Search

BackgroundThe COVID-19 outbreak has spread rapidly and hospitals are overwhelmed with COVID-19 patients. While analysis of nasal and throat swabs from patients is the main way to detect COVID-19, analyzing chest images could offer an alternative method to hospitals, where health care personnel and testing kits are scarce. Deep learning (DL), in particular, has shown impressive levels of performance when analyzing medical images, including those related to COVID-19 pneumonia. ObjectiveThe goal of this study was to perform a systematic review with a meta-analysis of relevant studies to quantify the performance of DL algorithms in the automatic stratification of COVID-19 patients using chest images. MethodsA search strategy for use in PubMed, Scopus, Google Scholar, and Web of Science was developed, where we searched for articles published between January 1 and April 25, 2020. We used the key terms “COVID-19,” or “coronavirus,” or “SARS-CoV-2,” or “novel corona,” or “2019-ncov,” and “deep learning,” or “artificial intelligence,” or “automatic detection.” Two authors independently extracted data on study characteristics, methods, risk of bias, and outcomes. Any disagreement between them was resolved by consensus. ResultsA total of 16 studies were included in the meta-analysis, which included 5896 chest images from COVID-19 patients. The pooled sensitivity and specificity of the DL models in detecting COVID-19 were 0.95 (95% CI 0.94-0.95) and 0.96 (95% CI 0.96-0.97), respectively, with an area under the receiver operating characteristic curve of 0.98. The positive likelihood, negative likelihood, and diagnostic odds ratio were 19.02 (95% CI 12.83-28.19), 0.06 (95% CI 0.04-0.10), and 368.07 (95% CI 162.30-834.75), respectively. The pooled sensitivity and specificity for distinguishing other types of pneumonia from COVID-19 were 0.93 (95% CI 0.92-0.94) and 0.95 (95% CI 0.94-0.95), respectively. The performance of radiologists in detecting COVID-19 was lower than that of the DL models; however, the performance of junior radiologists was improved when they used DL-based prediction tools. ConclusionsOur study findings show that DL models have immense potential in accurately stratifying COVID-19 patients and in correctly differentiating them from patients with other types of pneumonia and normal patients. Implementation of DL-based tools can assist radiologists in correctly and quickly detecting COVID-19 and, consequently, in combating the COVID-19 pandemic.

IntroductionThe amount of information being uploaded onto social video platforms, such as YouTube, Vimeo, and Veoh, continues to spiral, making it increasingly difficult to discern reliable health information from misleading content. There are thousands of YouTube videos promoting misleading information about anorexia (eg, anorexia as a healthy lifestyle). ObjectiveThe aim of this study was to investigate anorexia-related misinformation disseminated through YouTube videos. MethodsWe retrieved YouTube videos related to anorexia using the keywords anorexia, anorexia nervosa, proana, and thinspo on October 10, 2011.Three doctors reviewed 140 videos with approximately 11 hours of video content, classifying them as informative, pro-anorexia, or others. By informative we mean content describing the health consequences of anorexia and advice on how to recover from it; by pro-anorexia we mean videos promoting anorexia as a fashion, a source of beauty, and that share tips and methods for becoming and remaining anorexic. The 40 most-viewed videos (20 informative and 20 pro-anorexia videos) were assessed to gauge viewer behavior. ResultsThe interrater agreement of classification was moderate (Fleiss’ kappa=0.5), with 29.3% (n=41) being rated as pro-anorexia, 55.7% (n=78) as informative, and 15.0% (n=21) as others. Pro-anorexia videos were favored 3 times more than informative videos (odds ratio [OR] 3.3, 95% CI 3.3-3.4, P

Aims: Sex differences in long‐term post‐discharge clinical outcomes in Asian patients hospitalized for acute decompensated heart failure (HF) persist despite the world‐wide implementation of guideline‐directed medical therapy for decades. The present study aims to elucidate the puzzling dilemma and to depict the directions of solution. Methods and results: Between 2011 and 2020, a total of 12 428 patients (6518 men and 5910 women, mean age 73.50 ± 14.85) hospitalized for acute decompensated HF were retrospectively enrolled from a university HF cohort. Compared with men, women hospitalized for acute decompensated HF were older in age (76.40 ± 13.43 vs. 71.20 ± 15.67 years old, P < 0.0001) with more coexisting hypertension, diabetes, hyperlipidaemia and moderate to severe chronic kidney disease, but less with ischaemic heart disease, cerebrovascular disease and chronic obstructive pulmonary disease (P < 0.0001). In echocardiography measurement parameters, women had smaller left ventricular and left atrial dimensions, higher left ventricular mass index, higher left ventricular ejection fraction (LVEF) and more in HF with preserved ejection fraction (EF) category (LVEF > 50%) than men (P < 0.0001). In HF therapy, women compared with men received more guideline‐directed medical HF therapies including angiotensin‐converting enzyme inhibitors, angiotensin receptor blockers, angiotensin receptor–neprilysin inhibitors and sodium–glucose cotransporter‐2 inhibitors, but similar beta‐blockers and mineralocorticoid receptor antagonists (P < 0.0001). Post‐discharge long‐term clinical outcomes after multivariate‐adjusted analysis revealed that women compared with men had lower all‐cause mortality [adjusted hazard ratio (aHR): 0.89, 95% confidence interval (CI): 0.84–0.93], lower cardiovascular mortality (aHR: 0.89, 95% CI: 0.80–0.99) and lower 1 year mortality (aHR: 0.91, 95% CI: 0.84–0.99) but similar HF rehospitalization rate (aHR: 1.02, 95% CI: 0.95–1.09) over 8 years of follow‐up. The superiority of women over men in all‐cause mortality was shown in HF with preserved EF (>50%) and HF with mildly reduced EF (40%–50%), but not in HF with reduced EF (<40%) category. Subgroup forest plot analysis showed body mass index, coexisting hypertension and chronic obstructive pulmonary disease as significant interacting factors. Conclusions: With more coronary risk factors and medical comorbidities, less cardiac remodelling and better adherence to guideline‐directed HF therapy, women hospitalized for acute decompensated HF demonstrated superiority over men in long‐term post‐discharge clinical outcomes, including all‐cause mortality, cardiovascular mortality and 1 year mortality, and mainly in HF with preserved and mid‐range EF categories, in the Asian HF cohort. [ABSTRACT FROM AUTHOR]

Recent studies propose fallopian tubes as the tissue origin for many ovarian epithelial cancers. To further support this paradigm, we assessed whether salpingectomy for treating ectopic pregnancy had a protective effect using the Taiwan Longitudinal National Health Research Database. We identified 316 882 women with surgical treatment for ectopic pregnancy and 3 168 820 age- and index-date-matched controls from 2000 to 2016. In a nested cohort, 91.5% of cases underwent unilateral salpingectomy, suggesting that most surgically managed patients have salpingectomy. Over a follow-up period of 17 years, the ovarian carcinoma incidence was 0.0069 (95% confidence interval [CI] = 0.0060 to 0.0079) and 0.0089 (95% CI = 0.0086 to 0.0092) in the ectopic pregnancy and the control groups, respectively (P  < .001). After adjusting the events to per 100 person-years, the hazard ratio (HR) in the ectopic pregnancy group was 0.70 (95% CI = 0.61 to 0.80). The risk reduction occurred only in epithelial ovarian cancer (HR = 0.73, 95% CI = 0.63 to 0.86) and not in non-epithelial subtypes. These findings show a decrease in ovarian carcinoma incidence after salpingectomy for treating ectopic pregnancy. [ABSTRACT FROM AUTHOR]

Backgrounds: Influenza vaccination could decrease the risk of major cardiac events in patients with hypertension. However, the vaccine's effects on decreasing the risk of chronic kidney disease (CKD) development in such patients remain unclear. Methods: We retrospectively analysed the data of 37,117 patients with hypertension (≥55 years old) from the National Health Insurance Research Database during 1 January 2001 to 31 December 2012. After a 1:1 propensity score matching by the year of diagnosis, we divided the patients into vaccinated (n = 15,961) and unvaccinated groups (n = 21,156). Results: In vaccinated group, significantly higher prevalence of comorbidities such as diabetes, cerebrovascular disease, dyslipidemia, heart and liver disease were observed compared with unvaccinated group. After adjusting age, sex, comorbidities, medications (anti-hypertensive agents, metformin, aspirin and statin), level of urbanization and monthly incomes, significantly lower risk of CKD occurrence was observed among vaccinated patients in influenza season, non-influenza season and all season (Adjusted hazard ratio [aHR]: 0.39, 95% confidence level [C.I.]: 0.33–0.46; 0.38, 95% C.I.: 0.31–0.45; 0.38, 95% C.I.: 0.34–0.44, respectively). The risk of hemodialysis significantly decreased after vaccination (aHR: 0.40, 95% C.I.: 0.30–0.53; 0.42, 95% C.I.: 0.31–0.57; 0.41, 95% C.I.: 0.33–0.51, during influenza season, non-influenza season and all season). In sensitivity analysis, patients with different sex, elder and non-elder age, with or without comorbidities and with or without medications had significant decreased risk of CKD occurrence and underwent hemodialysis after vaccination. Moreover, the potential protective effect appeared to be dose-dependent. Conclusions: Influenza vaccination decreases the risk of CKD among patients with hypertension and also decrease the risk of receiving renal replacement therapy. Its potential protective effects are dose-dependent and persist during both influenza and noninfluenza seasons. [ABSTRACT FROM AUTHOR]