http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
윤웅창(Woongchang Yoon),장병탁(Byoung-Tak Zhang) 한국정보과학회 2009 한국정보과학회 학술발표논문집 Vol.36 No.2C
이미지와 텍스트 또는 텍스트와 오디오와 같이 하나 이상의 모달리티 간의 상호작용을 이용한 데이터의 활용이 증가하고 있다. 특히 텍스트를 이용한 이미지 검색이나 카테고리 분류가 점차 늘어나고 있는 상황에서 이미지를 자동으로 분류하고 주석을 달아 주는 것이 필요한 일되었다. 본 연구에서는 대용량 이미지로 유명한 flickr라는 이미지 전문 웹사이트에서 이미지와 이미지에 달려있는 주석를 수집하여 핵심어 및 시각 언어 추출 과정을 통해서 얻어진 정보를 바탕으로 텍스트-이미지 간의 관련정보와 진화적 하이퍼 네트워크 모델을 이용한 기계 학습 결과로 자동으로 이미지에 주석을 달아주는 연구를 수행 하였다.
랜덤 하이퍼그래프 메모리 모델을 이용한 문장생성에 멀티 모달리티가 미치는 영향
윤웅창(Woongchang Yoon),장병탁(Byoung-Tak Zhang) 대한전자공학회 2010 대한전자공학회 학술대회 Vol.2010 No.6
In this paper, we introduce a generation method based on the random hypergraph memory model using multimodal cue. In order to demonstrate the effects of multi-modal cue on sentence generation, we compare experimental results of the sentence generation given only text or image cues and image-text multi-modal pair cues. For experiments, we use TV drama screen shots and script data. The image-text corpus are converted into a set of weighted hyperedge. Our aim is to generate sentences using cues based on these set of hyperedges. It is our experiment"s result show that language generation is helped by multi-modality data more than only text or image data. We accomplish to simulate effects of supporting memory improvement by multi-modal data set without templates or rote learning.
랜덤 하이퍼그래프 모델을 이용한 순차적 멀티모달 데이터에서의 문장 생성
윤웅창(Woongchang Yoon),장병탁(Byoung-Tak Zhang) 한국정보과학회 2010 한국정보과학회 학술발표논문집 Vol.37 No.1C
인간의 학습과 기억현상에 있어서 멀티모달 데이터를 사용하는 것은 단순 모달리티 데이터를 사용하는 것에 비해서 향상된 효과를 보인다는 여러 연구 결과가 있어왔다. 이 논문에서는 인간의 순차적인 정보처리와 생성현상을 기계에서의 시뮬레이션을 통해서 기계학습에 있어서도 동일한 현상이 나타나는지에 대해서 알아보고자 하였다. 이를 위해서 가중치를 가진 랜덤 하이퍼그래프 모델을 통해서 순차적인 멀티모달 데이터의 상호작용을 하이퍼에지들의 조합으로 나타내는 것을 제안 하였다. 이러한 제안의 타당성을 알아보기 위해서 비디오 데이터를 이용한 문장생성을 시도하여 보았다. 이전 장면의 사진과 문장을 주고 다음 문장의 생성을 시도하였으며, 단순 암기학습이나 주어진 룰을 통하지 않고 의미 있는 실험 결과를 얻을 수 있었다. 단순 텍스트와 텍스트-이미지 쌍의 단서를 통한 실험을 통해서 멀티 모달리티가 단순 모달리티에 비해서 미치는 영향을 보였으며, 한 단계 이전의 멀티모달 단서와 두 단계 및 한 단계 이전의 멀티 모달 단서를 통한 실험을 통해서 순차적 데이터의 단계별 단서의 차이에 따른 영향을 알아볼 수 있었다. 이를 통하여 멀티 모달리티가 시공간적으로 미치는 기계학습에 미치는 영향과 순차적 데이터의 시간적 누적에 따른 효과가 어떻게 나타날 수 있는지에 대한 실마리를 제공할 수 있었다고 생각된다.
Multi-biomarker panel prediction model for diagnosis of pancreatic cancer
Doo-Ho LEE,Woongchang YOON,Areum LEE,Youngmin HAN,Yoonhyeong BYUN,Jae Seung KANG,Hongbeom KIM,Wooil KWON,Young-Ah SUH,Yonghwan CHOI,Junghyun NAMKUNG,Sangjo HAN,Sung Gon YI,Jin Seok HEO,In Woong HAN,Jo 한국간담췌외과학회 2021 Annals of hepato-biliary-pancreatic surgery Vol.25 No.-
Diagnostic model for pancreatic cancer using a multi-biomarker panel
Yoo Jin Choi,Woongchang Yoon,Areum Lee,Youngmin Han,Yoonhyeong Byun,Jae Seung Kang,Hongbeom Kim,Wooil Kwon,Young-Ah Suh,Yongkang Kim,Seungyeoun Lee,Junghyun Namkung,Sangjo Han,Yonghwan Choi,Jin Seok H 대한외과학회 2021 Annals of Surgical Treatment and Research(ASRT) Vol.100 No.3
Purpose: Diagnostic biomarkers of pancreatic ductal adenocarcinoma (PDAC) have been used for early detection to reduce its dismal survival rate. However, clinically feasible biomarkers are still rare. Therefore, in this study, we developed an automated multi-marker enzyme-linked immunosorbent assay (ELISA) kit using 3 biomarkers (leucine-rich alpha-2-glycoprotein [LRG1], transthyretin [TTR], and CA 19-9) that were previously discovered and proposed a diagnostic model for PDAC based on this kit for clinical usage. Methods: Individual LRG1, TTR, and CA 19-9 panels were combined into a single automated ELISA panel and tested on 728 plasma samples, including PDAC (n = 381) and normal samples (n = 347). The consistency between individual panels of 3 biomarkers and the automated multi-panel ELISA kit were accessed by correlation. The diagnostic model was developed using logistic regression according to the automated ELISA kit to predict the risk of pancreatic cancer (high-, intermediate-, and low-risk groups). Results: The Pearson correlation coefficient of predicted values between the triple-marker automated ELISA panel and the former individual ELISA was 0.865. The proposed model provided reliable prediction results with a positive predictive value of 92.05%, negative predictive value of 90.69%, specificity of 90.69%, and sensitivity of 92.05%, which all simultaneously exceed 90% cutoff value. Conclusion: This diagnostic model based on the triple ELISA kit showed better diagnostic performance than previous markers for PDAC. In the future, it needs external validation to be used in the clinic.
Yang, Jinsung,Kim, Min Ju,Yoon, Woongchang,Kim, Eun Young,Kim, Hyunjin,Lee, Yoonjeong,Min, Boram,Kang, Kyung Shin,Son, Jin H.,Park, Hwan Tae,Chung, Jongkyeong,Koh, Hyongjong Public Library of Science 2017 PLoS genetics Vol.13 No.8
<▼1><P><I>DJ-1</I> is one of the causative genes for early onset familiar Parkinson’s disease (PD) and is also considered to influence the pathogenesis of sporadic PD. DJ-1 has various physiological functions which converge on controlling intracellular reactive oxygen species (ROS) levels. In RNA-sequencing analyses searching for novel anti-oxidant genes downstream of DJ-1, a gene encoding NADP<SUP>+</SUP>-dependent isocitrate dehydrogenase (IDH), which converts isocitrate into α-ketoglutarate, was detected. Loss of <I>IDH</I> induced hyper-sensitivity to oxidative stress accompanying age-dependent mitochondrial defects and dopaminergic (DA) neuron degeneration in <I>Drosophila</I>, indicating its critical roles in maintaining mitochondrial integrity and DA neuron survival. Further genetic analysis suggested that DJ-1 controls IDH gene expression through nuclear factor-E2-related factor2 (Nrf2). Using <I>Drosophila</I> and mammalian DA models, we found that IDH suppresses intracellular and mitochondrial ROS level and subsequent DA neuron loss downstream of DJ-1. Consistently, trimethyl isocitrate (TIC), a cell permeable isocitrate, protected mammalian <I>DJ-1</I> null DA cells from oxidative stress in an IDH-dependent manner. These results suggest that isocitrate and its derivatives are novel treatments for PD associated with <I>DJ-1</I> dysfunction.</P></▼1><▼2><P><B>Author summary</B></P><P>The molecular pathogenesis of Parkinson’s disease (PD) is still elusive even though many causative genes for the disease have been identified. In this study, we demonstrated that isocitrate dehydrogenase (IDH), the enzyme responsible for converting isocitrate into α-ketoglutarate, is critical for the pathogenesis of PD by providing NADPH as a reducing power in the cell. <I>IDH</I> mutant animals showed increased reactive oxygen species (ROS) levels and phenotypes related to PD including dopaminergic (DA) neuron degeneration and locomotor defects. Conversely, elevating IDH function either by overexpression or treating a cell-permeable derivative of isocitrate, trimethyl isocitrate (TIC), made DA cells resist oxidative stress and reduce ROS level, thereby suppressing PD phenotypes induced by <I>DJ-1</I> mutations. These results demonstrate that IDH protects DA neurons from ROS at the downstream of DJ-1 and cell-permeable isocitrates can be novel treatments for PD.</P></▼2>
Chung, Won-Hyong,Jeong, Namhee,Kim, Jiwoong,Lee, Woo Kyu,Lee, Yun-Gyeong,Lee, Sang-Heon,Yoon, Woongchang,Kim, Jin-Hyun,Choi, Ik-Young,Choi, Hong-Kyu,Moon, Jung-Kyung,Kim, Namshin,Jeong, Soon-Chun Oxford University Press 2014 DNA research Vol.21 No.2
<P>Despite the importance of soybean as a major crop, genome-wide variation and evolution of cultivated soybeans are largely unknown. Here, we catalogued genome variation in an annual soybean population by high-depth resequencing of 10 cultivated and 6 wild accessions and obtained 3.87 million high-quality single-nucleotide polymorphisms (SNPs) after excluding the sites with missing data in any accession. Nuclear genome phylogeny supported a single origin for the cultivated soybeans. We identified 10-fold longer linkage disequilibrium (LD) in the wild soybean relative to wild maize and rice. Despite the small population size, the long LD and large SNP data allowed us to identify 206 candidate domestication regions with significantly lower diversity in the cultivated, but not in the wild, soybeans. Some of the genes in these candidate regions were associated with soybean homologues of canonical domestication genes. However, several examples, which are likely specific to soybean or eudicot crop plants, were also observed. Consequently, the variation data identified in this study should be valuable for breeding and for identifying agronomically important genes in soybeans. However, the long LD of wild soybeans may hinder pinpointing causal gene(s) in the candidate regions.</P>