Yonsei University

About: Yonsei University is a education organization based out in Seoul, South Korea. It is known for research contribution in the topics: Population & Cancer. The organization has 50162 authors who have published 106172 publications receiving 2279044 citations. The organization is also known as: Yonsei.

Fast Video Object Segmentation by Reference-Guided Mask Propagation

Abstract: We present an efficient method for the semi-supervised video object segmentation. Our method achieves accuracy competitive with state-of-the-art methods while running in a fraction of time compared to others. To this end, we propose a deep Siamese encoder-decoder network that is designed to take advantage of mask propagation and object detection while avoiding the weaknesses of both approaches. Our network, learned through a two-stage training process that exploits both synthetic and real data, works robustly without any online learning or post-processing. We validate our method on four benchmark sets that cover single and multiple object segmentation. On all the benchmark sets, our method shows comparable accuracy while having the order of magnitude faster runtime. We also provide extensive ablation and add-on studies to analyze and evaluate our framework.

Novel Acquired Metallo-β-Lactamase Gene, blaSIM-1, in a Class 1 Integron from Acinetobacter baumannii Clinical Isolates from Korea

Abstract: Carbapenem resistance mediated by acquired carbapenemase genes has been increasingly reported, particularly for clinical isolates of Pseudomonas aeruginosa and Acinetobacter spp. Of 1,234 nonduplicate isolates of carbapenem-resistant Pseudomonas spp. and Acinetobacter spp. isolated at a tertiary-care hospital in Seoul, Korea, 211 (17%) were positive for metallo-β-lactamase (MBL). Of these, 204 (96%) had either the blaIMP-1 or blaVIM-2 allele. In addition, seven Acinetobacter baumannii isolates were found to have a novel MBL gene, which was designated blaSIM-1. The SIM-1 protein has a pI of 7.2, is a new member of subclass B1, and exhibits 64 to 69% identity with the IMP-type MBLs, which are its closest relatives. All SIM-1-producing isolates exhibited relatively low imipenem and meropenem MICs (8 to 16 μg/ml) and had a multidrug resistance phenotype. Expression of the cloned blaSIM-1 gene in Escherichia coli revealed that the encoded enzyme is capable of hydrolyzing a broad array of β-lactams, including penicillins, narrow- to expanded-spectrum cephalosporins, and carbapenems. The blaSIM-1 gene was carried on a gene cassette inserted into a class 1 integron, which included three additional cassettes (arr-3, catB3, and aadA1). The strains were isolated from sputum and urine specimens from patients with pneumonia and urinary tract infections, respectively. All patients had various underlying diseases. Pulsed-field gel electrophoresis of SmaI-digested genomic DNAs showed that the strains belonged to two different clonal lineages, indicating that horizontal transfer of this gene had occurred and suggesting the possibility of further spread of resistance in the future.

Aberrant CFTR-dependent HCO3- transport in mutations associated with cystic fibrosis.

Abstract: Cystic fibrosis (CF) is a disease caused by mutations in the cystic fibrosis transmembrane conductance regulator (CFTR). Initially, Cl- conductance in the sweat duct was discovered to be impaired in CF, a finding that has been extended to all CFTR-expressing cells. Subsequent cloning of the gene showed that CFTR functions as a cyclic-AMP-regulated Cl- channel; and some CF-causing mutations inhibit CFTR Cl- channel activity. The identification of additional CF-causing mutants with normal Cl- channel activity indicates, however, that other CFTR-dependent processes contribute to the disease. Indeed, CFTR regulates other transporters, including Cl(-)-coupled HCO3- transport. Alkaline fluids are secreted by normal tissues, whereas acidic fluids are secreted by mutant CFTR-expressing tissues, indicating the importance of this activity. HCO3- and pH affect mucin viscosity and bacterial binding. We have examined Cl(-)-coupled HCO3- transport by CFTR mutants that retain substantial or normal Cl- channel activity. Here we show that mutants reported to be associated with CF with pancreatic insufficiency do not support HCO3- transport, and those associated with pancreatic sufficiency show reduced HCO3- transport. Our findings demonstrate the importance of HCO3- transport in the function of secretory epithelia and in CF.

Behavioral and Emotional Problems Reported by Parents of Children Ages 6 to 16 in 31 Societies

Abstract: This study compared parents' ratings of behavioral and emotional problems on the Child Behavior Checklist (Achenbach, 1991;Achenbach & Rescorla, 2001) for general population samples of children age...

Fully Deep Blind Image Quality Predictor

Abstract: In general, owing to the benefits obtained from original information, full-reference image quality assessment (FR-IQA) achieves relatively higher prediction accuracy than no-reference image quality assessment (NR-IQA). By fully utilizing reference images, conventional FR-IQA methods have been investigated to produce objective scores that are close to subjective scores. In contrast, NR-IQA does not consider reference images; thus, its performance is inferior to that of FR-IQA. To alleviate this accuracy discrepancy between FR-IQA and NR-IQA methods, we propose a blind image evaluator based on a convolutional neural network (BIECON). To imitate FR-IQA behavior, we adopt the strong representation power of a deep convolutional neural network to generate a local quality map, similar to FR-IQA. To obtain the best results from the deep neural network, replacing hand-crafted features with automatically learned features is necessary. To apply the deep model to the NR-IQA framework, three critical problems must be resolved: 1) lack of training data; 2) absence of local ground truth targets; and 3) different purposes of feature learning. BIECON follows the FR-IQA behavior using the local quality maps as intermediate targets for conventional neural networks, which leads to NR-IQA prediction accuracy that is comparable with that of state-of-the-art FR-IQA methods.