• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.4
• /
• pp.826-842
• /
• 2024

As 5G and AI continue to develop, there has been a significant surge in the healthcare industry. The COVID-19 pandemic has posed immense challenges to the global health system. This study proposes an FL-supported edge computing model based on federated learning (FL) for predicting clinical outcomes of COVID-19 patients during hospitalization. The model aims to address the challenges posed by the pandemic, such as the need for sophisticated predictive models, privacy concerns, and the non-IID nature of COVID-19 data. The model utilizes the FATE framework, known for its privacy-preserving technologies, to enhance predictive precision while ensuring data privacy and effectively managing data heterogeneity. The model's ability to generalize across diverse datasets and its adaptability in real-world clinical settings are highlighted by the use of SHAP values, which streamline the training process by identifying influential features, thus reducing computational overhead without compromising predictive precision. The study demonstrates that the proposed model achieves comparable precision to specific machine learning models when dataset sizes are identical and surpasses traditional models when larger training data volumes are employed. The model's performance is further improved when trained on datasets from diverse nodes, leading to superior generalization and overall performance, especially in scenarios with insufficient node features. The integration of FL with edge computing contributes significantly to the reliable prediction of COVID-19 patient outcomes with greater privacy. The research contributes to healthcare technology by providing a practical solution for early intervention and personalized treatment plans, leading to improved patient outcomes and efficient resource allocation during public health crises.

• Atmosphere
• /
• v.34 no.3
• /
• pp.233-255
• /
• 2024

This study analyzed the distribution of Aerosol Robotic Network (AERONET) Version 3 Level 2.0 data, spanning over two decades, across South Korea and its six administrative regions (Seoul metropolitan area, Chungcheong, Jeolla, Gangwon, Gyeongsang, and Jeju). The research assessed long-term trends in aerosol optical depth (AOD) and mass concentration of particulate matter (i.e., PM₁₀ and PM_2.5), using data from the AERONET direct sun product and AirKorea, respectively. Additionally, eight aerosol types were identified using the scattering Ångström exponent and absorption Ångström exponent from the AERONET inversion product. The study further explored their domestic and regional distributions. Findings indicated that AERONET data were predominantly concentrated in the western regions of South Korea, including the Seoul metropolitan area, Chungcheong, and Jeolla, with a higher frequency of data in spring, thus demonstrating spatial and temporal heterogeneity. The annual average AOD exhibited a declining trend of -0.006 yr^-1. Similarly, PM₁₀ and PM_2.5 mass concentrations decreased by -1.324 ㎍ m^-3 yr^-1 and -1.335 ㎍ m^-3 yr^-1, respectively. These trends in AOD and PM₁₀ (PM_2.5) demonstrated positive correlations, with correlation coefficients of 0.674 (0.753) and statistically significant low p-values of 0.00058 (0.03), respectively. The analysis also revealed that aerosols in South Korea predominantly consisted of black carbon (BC) or BC-mixed types (84.09%), with a notable presence of smaller, less absorbent aerosol types (13.11%).

• Journal of the Korean Data and Information Science Society
• /
• v.19 no.3
• /
• pp.937-949
• /
• 2008

The purpose of this meta-analysis was to investigate the anti-hepatotoxic effect of ginseng in rats induced with CC14 or TCDD, the toxicities that cause liver damages. Primary studies were collected from the ScienceDirect database, the DBpia, and the KISS. The data on the effect factors in plasma and in enzyme are listed as many as possible: The effect factors were alanine transaminase(ALT), aspartate transaminase(AST), liver aminopyrine N-demethylase(AD), liver aniline hydroxylase(AH), liver 3,4-Methylenedioxyamphetamine(liver MDA), cytochrome P450(P450), serum alkaline phosphatase(ALP), serum lactate dehydrogenase(LDH), cytochrome b5(Cyto b5), glutathione reductase (GR), Liver glutathione S-transferase(GST), liver glutamyltransferase (GT), Liver($\gamma$-GCS), serum liver 3,4-Methylenedioxyamphetamine(serum MDA), serum sorbitol dehydrogenase(SDH), serum total protein(TP), and serum $\gamma$-glutamyltransferase($\gamma$-GT). In order to investigate the effect of ginseng, the standard mean difference(HG) between the group of rats induced with toxicity(RH) and the group of rats induced with ginseng(RHG) were combined, and the significance of HGs were tested. The combined HGs checked the biases caused by heterogeneity among studies and the publication biases. Then they were adjusted by using the random effect model and trim and fill method. Although the publication biases were assumed, among all plasma factors the HGs of ALT, AST, serum MDA, SDH, TP, and $\gamma$-GT were significant, and among all enzyme factors the HGs of liver MDA, Cyto b5, GR, GST, and GT were significant. The treatment with ginseng significantly affected the plasma and enzyme levels in rats induced with toxicity.

• Journal of Environmental Science International
• /
• v.22 no.1
• /
• pp.119-129
• /
• 2013

We compared equilibrium evaporation($E_{equili}$) eddy-covariance($E_{eddy}$) with soil moisture data($E_{SMseries}$) which were measured with a 2 hours sampling interval at three points for a humid forest hillslope from May 5th to May 31th in 2009. Accumulations of $E_{eddy}$, $E_{equili}$ for the study period were estimated as 2.52, 3.28 mm and those of $E_{SMseries}$ were ranged from 1.91 to 2.88 mm. It suggested that the eddy-covariance method considering the spatial heterogeneity of soil evaporation is useful to evaluate the soil evaporation. Method A, B and C were proposed using mean meterological data and daily moisture variation and the computations were compared to eddy-covariance method and equilibrium evaporation. The methods using soil moisture data can describe the variations of soil evaporation from eddy-covariance through simple moving average analysis. Method B showed a good matched with eddy-covariance method. This indicated that Dry Surface Layer (DSL) at 14:00 which was used for method B is important variable for the evaluation of soil evaporation. The total equilibrium evaporation was not significantly different to those of the others. However, equilibrium evaporation showed a problem in estimating soil evaporation because the temporal tendency of $E_{equili}$ was not related with the those of the other methods. The improved understanding of the soil evaporation presented in this study will contribute to the understandings of water cycles in a forest hillslope.

• Journal of Korea Technology Innovation Society
• /
• v.8 no.1
• /
• pp.135-153
• /
• 2005

With an explosive growth of internet and a change of consumer's demand, Post PC is emerging as a substitute for the existing desktop computer and laptop computer. Post PC is a new computer which is portable with wireless networking, cheaper than the existing PCs and specialized in the specific function that consumers want. Furthermore, the shape and the ability of Post PC are dependent on its specific function. The rudimentary Post PCs such as PDA, web-pad and smart-phone are already usable. And in the future, owing to the progress of technologies, the heterogeneity of relate companies and the low entry barrier, various Post PCs will be developed. In this paper, we analyze consumers' willingness-to-pay for Post PC with the attributes such as the portability, the speed of CPU, the data input method (using key board, or notepad etc), the size of monitor and price. To estimating consumers' willingness-to-pay for the attributes of Post PC, we apply a contingent ranking method, which makes the respondents rank hypothetical Post the portability, the speed of CPU, the data input method (using key board, or notepad etc),the size of monitor and price. To estimating consumers' willingness-to-pay for the attributes of Post PC, we apply a contingent ranking method, which makes the respondents rank hypothetical Post PC alternatives featuring various combinations of attributes, to a survey data collected in Korea. Using the estimated willingness-to-pay, we predict the shape and the ability of future's Post PC and draw policy implications for the national and company level R&D strategies.

• Research in Community and Public Health Nursing
• /
• v.31 no.2
• /
• pp.212-221
• /
• 2020

Purpose: This study aims to identify the types of health-related quality of life (QoL) based on the EuroQoL 5 Dimensions among community older people and predict the factors affecting these types. Methods: This study used data from the 2016 Korea Health Panel Survey, whose participants included 3,848 older people. The data were analyzed using the software jamovi 1.2.17 and Mplus 8.2 for latent class analysis. Results: The subgroups of the older people's health-related QoL were identified as three latent classes: General stable type (43.9%), pain-related low type (35.0%), and general low type (21.1%). The types and characteristics of health-related QoL among the latent classes differed. Comparing the difference between the general low type and general stable type, the subjects showed higher probability of belonging to the general stable type when they were men, younger, higher education level, employment, better subjective health, lower BMI and stress level, and no suicidal ideation. A comparison between the general low type and the pain-related low type showed that the subjects were more likely to be classified as the pain-related low type when they were younger, higher education, employment, and better subjective health. Conclusion: The results showed a significant heterogeneity in the types of health-related QoL among community older people, and the predictors for each type were not the same. These findings present basic data for cultivating nursing interventions that enhance health-related QoL.

• Journal of Internet Computing and Services
• /
• v.11 no.2
• /
• pp.41-48
• /
• 2010

It is essential to guarantee a smooth communication and data exchange in a PHD(Personal Healthcare Device) network for applications providing U-health services. In spite of that, most of PHDs are heterogeneous, so the heterogeneity of their protocols makes it difficult to develop an integrated gateway sending sensed healthcare data to U-health service providers. To solve this problem, we suggest the design and implementation of a device adapter model based on dynamic managed module in this paper. Our model were implemented to work on the OSGi-based gateway middleware and to have interoperability in connection with the HL7 system that is the standard of the Healthcare Information systems. In addition, our model has an architecture supporting a communication based on the object serialization in order to provide extensibility in the functional aspect of applications. Through the experiment on a test-bed which is an implementation of the device adapter module for electrocardiogram and blood-pressure/blood-sugar device having one channel, we have confirmed the accuracy of sensing and sending data.

• Journal of Genetic Medicine
• /
• v.13 no.1
• /
• pp.20-25
• /
• 2016

Purpose: Xeroderma pigmentosum (XP) is rare autosomal recessive genetic disorder of DNA repair in which the ability to repair damage caused by ultraviolet light is deficient. We reported the first molecularly confirmed Korean patient of XP by targeted exome sequencing. The prevalence of XP included all subtype and carrier frequency of XP-A the using public data were estimated for the first time in South Korea. Materials and Methods: We described a 4-year-old Korean girl with clinical diagnosis of XP. We performed targeted exome sequencing in the patient for genetic confirmation considering disease genetic heterogeneity and for differential diagnosis. We verified a carrier frequency of c.390-1G>C in XPA gene known as mutational hot spot using Korean Reference Genome Data Base. We estimated the period prevalence of all subtypes of XP based on claims data of the Health Insurance Review and Assessment Service in South Korea. Results: We identified homozygous splicing mutation of XPA (c.390-1G>C) in the patient. The carrier frequency of risk for XPA (c.390-1G>C) was relatively high 1.608 e-03 (allele count 2/1244). The prevalence of XP in South Korea was 0.3 per million people. Conclusion: We expect that c.390-1G>C is hot spot for the mutation of XPA and possible founder variant in South Korea. However, the prevalence in South Korea was extremely low compared with Western countries and Japan.

• The Journal of Information Systems
• /
• v.19 no.2
• /
• pp.23-36
• /
• 2010

The focus of this study is to analyze between South Korea and China in terms of diversity in Management Information Systems research. As a result, there is herding phenomenon regarding researches in both of the countries compared to other previous studies. There also seemed to be lack of in-depth study of basic theories and connection to related field of study considering the broad topics of management information, unlike other international journals that pursue diversity in control referencing various methodologies, analysis units and many citations. Therefore, both South Korea and China should stabilize theoretical base of MIS through securing independent field of MIS along with founding the fundamentals by searching in diversified fields, methodologies and analysis units with focused topics. The upshot here is that with respect to diversity in MIS from both countries, to alleviate herding phenomenon, enforcing accuracy of data collection and reducing convenience pursuit should be implemented. Also as criteria to select a topic, different kinds of consulting concepts and taking social issues into consideration that helps tool development and analysis power should be done. Through varied methodologies, tool development and analysis power should be assisted and analysis unit should be shifted to organization unit for Korea and team or individual for China only to augment the accessibility. As a limitation for the paper is that the data used in this analysis is secondary data. In addition, although time period used in both countries were the same, object of analysis had homogeneity in Korean case while Chinese one having heterogeneity from 19 different journals. For the future studies, multicultural comparison or time series analysis and their comparison in deeper approach with regard to object of analysis and methodologies can contribute to further MIS diversity.

• Journal of KIISE:Databases
• /
• v.31 no.4
• /
• pp.362-372
• /
• 2004

One of the most difficult problems in building a distributed GIS lies in the heterogeneity of spatial databases. In particular, positional mismatches between spatial databases, which arise due to several reasons, may incur incorrect query results. They result in unreliable outputs of query processing. One simple solution is to correct positional data in spatial databases at each site, according to the most accurate one. This solution is however not practical in cases where the autonomy of each database should be respected. In this paper, we propose a spatial query processing method without correcting positional data in each spatial database. Instead of correcting positional data, we dynamically transform a given query region or position onto each space where spatial objects of each site are located. Our proposed method is based on an elastic transformation method by using delaunay triangulation. Accuracy of this method is proved mathematically, and is confirmed by an experiment. Moreover, we implemented using common use database system for usefulness verification of this method.