disease: a feasibility study.

Automatic computer-aided diagnosis (CAD) system has been widely used as an assisting tool for mass screening and risk assessment of infectious pulmonary diseases (PDs). However, such a system still lacks clinical acceptability and trust due to the integration gap between the patient's metadata, radiologist feedback, and the CAD system. This paper proposed three integration frameworks, namely-direct integration (DI), rule-based integration (RBI), and weight-based integration (WBI). The proposed framework helps clinicians diagnose lung inflammation and provide an end-to-end robust diagnostic system. Initially, the feasibility of integrating patients' symptoms, clinical pathologies, and radiologist feedback with CAD system to improve the classification performance is investigated. Subsequently, the patient's metadata and radiologist feedback are integrated with the CAD system using the proposed integration frameworks. The proposed method's performance is evaluated using a private dataset consisting of 70 chest X-ray (CXR) images (31 COVID-19, 14 other diseases, and 25 normal). The obtained results reveal that the proposed WBI achieved the highest classification performance (accuracy = 98.18%, F1 score = 97.73%, and Matthew's correlation coefficient = 0.969) compared to DI and RI. The generalization capability of the proposed framework is also verified from an external validation set. Furthermore, the Friedman average ranking and Shaffer and Holm post hoc statistical methods reveal the obtained results' statistical significance. Methodological diagram of proposed integration frameworks.

Mandating Limits on Workload, Duty, and Speed in Radiology.

Research has not yet quantified the effects of workload or duty hours on the accuracy of radiologists. With the exception of a brief reduction in imaging studies during the 2020 peak of the COVID-19 pandemic, the workload of radiologists in the United States has seen relentless growth in recent years. One concern is that this increased demand could lead to reduced accuracy. Behavioral studies in species ranging from insects to humans have shown that decision speed is inversely correlated to decision accuracy. A potential solution is to institute workload and duty limits to optimize radiologist performance and patient safety. The concern, however, is that any prescribed mandated limits would be arbitrary and thus no more advantageous than allowing radiologists to self-regulate. Specific studies have been proposed to determine whether limits reduce error, and if so, to provide a principled basis for such limits. This could determine the precise susceptibility of individual radiologists to medical error as a function of speed during image viewing, the maximum number of studies that could be read during a work shift, and the appropriate shift duration as a function of time of day. Before principled recommendations for restrictions are made, however, it is important to understand how radiologists function both optimally and at the margins of adequate performance. This study examines the relationship between interpretation speed and error rates in radiology, the potential influence of artificial intelligence on reading speed and error rates, and the possible outcomes of imposed limits on both caseload and duty hours. This review concludes that the scientific evidence needed to make meaningful rules is lacking and notes that regulating workloads without scientific principles can be more harmful than not regulating at all.

Virtual conferences: results of an international survey on radiologist preferences and perspectives.

BACKGROUND: We explored perceptions and preferences regarding the conversion of in-person to virtual conferences as necessitated by travel and in-person meeting restrictions. METHODS: A 16-question online survey to assess preferences regarding virtual conferences during the COVID-19 pandemic and future perspectives on this subject was disseminated internationally online between June and August 2020. FINDINGS: A total of 508 responses were received from 73 countries. The largest number of responses came from Italy and the USA. The majority of respondents had already attended a virtual conference (80%) and would like to attend future virtual meetings (97%). The ideal duration of such an event was 2-3 days (42%). The preferred time format was a 2-4-h session (43%). Most respondents also noted that they would like a significant fee reduction and the possibility to attend a conference partly in-person and partly online. Respondents indicated educational sessions as the most valuable sections of virtual meetings. The reported positive factor of the virtual meeting format is the ability to re-watch lectures on demand. On the other hand, the absence of networking and human contact was recognized as a significant loss. In the future, people expressed a preference to attend conferences in person for networking purposes, but only in safer conditions. CONCLUSIONS: Respondents appreciated the opportunity to attend the main radiological congresses online and found it a good opportunity to stay updated without having to travel. However, in general, they would prefer these conferences to be structured differently. The lack of networking opportunities was the main reason for preferring an in-person meeting. KEY POINTS: • Respondents appreciated the opportunity to attend the main radiological meetings online, considering it a good opportunity to stay updated without having to travel. • In the future, it is likely for congresses to offer attendance options both in person and online, making them more accessible to a larger audience. • Respondents indicated that networking represents the most valuable advantage of in-person conferences compared to online ones.

COVID-19 pneumonia chest radiographic severity score: variability assessment among experienced and in-training radiologists and creation of a multireader composite score database for artificial intelligence algorithm development.

OBJECTIVE: The purpose was to evaluate reader variability between experienced and in-training radiologists of COVID-19 pneumonia severity on chest radiograph (CXR), and to create a multireader database suitable for AI development. METHODS: In this study, CXRs from polymerase chain reaction positive COVID-19 patients were reviewed. Six experienced cardiothoracic radiologists and two residents classified each CXR according to severity. One radiologist performed the classification twice to assess intraobserver variability. Severity classification was assessed using a 4-class system: normal (0), mild (1), moderate (2), and severe (3). A median severity score (Rad Med) for each CXR was determined for the six radiologists for development of a multireader database (XCOMS). Kendal Tau correlation and percentage of disagreement were calculated to assess variability. RESULTS: A total of 397 patients (1208 CXRs) were included (mean age, 60 years SD ± 1), 189 men). Interobserver variability between the radiologists ranges between 0.67 and 0.78. Compared to the Rad Med score, the radiologists show good correlation between 0.79-0.88. Residents show slightly lower interobserver agreement of 0.66 with each other and between 0.69 and 0.71 with experienced radiologists. Intraobserver agreement was high with a correlation coefficient of 0.77. In 220 (18%), 707 (59%), 259 (21%) and 22 (2%) CXRs there was a 0, 1, 2 or 3 class-difference. In 594 (50%) CXRs the median scores of the residents and the radiologists were similar, in 578 (48%) and 36 (3%) CXRs there was a 1 and 2 class-difference. CONCLUSION: Experienced and in-training radiologists demonstrate good inter- and intraobserver agreement in COVID-19 pneumonia severity classification. A higher percentage of disagreement was observed in moderate cases, which may affect training of AI algorithms. ADVANCES IN KNOWLEDGE: Most AI algorithms are trained on data labeled by a single expert. This study shows that for COVID-19 X-ray severity classification there is significant variability and disagreement between radiologist and between residents.

Clinical importance of second-opinion interpretations of abdominal imaging studies in a cancer hospital and its impact on patient management.

PURPOSE: The purpose of this retrospective study was to evaluate the quality of outside hospital imaging and associated reports submitted to us for reinterpretation related to clinical care at our tertiary cancer center. We compared the initial study interpretations to that of interpretations performed by subspecialty-trained abdominal radiologists at our center and whether this resulted in a change in inpatient treatment. MATERIALS AND METHODS: We performed an institutional review board-approved retrospective single-institution study of 915 consecutive outside computed tomography (CT) and magnetic resonance (MR) abdominal imaging studies that had been submitted to our institution between August 1, 2020 and November 30, 2020. The assessed parameters included the quality and accuracy of the report, the technical quality of the imaging compared to that at our institution, the appropriateness of the imaging for staging or restaging, usage of oral and IV contrast, and CT slice thickness. Clinical notes, pathologic findings, and subsequent imaging were used to establish an accurate diagnosis and determine the effect on clinical treatment. Discrepancies between the initial and secondary interpretations were identified independently by a panel of radiologists to assess changes in treatment. The impact of discrepancies on treatment was evaluated based on current treatment guidelines. RESULTS: Of 744 CT (81%) and 171 MR (19%) outside imaging studies, 65% had suboptimal quality compared to the images at our institution, and 31% were inappropriate for oncological care purposes. Only 21% of CT studies had optimal slice thickness of <3 mm. Of 375 (41%) outside reports, 131 (34%) had discrepancies between secondary and initial interpretations. Of the 88 confirmed discrepant studies, 42 patients (48%) had a change in treatment based on the secondary interpretation. CONCLUSIONS: Imaging studies from outside institutions have variable image quality and are often inadequate for oncologic imaging. The secondary interpretations by subspecialty-trained radiologists resulted in treatment change.

Physicians' preferences and willingness to pay for artificial intelligence-based assistance tools: a discrete choice experiment among german radiologists.

BACKGROUND: Artificial Intelligence (AI)-based assistance tools have the potential to improve the quality of healthcare when adopted by providers. This work attempts to elicit preferences and willingness to pay for these tools among German radiologists. The goal was to generate insights for tool providers and policymakers regarding the development and funding of ideally designed and priced tools. Ultimately, healthcare systems can only benefit from quality enhancing AI when provider adoption is considered. METHODS: Since there is no established market for AI-based assistance tools in radiology yet, a discrete choice experiment was conducted. Respondents from the two major German professional radiology associations chose between hypothetical tools composed of five attributes and a no-choice option. The attributes included: provider, application, quality impact, time savings and price. A conditional logit model was estimated identifying preferences for attribute levels, the no-choice option, and significant subject-related interaction effects. RESULTS: 114 respondents were included for analysis of which 46% were already using an AI-based assistance tool. Average adoption probability for an AI-based tool was 81% (95% CI 77.1% - 84.4%). Radiologists preferred a tool that assists in routine diagnostics performing at above-radiologist-level quality and saves 50% in diagnostics time at a price-point of €3 per study. The provider is not a significant factor in the decisions. Time savings were considered more important than quality improvements (i.e., detecting more anomalies). CONCLUSIONS: Radiologists are overall willing to invest in AI-based assistance tools. Development, funding, and research regarding these tools should, however, consider providers' preferences for features of immediate everyday and economic relevance like time savings to optimize adoption.