• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.18 no.12
• /
• pp.581-588
• /
• 2017

This study was conducted to determine which factors would predict metabolic syndrome (MetS) perception and exercise by applying a machine learning classifier, or Extreme Gradient Boosting algorithm (XGBoost) from July 2014 to December 2015. Data were obtained from the Korean Community Health Survey (KCHS), representing different community-dwelling Korean adults 19 years and older, from 2009 to 2013. The dataset includes 370,430 adults. Outcomes were categorized as follows based on the perception of MetS and physical activity (PA): Stage 1 (no perception, no PA), Stage 2 (perception, no PA), and Stage 3 (perception, PA). Features common to all questionnaires for the last 5 years were selected for modeling. Overall, there were 161 features, categorical except for age and the visual analogue scale (EQ-VAS). We used the Extreme Boosting algorithm in R programming for a model to predict factors and achieved prediction accuracy in 0.735 submissions. The top 10 predictive factors in Stage 3 were: age, education level, attempt to control weight, EQ mobility, nutrition label checks, private health insurance, EQ-5D usual activities, anti-smoking advertising, EQ-VAS, education in health centers for diabetes, and dental care. In conclusion, the results showed that XGBoost can be used to identify factors influencing disease prevention and management using healthcare bigdata.

• Journal of agricultural medicine and community health
• /
• v.6 no.1
• /
• pp.13-24
• /
• 1981

This study was conducted with 478 farm households located in Pyongtaek Gun of Gyonggi Do, Chongju City and Chongwon Gun of Chung Buk Do, and Hongchon Gun of Gangwon Do for the period of 10, September through 20, October 1980. It dealt with general features of the farmhouseholds, their practices of handling and using pescides, and relationship between safe use of pesticides and poisoning experiences. The results of the study are summarized as follows: 1. Of the total, 63.2% purchased pesticides mainly through the Agricultural Coops, and 95.4% transported them home separately from other goods. 2 Pesticides were sold contained mainly in glass bottles and paper bags. 3. Mixing of the respodents (87.5%) stored pesticides in sheds, warehouses and boxes. 4. Mixing containers and measuring tools were not used by the majority of the group studied. This indicates that there exist serious problems of safety and accuracy with regard to use of pesticides. 5. As to pesticide protective devices, gloves, respirators, goggles and boots were not used in general. Even in case of using them, their quality was not satisfactory. 6. Among the interviewees, 4.2% did not read pesticide label, 33.9% did not wash immediately when pesticide was spilled on the skin, 48.4% did not care the wind direction during pesticide application, and 17.0% did not wash hands when smoking a cigarette while pesticide is being handled. 7. Of the total, 40.4% were found to harvest vegetables within 10 days from the date of pesticide application, which indicates their unawareness of possible hazards by pesticide residues. 8. Empty pesticide containers or bags were not properly disposed of by majority of the respondents. Furthermore, impacts on the environment by the misuse or abuse of pesticides were not understood by 67.6% 9. Of the interviewees, 48.7% were found to have not received instructions on safe use of pesticides. 10. The rate of pesticidal poisoning experience was about 28.0% if minor symptoms excluded, and it rose to about 44.0% if included. Meanwhile, the rate of pesticidal poisoning tends to increase with the size of farm. 11. The study failed to show a statistically significant relationship between the rate of pesticidal poisoning and use of mixing container and/or measuring tools. However, use of gloves showed a statistical significance on the rate of pesticidal poisoning. 12. Among the poisoned cases, 19.0% were cared for in either hospitals or clinics.

• Journal of Korean Society of Transportation
• /
• v.22 no.6
• /
• pp.121-131
• /
• 2004

The K-shortest path algorithms are largely classified into two groups: oneis for finding loopless path (simple path), another loop paths. In terms of cimputational complexities, in general the loop-paths-finding ones are considered more efficient and easier to be handled than the loopless-paths-finding. The entire path deletion methods have been known as the best efficient algorithms among the proposed K-shortest path algorithms. These algorithms exploit the K-th network transformation to prevent the same path, which was already selected as the (K-1)th path, from being redetected. Nevertheless, these algorithms have a critical limitationto be applied in the practical traffic networks because the loops, in which the same modes and links can be unlimitedly repeated, are not preventable. This research develops a way to be able to selectively control loop-paths by applying link-label. This research takes an advantage of the link-based shortest path algorithms that since the algorithms can take care of two links simultaneouslyin the searching process, the generation of loops can be controlled in the concatenation process of the searched link and the preceded link. In concatenation of two links, since the precede link can be treated a sub-shortest to this link from the origination, whether both the node and the link of the searched link were already existed or not can be evaluated. Terefore, both the node-loopless path, in which the same node is not appeared, and the link-loopless, in which the same link is not appeared, can be separately controlled. Especially, the concept of the link-loopless path is expended to take into consideration reasonable route choice behaviors such as U-Turn, P-Turn, and Turn-Penalty, which are frequently witnessed in urban traffic network with intersections. The applicability of the proposed method is verified through case studies.

• Korean Journal of Acupuncture
• /
• v.31 no.1
• /
• pp.5-13
• /
• 2014

Objectives : This study aims to evaluate the effectiveness and safety of acupuncture and explore the appropriate number of treatment for postmenopausal women diagnosed with prehypertension and stage 1 hypertension. Methods : A 4-arm randomized open label pilot trial will be performed at 2 centers. Sixty participants will be divided into 2 treatment groups and 2 control groups. Treatment groups will receive acupuncture at 8 points(bilateral GB20, LI11, ST36, SP6) for 4 weeks(treatment group A, 10 total sessions) or 8 weeks(treatment group B, 20 total sessions), while maintaining usual care. Control groups will not receive acupuncture but will be under usual care for 16 weeks(control group C) or 20 weeks(control group D). Each patient's living habits will be corrected and drugs that may affect blood pressure(BP) will be prohibited. Treatment group A and control group C will be evaluated at 4, 8, 12, and 16 weeks after randomization, while treatment group B and control group D will be evaluated at 4, 8, 12, 16, and 20 weeks after randomization. The major outcome variable is the magnitude of change in diastolic BP levels at 4 weeks after randomization; auxiliary outcome variables are (1) diastolic BP change at 8, 16, and 20 weeks, (2) systolic BP change, (3) BP control rate, (4) lipid profiles, and (5) high-sensitivity C-reactive protein. Patient safety will be assessed at every visit. Results and Conclusions : The study findings may help develop evidence for the effectiveness and safety of acupuncture for BP control.

• Pediatric Infection and Vaccine
• /
• v.20 no.3
• /
• pp.139-146
• /
• 2013

Objective: As per the requirement of Korean Food and Drug Administration, this post-marketing surveillance was conducted in Korea to evaluate the safety and reactogenicity of Poliorix$^{TM}$ following its introduction in 2006. Methods: In this open, multicenter study, the vaccine was administered as per the current practice of Korean doctors and in reference to the guidebook by the Korean Pediatric Society and as indicated in the Korean label which was as follows - for primary vaccination three doses were given to infants at ages 2, 4 and 6 months whereas, for the booster dose a single dose was given to children aged 4-6 years. Safety data during this six year surveillance was collected using diary cards which were distributed to the parents to record adverse events. Results: A total of 639 subjects were enrolled into the study. Of these, 617 subjects and 22 subjects received the vaccine as a primary and booster dose, respectively. At least one unsolicited symptom was reported in 11.4% (73/639) of the subjects during the 7-day follow-up period; upper respiratory tract infection (2.5%;16/639) was the most frequently reported unsolicited symptom. One subject reported at least one unsolicited symptom (gastroenteritis) of grade 3 intensity within the 31-day post-vaccination period. Approximately 1.7% (11/639) of subjects reported 13 serious adverse events (SAEs). All SAEs were resolved by the end of the study. Conclusion: In Korea, primary and booster vaccination with Poliorix$^{TM}$ was well-tolerated in healthy subjects when administered according to the prescribing information as part of routine clinical practice.

• Journal of Korean Medical Science
• /
• v.33 no.52
• /
• pp.346.1-346.11
• /
• 2018

Background: To evaluate the therapeutic benefits of the treat-to-target (T2T) strategy for Asian patients with early rheumatoid arthritis (RA) in Korea. Methods: In a 1-year, multicenter, open-label strategy trial, 346 patients with early RA were recruited from 20 institutions across Korea and stratified into 2 groups, depending on whether they were recruited by rheumatologists who have adopted the T2T strategy (T2T group) or by rheumatologists who provided usual care (non-T2T group). Data regarding demographics, rheumatoid factor titer, anti-cyclic citrullinated peptide antibody titer, disease activity score of 28 joints (DAS28), and Korean Health Assessment Questionnaire (KHAQ) score were obtained at baseline and after 1 year of treatment. In the T2T group, the prescription for disease-modifying antirheumatic drugs was tailored to the predefined treatment target in each patient, namely remission (DAS28 < 2.6) or low disease activity (LDA) ($2.6{\leq}DAS28$ < 3.2). Results: Data were available for 163 T2T patients and 162 non-T2T patients. At the end of the study period, clinical outcomes were better in the T2T group than in the non-T2T group (LDA or remission, 59.5% vs. 35.8%; P < 0.001; remission, 43.6% vs. 19.8%; P < 0.001). Compared with non-T2T, T2T was also associated with higher rate of good European League Against Rheumatism response (63.0% vs. 39.8%; P < 0.001), improved KHAQ scores (-0.38 vs. -0.13; P = 0.008), and higher frequency of follow-up visits (5.0 vs. 2.0 visits/year; P < 0.001). Conclusion: In Asian patients with early RA, T2T improves disease activity and physical function. Setting a pre-defined treatment target in terms of DAS28 is recommended.

• Journal of Intelligence and Information Systems
• /
• v.28 no.2
• /
• pp.127-146
• /
• 2022

Recently, as word embedding has shown excellent performance in various tasks of deep learning-based natural language processing, researches on the advancement and application of word, sentence, and document embedding are being actively conducted. Among them, cross-language transfer, which enables semantic exchange between different languages, is growing simultaneously with the development of embedding models. Academia's interests in vector alignment are growing with the expectation that it can be applied to various embedding-based analysis. In particular, vector alignment is expected to be applied to mapping between specialized domains and generalized domains. In other words, it is expected that it will be possible to map the vocabulary of specialized fields such as R&D, medicine, and law into the space of the pre-trained language model learned with huge volume of general-purpose documents, or provide a clue for mapping vocabulary between mutually different specialized fields. However, since linear-based vector alignment which has been mainly studied in academia basically assumes statistical linearity, it tends to simplify the vector space. This essentially assumes that different types of vector spaces are geometrically similar, which yields a limitation that it causes inevitable distortion in the alignment process. To overcome this limitation, we propose a deep learning-based vector alignment methodology that effectively learns the nonlinearity of data. The proposed methodology consists of sequential learning of a skip-connected autoencoder and a regression model to align the specialized word embedding expressed in each space to the general embedding space. Finally, through the inference of the two trained models, the specialized vocabulary can be aligned in the general space. To verify the performance of the proposed methodology, an experiment was performed on a total of 77,578 documents in the field of 'health care' among national R&D tasks performed from 2011 to 2020. As a result, it was confirmed that the proposed methodology showed superior performance in terms of cosine similarity compared to the existing linear vector alignment.