• 한국멀티미디어학회논문지
• /
• 제19권8호
• /
• pp.1516-1529
• /
• 2016

While global manufacturing is becoming more competitive due to variety of customer demand, increase in production cost and uncertainty in resource availability, the future ability of manufacturing industries depends upon the implementation of Smart Factory. With the convergence of new information and communication technology, Smart Factory enables manufacturers to respond quickly to customer demand and minimize resource usage while maximizing productivity performance. This paper presents the development of a big data analytics platform architecture for Smart Factory. As this platform represents a conceptual software structure needed to implement data-driven decision-making mechanism in shop floors, it enables the creation and use of diagnosis, prediction and optimization models through the use of data analytics and big data. The completion of implementing the platform will help manufacturers: 1) acquire an advanced technology towards manufacturing intelligence, 2) implement a cost-effective analytics environment through the use of standardized data interfaces and open-source solutions, 3) obtain a technical reference for time-efficiently implementing an analytics modeling environment, and 4) eventually improve productivity performance in manufacturing systems. This paper also presents a technical architecture for big data infrastructure, which we are implementing, and a case study to demonstrate energy-predictive analytics in a machine tool system.

• ETRI Journal
• /
• 제44권1호
• /
• pp.135-146
• /
• 2022

The importance of big data analytics has become apparent with the increasing volume of data on the Internet. The amount of data will increase even more with the widespread use of Internet of Things (IoT). One of the most important application areas of the IoT is healthcare. This study introduces new real-time data analytics architecture for an IoT-based smart healthcare system, which consists of a wireless sensor network and a radio-frequency identification technology in a vertical domain. The proposed platform also includes high-performance data analytics tools, such as Kafka, Spark, MongoDB, and NodeJS, in a horizontal domain. To investigate the performance of the system developed, a diagnosis of Wolff-Parkinson-White syndrome by logistic regression is discussed. The results show that the proposed IoT data analytics system can successfully process health data in real-time with an accuracy rate of 95% and it can handle large volumes of data. The developed system also communicates with a riverbed modeler using Transmission Control Protocol (TCP) to model any IoT-enabling technology. Therefore, the proposed architecture can be used as a time-saving experimental environment for any IoT-based system.

• 한국시뮬레이션학회논문지
• /
• 제8권1호
• /
• pp.89-99
• /
• 1999

The major objective of this research is to propose a design architecture for autonomous defense systems for supporting highly intelligent behavior by combining decision, perception, and action components. Systems with such high levels of autonomy are critical for advanced battlefield missions. By integrating a plenty of advanced modeling concepts such as system entity structure, endomorphic modeling, engine-based modeling, and hierarchical encapsulation & abstraction principle, we have proposed four layered design methodology for autonomous defense systems that can support an intelligent behavior under the complicated and unstable warfare. Proposed methodology has been successfully applied to a design of autonomous tank systems capable of supporting the autonomous planning, sensing, control, and diagnosis.

• 감성과학
• /
• 제13권3호
• /
• pp.429-440
• /
• 2010

자폐스펙트럼장애는 91명당 1명이 문제를 보인다고 보고되었으며 이는 우리나라 암 발생률보다도 높은 수준이지만, 국내에서는 제대로 된 진단 및 치료가 이루어지고 있지 않아 사회적으로 심각한 복지 사각지대에 있다. 자폐스펙트럼장애 아동의 경우 자신의 정서 상태를 제대로 표현하지 못하여 기존의 치료 및 교육 방법에 제약이 많기 때문에, 장애 아동의 정서를 실시간으로 인지하여 활용할 수 있는 새로운 개념의 치료 시스템의 개발이 요구된다. 본 논문에서는 생체신호 감지를 통한 정서 상태 인지 기술 및 가상 감성 에이전트(agent) 모델링 기술을 포함한 감성 컴퓨팅 기술에 대하여 기술 동향 분석 결과를 제시하고, 새로운 자폐스펙 트럼장애 진단 및 치료를 위한 시스템 구성방식을 제안한다. 제안된 시스템은 뇌파 등의 생체신호로 부터 자폐아동의 정서 상태를 인지하기 위한 생체신호 감지 모듈, 가상현실 환경에서 자폐아동이 감성 에이전트와 사회적 상호작용을 하면서 다양한 진단 및 치료 시나리오를 수행하기 위한 가상 감성 에이전트 환경 모듈, 역감 교류(haptic interface)장치를 이용하여 자폐아동의 행동 입력이 가능하게 하고, 힘 반력을 자폐아동에게 전달해 주기 위한 역감 교류 모듈로 구성된다. 본 논문에서 제안한 시스템을 통해 자폐스펙트럼장애를 객관적으로 진단할 수 있으며, 생활 속에서 지속적인 치료를 가능하게 할 수 있을 것으로 기대된다.

• 의료ㆍ복지 건축 : 한국의료복지건축학회 논문집
• /
• 제22권4호
• /
• pp.7-17
• /
• 2016

Purpose: As our population ages and becomes an elderly society the number of elderly care hospitals is rapidly increasing. Because physical functions and spatial perception in the elderly decrease with age, these hospitals require more systematic and intelligent space designs. The design of these spaces are even more complex because they must accommodate medical programs to treat various different diseases and ailments and also because there are many first time patients and irregular short term patients that seek out outpatient treatment services. Also by analyzing the spatial configuration systems and systematic relationships between each of the functional spaces of the outpatient treatment service departments for hospitals specialized in care for the elderly by focusing on the hallway and corridor systems of these hospitals, the according characteristics and trends were examined. Methods: Based on preceding research, the types of hallway and corridor systems of these hospitals were categorized into five types, including gallery corridors, middle corridors, hall-type, mixed type and cyclic type corridors, and into six types according to function including by medical diagnosis, patient registration, examination, administration and convenience and shared common space to derive any interconnecting relationships between the corridor systems. Also by comprehensively examining the types and combined utilization of the corridor types and the integration and the intelligibility of the space syntax, any trends within the corridor system were derived. The elderly care hospitals examined in this research study were twelve hospitals that opened after the year 2000 in Korea with more that 150 sick beds with areas larger than $1000m^2$ and with all outpatient medical service related rooms located entirely on a single floor of the hospital. Results: The following results could be confirmed based on this research study. 1) The spaces where medical diagnosis and examination occurred were adjacent, and the movement lines for first time patients and re-visiting patients were taken into consideration by separating the treatment space. 2) This research study confirmed that the larger the size of the hospital was, there were more detailed categorizations of treatment services and that there was a tendency for treatment areas to be separated and independent from examination areas. 3) There was a tendency for integration and intelligibility to decrease the more complex and diverse the combination of hall types designed into the corridor systems of these hospitals was. cyclic type corridors dramatically decreased the intelligibility of the corridor systems of these hospitals. 4) The priority rank of these spaces were confirmed to be highest in the order of registration, diagnosis, examination, treatment, administration and shared common spaces. However it was confirmed for the local integration that the diagnosis scope had the highest priority rank. Implications: There were exceptional cases confirmed where the number of unit spaces did not have an absolute effect on integration and intelligibility. These results can be interpreted to mean that this can be overcome through efficient architectural planning.

• 한국조경학회지
• /
• 제30권5호
• /
• pp.98-106
• /
• 2002

This study was designed to present a river model with the aim of restoring the ecosystem and improving the landscape along the urban rivers on the basin of the Namhanl river, a core life channel for the National Capital regin. The revelation of botanical status, transition trend and correlation of plants might lead to providing the urban river restoration projects and ecological river formation projects with basic data for a model of ideal aquatic ecology and landscape. The outcomes of this study could be summed up as follows: 1. The plant communities of river flora found on the basin of the Namhan river could be categorized largely into 39 plant communities 2. Most diverse plants were distributed in the rivers lower reaches such as Unsim-ri where the protected zone of Paldang reservoir for city water borders the body of Jodae swamp where natural streams flow nearby. 3. One of the greatest threats to the biomass of the River Namhan is that the communities of such invasive alien plants as Panicum dichotomiflorum and Ambrosia artemisiifolia var. elatior dominate most parts of the area, a fact that has resulted in a reduced variety of plants and will, sooner or later, be likely to cause an ecological imbalance in the hitherto healthy Aquatic plant life. It is highly advisable to gradually diversify the species of trees and to return the plants bark to their original state since, besides the naturalized plants, plantations afforested with Erigeron canadensis and Erigeron annuus stocks in buckwheat field, Robinia Pseudo-accacia in riverside forest, Pinus rigida in terrestrial forest on the river basin and Larix leptolepis are anticipated to act as interceptors of normal migrations of the fluvial and terrestrial ecosystems. Finally, it seems also desirable to continue to explore the relationship between fluvial and terrestrial ecologies with the purpose of building up a model of natural streams in urban areas based on the surveyed factors for plant life, soil and landscape and, moreover, on the forecasting for overall influences derived from the rotation upon the ecosystem.

• 의료ㆍ복지 건축 : 한국의료복지건축학회 논문집
• /
• 제23권4호
• /
• pp.27-36
• /
• 2017

Purpose: The characteristics of spatial allocation planning in dentistry through examining the dental hospitals in Korea and comparison between them are necessary for the development of planning of the dental healthcare system. This study has been started to provide basic informations such as zoning, allocation distribution, and space configuration for the planning of dental hospital architecture. Methods: Literature review of dental care departments and investigation on current status of dental hospital in Korea have been conducted. The spatial allocation and space configuration of eleven dental hospitals have been analyzed. Results: The result of this study can be summarized in three points. The first one is that dental hospitals in Korea are consisted with eight to eleven dental care departments and they are divided with the horizontal allocation type with three to four departments in a floor for the spatial communication or the vertical allocation type with a department in each floor for the independent space. The second one is that oral medicine and oral maxillofacial radiology are located near the main entrance, orthodontics and pedodontic dentistry in lower level, prosthodontics in upper level, and conservative dentistry and periodontics have no specific spatial consideration. The third one is that the factors to consider the allocation planning are zoning for examination & diagnosis, basic practice, adolescence, surgery, circulations for patient, dentist, staff, different access for department like as easy access for reception and pedodontic dentistry, enclosure space for prosthodontics and surgery, frequency of visit and treatment care time, and change of treatment concept from treatment department to disease control corporative practice. Implications: This study is the starting point for the research of spatial configuration in dentistry and it is necessary to analyze the architectural planning to develop the dental healthcare system.

• 한국방사선학회논문지
• /
• 제16권5호
• /
• pp.661-672
• /
• 2022

COVID-19를 발생시키는 SARS-CoV2 바이러스가 발생한 후 전염병은 전 세계로 확산되며, 감염 사례와 사망자의 수가 빠르게 증가함에 따라 의료자원의 부족 문제가 야기되었다. 이것을 해결하려는 방법으로 인공지능을 활용한 흉부 X-ray 검사가 일차적인 진단 방법으로 관심을 받게 되었다. 본 연구에서는 인공지능을 통한 COVID-19 판독 방식들에 대해 종합적으로 분석하는 것에 목적을 두고 있다. 이 목적을 달성하기 위해 292개의 논문을 일련의 분류 방법을 거처 수집했다. 이러한 자료들을 토대로 Accuracy, Precision, Area Under Curve(AUC), Sensitivity, Specificity, F1-score, Recall, K-fold, Architecture, Class를 포함한 성능 측정정보를 분석했다. 그 결과로 평균 Accuracy, Precision, AUC, Sensitivity, Specificity 값은 각각 95.2%, 94.81%, 94.01%, 93.5%, 93.92%로 도출되었다. 연도별 성능 측정정보는 점차 증가하는 값을 나타냈고 이 외에도 Class 수, 이미지 데이터 수에 따른 변화율, Architecture 사용 비율, K-fold에 관한 연구를 진행했다. 현재 인공지능을 활용한 COVID-19의 진단은 독자적으로 사용되기에는 여러 문제가 존재하지만, 의사의 보조수단으로써 사용됨에는 부족함이 없을 것으로 예상된다.

• The Korean Journal of Physiology and Pharmacology
• /
• 제24권1호
• /
• pp.89-99
• /
• 2020

Although microscopic analysis of tissue slides has been the basis for disease diagnosis for decades, intra- and inter-observer variabilities remain issues to be resolved. The recent introduction of digital scanners has allowed for using deep learning in the analysis of tissue images because many whole slide images (WSIs) are accessible to researchers. In the present study, we investigated the possibility of a deep learning-based, fully automated, computer-aided diagnosis system with WSIs from a stomach adenocarcinoma dataset. Three different convolutional neural network architectures were tested to determine the better architecture for tissue classifier. Each network was trained to classify small tissue patches into normal or tumor. Based on the patch-level classification, tumor probability heatmaps can be overlaid on tissue images. We observed three different tissue patterns, including clear normal, clear tumor and ambiguous cases. We suggest that longer inspection time can be assigned to ambiguous cases compared to clear normal cases, increasing the accuracy and efficiency of histopathologic diagnosis by pre-evaluating the status of the WSIs. When the classifier was tested with completely different WSI dataset, the performance was not optimal because of the different tissue preparation quality. By including a small amount of data from the new dataset for training, the performance for the new dataset was much enhanced. These results indicated that WSI dataset should include tissues prepared from many different preparation conditions to construct a generalized tissue classifier. Thus, multi-national/multi-center dataset should be built for the application of deep learning in the real world medical practice.

• KIEAE Journal
• /
• 제16권4호
• /
• pp.71-77
• /
• 2016

Purpose : The energy saving in a residential building (apartment) sector is known as one of the effective solution of energy reduction. In South Korea, the government has recently reinforced regulations associated with the energy performance of buildings. However, there is a lack of research on the methods for the energy performance diagnosis that is used to analyze the wall thermal performance of the existing apartments. Because a reliable diagnosis is necessary to save the building energy, this study analyzed wall thermal performance of an existing apartment in Seoul. Method : This paper applied two methods for analysis of the thermal insulation performance; HFM(Heat Flow Meter) method and ASTR(Air-Surface Temperature Ratio) method. The HFM method is suggested by ISO9869-1 code to measure the thermal performance. The ASTR method is proposed by this study for the simplified In-situ measurement and it uses three temperature data (interior wall surface, interior and exterior air) and the overall heat transfer coefficient. This study conducted the experiment of an existing apartment in Seoul using these methods and analyzed the results. Furthermore, the energy simulation tool of the building was used to suggest retrofit of the building based on the results of measurements. Result : The error rate of HFM method and ASTR method was analyzed in about 17 to 20%. As the results of comparison between the initial design values of the wall and the measured values, the 26% degradation of insulation thermal performance was measured. Lastly, the energy simulation tool of the building shows 10.8% energy savings in accordance with the construction of suggested retrofit.