• Clinical Nutrition Research
• /
• v.12 no.2
• /
• pp.126-137
• /
• 2023

Hemodialysis (HD) patients face a common problem of malnutrition due to poor appetite. This study aims to verify the appetite alteration model for malnutrition in HD patients through quantitative data and the International Classification of Functioning, Disability, and Health (ICF) framework. This study uses the Mixed Method-Grounded Theory (MMGT) method to explore various factors and processes affecting malnutrition in HD patients, create a suitable treatment model, and validate it systematically by combining qualitative and quantitative data and procedures. The demographics and medical histories of 14 patients were collected. Based on the theory, the research design is based on expansion and confirmation sequence. The usefulness and cut-off points of the creatinine index (CI) guidelines for malnutrition in HD patients were linked to significant categories of GT and the domain of ICF. The retrospective CIs for 3 months revealed patients with 3 different levels of appetite status at nutrition assessment and 2 levels of uremic removal. In the same way, different levels of dry mouth, functional support, self-efficacy, and self-management were analyzed. Poor appetite, degree of dryness, and degree of taste change negatively affected CI, while self-management, uremic removal, functional support, and self-efficacy positively affected CI. This study identified and validated the essential components of appetite alteration in HD patients. These MM-GT methods can guide the selection of outcome measurements and facilitate the perspective of a holistic approach to self-management and intervention.

• Progress in Medical Physics
• /
• v.27 no.3
• /
• pp.139-145
• /
• 2016

With the development in field of industry and medicine, new machines and techniques are being launched. Moreover, the complexity of the techniques is associated to an increasing risk of incident. Especially, a small error in radiotherapy can lead to a serious patient-related incident, risk management is necessary in radiotherapy in order to reduce the risk of incident. However, in field of radiotherapy, there are no legally binding clauses for risk management and there is an absence of risk management systems at an institutional level. Therefore, we analyzed institutional status of risk management, reporting & classification systems, and risk assessment & analysis in 31 countries. For risk management and reporting systems, 65% of countries investigated had legislation or regulations; however, only 35% of countries used classification systems. It was found that 43% more countries had legislation for risk management in healthcare than those for radiotherapy; 19% more countries had reporting systems for healthcare than those for radiotherapy. For classification systems, 60% more countries had legislation, recommendation, and guidelines in the field of radiotherapy than those for healthcare. Recently, international institutes have published several reports for risk management and patient safety in radiotherapy, owing to which, countries adopting risk management for radiotherapy will gradually increase. Before adopting risk management in Korea, we should precisely understand the procedures and functions of risk management, in order to increase efficiency of risk management because classification & reporting system and risk assessment & analysis are connected organically, and institutional management is needed for high quality of risk management in Korea.

• The Journal of Society for e-Business Studies
• /
• v.26 no.3
• /
• pp.97-117
• /
• 2021

Predicting term deposit subscriptions is one of representative financial marketing in banks, and banks can build a prediction model using various customer information. In order to improve the classification accuracy for term deposit subscriptions, many studies have been conducted based on machine learning techniques. However, even if these models can achieve satisfactory performance, utilizing them is not an easy task in the industry when their decision-making process is not adequately explained. To address this issue, this paper proposes an explainable scheme for term deposit subscription forecasting. For this, we first construct several classification models using decision tree-based ensemble learning methods, which yield excellent performance in tabular data, such as random forest, gradient boosting machine (GBM), extreme gradient boosting (XGB), and light gradient boosting machine (LightGBM). We then analyze their classification performance in depth through 10-fold cross-validation. After that, we provide the rationale for interpreting the influence of customer information and the decision-making process by applying Shapley additive explanation (SHAP), an explainable artificial intelligence technique, to the best classification model. To verify the practicality and validity of our scheme, experiments were conducted with the bank marketing dataset provided by Kaggle; we applied the SHAP to the GBM and LightGBM models, respectively, according to different dataset configurations and then performed their analysis and visualization for explainable term deposit subscriptions.

• Journal of Environmental Health Sciences
• /
• v.45 no.4
• /
• pp.370-380
• /
• 2019

Objective: This study was performed to investigate the distribution of individual exposure characteristics according to an exposure assessment classification for humidifier disinfectant and to identify the factors that influence assessment classification. Methods: We examined the exposure characteristics of 4,482 subjects who applied for the 4-1 and 4-2 assessments of environmental exposure to humidifier disinfectant conducted by the Korea Environmental Industry & Technology Institute (KEITI). Environmental exposure assessment classification was assessed using the following seven criteria: 1) Distance from humidifier to face; 2) Spray direction; 3) Time used, daytime 4) Time used, during sleep; 5) Time used, cumulative; 6) Exposure intensity; and 7) Cumulative exposure level. Each criteria was then classified as 'high' or low'. When participants answered for more than four criteria, exposure assessment was determined as 'definite,' 'probable,' or 'possible' depending on the ratio of 'high' responses. If participants' responses were inconsistent, exposure assessment was listed as 'unlikely.' If participants answered for less than four criteria, exposure assessment was considered 'indeterminate.' Results: For the exposure assessment classes, definite was assigned to 38.5% (1,725 subjects), probable assigned to 32.9% (1,474 subjects), 25.0% (1,122 subjects) were assigned to as possible, unlikely assigned to 0.1% (3 subjects), and indeterminate assigned to 3.5% (158 subjects). Overall, participants who used 'Oxy Ssakssak New Gaseupgi Dangbun,' 'Aekyung Gaseupgi Mate,' 'Homeplus Gaseupgi Chungjungje,' and 'E-Mart Gaseupgi Salgyunje' totaled 2,996, 557, 176, and 162 subjects, respectively. There was a statistical difference in the type of humidifier disinfectant products between high-exposed and low-exposed participants. Based on the assessment criteria of humidifier disinfectant exposure, subjects were likely to be in the highly exposed classes (definite and probable) when the subjects were exposed 1) for more than ten hours per day and 2) for more than four hours at night 3) when the total cumulative exposure time was higher than the average, 4) when the direction of humidifier spray was toward the face, 5) when the respiratory position was less than 1 meter of distance from the humidifier, 6) when the concentration of indoor contaminants (ug/m3) was higher than the average exposure intensity, and 7) when overall exposure level ($ug/m3^*hr$) was higher than the average exposure level. Conclusion: This study suggests that each exposure assessment criteria was able to appropriately estimate cumulative exposure levels.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.19 no.6
• /
• pp.96-106
• /
• 2020

To solve the limitation of traffic monitoring that occur from embedded sensor such as loop and piezo sensors, the thermal imaging camera was installed on the roadside. As the length of Class 1(passenger car) is getting longer, it is becoming difficult to classify from Class 3(2-axle truck) by using an embedded sensor. The collected images were labeled to generate training data. A total of 17,536 vehicle images (640x480 pixels) training data were produced. CNN (Convolutional Neural Network) was used to achieve vehicle classification based on thermal image. Based on the limited data volume and quality, a classification accuracy of 97.7% was achieved. It shows the possibility of traffic monitoring system based on AI. If more learning data is collected in the future, 12-class classification will be possible. Also, AI-based traffic monitoring will be able to classify not only 12-class, but also new various class such as eco-friendly vehicles, vehicle in violation, motorcycles, etc. Which can be used as statistical data for national policy, research, and industry.

• Korean Journal of Social Welfare Studies
• /
• v.42 no.2
• /
• pp.123-153
• /
• 2011

This study tries to categorize innovation types for social welfare organizations and investigate the level of innovation in each type in practical field. Firstly, this study scrutinizes the concept and classification's criterias of innovation. Secondly, this study reviews not only classification of innovation in profit organization but also several researches of innovation in service industry and public sectors, and finally, this study makes a suggestion of innovations' classification that is applicable for social welfare organizations. Based on this suggestion, fifteen questions are designed to ask the innovative activities in the organizations. And total 496 respondents from 116 organizations answered these questionnaire. The outcomes of this survey were substantiated by second data through converted procedures to mean value of organizations. Consquently, service innovation, administrative innovation and human resource innovation, proposed based on theoretical review, were subdivided into six categories such as service innovation, structural innovation, internal and efficiency innovation, marketing and communication innovation, external and employment innovation and evalution and mission innovation. The mean value of service(mean=14.7) and marketing innovation(mean=13.3) are higher than other type of innovations, which shows the aspect of innovative activities in social welfare organizations. Based on this result, we can get the directions of following study in investigating innovation of social welfare organization.

• KIPS Transactions on Software and Data Engineering
• /
• v.8 no.5
• /
• pp.193-204
• /
• 2019

The explosion of data due to the improvement of sensor technology and computing performance has become the basis for analyzing the situation in the industrial fields, and various attempts to detect events based on such data are increasing recently. In particular, sound signals collected from sensors are used as important information to classify events in various application fields as an advantage of efficiently collecting field information at a relatively low cost. However, the performance of sound-event classification in the field cannot be guaranteed if noise can not be removed. That is, in order to implement a system that can be practically applied, robust performance should be guaranteed even in various noise conditions. In this study, we propose a system that can classify the sound event after generating the enhanced sound signal based on the deep learning algorithm. Especially, to remove noise from the sound signal itself, the enhanced sound data against the noise is generated using SEGAN applied to the GAN with a VAE technique. Then, an end-to-end based sound-event classification system is designed to classify the sound events using the enhanced sound signal as input data of CNN structure without a data conversion process. The performance of the proposed method was verified experimentally using sound data obtained from the industrial field, and the f1 score of 99.29% (railway industry) and 97.80% (livestock industry) was confirmed.

• The Journal of Society for e-Business Studies
• /
• v.26 no.1
• /
• pp.1-28
• /
• 2021

FinTech has been discussed as an important business area towards technology-driven financial innovation. The term fintech is a combination of finance and technology, which means ICT technology currently associated with all finance areas. The popularity of the fintech industry has significantly increased over time, with full investment and support for numerous startups. Therefore, both academia and practice tried to analyze the trend of the fintech area. Despite the fact, however, previous research has limitations in terms of collecting relevant databases for fintech and identifying proper application areas. In response, this study proposed a new method for analyzing the trend of Fintech fields by expanding Fintech's terminology and using network analysis and topic modeling. A new Fintech terminology list was created and a total of 18,341 patents were collected from USPTO for 10 years. The co-classification analysis and network analysis was conducted to identify the technological trends of patent classification. In addition, topic modeling was conducted to identify the trends of fintech in order to analyze the contents of fintech. This study is expected to help both managers and investors who want to be involved in technology-driven financial services seize new FinTech technology opportunities.

• Journal of Navigation and Port Research
• /
• v.46 no.3
• /
• pp.269-279
• /
• 2022

With the development of artificial intelligence, artificial intelligence is being introduced to automate systems throughout the industry. In the maritime industry, artificial intelligence is being applied step by step, through the paradigm of autonomous ships. In line with this trend, ABS and DNV have published guidelines for autonomous vessels. However, there is a possibility that the risk of artificial intelligence has not been sufficiently considered, as the classification guidelines describe the requirements from the perspective of ship operation and marine service. Thus in this study, using the standards established by the ISO/ IEC JTC1/SC42 artificial intelligence division, classification requirements are classified as the causes of risk, and a measure that can evaluate risks through the combination of risk causes and artificial intelligence metrics want to use. Through the combination of the risk causes of artificial intelligence proposed in this study and the characteristics to evaluate them, it is thought that it will be beneficial in defining and identifying the risks arising from the introduction of artificial intelligence into the marine system. It is expected that it will enable the creation of more detailed and specific safety requirements for autonomous ships.

• Journal of Radiation Industry
• /
• v.17 no.1
• /
• pp.19-32
• /
• 2023

The amount of radioactive waste is expected to dramatically increase with decommissioning of nuclear power plants such as Kori-1, the first nuclear power plant in South Korea. Accurate nuclide analysis is necessary to manage the radioactive wastes safely, but research on verification of radionuclide analysis has yet to be well established. This study aimed to develop the technology that can verify the results of radionuclide analysis based on artificial intelligence. In this study, we propose an anomaly detection algorithm for inspecting the analysis error of radionuclide. We used the data from 'Updated Scaling Factors in Low-Level Radwaste' (NP-5077) published by EPRI (Electric Power Research Institute), and resampling was performed using SMOTE (Synthetic Minority Oversampling Technique) algorithm to augment data. 149,676 augmented data with SMOTE algorithm was used to train the artificial neural networks (classification and anomaly detection networks). 324 NP-5077 report data verified the performance of networks. The anomaly detection algorithm of radionuclide analysis was divided into two modules that detect a case where radioactive waste was incorrectly classified or discriminate an abnormal data such as loss of data or incorrectly written data. The classification network was constructed using the fully connected layer, and the anomaly detection network was composed of the encoder and decoder. The latter was operated by loading the latent vector from the end layer of the classification network. This study conducted exploratory data analysis (i.e., statistics, histogram, correlation, covariance, PCA, k-mean clustering, DBSCAN). As a result of analyzing the data, it is complicated to distinguish the type of radioactive waste because data distribution overlapped each other. In spite of these complexities, our algorithm based on deep learning can distinguish abnormal data from normal data. Radionuclide analysis was verified using our anomaly detection algorithm, and meaningful results were obtained.