[시와사람_이선미_선정작] 미운 꽃 외 1편

이선미 고요아침 2016 열린시학 Vol.21 No.3

[시와사람_이선미_신작] 물 소리를 듣다

이선미 고요아침 2016 열린시학 Vol.21 No.3

시와사람_이선미_선정소감

이선미 고요아침 2016 열린시학 Vol.21 No.3

보건의료서비스연구에서의 베이시안 네트워크 기법 활용

이선미 대한의료정보학회 2006 Healthcare Informatics Research Vol.12 No.1

Objective: To explore the feasibility of using the Bayesian network approach to study health outcomes and evaluate its predictive performance. Methods: The Human immuno-deficiency virus Cost and Services Utilization Study (HCSUS) baseline dataset consisting of 2,864 human immuno-deficiency virus positive adults was used. The Hugin Researcher 6.2TM was used to develop the Bayesian network and Na ve Bayes models. The SAS/STAT PROC LOGISTIC was used to develop the logistic regressions.Results: The area under the receiver operating characteristic curve of the Bayesian network model was statistically higher than that of the Na ve Bayes model, but no higher than that of the logistic regression model using the 8 variables from a previous study. In a second analysis using the 10 most influential predictors discovered by the Bayesian network approach, the Na ve Bayes and the logistic regression performance improved. Conclusion: The BN approaches contributed to the discovery of additional influential predictors that lead to an increase of the models' predictive performance. When attempting to discover unknown relationships that might be missed by traditional analysis methods alone, the use of the Bayesian network as complementary methods may add value. (Journal of Korean Society of Medical Informatics 12-1,71-81, 2006)

HIV 환자 입원 예측 연구를 위한 베이시안 네트워크 활용

이선미 대한의료정보학회 2004 Healthcare Informatics Research Vol.10 No.3

Objective: The purpose of this study was to explore the potential application of a Bayesian network, an emerging data mining technique, in predicting outcomes using large healthcare databases. Methods: The HIV Cost and Services Utilization Study(HCSUS) dataset, consisting of 2,864 HIV positive adults in the US, was used. A total of 35 variables were selected including one output variable defined as more than one hospitalization in six months representing a sub-optimal pattern of healthcare utilization in HIV care. The HUGIN Researcher 6.2 was used to develop a Bayesian network model with two learning algorithms: 1) Necessary Path Condition(NPC) to construct a Bayesian network structure, and 2) Expectation-Maximization(EM) algorithm to estimate parameters. Results: The area under the Receiver Operating Characteristic(ROC) curve was .72. The accuracy of the prediction model was .66. Sensitivity and specificity were .65 and .66, respectively.Conclusion: The Bayesian network showed fair performance in this prediction problem. This study provides researchers new insight into working with large sets of data, which continue to grow in a number of cases and variables. The repeated testing and refinement of the Bayesian network modeling techniques with other health outcomes in large databases is recommended.

Study on the diagnosis of disturbed forest ecosystem in the Republic of Korea: in case of Daegwallyeong and Chupungryeong

이선미,차재규,문호경 한국생태학회 2017 Journal of Ecology and Environment Vol.41 No.7

Background: Baekdudaegan was designated in 2005 as a protected area to prevent destruction and conserve. However, there are many disturbed and destroyed areas. The total disturbed area amounts to 25.9 km2 (0.94%), including 13.4 km2 (0.49%) in the core area and 12.5 km2 (0.45%) in the buffer area. This study aims to classify the vegetation types established in the disturbed areas and diagnose the current conditions for ecological restoration in the forest ecosystem. Methods: We surveyed the vegetation in the disturbed areas of Daegwallyeong and Chupungryeong and the surrounding natural areas. The survey conducted from July to September 2015 targeted a total of 54 quadrats by Braun-Blanquet method (Daegwallyeong, 22; Chupungryeong, 32). We also investigated the height and coverage of each layer. We classified the vegetation types based on the field data and analyzed the ratio of life form and the exotic plants, species richness, and vegetation index (Hcl). The Normalized Difference Vegetation Index (NDVI) was calculated from rapideye satellite imagery in 2014 and 2015. Results: Vegetation types were classified into 11 groups according to the criteria that included successional sere or plantation at first, followed by developmental stage and origins. As a result of the analysis of the survey data, species richness, vegetation index (Hcl), ratio of tree plants, and the NDVI tended to increase, while the ratio of the exotic plants tended to decrease with the time since disturbance. These indicators had the classified values according to the vegetation types with time since the disturbance. Conclusions: These indicators can be effectively used to diagnose the conditions of the present vegetation in the disturbed area of the Baekdudaegan area. In addition, the NDVI might be effective for the diagnosis of the disturbed status instead of the human efforts based on the higher spatial resolution of satellite imagery. Appropriate diagnosis of the disturbed forests in the Baekdudaegan area considering the established vegetation types is essential for the elaboration of restoration plans. In addition, restoration target and level should be different according to the disturbed status of restoration site.

임상에서의 데이터 마이닝 개념과 원칙

이선미,박래웅 대한의료정보학회 2009 Healthcare Informatics Research Vol.15 No.2

Recently, many hospitals have been adopting clinical data warehouses (CDW) as well as electronic medical records. These new hospital information systems are inevitably introducing very large amounts of clinical data that might be useful for further analysis. However, the electronic clinical data in the CDW are usually byproducts of clinical practice rather than the product of research. Therefore, they include inconsistent and sometimes erroneous information that might not have the specific context of the clinical situations. Data miners usually have various academic backgrounds such as electronics, informatics, statistics, biomedicine, and public health. If the complex situations surrounding the clinical data are not well understood, investigators performing data mining in clinical fields may have problems assessing the information they are confronted with. Here, we would like to introduce some basic concepts on the principles of data mining in clinical fields including legal and ethical considerations as well as technical concerns.

Dyslipidemia is Positively Associated with Cervical Cancer in Korea: Korean National Health and Nutrition Examination Survey 2010-2020

이선미,이상훈,조현웅,민경진,홍진화,송재윤,이재관,이낙우 대한산부인과학회 2022 대한산부인과학회 학술대회 Vol.108 No.-

Objective: Several previous papers have reported that hypoxic environments induced by hypertension, Insulin like growth factor-1 (IGF-1) overexpressed by hyperinsulinemia, hyperactivity of fatty acid metabolism, stimulation of cell growth signaling factors, and formation of carcinogen by dyslipidemia contribute to tumor cell proliferation. Therefore, we hypothesized that hypertension, diabetes, and dyslipidemia are positively correlated with cervical cancer; the purpose of this study was to evaluate this correlation. Methods: We included 25,055 participants from the Korean National Health and Nutrition Examination Survey between 2010 and 2020. The participants were divided into the non-cervical cancer and cervical cancer groups. For comparisons between the two groups, continuous variables were analyzed using Student’s t-test, and categorical variables were analyzed using the chi-squared test. The odds ratios (ORs) for cervical cancer were obtained using univariate logistic regression, and meaningful confounding variables were adjusted using multivariate logistic regression. Results: A total of 24,744 females were included in the non-cervical cancer group, and 311 females were included in the cervical cancer group. In the cervical cancer group, dyslipidemia had an OR of 2.920 compared with that of the non-cervical cancer group (OR: 2.920, CI: 1.910-4.464, p <0.001) after adjusting for confounding variables in the multivariate logistic regression analysis. Conclusion: Among the three major chronic diseases, dyslipidemia was positively correlated with cervical cancer. Therefore, improving dyslipidemia might be associated with lowering the risk of cervical cancer later in life.

단편소설에 드러난 현실인식 연구 - 해방전을 중심으로

이선미 상허학회 1993 상허학보 Vol.1 No.-

치과병의원에 근무하는 치과위생사의 의사소통능력

이선미,윤정원,이정화 한국치위생학회 2020 한국치위생학회지 Vol.20 No.3

Objectives: The purpose of this study was to recognize the importance of dental hygienist's communication competence and to establish the basis for developing a program to improve communication competence. Methods: A survey was conducted on members who underwent participatory training in some trials in 2019, and a total of 348 people were analyzed. Results: Total communication competence was 3.63 points. The highest area was 3.88 points for understanding others' standards, and the lowest area was 3.27 points for escaping stereotypes. Most areas of communication showed statistically significant positive correlations, but in some areas, there was an inverse correlation. As a dental hygienist, it is necessary to provide systematic education for improving communication competence from college students. Conclusions: In addition, it is necessary to develop and implement a training program for improving communication competence of dental hygienists.