http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Deep Generative Convolutional Variational Autoencoder for Identification of Wafer Map
Ho Sun Shon(손호선),Erdenebileg Batbaatar,Kyung Hee Lee(이경희),Wan-Sup Cho(조완섭),Seong Gon Choi(최성곤) 한국통신학회 2021 한국통신학회 학술대회논문집 Vol.2021 No.2
In this paper, we develop a novel Variational Autoencoder (VAE) to identify wafer map defect patterns. The Deep Generative Convolutional Network (DGCN) is used as an encoder to approximate distribution for the latent features linked to generative models. And the Deep Generative Deconvolutional Network (DGDN) is used as a decoder of the wafer map defect patterns. Firstly, we pre-trained unsupervised VAE which consists of DGCN and DGDN. Secondly, we fine-tuned DGCN with neural network classification as a predictor. In addition to enforcing the deep feature consistent principle thus ensuring the VAE output and its corresponding input images to have similar deep features. We present experimental results to show that the VAE trained with our new method outperforms the state-of-the-art in identifying wafer map defect patterns. We also show that our method can learn powerful embeddings of input wafer map images, which can be used to achieve defect pattern manipulation.
Ho Sun Shon,Jang Whan Bae,Kyoung Ok Kim,Eun Jong Cha,Kyung Ah Kim 충북대학교 동물의학연구소 2017 Journal of Biomedical and Translational Research Vol.18 No.3
Recently, N-terminal pro-brain natriuretic peptide (NTproBNP) has been widely used in the areas of diagnosis, monitoring treatment efficiency, and prognosis for various heart diseases, especially heart failure (HF). In this paper, we try to estimate the prognostic significance of NT-proBNP as a risk evaluation marker in Non-ST-segment Elevation Myocardial Infarction (NSTEMI) patients. We selected NSTEMI patients who underwent percutaneous coronary intervention (PCI) primarily using a drug-eluting stent within 24 h after the onset of chest pain. We compared incidences of major adverse cardiac events (MACE) including death, myocardial infarction (MI), stent thrombosis (ST), and target vessel revascularization (TVR) in two patient groups according to a high or low serum concentration of NT-proBNP, which was measured in the emergency room (ER). We intend to minimize selection bias selecting comparing groups, considering covariate of observed variables together using propensity score matching (PSM) and propensity score weighting (PSW) based on propensity score (PS) to control the difference in baseline characteristics between high- and low NT-proBNP groups. We found that as the log NT-proBNP value increases by 1 through a hazard function of COX’s analysis, the risk of MACE increases by 1.312 times. This result indicated that the NT-proBNP level on ER admission can be used as a significant prognostic indicator to estimate 1 year of MACE in NSTEMI patients who were treated with PCI within 24 h after the onset of chest pain.
Kang, Ho Won,Park, Hongyong,Seo, Sung Pil,Byun, Young Joon,Piao, Xuan-Mei,Kim, Sung Min,Kim, Won Tae,Yun, Seok-Joong,Jang, Wooyeong,Shon, Ho Sun,Ryu, Keun Ho,Lee, Sang-Cheol,Kim, Wun-Jae,Kim, Yong-Jun The Korean Academy of Medical Sciences 2019 JOURNAL OF KOREAN MEDICAL SCIENCE Vol.34 No.19
<P><B>Background</B></P><P>Little is known about epigenetic silencing of genes by promoter hypermethylation in renal cell carcinoma (RCC). The aim of this study was to identify prognostic methylation markers in surgically treated clear cell RCC (ccRCC).</P><P><B>Methods</B></P><P>Methylation patterns were assayed using the Infinium HumanMethylation450 BeadChip array on pairs of ccRCC and normal tissue from 12 patients. Using quantitative PSQ analysis, tumor-specific hypermethylated genes were validated in 25 independent cohorts and their clinical relevance was also verified in 152 independent cohorts.</P><P><B>Results</B></P><P>Using genome-wide methylation array, Zinc finger protein 278 (<I>ZNF278</I>), Family with sequence similarity 155 member A (<I>FAM155A</I>) and Dipeptidyl peptidase 6 (<I>DPP6</I>) were selected for tumor-specific hypermethylated genes in primary ccRCC. The promoter methylation of these genes occurred more frequently in ccRCC than normal kidney in independent validation cohort. The hypermethylation of three genes were associated with advanced tumor stage and high grade tumor in ccRCC. During median follow-up of 39.2 (interquartile range, 15.4–79.1) months, 22 (14.5%) patients experienced distant metastasis. Multivariate analysis identified the methylation status of these three genes, either alone, or in a combined risk score as an independent predictor of distant metastasis.</P><P><B>Conclusion</B></P><P>The promoter methylation of <I>ZNF278</I>, <I>FAM155A</I> and <I>DPP6</I> genes are associated with aggressive tumor phenotype and early development of distant metastasis in patients with surgically treated ccRCC. These potential methylation markers, either alone, or in combination, could provide novel targets for development of individualized therapeutic and prevention regimens.</P>