Explore the vast range of titles available.

In recent years, deep learning (DL) has been widely studied using various methods across the globe, especially with respect to training methods and network structures, proving highly effective in a wide range of tasks and applications, including image, speech, and text recognition. One important aspect of this advancement is involved in the effort of designing and upgrading neural architectures, which has been consistently attempted thus far. However, designing such architectures requires the combined knowledge and know-how of experts from each relevant discipline and a series of trial-and-error steps. In this light, automated neural architecture search (NAS) methods are increasingly at the center of attention; this paper aimed at summarizing the basic concepts of NAS while providing an overview of recent studies on the applications of NAS. It is worth noting that most previous survey studies on NAS have been focused on perspectives of hardware or search strategies. To the best knowledge of the present authors, this study is the first to look at NAS from a computer vision perspective. In the present study, computer vision areas were categorized by task, and recent trends found in each study on NAS were analyzed in detail.

Particulate matter has been increasing worldwide causing air pollution and serious health hazards. Owing to increased time spent indoors and lifestyle changes, assessing indoor air quality has become crucial. This study investigated the effect of watering and drought and illumination conditions (constant light, light/dark cycle, and constant dark) on particulate matter2.5 (PM2.5) removal and surface characterization of leaf in a botanical plant-based biofilter system. Using Ardisia japonica and Hedera helix as experimental plants in the plant-based biofilter system, PM2.5, volatile organic carbon, and CO2, as the evaluators of indoor air quality, were estimated using a sensor. Morphological and chemical changes of the leaf surface (i.e., roughness and wax) associated with PM2.5 removal were characterized via scanning electron microscopy, Fourier transform infrared spectroscopy, and atomic force microscopy. The highest PM2.5 removal efficiency, stomata closure, high leaf roughness, and wax layer were observed under drought with constant light condition. Consequently, PM2.5 removal was attributed to the combined effect of leaf roughness and wax by adsorption rather than stomatal uptake. These results suggest that operating conditions of indoor plant-based biofilter system such as watering (or drought) and illumination may be applied as a potential strategy for enhancing PM2.5 removal.

Protein assembly and biological interpretation of the assembled protein lists are critical steps in shotgun proteomics data analysis. Although most biological functions arise from interactions among proteins, current protein assembly pipelines treat proteins as independent entities. Usually, only individual proteins with strong experimental evidence, that is, confident proteins, are reported, whereas many possible proteins of biological interest are eliminated. We have developed a clique‐enrichment approach (CEA) to rescue eliminated proteins by incorporating the relationship among proteins as embedded in a protein interaction network. In several data sets tested, CEA increased protein identification by 8–23% with an estimated accuracy of 85%. Rescued proteins were supported by existing literature or transcriptome profiling studies at similar levels as confident proteins and at a significantly higher level than abandoned ones. Applying CEA on a breast cancer data set, rescued proteins coded by well‐known breast cancer genes. In addition, CEA generated a network view of the proteins and helped show the modular organization of proteins that may underpin the molecular mechanisms of the disease. We have developed a clique‐enrichment approach (CEA) to improve protein identification in shotgun proteomics by taking into consideration the functional relationship among proteins as embedded in protein interaction networks. CEA is an accurate approach that can be easily incorporated into routine shotgun proteomics protein assembly pipelines to improve protein identification. Applying CEA on a breast cancer data set rescued proteins coded by well‐known breast cancer genes. CEA creates a modular view of the identified proteins and thus facilitates the biological interpretation of shotgun proteomics data

Background Tuberculosis (TB) has been a major public health problem in South Korea. Although TB notification rate in Korea is gradually decreasing, still highest among the member countries of the Organization for Economic Cooperation and Development. To effectively control TB, understanding the TB epidemiology such as prevalence of latent tuberculosis infection (LTBI) and annual risk of TB infection (ARI) are important. This study aimed to identify the prevalence of LTBI and ARI among South Korean health care workers (HCWs) based on their interferon-gamma release assays (IGRA). Methods This was single center, cross-sectional retrospective study in a tertiary hospital in South Korea. We performed IGRA in HCWs between May 2017 and March 2018. We estimated ARI based on IGRA results. Logistic regression model was used to identify factors affecting IGRA positivity. Results A total of 3233 HCWs were analyzed. Median age of participants was 38.0 and female was predominant (72.6%). Overall positive rate of IGRA was 24.1% and IGRA positive rates age-group wise were 6.6%, 14.4%, 34.3%, and around 50% in the age groups 20s, 30s, 40s, and 50s and 60s, respectively. The ARIs was 0.26–1.35% between 1986 and 2005; rate of TB infection has gradually decreased in the last two decades. Multivariable analysis indicated that older age, healed TB lesion in x-ray, and male gender were risk factors for IGRA positivity, whereas working in high-risk TB departments was not. Conclusions Results showed that ARI in South Korean HCWs gradually decreased over two decades, although LTBI remained prevalent. Our results suggest that the LTBI test result of HCWs might be greatly affected by age, rather than occupational exposure, in intermediate TB burden countries. Thus, careful interpretation considering the age structure is required.

Background Pneumocystis jirovecii pneumonia (PCP) with acute respiratory failure can result in development of pneumothorax during treatment. This study aimed to identify the incidence and related factors of pneumothorax in patients with PCP and acute respiratory failure and to analyze their prognosis. Methods We retrospectively reviewed the occurrence of pneumothorax, including clinical characteristics and results of other examinations, in 119 non-human immunodeficiency virus patients with PCP and respiratory failure requiring mechanical ventilator treatment in a medical intensive care unit (ICU) at a tertiary-care center between July 2016 and April 2019. Results During follow up duration, twenty-two patients (18.5%) developed pneumothorax during ventilator treatment, with 45 (37.8%) eventually requiring a tracheostomy due to weaning failure. Cytomegalovirus co-infection (odds ratio 13.9; p  = 0.013) was related with occurrence of pneumothorax in multivariate analysis. And development of pneumothorax was not associated with need for tracheostomy and mortality. Furthermore, analysis of survivor after 28 days in ICU, patients without pneumothorax were significantly more successful in weaning from mechanical ventilator than the patients with pneumothorax (44% vs. 13.3%, p  = 0.037). PCP patients without pneumothorax showed successful home discharges compared to those who without pneumothorax ( p  = 0.010). Conclusions The development of pneumothorax increased in PCP patient with cytomegalovirus co-infection, pneumothorax might have difficulty in and prolonged weaning from mechanical ventilators, which clinicians should be aware of when planning treatment for such patients.

Background Healthcare-associated pneumonia (HCAP) has more similarities to nosocomial pneumonia than to community-acquired pneumonia (CAP). However, there have only been a few epidemiological studies of HCAP in South Korea. We aimed to determine the differences between HCAP and CAP in terms of clinical features, pathogens, and outcomes, and to clarify approaches for initial antibiotic management. Methods We conducted a retrospective, observational study of 527 patients with HCAP or CAP who were hospitalized at Severance Hospital in South Korea between January and December 2008. Results Of these patients, 231 (43.8%) had HCAP, and 296 (56.2%) had CAP. Potentially drug-resistant (PDR) bacteria were more frequently isolated in HCAP than CAP (12.6% vs. 4.7%; P = 0.001), especially in the low-risk group of the PSI classes (41.2% vs. 13.9%; P = 0.027). In-hospital mortality was higher for HCAP than CAP patients (28.1% vs. 10.8%, P < 0.001), especially in the low-risk group of PSI classes (16.4% vs. 3.1%; P = 0.001). Moreover, tube feeding and prior hospitalization with antibiotic treatment within 90 days of pneumonia onset were significant risk factors for PDR pathogens, with odds ratios of 14.94 (95% CI 4.62-48.31; P < 0.001) and 2.68 (95% CI 1.32-5.46; P = 0.007), respectively. Conclusions For HCAP patients with different backgrounds, various pathogens and antibiotic resistance of should be considered, and careful selection of patients requiring broad-spectrum antibiotics is important when physicians start initial antibiotic treatments.

We assessed the incidence and clinical characteristics of acute kidney injury (AKI) in critically ill patients infected with pandemic influenza A (H1N1) and its effect on clinical outcomes. We conducted a multicenter, retrospective, observational study of patients with pandemic H1N1-related critical illness admitted to intensive care units (ICUs) of 28 tertiary or referral hospitals in South Korea between September 1, 2009, and February 28, 2010. Outcomes were AKI within 72 hours after ICU admission and 30-day mortality. Acute kidney injury was defined according to the Risk, Injury, Failure, Loss, and End-stage renal failure criteria. Of the 221 patients, 50 (22.6%) developed AKI within 72 hours after ICU admission. Independent risk factors for AKI were age (odds ratio [OR], 1.05; P = .003), chronic kidney disease (OR, 14.82; P = .004), and Sequential Organ Failure Assessment score (OR, 1.45; P < .001). Age (OR, 1.04; P = .003), Sequential Organ Failure Assessment score (OR, 1.28; P = .012), state of immune suppression (OR, 4.09; P = .01), mechanical ventilation (OR, 18.24; P = .001), corticosteroid use (OR, 3.09; P = .007), and AKI (OR, 2.86; P = .035) were significantly associated with 30-day mortality. A significant number of patients with H1N1-related critical illness developed AKI within 72 hours of ICU admission, and this early development of AKI was associated with 30-day mortality.

The purpose of the study was to evaluate the utility of N-terminal pro–brain natriuretic peptide (NT-proBNP) as a marker of right ventricular (RV) dysfunction after open-lung approach (OLA) in patients with acute lung injury (ALI)/acute respiratory distress syndrome (ARDS). Twenty-seven patients with ALI/ARDS underwent OLA (2-minute steps of fixed pressure-controlled ventilation with progressive positive end-expiratory pressure levels up to 30 cm H2O, followed by stepwise decrement of positive end-expiratory pressure level by 2 cm H2O). Patients who showed a PaO2/FiO2 increase of more than 50% from baseline were defined as responders. Plasma NT-proBNP levels were taken immediately before OLA and 2 and 6 hours later. A minimum 30% increase in NT-proBNP level from baseline was considered significant. Right-over-left ventricular stroke work ratio and its percentage change did not differ between responders and nonresponders, whereas these values were higher in patients showing NT-proBNP increase (P < .05). The NT-proBNP percentage change correlated with right-over-left ventricular stroke work ratio percentage change (r = 0.83), pulmonary vascular resistance (r = 0.81), and RV ejection fraction (r = -0.79) and correlated with plateau pressure in nonresponders only (r = 0.82). In patients with ALI/ARDS, intraindividual NT-proBNP changes correlated with RV afterload following OLA, thereby serving as a potential marker for RV dysfunction after OLA.

Modelling biological networks is inherently data-driven and data-intensive. The combinatorial nature of this type of modelling, however, requires new methods capable of dealing with the enormous size and irregularity of the search. Searching via backtracking is one possible solution that avoids exhaustive searches by constraining the search space to the subspace of feasible solutions. Despite its wide use in many combinatorial optimization problems, there are currently few parallel implementations of backtracking capable of effectively dealing with the memory-intensive nature of the process and the extremely unbalanced loads present. In this paper, a parallel, scalable, and memory-efficient backtracking algorithm within the context of maximal clique enumeration is presented, and its applicability to large-scale biological networks aimed at studying the mechanisms for efficient bioethanol production is discussed.

Effects of the bulk inlet velocity on the characteristics of dual-inlet side-dump flows are numerically investigated. Non-reacting subsonic turbulent flow is solved by a preconditioned Reynolds-averaged Navier-Stokes equation system with low-Reynolds number k − ɛ turbulence model. The numerical method is properly validated with measured velocity distributions in the head dome and the combustor. With substantial increase in the bulk inlet velocity, general profiles of essential primary and secondary flows normalized by the bulk inlet velocity are quantitatively invariant to the changes in the bulk inlet velocity.