• Bulletin of the Korean Chemical Society
• /
• 제26권12호
• /
• pp.2007-2016
• /
• 2005

An artificial neural network (ANN) is successfully presented for prediction acidity constant (pKa) of various benzoic acids and phenols with diverse chemical structures using a nonlinear quantitative structure-property relationship. A three-layered feed forward ANN with back-propagation of error was generated using six molecular descriptors appearing in the multi-parameter linear regression (MLR) model. The polarizability term $(\pi_1)$, most positive charge of acidic hydrogen atom $(q^+)$, molecular weight (MW), most negative charge of the acidic oxygen atom $(q^-)$, the hydrogen-bond accepting ability $(\epsilon_B)$ and partial charge weighted topological electronic (PCWTE) descriptors are inputs and its output is pKa. It was found that properly selected and trained neural network with 205 compounds could fairly represent dependence of the acidity constant on molecular descriptors. For evaluation of the predictive power of the generated ANN, an optimized network was applied for prediction pKa values of 37 compounds in the prediction set, which were not used in the optimization procedure. Squared correlation coefficient $(R^2)$ and root mean square error (RMSE) of 0.9147 and 0.9388 for prediction set by the MLR model should be compared with the values of 0.9939 and 0.2575 by the ANN model. These improvements are due to the fact that acidity constant of benzoic acids and phenols in water shows nonlinear correlations with the molecular descriptors.

• Journal of Preventive Medicine and Public Health
• /
• 제43권1호
• /
• pp.42-49
• /
• 2010

Objectives: To compare the performance of three International Statistical Classification of Diseases, 10th Revision translations of the Charlson comorbidities when predicting in-hospital among patients with myocardial infarction (MI). Methods: MI patients ${\geq}20$ years of age with the first admission during 2006 were identified(n=20,280). Charlson comorbidities were drawn from Heath Insurance Claims Data managed by Health Insurance Review and Assessment Service in Korea. Comparisions for various conditions included (a) three algorithms (Halfon, Sundararajan, and Quan algorithms), (b) lookback periods (1-, 3- and 5-years), (c) data range (admission data, admission and ambulatory data), and (d) diagnosis range (primary diagnosis and first secondary diagnoses, all diagnoses). The performance of each procedure was measured with the c-statistic derived from multiple logistic regression adjusted for age, sex, admission type and Charlson comorbidity index. A bootstrapping procedure was done to determine the approximate 95% confidence interval. Results: Among the 20,280 patients, the mean age was 63.3 years, 67.8% were men and 7.1% died while hospitalized. The Quan and Sundararajan algorithms produced higher prevalences than the Halfon algorithm. The c-statistic of the Quan algorithm was slightly higher, but not significantly different, than that of other two algorithms under all conditions. There was no evidence that on longer lookback periods, additional data, and diagnoses improved the predictive ability. Conclusions: In health services study of MI patients using Health Insurance Claims Data, the present results suggest that the Quan Algorithm using a 1-year lookback involving primary diagnosis and the first secondary diagnosis is adequate in predicting in-hospital mortality.

• 가정과삶의질연구
• /
• 제24권5호
• /
• pp.223-237
• /
• 2006

This study explores how well parents and their children recognize the social attitudes of one another. Mothers, fathers and youths were asked to state their own opinion on various social issues then predict their children's, fathers' and mothers' responses(attributed attitudes). Empirical evaluation of the possible socialization consequences of actual versus attributed attitudes leads to a series of hypotheses. The data were collected from single students at a university in Seoul and their parents. Included in the seven social attitude were sexuality, educational, economic, political, ecological, religious and family issues. Analysis of the responses 98-110 triads, each consisting a mother, a father and a young adult child showed that both mothers and fathers were limited in their ability to gauge the attitudes of their children. Guided by attribution theory, this study tested several hypothesized relationships between the actual response of mother, the actual response of the father, the perceived response of the mother, the perceived response of the father and the actual response of the child. The theoretical model was tested with AMOS 5.0, utilizing path analysis, which is a form of structural equation modeling with manifest variables. Overall model fit was assessed by examining GFI, NFI, TLI, CFI and RMR. Results of the data analysis can be summarized as follows. First, the children perceived their mothers and fathers to be highly similar in their opinions and the actual responses of the mothers and the fathers were considerably correlated. Second, the fathers' responses whether attributed or actual were more predictive than the mothers' responses to their children's opinions. The alternative model suggests considerable support for the attribution theory. Indeed, within a family, the actual opinions of parents appear to have little direct bearing on the child's orientations, except when the actual orientations are perceived and reinterpreted by the children. It is not what parents think, but what their children think they think that predicts their offsprings' attitudes.

• 한국정보처리학회논문지
• /
• 제7권12호
• /
• pp.3821-3828
• /
• 2000

많은 소프트웨어 프로젝트는 시험이나 운영단계에서 고장 시간이나 고장 수 데이타 보다는 그룹 고장 데이타 (여러 고장 간격에서 또는 가변적인 시간 간격에서의 고장 들)가 수집된다. 본 논문은 그룹 고장 데이타에 대해 가변적인 미래의 시간에서 누적 고장 수를 예측할 수 있는 신경망 모델을 제시한다. 신경망의 입-출력으로 무엇을 선택하고 어떤 순서로 훈련을 수행하느냐에 따라 신경망의 예측력에 영향을 미친다. 따라서, 신경망의 입-출력에 대한 11개의 훈련제도가 고려되었으며, 모델의 성능을 평가하기 위해 다음 단계 평균 상대 예측 오차 (AE)와 정규화된 AE (NAE) 측도에 의해 최적의 훈련제도가 선택되고, 다른 잘 알려진 신경망 모델과 통계적 소프트웨어 신뢰성 성장 모델과 비교되었다. 실험 결과, 가변적인 미래의 시간 간격에서 누적 고장 수를 예측하기 위해서는 신경망 모델에 가변 시간간격 정보가 필요함을 보였다.

• 부모자녀건강학회지
• /
• 제7권2호
• /
• pp.144-159
• /
• 2004

Purpose: The purpose of this study is to explore mother's perception and practice of Taegyo. In addition, belief factors toward Taegyo oriented Korean mother-fetus interaction are identified in order to obtain baseline data for the development of Taegyo oriented program enhancing mother-fetus interaction. Method: The sample consisted to 186 pregnant women who visited public health center or hospital for prenatal care. Self-reporting questionnaire devised by investigator was used. Data were analyzed by SPSS win program and contents analysis was used. Result: Although mothers intended to a pregnancy, many of them were perceived negatively toward pregnancy. The study revealed that most of pregnant women continued cultural practices related to Taegyo, they focused on mother-fetus interaction behavior-such as listen to the music, reading a book, talk to fetus, stroke the fetus- with being altered traditional beliefs toward Taegyo or Taemong. Most of pregnant women had Taemong and believed a it's predictive functions. Beliefs toward mother-fetus interaction were classified to 6 factors, pregnancy, Taemong. fetus, practice behavior, infant's abilities and mother-fetus interaction behavior. The scores of belief toward mother-fetus interaction and each factors of pregnancy, fetus, practice behavior, infant's ability were statistically significant different depending upon husbands education. Also pregnancy factor in belief toward mother-fetus interaction was statistically significant different depending upon the number of children, child birth, experience of abortion, premature birth. Conclusion: Developing Taegyo oriented program enhancing the mother-fetus interaction need to include the elements of being facilitated maternal identity formation and maternal confidence for the pregnant women. In addition, it should be developed the program concerning with the characteristics of the pregnant women.

• 지능정보연구
• /
• 제7권1호
• /
• pp.135-147
• /
• 2001

본 연구에서는 퍼지신경망을 이용한 기업부실예측모형을 제안한다. 신경망은 탁월한 학습능력을 가진 것으로 알려져 있으나, 잡음이 심한 재무자료에 대해서는 종종 일관되지 못하고 기대에 미치지 못하는 예측성과를 보인다. 이는 연속형의 형태를 지닌 독립변수와 과다한 양의 원자료로부터 예측에 필요한 일정한 패턴을 찾기가 어렵기 때문이다. 이러한 문제점은 예측모형에서의 독립변수와 종속변수간의 인과관계를 신경망이 용이하게 찾아낼 수 있도록 독립변수의 형태를 변환함으로써 해결한 수 있다. 이러한 해결방법의 하나는 기존 신경망에 퍼지집합의 개념을 적용하여 신경망 학습에 사용될 자료를 퍼지화하고 이를 신경망에 학습시키는 것이다 입력자료를 퍼지화 함으로써 정보의 손실 없이도 신경망이 자료 내의 복잡한 관계를 용이하게 학습하는 것이 가능하다. 본 연구에서 제안된 퍼지신경망을 기업부도예측에 적용한 결과, 퍼지신경망이 기존의 신경망보다 우월한 예측성과를 나타내었다.

• 대한토목학회논문집
• /
• 제14권3호
• /
• pp.379-390
• /
• 1994

기존(旣存)의 지체모형(遲滯模型)은 지체(遲滯)의 분석(分析)을 위하여 무작위도착(無作爲到着)(Random Arrival)을 가정(假定)하고 있으므로 가로망상(街路網上)에서 군(群)의 형태를 이루는 교통류(交通流)에 대한 실질지체(實質遲滯)를 정확히 분석(分析)하지 못한다. 이러한 단점(短點)을 보완(補完)하기 위하여 미국(美國) 도로용량편람(道路容量便覽)(HCM)에서는 연동화보정계수(連動化補正係數)(PAF)를 사용하여 보다 정확한 지체(遲滯)의 분석을 시도(試圖)하고 있다. 그러나 미국(美國)의 도로용량편람(道路容量便覽)에서 제시(提示)하고 있는 연동화보정계수(連動化補正係數)는 차량군(車輛群)의 도착형태(到着形態), 차량군비(車輛群比) 등에서 많은 문제점(問題點)을 내포(內包)하고 있다. 본(本) 연구(硏究)는 이러한 문제점(問題點)에 대하여 미국(美國) NCHRP 연구결과(硏究結果)를 참고(參考)로 하여 우리나라 실정(實情)에 적용(適用)할 수 있는 교차점(交叉點) 지체모형(遲滯模型)과 연동화(連動化) 분석모형(分析模型)을 제안(提案)하고 기존(旣存)에 제시(提示)된 모형(模型)의 결과(結果)와 비교(比較)함으로서 각 모형(模型)들간의 차이점(差異點)과 예측력(豫測力)을 이론적(理論的)으로 검토(檢討)하여 본다.

• Asian Pacific Journal of Cancer Prevention
• /
• 제15권22호
• /
• pp.9865-9869
• /
• 2014

Background: Transthoracic fine needle aspiration (FNA) is one of several methods for establishing tissue diagnosis of lung lesions. Other tissue or cell sources for diagnosis include sputum, endobronchial biopsy, washing and brushing, endobronchial FNA, transthoracic core needle biopsy, biopsy from thoracoscopy or thoracotomy. The purpose of this study was to compare the sensitivity and specificity of FNA and other diagnostic tests in diagnosing lung lesions. Materials and Methods: The population included all patients undergoing FNA for lung lesions at Meir Medical Center from 2006 through 2010. Information regarding additional tissue tests was derived from the electronic archives of the Department of Pathology, patient records and files from the Department of Oncology. Sensitivity, specificity, diagnostic accuracy, and positive and negative predictive values were calculated for each test. Results: FNA was carried out in 245 patients. Malignant tumors were diagnosed in 190 cases (78%). They included adenocarcinoma (43%), squamous cell carcinoma (15%), non-small cell carcinoma, not otherwise specified (19%), neurondocrine tumors (7%), metastases (9%) and lymphoma (3%). The specificity of FNA for lung neoplasms was 100%; sensitivity and diagnostic accuracy were 87%. Conclusions: FNA is the most sensitive procedure for establishing tissue diagnoses of lung cancer. Combination with core needle biopsy increases the sensitivity. Factors related to the lesion (nature, degenerative changes, location) and to performance of all stages of test affect the ability to establish a diagnosis.

• Asian Pacific Journal of Cancer Prevention
• /
• 제15권23호
• /
• pp.10251-10254
• /
• 2015

Uterine cancer is the most common invasive gynaecological cancer in Australia. Early detection is a key predictive factor achieved by increasing public awareness and participation in screening. This observational study measures awareness of gynaecological malignancies, particularly uterine, among women in two rural areas of New South Wales, Australia. Patients presenting to gynaecology clinics in January to March 2014 were invited to complete a structured questionnaire. Women with a history of cancer and incomplete questionnaires were excluded. Of the 382 patients invited to participate, 329 (86%) responded with complete feedback. Most respondents were younger than than 50 years (66%) and married with at least 2 children (74%). The majority (94%) of participants had no awareness of uterine cancer and many (46%) were unable to identify common risk factors including obesity, diabetes and hypertension. The ability to identify risk factors was correlated to age, marital status and obesity. The study identifies poor awareness on uterine malignancies in two typical areas of rural Australia. Although external validity is limited by sociological factors, poor awareness of uterine cancer among rural patients in this study represents a valid public health concern. It is imperative to improve awareness of uterine cancer and available screening programs to facilitate early detection and cure.

• Asian Pacific Journal of Cancer Prevention
• /
• 제15권16호
• /
• pp.6983-6987
• /
• 2014

Background: Modified toluidine blue staining (MTBs) is a simple, inexpensive and time saving method to detect H. pylori in gastric biopsy specimens. As a metachromatic stain, it simultaneously highlights intestinal metaplasia, a gastric cancer precancerous lesion. The aim of this study was to assess the reliability of MTBs compared with hematoxylin-eosin (H&E) for H. pylori detection using immunoperoxidase staining as the gold standard. This technique would be beneficial for a routine diagnosis and confirmation of H. pylori eradication in developing countries where endoscopic-based approaches are dominant. Materials and Methods: Esophagogastroduodenoscopy with triple site gastric biopsies was undertaken in 207 dyspeptic patients at Thammasat University Hospital, Thailand between 1997 and 1999. H&E, MTBs and immunoperoxidase staining were applied to each specimen. The presence or absence of H. pylori with each stain was interpreted separately and the sensitivity, specificity, positive and negative predictive values of H&E and MTBs were calculated. Results: A total of 282 specimens from 207 patients were evaluated. Using immunoperoxidase staining, organisms were positive in 117 specimens (41%). MTBs proved almost equally sensitive as immunoperoxidase (99%) and significantly more sensitive than H&E (85%). It has comparable specificity (96% vs 96%), PPV (95% vs 94%), and NPV (99% vs 90%) to H&E, using immunoperoxidase staining as gold standard. MTBs compared with immunoperoxidase staining, is cheaper (2 USD vs 12 USD) and faster (20 min vs 16 hrs) compared to immunoperoxidase staining. Conclusions: MTBs is effective, economical and easy to use in daily practice for the detection of H. pylori in gastric biopsy specimens. In addition to saving time in evaluating H. pylori associated gastritis, with a high sensitivity and ability to demonstrate intestinal metaplasia, the technique may have a role in confirmation of H. pylori eradication for gastric cancer prevention in a developing country setting.