신생아 집중치료를 받은 극소저출생체중아 치료성적의 시기에 따른 변화

김성신,김문희,심재원,고선영,이은경,장윤실,강이석,박원순,이문향,이상일,이흥재,Kim, Shung Shin,Kim, Moon Hee,Shin, Jae Won,Ko, Sun Young,Lee, Eun Kyung,Chang, Yun Sil,Kang, I Seok,Park, Won Soon,Lee, Mun Hyang,Lee, Sang Il,Lee, Heung J 대한소아청소년과학회 2002 Clinical and Experimental Pediatrics (CEP) Vol.45 No.7

목 적: 출생 체중 1,500 그램 미만의 극소저출생체중아(very low birth weight infant, VLBWI)의 치료성적은 신생아 집중치료술 향상여부를 모니터 하는 주요한 지표로 이용되고 있다. 이에 본 연구자들은 개원이래 최근 7년간 삼성서울병원 신생아 중환자실의 VLBWI 치료성적의 변화양상을 기간별로 관찰 분석하였다. 방 법: 1994년 10월부터 2000년 12월까지 삼성서울병원 신생아 중환자실로 생후 3일 이내 입원한 VLBWI 374명을 대상으로 기간을 나누어(I기 : 이미 성적이 보고된 1994. 10-1996. 9, II기 : 1996. 10-1998. 12, III기 : 1999. 1-2000. 12) 각 기별, 출생체중별, 재태기간별, VLBWI의 발생빈도와 생존율, 유병률 및 생존기간 등에 관해 의무기록지를 통한 후향적조사를 실시하였다. 생존율은(생존아/총출생 VLBWI)${\times}100$으 로 산출하였고 생존아는 신생아 중환자실에서 생존하여 퇴원한 경우로 정의하였다. 결 과 : VLBWI의 발생빈도는 기간별로 차이가 없었고(I기 : 1.3%, II기 : 1.5%, III기 : 1.4%), 생존율은 III기에 의미 있게 향상되었다. 출생체중별 생존율에서 III기에서 750 gm 미만군(vs II기)과, 1,250-1,499 gm 군(vs I기)의 향상이, 1,000-1,249 gm 군에서는 II기(vs I기)의 향상이 의미 있었고 최저출생 생존아는 I기 624 gm($26^{+5}$주), II기 667 gm($25^{+6}$주), III기 480 gm($26^{+2}$주)였다. 재태기간별 생존율에서 III기의 25-26주군과 29-30군이 I, II기에 비해 의미 있게 향상되었고 최저출생 재태기간의 생존례는 I기 26주(970 gm), II기 $23^{+5}$주(791 gm), III기 $24^{+1}$주(740 gm)였다. VLBWI의 주요 유병률 중 III도 이상의 심한 뇌실 내 출혈의 빈도가 III기 5%로 I기 13%, II기 10.5%에 비해 의미 있게 감소되었고, 전체 사망 중 7일 이전 사망률이 III기(15.4%)에 I기(55.5%)에 비해 의의 있게 감소되었다. 결 론 : 최근 VLBWI의 생존율이 주요 유병률의 증가 없이 현저히 향상되었고, 특히 26주 이하, 750 gm 미만의 초극소미숙아의 생존율 향상이 두드러졌다. Purpose : The outcomes of infants weighing less than 1,500 gm(very low birth weight infant : VLBWI) reflect recent progress in neonatal intensive care. In this study, we analyzed changes over time in survival rate and morbidity of VLBWIs during the past seven years. Methods : A retrospective review of medical records was analyzed for VLBWIs admitted to the neonatal intensive care unit of Samsung Medical Center within three days from birth. We compared the outcomes of previous corresponding data(period I : Oct. 1994 to Sept. 1996), with the outcomes of period II(Oct. 1996 to Dec. 1998) and period III(Jan. 1999 to Dec. 2000). Results : As shown in Tables 1 and 3, the distribution of birth weight, gestational age(GA), gender, and inborn admissions did not change during the 7-year study. The overall survival rate of VLBWI increased significantly over time(period I : 72% vs period III : 88.3%, P<0.05). Between period I and period II, the birth weight-specific survival rate increased by 23.6%(75% vs 92.7%, P<0.05) for infants 1,000 to 1,249 gm. Between period II and period III, the birth weight-specific survival rate increased three times(20% vs 66.7%, P<0.05) for infants <750 gm. The survivors of lowest birth weight included infants at 624 gm(GA : $26^{+5}$ weeks), 667 gm(GA : $25^{+6}$ weeks) and 480 gm(GA : $26^{+2}$ weeks) in each period. The gestational age-specific survival rate in period III increased significantly in GA 25-26 weeks and 29-30 weeks(vs period I & period II, P<0.05). The survivors of lowest gestational age included infants at GA 26 weeks(970 gm), GA $23^{+5}$ weeks(791 gm) and GA $24^{+1}$ weeks(740 gm) in each period. The incidence of severe IVH(grade III, IV) and the early death rate(< seven days) decreased in period III(vs period I; P<0.05), reflecting improvements in neonatal intensive care. Conclusion : The survival rate of VLBWI continues to increase, particularly for BW < 750 gm, GA < 26 weeks. This increase in survival is not associated by any increase in major morbidities.

A Neuro-Fuzzy Approach to Integration and Control of Industrial Processes:Part I

김성신,Kim, Sung-Shin Korean Institute of Intelligent Systems 1998 한국지능시스템학회논문지 Vol.8 No.6

This paper introduces a novel neuro-fuzzy system based on the polynomial fuzzy neural network(PFNN) architecture. The PFNN consists of a set of if-then rules with appropriate membership functions whose parameters are optimized via a hybrid genetic algorithm. A polynomial neural network is employed in the defuzzification scheme to improve output performance and to select appropriate rules. A performance criterion for model selection, based on the Group Method of DAta Handling is defined to overcome the overfitting problem in the modeling procedure. The hybrid genetic optimization method, which combines a genetic algorithm and the Simplex method, is developed to increase performance even if the length of a chromosome is reduced. A novel coding scheme is presented to describe fuzzy systems for a dynamic search rang in th GA. For a performance assessment of the PFNN inference system, three well-known problems are used for comparison with other methods. The results of these comparisons show that the PFNN inference system outperforms the other methods while it exhibits exceptional robustness characteristics.

앉은 자세에서 글쓰기 작업 시 구부정한 자세와 곧게 세운 자세에 따른 목과 체간 신전근의 근활성도 변화 비교

김성신(Kim Sung-Shin),김규용(Kim Gyu-Yong) 대한치료과학회 2016 대한치료과학회지 Vol.8 No.1

Objective: The purpose of this study was to investigate whether slump and upright posture affect on the neck and trunk range of motion, and these erector muscle activity during writing task. Method: All participants performed a writing task at the same desk and chair in the upright and slump sitting posture. Motion analysis system and surface electromyogram (EMG) were used to analyzing the neck and trunk motion and spine erector muscles activity. Paired t-tests were used to compare neck and trunk angles and both muscles activity. Results: There are significant differences in neck flexion-extension, lateral flexion, and rotation (p<.05). Trunk flexion, rotation was significantly greater in slump posture. In both cervical and lumbar erector muscles, there are significant differences in slump and upright posture (p<.05). Lumbar muscle symmetry ratio was significantly higher in slump than upright sitting posture (p<.05). Conclusion: Slump sitting posture may lead to more joint motion and muscle activity in neck and trunk joints and spine erector muscles. These findings suggest that maintain the upright posture would be more effectiveness to joints and muscles for prevent musculoskeletal disorders.

게임을 이용한 초등영어 문자언어 지도가 읽기·쓰기에 미치는 효과

김성신(Kim, Sung-Shin),부경순(Boo, Kyung-Soon) 한국초등영어교육학회 2009 초등영어교육 Vol.15 No.2

This study examines the effects of presenting written English through games. The subjects were fifty-six fifth graders at an elementary school in Incheon. The experimental group was shown written English through games while the control group received lessons focusing on spoken language in accordance with the English curriculum. To get the results of the study, written tests, questionnaires, and observation journals were used. The results showed that presenting written English had a positive effect on the experimental group. In terms of affective factors between the two groups, the experimental group had a more positive effect especially in the higher and middle levels. Most students participated actively in the games and the activities strengthened their relationships with the benefits of interest, suspense and cooperative spirit. The study leads to the conclusion that presenting written English through games is effective in the improvement of reading, writing and affective domains. Therefore, more written games and activities should be developed and used effectively to present written English. This would help students to have a positive attitude towards written English.

유전자 알고리즘을 이용한 IAWQ - Activated Sludge Model No.1의 Calibration과 불확실성 해석

김상현(Sang Hyun Kim),이현진(Hyun Jin Lee),김창원(Chang Won Kim),김성신(Sung Shin Kim),우혜진(Hye Jin Woo) 大韓環境工學會 2001 대한환경공학회지 Vol.23 No.3

확률 함수와 상황 전파 네트워크를 결합한 상황 데이터 생성 모델

천성표,김성신,Cheon, Seong-Pyo,Kim, Sung-Shin 한국정보통신학회 2009 한국정보통신학회논문지 Vol.13 No.7

일상적인 상황이 주로 시간을 고려한 일정에 따라서만 발생한다고 가정하면 각 상황별 확률분포함수에 기반한 데이터 생성법은 효과적이다. 하지만, 사람들은 상황을 인식하거나 판단할 때, 이전 상황을 고려하여 현재 상황을 미루어 짐작하거나 결정하는 경우도 흔하다. 본 논문에서 제안한 상황 전파 네트워크는 상황 인식 및 결정에서 이전 상황이 현재 상황에 미치는 영향을 고려하는 것과 유사하게 상황발생 확률분포함수를 보상해 줌으로써 상황의 전개가 연속적이면서 자연스러운 흐름을 갖도록 해준다. 본 논문에서는 평범한 직장인이 집이라는 공간에서 발생시키는 상황 데이터를 생성하는 모델을 제안하였고, 모의실험을 통해 상황 전파 네트워크가 상황의 자연스러운 흐름에 얼마나 영향을 기여하는지 확인하였으며, 특히, 순차적 상황 모호성과 충돌성 지표를 새롭게 정의하여 제안한 상황 데이터 생성 모델을 평가했다. 결론적으로 상황 전파 네트워크를 결합시켰을 경우, 확률 모델만을 사용했을 때보다 순차적 상황 모호성은 6.45% 상황 충돌성은 4.60% 감소함을 확인했다. Probabilistic distribution functions based data generation method is very effective. Probabilistic distribution functions are defined under the assumption that daily routine contexts are mainly depended on a time-based schedule. However, daily life contexts are frequently determined by previous contexts because contexts have consistency and/or sequential flows. In order to refect previous contexts effect, a situation propagation network is proposed in this paper. As proposed situation propagation network make parameters of related probabilistic distribution functions update, generated contexts can be more realistic and natural. Through the simulation study, proposed context-data generation model generated general outworker's data about 11 daily contexts at home. Generated data are evaluated with respect to reduction of ambiguity and confliction using newly defined indexes of ambiguity and confliction of sequential contexts. In conclusion, in case of combining situation propagation network with probabilistic distribution functions, ambiguity and confliction of data can be reduced 6.45% and 4.60% respectively.

주가지수 예측을 위한 퍼지 시스템과 지식 정보의 적용

김해균(Kim Hae-Gyun),배현(Bae Hyeon),김성신(Kim Sung-Shin) 제어로봇시스템학회 2001 제어로봇시스템학회 국제학술대회 논문집 Vol.2001 No.10

로봇 보조 보행시 이중과제 훈련이 만성뇌졸중환자의 균형 및 보행능력에 미치는 영향

김민규(Kim Min Kyu),정신호(Chung Shin Ho),김성신(Kim Sung Shin) 대한치료과학회 2016 대한치료과학회지 Vol.8 No.2

The purpose of this study was to investigate the effect of robot assisted gait training with dual task on balance and gait in patients with chronic stroke. Method: Nineteen in patients with chronic stroke selected by inclusion criteria. Participants were randomly allocated to two group: Dual task oriented robot assisted gait training(n=9) and robot assisted gait training(n=10). Two groups trained for three session thirty minutes per week during eight weeks. Gaitview, Berg balance scale(BBS), Timed Up and Go(TUG), 10 M Walking Test(10WT), Walkbot were used as evaluation tool. The data was analyzed independent sample t-test and repeated measure ANOVA on the SPSS 22.0 program for descriptive statistics. The level for statistical significance was set at p<.05 for all data. Results: There were significantly increased by dual task oriented robot assisted gait training that balance and gait ability(p<.05). There were significantly increased by robot assisted gait training that gait ability(p<.05). There were significantly increased by dual task oriented robot assisted gait training more than robot assist gait training that balance and gait ability(p<.05). Conclusion: Dual task oriented robot assisted gait training improves balance and gait ability for the chronic stroke patients.

영역 확장 기법과 오류 역전파 알고리즘을 이용한 자궁경부 세포진 영역 분할 및 인식

김광백,김성신,Kim Kwang-Baek,Kim Sung-Shin 한국정보통신학회 2006 한국정보통신학회논문지 Vol.10 No.6

자궁 경부 세포진 영상의 핵 영역 분할은 자궁 경부암 자동화 검색 시스템의 가장 어렵고도 중요한 분야로 알려져 있다. 자궁 경부 세포진 영상은 배경과 세포의 영역이 확실히 구분되지 않는 경우가 많기 때문에 이들을 확실히 구분하는 것이 매우 중요하다. 본 논문에서는 이러한 문제점을 해결하기 위해 자궁 경부 세포진 영상에서 Region growing 기법을 적용하여 세포 영상을 분할한다. Region growing 기법은 화소간의 유사도를 측정하여 영역을 확장하여 분할하는 방법이다. 세포와 배경이 분할된 영상을 일정 임계값을 이용하여 영상을 이진화 한 후, 8방향 윤곽선 추적 알고리즘을 이용해 세포 영역을 추출한다. 추출된 세포 영역을 원 영상인 RGB 컬러로 변환한 후에 K-means 알고리즘을 적용하여 각 세포 영역의 RGB 화소를 R, G, B 채널로 각각 분리하여 클러스터링 한다. 클러스터링된 각 각의 R, G, B 채널의 클러스터 값을 이용하여 HSI 모델로 변환시킨 후에 세포핵 영역의 Hue 정보를 추출한다. 추출된 세포핵의 특징을 오류 역전파 알고리즘을 적용하여 정상 세포와 비정상 세포를 분류하고 인식한다. The classification of the background and cell areas is very important research area because of the ambiguous boundary. In this paper, the region of cell is extracted from an image of uterine cervical cytodiagnosis using the region growing method that increases the region of interest based on similarity between pixels. Segmented image from background and cell areas is binarized using a threshold value. And then 8-directional tracking algorithm for contour lines is applied to extract the cell area. First, the extracted nucleus is transformed to RGB color that is the original image. Second, the K-means clustering algorithm is employed to classify RGB pixels to the R, G, and B channels, respectively. Third, the Hue information of nucleus is extracted from the HSI models that is the transformation of the clustering values in R, G, and B channels. The backpropagation algorithm is employed to classify and identify the normal or abnormal nucleus.

Selection of data set with fuzzy entropy function

이상혁(Sang-Hyuk Lee),천성표(Seong-Pyo Cheon),김성신(Sung-Shin Kim) 한국지능시스템학회 2004 한국지능시스템학회 학술발표 논문집 Vol.14 No.1

In this literature, the selection of data set among the universe set is carried out with the fuzzy entropy function. By the definition of fuzzy entropy, we have proposed the fuzzy entropy function and the proposed fuzzy entropy function is proved through the definition. The proposed fuzzy entropy function calculate the certainty or uncertainty value of data set, hence we can choose the data set that satisfying certain bound or reference. Therefore the reliable data set can be obtained by the proposed fuzzy entropy function. With the simple example we verify that the proposed fuzzy entropy function select reliable data set.