• Journal of Biomedical Engineering Research
• /
• v.43 no.4
• /
• pp.185-192
• /
• 2022

Early detection of mild cognitive impairment can help prevent the progression of dementia. The purpose of this study was to design and validate a machine learning model that automatically differential diagnosed patients with mild cognitive impairment and identified cognitive decline characteristics compared to a control group with normal cognition using resting-state quantitative electroencephalogram (qEEG) with eyes closed. In the first step, a rectified signal was obtained through a preprocessing process that receives a quantitative EEG signal as an input and removes noise through a filter and independent component analysis (ICA). Frequency analysis and non-linear features were extracted from the rectified signal, and the 3067 extracted features were used as input of a linear support vector machine (SVM), a representative algorithm among machine learning algorithms, and classified into mild cognitive impairment patients and normal cognitive adults. As a result of classification analysis of 58 normal cognitive group and 80 patients in mild cognitive impairment, the accuracy of SVM was 86.2%. In patients with mild cognitive impairment, alpha band power was decreased in the frontal lobe, and high beta band power was increased in the frontal lobe compared to the normal cognitive group. Also, the gamma band power of the occipital-parietal lobe was decreased in mild cognitive impairment. These results represented that quantitative EEG can be used as a meaningful biomarker to discriminate cognitive decline.

• Design & Manufacturing
• /
• v.16 no.4
• /
• pp.1-6
• /
• 2022

Fine blanking is a press processing technology that can process most of the product thickness into a smooth surface with a single stroke. In this fine blanking process, shear is an essential step. The punches and dies used in the shear are subjected to impacts of tens to hundreds of gravitational accelerations, depending on the type and thickness of the material. Therefore, among the components of the fine blanking mold (dies), punches and dies are the parts with the shortest lifespan. In the actual production site, various types of tool damage occur such as wear of the tool as well as sudden punch breakage. In this study, machine learning algorithms were used to predict these problems in advance. The dataset used in this paper consisted of the signal of the vibration sensor installed in the tool and the measured burr size (tool wear). Various features were extracted so that artificial intelligence can learn effectively from signals. It was trained with 5 features with excellent distinguishing performance, and the SVM algorithm performance was the best among 33 learning models. As a result of the research, the vibration signal at the time of imminent tool replacement was matched with an accuracy of more than 85%. It is expected that the results of this research will solve problems such as tool damage due to accidental punch breakage at the production site, and increase in maintenance costs due to prediction errors in punch exchange cycles due to wear.

• Plant Journal
• /
• v.18 no.4
• /
• pp.48-57
• /
• 2023

In this study, I discussed a way of the improved operation criteria to detect combustion instability in advance F-Class Gas Turbine, which adopts lean pre-mixed combustion system. The data of 16 blades path temperature thermocouple installed radially at the gas Turbine exit were collected to analyze the variation of individual blade path temperature. The cumulative variation in individual blade path temperature for one week under normal combustion conditions was confirmed to be up to 26℃. On the other hand, in the event of combustion instability, the symptoms of increased temperature variations in the individual thermocouple were mostly seen from a few days ago. Based on the results of this study, it is deemed appropriate to inspect and maintain in Ulsan Thermal Power Gas Turbine when the individual blade path temperature exceeds 50℃ of the cumulative variation for 10 days.

• Asia-pacific Journal of Multimedia Services Convergent with Art, Humanities, and Sociology
• /
• v.7 no.9
• /
• pp.317-325
• /
• 2017

In this paper, we try to classify Korean hedge sentences, which are regarded as not important since they express uncertainties or personal assumptions. Through previous researches to English language, we found dependency information of words has been one of important features in hedge classification, but not used in Korean researches. Additionally, we found that word embedding vectors include the word usage information. We assume that the word usage information could somehow represent the dependency information. Therefore, we utilized word embedding and neural networks in hedge sentence classification. We used more than one and half million sentences as word embedding dataset and also manually constructed 12,517-sentence hedge classification dataset obtained from online news. We used SVM and CRF as our baseline systems and the proposed system outperformed SVM by 7.2%p and also CRF by 1.2%p. This indicates that word usage information has positive impacts on Korean hedge classification.

• Journal of Software Assessment and Valuation
• /
• v.15 no.2
• /
• pp.121-128
• /
• 2019

Computing systems have various vulnerabilities to cyber attacks. In particular, various cyber attacks that are intelligent in the information society have caused serious social problems and economic losses. Traditional security systems are based on misuse-based technology, which requires the continuous updating of new attack patterns and the real-time analysis of vast amounts of data generated by numerous security devices in order to accurately detect. However, traditional security systems are unable to respond through detection and analysis in real time, which can delay the recognition of intrusions and cause a lot of damage. Therefore, there is a need for a new security system that can quickly detect, analyze, and predict the ever-increasing cyber security threats based on machine learning and big data analysis models. In this paper, we present a IDS model that combines machine learning and big data technology.

• Journal of the Korea Society of Computer and Information
• /
• v.29 no.2
• /
• pp.87-99
• /
• 2024

This study aims to identify the competencies required for webtoon artists to survive in the future era of AI commercialization. It explores the current and future use of AI in webtoons, and predicts the role of artists in the future webtoon industry. The study finds that AI will replace human workers in some areas, but human empathy-related fields can be sustained. Artist roles like story projectors, Visual directors, and AI editors were identified as potential models for the changing role of artists. To address terminology ambiguity, a three-step AI categorization mechanical type AI, humanoid type AI, and transcendent type AI was proposed for a more realistic separation of AI capabilities. The researcher suggested these findings as guidelines for developing skills in emerging artists or re-skilling existing ones, emphasizing collaboration with AI for mutual growth rather than a negative acceptance of new technology.

• Journal of Korea Society of Industrial Information Systems
• /
• v.9 no.4
• /
• pp.56-67
• /
• 2004

This study analyzes human error of workers performing simple repetitive tasks, and in order to prepare preventative measures, 486 people were used as subjects. The results of the study are like the following. First, the biggest cause of human error showed to be the worker himself in $77.8\%$ of the cases, machinery showed to be the cause in $16.3\%$ of the cases and management showed to be the cause in $6.0\%$ of the cases. The results show that most of the human error occurred due to the worker performing simple repetitive tasks and the human errors showed to be caused more by bad ergonomics and long hours rather than by problems with machinery. In addition, the area with the highest rate of human error showed to be the Human Information Processing System with Task Input Error being the highest with $46.9\%$, followed by Judgement and Memory Error with $36.4\%$ and Recognition Verification Error with $16.7\%$. Although fully automated tasks may reduce the rate of human error we must focus on lowering the rate of problems arising from spontaneous errors caused by workers performing simple repetitive tasks by continuously renewing plans and budgets in order to standardize tasks by incorporating cyclic positioning according to experience and positional exchange and by inspecting the workplace to increase efficiency of the workers.

• Tuberculosis and Respiratory Diseases
• /
• v.40 no.1
• /
• pp.29-34
• /
• 1993

Background: The development of the flexible fiberoptic broncoscope by Ikeda was an important technologic advance in the diagnosis and management of patients with pulmonary disease. But, cross contamination related to fiberoptic bronchoscope was reported in cases involving tubercle bacilli, MOTT and other agents. Therefore, cleaning and disinfecting of fiberoptic bronchoscope requires careful attention. Methods: From September 1991 to May 1992, medical records of all patients with positive culture for MOTT in bronchial washing specimens were reviewed. Also to evaluate bactericidal effect of 2% glutaraldehyde, culture was performed after inoculum of MOTT, Serratia marsescens and Pseudomonas aeruginosa to the disinfectant solution. Results: In 2% alkaline glutaraldehyde, MOTT was not survived only after 30 minute exposure, but P. aeruginosa and S. marsescens were rapidly inactivated with no survivors after exposure to 2% glutaraldehyde. Since vigorous mechanical cleansing and more than 30 minute of contact time within washing machine, no more outbreak was observed. Conclusions: It is also very important that bronchoscopes must be meticulously cleaned after each procedure and more than 30 minute exposure would be required for eradication of MOTT with 2% glutaraldehyde. However even the most strictly applied infection control measures cannot exclude contamination completly and clinicians have to stay alert to this possibility. Prompt detection of pseudoepidemics is possible if abrupt increase in isolation rates, especially if they involve unusual or generally nonpathogenic organisms, are readily recognized.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.41 no.9
• /
• pp.825-837
• /
• 2017

This paper describes a novel type of a walking rehabilitation robot that applies robot technologies to crutches used by patients with walking difficulties in the lower body. The primary features of the developed robot are divided into three parts. First, the developed robot is worn on the patient's chest, as opposed to the conventional elbow crutch that is attached to the forearm; hence, it can effectively disperse the patient's weight throughout the width of the chest, and eliminate the concentrated load at the elbow. Furthermore, it allows free arm motion during walking. Second, the developed robot can recognize the walking intention of the patient from the magnitude and direction of the ground reactive forces. This is done using three-axis force sensors attached to the feet of the robot. Third, the robot can perform a stair walking function, which can change vertical movement trajectories in order to step up and down a single stair according to the floor height. Consequently, we experimentally showed that the developed robot can effectively perform walking rehabilitation assistance by perceiving the walking intention of the patient. Moreover we quantitatively verified muscle power assistance by measuring the electromyography (EMG) signals of the muscles of the lower limb.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.36 no.4
• /
• pp.457-463
• /
• 2012

The basilar membrane (BM), one of organs of cochlea, has the specific positions of the maximum amplitude at each of related frequencies. This phenomenon is due to the geometry of BM. In this study, as the part of the research for the development of fully implantable artificial cochlea which is based on polymer membrane, parametric studies are performed to suggest the desirable artificial basilar membrane model which can detect wider range of frequency separation. The vibro-acoustic characteristics of the artificial basilar membrane are predicted through finite element analysis using commercial software Abaqus. Simulation results are verified by comparing with experimental results. Various geometric shapes of the BM and residual stress effects on the BM are investigated through the parametric study to enable a wider detectable frequency separation range.