http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
The Effects of Uncertainties on Network Embeddedness and the Mediating Effect of Information Sharing
( Hyunseung Kang ),( Junhong Min ),( Jung Kook Lee ) 서암순창장학회 2016 Journal of Marketing Thought Vol.2 No.4
Conceptual model that both technology and volume uncertainty promote exchange partners to rely on the network norm of information sharing which is the necessary ingredient of the network embeddedness. Data was collected from the 143 manufacturers in high-tech market in which triadic relationships among the manufacturers (seller), their first vendors (first buyers), and the second vendors (customers of the first buyers) in high-tech markets were particu-larly focused. Results from the structural equation model and multiple regression analysis reveal that while the tech-nological uncertainty has a positive effect on the network norm of information sharing, the volume uncertainty is not statistically significant. In addition, we find that there existsthe mediator effect of the network norm of information sharing in the relation between the uncertainties and the network embeddedness.
A RGB-D Vision based Indoor SLAM using 2.5D Map by Multiple UAVs
Hyunseung. Kang,Kyomun. Ku,Jaehong. Shim 제어로봇시스템학회 2021 제어로봇시스템학회 국제학술대회 논문집 Vol.2021 No.10
This paper presents an approach to build an indoor 2.5D map with multiple UAVs(Unmanned Aerial Vehicles). In an indoor environment, GPS system could not be used so each UAV adopted a tracking camera(Intel RealSense T265) to localize itself and 3D data for map building is acquired by the stereo depth camera(Intel RealSense D435). However, the raw data of the above sensors can have significant noise and large size of data, so the octree filter is applied as a solution. Usually indoor aerial robots have not enough flying time to build entire map of the floor so this paper suggested an improved way to build entire map in 2.5D with multiple UAVs. Each UAV can build a map for part of the global floor and each 3D local maps are converted to 2.5D, then merged together. To make a global map from several local maps, feature points are extracted that can be detected in 2.5D of the indoor environment. Through a series of experiments, the proposed method creates a map that takes into account obstacles and complex structures in the indoor environment, and shows that the ground robot can autonomously drive using this map.
Multimodal imaging을 통한 다초점맥락막염과 점상내측맥락막병증의 비교, 그리고 그 정의에 대한 고찰
Hyunseung Kang,Suk Ho Byeon 한국망막학회 2016 Journal of Retina Vol.1 No.2
Multifocal choroiditis (MFC) has been defined as a disease characterized by multifocal choroidal inflammatory lesions occurring predominantly in young myopic women, idiopathic in origin, and not associated with any systemic disorder. However, there are numerous diseases that share similar features and definitions, such as “punctate inner choroidopathy (PIC),” complicating the differential diagnosis. Many of the present authors regard MFC and PIC to be the same disorder and include them under the single diagnosis “idiopathic MFC” regardless of the presence of ocular inflammation. In addition, there are many diagnostic terms that overlap or duplicate idiopathic MFC, again complicating the differential diagnosis. Recent advancement in technology has allowed a technique called ‘multimodal imaging,’ which has produced new findings about previously little-known diseases. These new findings have shown that many cases of what were diagnosed as idiopathic choroidal neovascularization (CNV) or myopic CNV were actually secondary to unrecognized MFC/PIC. In other words, the incidence of MFC/PIC was underestimated. Recent findings with multimodal imaging suggest a reappraisal of the pathophysiology of these diseases. Herein, we introduce recent reports of MFC/PIC based on multimodal imaging. ‘Multifocal choroiditis (MFC)’는 주로 근시를 가진 젊은 여성에서 다초점의 맥락막 염증병변이 발생하는 질환으로 정의되어 왔다. 이용어는 원래 특발성으로 발생하는 경우를 지칭하지만 유사한 망막소견을 보이는 “Punctate inner choroidopathy (PIC)” 및 여러 가지 감염성 질환들에서 여러 연구자들이 사용한 다양한 병명들은 MFC를 이해하는 데 혼란을 주고 있다. 현재 많은 저자들이 안구내 염증반응 유무에 관계없이 MFC와 PIC를 같은 스펙트럼 질환이라고 보고 있으며, “idiopathic MFC”라는 포괄적인 개념의 진단명으로 묶어 사용하고 있다. 또한 imaging상 병변의 특징이 기존에 idiopathic choroidal neovascularization (CNV) 혹은 myopic CNV 등으로 진단되었던 환자들에서 드물지 않게 발견되면서, 임상적으로 실제 MFC/PIC의 빈도가 과소평가되었을 가능성도 제시되었다. 이렇게 Retinal imaging의 발전은 기존에 비슷한 임상양상으로 인하여 구분되지 못하였던 경우나, 혹은 오히려 같은 질환군임에도 다양한 임상양상으로 인하여 다양한 질환으로 오인되었던 경우에 있어서 질환의 정확한 진단과 개념 정립에 기여하고 있다.
Kubeflow 환경에서 CPU 집약적인 작업을 위한 컨테이너 수에 따른 연산 시간 비교 및 분석
정현승 ( Hyunseung Jung ),강태신 ( Taeshin Kang ),유헌창 ( Heonchang Yu ),강지훈 ( Jihun Kang ) 한국정보처리학회 2023 한국정보처리학회 학술대회논문집 Vol.30 No.2
머신 러닝의 수요가 증가함에 따라, 머신 러닝 워크플로우의 배포 수요도 증가했다. Kubeflow를 통해 머신 러닝 배포를 편리하게 할 수 있으며, Kubeflow Pipelines에서는 하나의 작업을 여러 컨테이너로 분산시켜서 연산하는 것이 가능하다. 하지만 컨테이너 수를 많이 늘릴수록 반드시 성능이 향상되는 것은 아니다. 따라서, 본 연구에서는 성능 향상의 한계를 제공하는 원인을 분석하기 위해서, Kubeflow에서 CPU 집약적인 작업을 여러 컨테이너로 분산시켜서 연산을 수행하였다. 컨테이너 수에 따른 연산 완료 시간을 비교 및 분석한 결과, 컨테이너 수가 증가할수록 연산 속도 향상이 빨라지나, 어느 시점을 지나면 속도가 다시 완만하게 줄어드는 현상을 확인하였다. 이는 리소스 제한으로 인해 모든 컨테이너가 동시에 스케줄링 되지 못한 것이 가장 큰 원인으로 분석하였다.
Long working hours and cardiovascular disease: a meta-analysis of epidemiologic studies.
Kang, Mo-Yeol,Park, Hyunseung,Seo, Jeong-Cheol,Kim, Donghoon,Lim, Youn-Hee,Lim, Sinye,Cho, Soo-Hun,Hong, Yun-Chul Lippincott Williams & Wilkins 2012 Journal of Occupational and Environmental Medicine Vol.54 No.5
<P>To conduct a meta-analysis from published studies to evaluate the relationship between long working hours and the risk of cardiovascular disease (CVD).</P>
Kang, Sungmin,Lee, Ho Won,Jeon, Young Hyun,Singh, Thoudam Debraj,Choi, Yun Ju,Park, Ji Young,Kim, Jun Sung,Lee, Hyunseung,Hong, Kwan Soo,Lee, Inkyu,Jeong, Shin Young,Lee, Sang-Woo,Ha, Jeoung-Hee,Ahn, Elsevier Science 2015 Molecular imaging and biology Vol.17 No.5
<P>This study aimed to track the migration of primary macrophages labeled with near-infrared (NIR) fluorescent magnetic nanoparticles toward chemically induced acute inflammatory lesions in mice and to visualize the effect of anti-inflammatory drugs on macrophage migration using combined fluorescence and magnetic resonance imaging (FLI/MRI).</P>
( Hyunseung Nam ),( Jae Hwa Cho ),( Eun Young Choi ),( Youjin Chang ),( Won-il Choi ),( Jae Joon Hwang ),( Jae Young Moon ),( Kwangha Lee ),( Sei Won Kim ),( Hyung Koo Kang ),( Yun Su Sim ),( Tai Sun 대한결핵 및 호흡기학회 2019 Tuberculosis and Respiratory Diseases Vol.82 No.3
Background: Data on noninvasive ventilation (NIV) use in intensive care units (ICUs) are very limited in South Korea. Methods: A prospective observational study was performed in 20 ICUs of university-affiliated hospitals from June 2017 to February 2018. Adult patients (age>18 years) who were admitted to the ICU and received NIV treatment for acute respiratory failure were included. Results: A total of 156 patients treated with NIV were enrolled (mean age, 71.9±11.6 years). The most common indications for NIV were acute hypercapnic respiratory failure (AHRF, n=89) and post-extubation respiratory failure (n=44). The main device for NIV was an invasive mechanical ventilator with an NIV module (61.5%), and the majority of patients (87.2%) used an oronasal mask. After the exclusion of 32 do-not-resuscitate patients, NIV success rate was 68.5% (85/124); ICU and hospital mortality rates were 8.9% and 15.3%, respectively. However, the success rate was lower in patients with de novo respiratory failure (27.3%) compared to that of patients with AHRF (72.8%) or post-extubation respiratory failure (75.0%). In multivariate analysis, immunocompromised state, de novo respiratory failure, post-NIV (2 hours) respiratory rate, NIV mode (i.e., non-pressure support ventilation mode), and the change of NIV device were significantly associated with a lower success rate of NIV. Conclusion: AHRF and post-extubation respiratory failure were the most common indications for NIV in Korean ICUs. Overall NIV success was achieved in 68.5% of patients, with the lowest rate in patients with de novo respiratory failure.
Delay-Efficient Energy-Minimized Data Collection With Dynamic Traffic in WSNs
Kang, Byungseok,Nguyen, Phuc,Choo, Hyunseung IEEE 2018 IEEE SENSORS JOURNAL Vol.18 No.7
<P>Data collection is one of the most important applications in wireless sensor networks where sensed data are gathered from sensor nodes to the base station. Reporting redundant data leads to the wastage of time and energy, the sensors therefore report only meaningful information to the base station, which are independent and distinct with the lasted ones. This assumption leads to unpredictable changes of data traffic over different sampling intervals in data collection process. In this paper, we first formulate the tight constraints of the problem and then propose a delay-efficient traffic adaptive (DETA) scheme for collecting data from sensor nodes with minimum energy consumption. The DETA scheme minimizes data collection delay by constructing delay-efficient, collision-free schedule, and by using a special mechanism to enable every node to self-adapt with the changes of data traffic. We also conducted simulations to evaluate the performance of the proposed scheme, and the results shown that the proposed scheme significantly decreases data collection delay and energy consumption compared with the existing schemes.</P>