ObjectivesLung ultrasound (LUS) is a portable, low-cost respiratory imaging tool but is challenged by user dependence and lack of diagnostic specificity. It is unknown whether the advantages of LUS implementation could be paired with deep learning (DL) techniques to match or exceed human-level, diagnostic specificity among similar appearing, pathological LUS images.DesignA convolutional neural network (CNN) was trained on LUS images with B lines of different aetiologies. CNN diagnostic performance, as validated using a 10% data holdback set, was compared with surveyed LUS-competent physicians.SettingTwo tertiary Canadian hospitals.Participants612 LUS videos (121 381 frames) of B lines from 243 distinct patients with either (1) COVID-19 (COVID), non-COVID acute respiratory distress syndrome (NCOVID) or (3) hydrostatic pulmonary edema (HPE).ResultsThe trained CNN performance on the independent dataset showed an ability to discriminate between COVID (area under the receiver operating characteristic curve (AUC) 1.0), NCOVID (AUC 0.934) and HPE (AUC 1.0) pathologies. This was significantly better than physician ability (AUCs of 0.697, 0.704, 0.967 for the COVID, NCOVID and HPE classes, respectively), p<0.01.ConclusionsA DL model can distinguish similar appearing LUS pathology, including COVID-19, that cannot be distinguished by humans. The performance gap between humans and the model suggests that subvisible biomarkers within ultrasound images could exist and multicentre research is merited.

ObjectivesLung ultrasound (LUS) is a portable, low cost respiratory imaging tool but is challenged by user dependence and lack of diagnostic specificity. It is unknown whether the advantages of LUS implementation could be paired with deep learning techniques to match or exceed human-level, diagnostic specificity among similar appearing, pathological LUS images.DesignA convolutional neural network was trained on LUS images with B lines of different etiologies. CNN diagnostic performance, as validated using a 10% data holdback set was compared to surveyed LUS-competent physicians.SettingTwo tertiary Canadian hospitals.Participants600 LUS videos (121,381 frames) of B lines from 243 distinct patients with either 1) COVID-19, Non-COVID acute respiratory distress syndrome (NCOVID) and 3) Hydrostatic pulmonary edema (HPE).ResultsThe trained CNN performance on the independent dataset showed an ability to discriminate between COVID (AUC 1.0), NCOVID (AUC 0.934) and HPE (AUC 1.0) pathologies. This was significantly better than physician ability (AUCs of 0.697, 0.704, 0.967 for the COVID, NCOVID and HPE classes, respectively), p < 0.01.ConclusionsA deep learning model can distinguish similar appearing LUS pathology, including COVID-19, that cannot be distinguished by humans. The performance gap between humans and the model suggests that subvisible biomarkers within ultrasound images could exist and multi-center research is merited.

IntroductionBleeding during cardiac surgery is associated with increased morbidity and mortality. Tranexamic acid is an antifibrinolytic with proven efficacy in major surgeries. Current clinical practice guidelines recommend intraoperative use in cardiac procedures. However, several complications have been reported with tranexamic acid including seizures. This review intends to summarise the evidence examining the efficacy and safety of tranexamic acid in patients undergoing cardiac surgery.Methods/designWe will search MEDLINE, Embase, PubMED, ACPJC, CINAHL and the Cochrane trial registry for eligible randomised controlled trials, the search dates for all databases will be from inception until 1 January 2019, investigating the perioperative use of topical and/or intravenous tranexamic acid as a stand-alone antifibrinolytic agent compared with placebo in patients undergoing open cardiac surgery. We categorised outcomes as patient critical or patient important. Selected patient-critical outcomes are: mortality (intensive care unit, hospital and 30-day endpoints), reoperation within 24 hours, postoperative bleeding requiring transfusion of packed red blood cells, myocardial infarction, stroke, pulmonary embolism, bowel infarction, upper or lower limb deep vein thrombosis and seizures. Those outcomes, we perceived as clinical experts to be most patient valued and patients were not involved in outcomes selection process. We will not apply publication date, language, journal or methodological quality restrictions. Two reviewers will independently screen and identify eligible studies using predefined eligibility criteria and then review full reports of all potentially relevant citations. A third reviewer will resolve disagreements if consensus cannot be achieved. We will present the results as relative risk with 95% CIs for dichotomous outcomes and as mean difference or standardised mean difference for continuous outcomes with 95% CIs. We will assess the quality of evidence using the Grading of Recommendations, Assessment, Development and Evaluation approach.Ethics and disseminationFormal ethical approval is not required as primary data will not be collected. The results will be disseminated through a peer-reviewed publicationTrial registration numberCRD42018105904

No abstract