IMPORTANCEMost early lung cancers present as pulmonary nodules on imaging, but these can be easily missed on chest radiographs. OBJECTIVE To assess if a novel artificial intelligence (AI) algorithm can help detect pulmonary nodules on radiographs at different levels of detection difficulty. DESIGN, SETTING, AND PARTICIPANTSThis diagnostic study included 100 posteroanterior chest radiograph images taken between 2000 and 2010 of adult patients from an ambulatory health care center in Germany and a lung image database in the US. Included images were selected to represent nodules with different levels of detection difficulties (from easy to difficult), and comprised both normal and nonnormal control.EXPOSURES All images were processed with a novel AI algorithm, the AI Rad Companion Chest X-ray. Two thoracic radiologists established the ground truth and 9 test radiologists from Germany and the US independently reviewed all images in 2 sessions (unaided and AI-aided mode) with at least a 1-month washout period. MAIN OUTCOMES AND MEASURESEach test radiologist recorded the presence of 5 findings (pulmonary nodules, atelectasis, consolidation, pneumothorax, and pleural effusion) and their level of confidence for detecting the individual finding on a scale of 1 to 10 (1 representing lowest confidence; 10, highest confidence). The analyzed metrics for nodules included sensitivity, specificity, accuracy, and receiver operating characteristics curve area under the curve (AUC). RESULTSImages from 100 patients were included, with a mean (SD) age of 55 (20) years and including 64 men and 36 women. Mean detection accuracy across the 9 radiologists improved by 6.4% (95% CI, 2.3% to 10.6%) with AI-aided interpretation compared with unaided interpretation. Partial AUCs within the effective interval range of 0 to 0.2 false positive rate improved by 5.6% (95% CI, −1.4% to 12.0%) with AI-aided interpretation. Junior radiologists saw greater improvement in sensitivity for nodule detection with AI-aided interpretation as compared with their senior counterparts (12%; 95% CI, 4% to 19% vs 9%; 95% CI, 1% to 17%) while senior radiologists experienced similar improvement in specificity (4%; 95% CI, −2% to 9%) as compared with junior radiologists (4%; 95% CI, −3% to 5%). CONCLUSIONS AND RELEVANCEIn this diagnostic study, an AI algorithm was associated with improved detection of pulmonary nodules on chest radiographs compared with unaided interpretation for different levels of detection difficulty and for readers with different experience.
BackgroundSocial networking sites, in particular Facebook, are not only predominant in students’ social life but are to varying degrees interwoven with the medical curriculum. Particularly, Facebook groups have been identified for their potential in higher education. However, there is a paucity of data on user types, content, and dynamics of study-related Facebook groups.ObjectiveThe aim of this study was to identify the role of study-related Facebook group use, characterize medical students that use or avoid using Facebook groups (demographics, participation pattern, and motivation), and analyze student posting behavior, covered topics, dynamics, and limitations in Facebook groups with regards to educational usage.MethodsUsing a multi-method approach (interviews, focus groups, and qualitative and quantitative analysis of Facebook posts), we analyzed two representative Facebook groups of medical preclinical semesters at Ludwig-Maximilians-University (LMU) Munich. Facebook primary posts and replies over one semester were extracted and evaluated by using thematic content analysis. We developed and applied a coding scheme for studying the frequency and distribution of these posts. Additionally, we interviewed students with various degrees of involvement in the groups, as well as “new minorities,” students not registered on Facebook.ResultsFacebook groups seem to have evolved as the main tool for medical students at LMU to complement the curriculum and to discuss study-related content. These Facebook groups are self-organizing and quickly adapt to organizational or subject-related challenges posed by the curriculum. A wide range of topics is covered, with a dominance of organization-related posts (58.35% [6916/11,853] of overall posts). By measuring reply rates and comments per category, we were able to identify learning tips and strategies, material sharing, and course content discussions as the most relevant categories. Rates of adequate replies in these categories ranged between 78% (11/14) and 100% (13/13), and the number of comments per post ranged from 8.4 to 13.7 compared with the average overall reply rate of 68.69% (1167/1699) and 3.9 comments per post. User typology revealed social media drivers (>30 posts per semester) as engines of group function, frequent users (11-30 posts), and a majority of average users acting rather as consumers or lurkers (1-10 posts).ConclusionsFor the moment, the medical faculty has no active involvement in these groups and therefore no influence on accuracy of information, professionalism, and ethical issues. Nevertheless, faculty could in the future benefit by extracting relevant information, identifying common problems, and understanding semester-related dynamics.
In patients with atrial fibrillation, oral anticoagulation with oral thrombin inhibitors (OTIs), in contrast to vitamin K antagonists (VKAs), associates with a modest increase in acute coronary syndromes (ACSs). Whether this observation is causatively linked to OTI treatment and, if so, whether OTI action is the result of a lower antithrombotic efficacy of OTI compared to VKA or reflects a yet undefined prothrombotic activity of OTI remain unclear. We analyzed platelet function in patients receiving OTI or dose-adapted VKA under static and flow conditions. In vivo, we studied arterial thrombosis in OTI-, VKA-, and vehicle-treated mice using carotid ligation and wire injury models. Further, we examined thrombus formation on human atherosclerotic plaque homogenates under arterial shear to address the relevance to human pathology. Under static conditions, aggregation in the presence of ristocetin was increased in OTI-treated blood, whereas platelet reactivity and aggregation to other agonists were only marginally affected. Under flow conditions, firm platelet adhesion and thrombus formation on von Willebrand factor, collagen, and human atherosclerotic plaque were increased in the presence of OTI in comparison to VKA. OTI treatment was associated with increased thrombus formation in injured carotid arteries of mice. Inhibition or ablation of GPIbα-thrombin interactions abolished the effect of OTI on thrombus formation, suggesting a mechanistic role of the platelet receptor GPIbα and its thrombin-binding site. The effect of OTI was also abrogated in the presence of aspirin. In summary, OTI treatment has prothrombotic activity that might contribute to the increase in ACS observed clinically in patients.
Objectives We hypothesized that published performances of algorithms for artificial intelligence (AI) pneumothorax (PTX) detection in chest radiographs (CXRs) do not sufficiently consider the influence of PTX size and confounding effects caused by thoracic tubes (TTs). Therefore, we established a radiologically annotated benchmarking cohort (n = 6446) allowing for a detailed subgroup analysis. Materials and Methods We retrospectively identified 6434 supine CXRs, among them 1652 PTX-positive cases and 4782 PTX-negative cases. Supine CXRs were radiologically annotated for PTX size, PTX location, and inserted TTs. The diagnostic performances of 2 AI algorithms (“AI_CheXNet” [Rajpurkar et al], “AI_1.5” [Guendel et al]), both trained on publicly available datasets with labels obtained from automatic report interpretation, were quantified. The algorithms' discriminative power for PTX detection was quantified by the area under the receiver operating characteristics (AUROC), and significance analysis was based on the corresponding 95% confidence interval. A detailed subgroup analysis was performed to quantify the influence of PTX size and the confounding effects caused by inserted TTs. Results Algorithm performance was quantified as follows: overall performance with AUROCs of 0.704 (AI_1.5) / 0.765 (AI_CheXNet) for unilateral PTXs, AUROCs of 0.666 (AI_1.5) / 0.722 (AI_CheXNet) for unilateral PTXs smaller than 1 cm, and AUROCs of 0.735 (AI_1.5) / 0.818 (AI_CheXNet) for unilateral PTXs larger than 2 cm. Subgroup analysis identified TTs to be strong confounders that significantly influence algorithm performance: Discriminative power is completely eliminated by analyzing PTX-positive cases without TTs referenced to control PTX-negative cases with inserted TTs. Contrarily, AUROCs increased up to 0.875 (AI_CheXNet) for large PTX-positive cases with inserted TTs referenced to control cases without TTs. Conclusions Our detailed subgroup analysis demonstrated that the performance of established AI algorithms for PTX detection trained on public datasets strongly depends on PTX size and is significantly biased by confounding image features, such as inserted TTS. Our established, clinically relevant and radiologically annotated benchmarking cohort might be of great benefit for ongoing algorithm development.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.