• 지질공학
• /
• 제26권3호
• /
• pp.371-382
• /
• 2016

해안지역의 염수침입은 지하수내 다양한 지화학 성분들에 의해서 탐지될 수 있다. 그러나 불충분한 지하수 자료는 염수침입의 인지를 어렵게 한다. 확률밀도함수는 제한된 수질자료를 이용하여 더 넓은 범위의 확률적 예측이 가능하며 특성화된 확률밀도 분포 도출을 통하여 해안지역의 염수침입을 효과적으로 판단할 수 있다. 본 연구에서는 부산시의 해안지역 지하수, 해안유출수, 하천수의 화학성분에 대해서 확률통계기법을 적용함으로써 염수침입을 평가하였다. 이 연구에 의해서, 해수 중에 농도가 높은 항목인 Na⁺, Mg²⁺, K⁺, SO₄²⁻, Cl⁻ 은 염수의 영향유무 판별의 유용한 지시자인 것으로 판명된다. 한편, 지하수, 해안유출수 및 하천수에서 유사한 확률분포형태를 보이는 항목인 Si⁴⁺, Fe²⁺, NO₃⁻, PO₄³⁻은 염수영향에 대한 지시자로서 적당하지 않은 것으로 판단된다.

• Biomolecules & Therapeutics
• /
• 제15권4호
• /
• pp.252-260
• /
• 2007

As it is needed to assay possible feasibility of extrapolation between in vivo and in vitro systems and to develop a new in vitro method for toxicity testing, we investigated global gene expression from both animal and cell line treated with thioacetamide (TAA) and compared between in vivo and in vitro genomic profiles. For in vivo study, mice were orally treated with TAA and sacrificed at 6 and 24 h. For in vitro study, TAA was administered to a mouse hepatic cell line, BNL CL.2 and sampling was carried out at 6 and 24 h. Hepatotoxicity was assessed by analyzing hepatic enzymes and histopathological examination (in vivo) or lactate dehydrogenase (LDH) assay and morphological examination (in vitro). Global gene expression was assessed using microarray. In high dose TAA-treated group, there was centrilobular necrosis (in vivo) and cellular toxicity with an elevation of LDH (in vitro) at 24 h. Statistical analysis of global gene expression identified that there were similar numbers of altered genes found between in vivo and in vitro at each time points. Pathway analysis identified several common pathways existed between in vivo and in vitro system such as glutathione metabolism, bile acid biosynthesis, nitrogen metabolism, butanoate metabolism for hepatotoxicty caused by TAA. Our results suggest it may be feasible to develop toxicogenomics biomarkers by comparing in vivo and in vitro genomic profiles specific to TAA for application to prediction of liver toxicity.

• Molecular & Cellular Toxicology
• /
• 제3권3호
• /
• pp.177-184
• /
• 2007

As a possible feasibility of the extrapolation between in vivo and in vitro systems, we investigated the global gene expression from both mouse liver and mouse hepatic cell line treated with hepatotoxic chemical, acetaminophen (APAP), and compared between in vivo and in vitro genomic profiles. For in vivo study, mice were orally treated with APAP and sacrificed at 6 and 24 h. For in vitro study, APAP were administered to a mouse hepatic cell line, BNL CL.2 and sampling was carried out at 6 and 24 h. Hepatotoxicity was assessed by analyzing hepatic enzymes and histopathological examination (in vivo) or lactate dehydrogenase (LDH) assay and morphological examination (in vitro). Global gene expression was assessed using microarray. In high dose APAPtreated group, there was centrilobular necrosis (in vivo) and cellular toxicity with the elevation of LDH (in vitro) at 24 h. Statistical analysis of global gene expression identified that there were similar numbers of altered genes found between in vivo and in vitro at each time points. Pathway analysis identified glutathione metabolism pathway as common pathways for hepatotoxicty caused by APAP. Our results suggest it may be feasible to develop toxicogenomics biomarkers or profiles by comparing in vivo and in vitro genomic profiles specific to this hepatotoxic chemical for application to prediction of liver toxicity.

• 응용통계연구
• /
• 제32권5호
• /
• pp.693-702
• /
• 2019

딥러닝은 대용량의 데이터의 분류 및 예측하는 방법으로 각광받고 있다. 데이터의 양이 많아지면서 신경망의 구조는 더 깊어 지고 있다. 이때 초기값이 지나치게 클 경우 층이 깊어 질수록 활성화 함수의 기울기가 매우 작아지는 포화(Saturation)현상이 발생한다. 이러한 포화현상은 가중치의 학습능력을 저하시키는 현상을 발생시키기 때문에 초기값의 중요성이 커지고 있다.이런 포화현상 문제를 해결하기 위해 Glorot과 Bengio (2010)과 He 등 (2015) 층과 층 사이에 데이터가 다양하게 흘러야 효율적인 신경망학습이 가능하고 주장했다. 데이터가 다양하게 흐르기 위해서는 각 층의 출력에 대한 분산과 입력에 대한 분산이 동일해야 한다고 제안했다. Glorot과 Bengio (2010)과 He 등 (2015)는 각 층별 활성화 값의 분산이 같다고 가정해 초기값을 설정하였다. 본 논문에서는 절단된 코쉬 분포와 절단된 정규분포를 활용하여 초기값을 설정하는 방안을 제안한다. 출력에 대한 분산과 입력에 대한 분산의 값을 동일하게 맞춰주고 그 값이 절단된 확률분포의 분산과 같게 적용함으로써 큰 초기값이 나오는 걸 제한하고 0에 가까운 값이 나오도록 분포를 조정하였다. 제안된 방법은 MNIST 데이터와 CIFAR-10 데이터를 DNN과 CNN 모델에 각각 적용하여 실험함으로써 기존의 초기값 설정방법보다 모델의 성능을 좋게 한다는 것을 보였다.

• Child Health Nursing Research
• /
• 제7권3호
• /
• pp.280-297
• /
• 2001

The purpose of the study was to develop and test the model for the quality of life in mothers of children with nephrotic syndrome. A hypothetical model was constructed on the basis of previous studies and a review of literature. The conceptual framework was built around ten constructs. Exogenous variables included in this model were mother's health, father's health, marital intimacy, mother's attitude on children, economic state, side effect of steroid, severity of illness and social support. Endogenous variables were mother's burden and quality of life. Empirical data for testing the hypothetical model were collected by using a self-report questionnaire from 152 mothers of children with nephrotic syndrom at the outpatient clinics and in the hospital. The data was collected from May, 1999 to August, 1999.Reliability of the seven instruments was tested with Cronbach's alpha which ranged from 0.71 - 0.92.For the data analysis, SPSS 8.0 WIN program and LISREL 8.20 WIN program were used for descriptive statistics and covariance structural analysis. The results of covariance structural analysis were as follow :1. The hypothetical model showed a good fit with the empirical data. [x2 = .56, df = 3, p = .90(p>.05 ), GFI = .99, AGFI = .99, RMSR = .005.] 2. For the parsimony of model, a modified model was constructed by deleting 1 variable and excluding 2 paths according to the criteria of statistical significance and meaning.3. The modified model also showed a good fit with the data[x2 = 2.83, df = 7, p = .90( p>.05 ), GFI = 1.00, AGFI = .97, RMSR = .011].The result of the testing of the hypothesis were as follows : 1. Mother's health(γ21 = .26, t = 4.16), father's health(γ22 = .19, t = 2.92), marital intimacy(γ23 = .26, t = 4.13) and social support(γ28 = .12, t = 2.03) had a significant direct effect on the quality of life.2. Mother's burden(β21 = -.20, t = -3.10) had a significant negative direct effect on the quality of life.3. Mother's attitude on children(γ14 = -.34, t = .-4.57), mother's health(γ11 = -.22, t = -2.96) and side effect of steroid (γ16 = -.23, t = .-2.69) had a significant direct negative effect on the burden. The result of this study showed that mother's health, marital intimacy, mother's burden, father's health, and social support had a significant direct effect on the quality of life. Mother's attitude on children, mother's health, and side effect of steroid had a significant direct effect on mother's burden. These six variables, mother's health, marital intimacy, father's health, social support, mother's attitude on children and side effect of steroid were identified as relatively important variables. The results of this study suggest, it needed to determine the nursing intervention will alleviate mother's burden and promote a greater quality of life in mothers of children with nephrotic syndrom.

• 대한토목학회논문집
• /
• 제31권5B호
• /
• pp.449-457
• /
• 2011

최근 기후변화/변동으로 인한 집중호우가 증가하여 수문기상재해에 따른 피해가 증가하고 있다. 미래의 발생가능한 극한 강우사상에 대응하기 위해, 일반적으로 수문학적 빈도해석을 이용하여 목표연도의 설계 강우량을 산정한다. 이것은 수문빈도 해석에 적용된 강우자료가 정상성임을 가정하여 설계 강우량을 산정하는 것이다. 하지만, 최근 관측된 강우자료를 살펴보면, 통계적 특성이 시간에 따라 변하는 경우가 있다. 본 연구는 연최대강우량의 회귀직선에 대한 잔차의 수문학적 빈도해석을 바탕으로, 가까운 미래로 설정된 목표연도의 확률강우량을 산정하는 방법을 제안하였다. 현재까지의 관측자료를 기초로 선형회귀식의 추세선을 이용하여 잔차 시계열을 생성하고, 잔차에 대한 확률밀도함수를 추정한 후, 추세선의 증가 및 감소 경향을 고려하여 확률강우량을 산정하였다. 14개의 강우관측지점에 적용한 결과, 증가경향을 보이는 경우에는 현시점까지의 자료에 대한 선형회귀식을 산정한 후, 목표연도까지 연장했을 때의 추세요소를 산정한 방법이 보다 적합한 확률강우량을 산정하는 것으로 나타났다. 이러한 결과는 정상성을 바탕으로 추정한 확률강우량과 비교했을 때, 5-25%의 예측편차가 1-22% 정도로 감소하였다.

• Asian Pacific Journal of Cancer Prevention
• /
• 제15권2호
• /
• pp.1005-1010
• /
• 2014

Background: A relation between abdominal obesity and colorectal tumor development has been reported repeatedly, and is believed to be more remarkable in man than in women. However, the details vary depending on scientific reports. This may be due at least partly to the selected surface anthropometric index in addition to the influence of gender and ethnic groups. To cope with this, we considered a new index of abdominal obesity and evaluated its risk prediction potential. Materials and Methods: Six hundred ninety five Japanese (262 women and 433 men) who had a colonoscopy were studied. The new index was named as waist circumference to height index (WHI) and was calculated by the formula of waist circumference (cm)/height (m)/height (m). Biochemical and lifestyle factors were investigated preceding the colonoscopy. Statistical analysis was performed using SPSS for Windows. Results: Increase of WHI was associated with altered metabolism of carbohydrate and lipid in both women and men. WHI was positively related with the development of colon tumor of women, while not with that of men. Logistic regression analysis performed for stratified age groups (45-54, 55-64 and 65-74 years) showed that WHI significantly increased odds ratio to 1.31 (CI 1.05-1.64 p=0.01) in women of 55-65 years. In contrast, in men this index WHI reduced the odds ratio insignificantly, while low density lipoprotein and triglyceride significantly increased the odds ratio to 1.01 (CI 1.00-1.03 p=0.02) in the 55-65 year group and to 1.02 (CI 1.00-1.03 p=0.02) in the 45-55 year group. Conclusions: In Japanese the risk factors for colon tumor development are different between women and men. WHI is a simple and efficient predictor of colon tumor risk in Japanese women and may be used to select those who should have colonoscopy.

• Asian-Australasian Journal of Animal Sciences
• /
• 제33권1호
• /
• pp.44-52
• /
• 2020

Objective: Minerals are one of the important nutrients for supporting the growth of sheep grazing in the highland, northwest of China. The experiment was conducted to investigate the relationship of both macro and micro minerals in sheep grazing in the highlands of six districts located in the Qilian Mountain of China. Methods: Samples of herbage (n = 240) and soil (n = 240) were collected at random in a "W" shape across the area designated for harvesting from 24 farms, where the sheep commonly graze in October (winter) for mineral analyses. In addition, serum samples were taken via jugular vein from 20 sheep per farm from 24 farms (n = 480 samples in total) for serum minerals analyses. Mean values of macro and micro minerals were statistically compared among districts and the correlations among soil-plant-animal were statistically analyzed and correlations were regressed, as well. Results: The results revealed that there were variations for both macro and micro minerals among districts. Statistical analysis of the correlation coefficients between herbage and sheep were significantly different for most of the minerals but not for P, Cu, and Se. Many correlation regression coefficients were found significantly different among minerals of herbage, soil, and sheep serum especially those of K, Na, Fe, Mn, and Zn (between herbage and sheep serum), and Fe and Mn (between herbage and soil), Na, Fe, Mn, and Zn (between soil and sheep serum), respectively. The regression coefficient equations derived under this experiment for prediction of Ca (R² = 0.618), K (R² = 0.803), Mg (R² = 0.767), Na (R² = 0.670), Fe (R² = 0.865),Zn (R² = 0.950), Mn (R² = 0.936), and Se (R² = 0.630), resulted in significant R² values. Conclusion: It is inferred that the winter herbage minerals in all the districts were below the recommended levels for macro minerals which indicated there would be some mineral deficiencies in sheep grazing the herbage in these regions. Supplemental minerals may therefore play an important role in balancing the minerals available from the herbage in winter and would lead to increased productivity in sheep on the highland areas of China. These findings could be potentially applied to the other regions for improving the livestock productivity.

• Asian Pacific Journal of Cancer Prevention
• /
• 제14권2호
• /
• pp.929-934
• /
• 2013

Aim: To analyze the significance of different clinical factors for prognostic prediction in diffuse large B-cell lymphoma (DLBCL) patients. Methods: Two hundred and twenty-seven DLBCL patients were retrospectively reviewed. Patients were managed with cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP) regimen or rituximab plus the CHOP (RCHOP) regimen. Results: Lactate dehydrogenase (LDH), ${\beta}2$-microglobulin (${\beta}2$-M), B symptoms, Ann Arbor stage and genetic subtypes were statistically relevant in predicting the prognosis of the overall survival (OS). In the CHOP group, the OS in patients with germinal center B-cell-like (GCB)(76.2%) was significantly higher than that of the non-GCB group (51.9%, P=0.032). With RCHOP management, there was no statistical difference in OS between the GCB (88.4%) and non-GCB groups (81.9%, P=0.288). Conclusion: Elevated LDH and ${\beta}2$-M levels, positive B symptoms, Ann Arbor stage III/IV, and primary nodal lymphoma indicate an unfavorable prognosis of DLBCL patients. Patients with GCB-like DLBCL have a better prognosis than those with non-GCB when treated with the CHOP regimen. The RCHOP treatment with the addition of rituximab can improve the prognosis of patients with DLBCL.

• Asian Pacific Journal of Cancer Prevention
• /
• 제14권2호
• /
• pp.753-757
• /
• 2013

Objective: To evaluate the relationships between lymph node ratio (LNR, the ratio of positive lymph nodes in excised axillary lymph nodes) and disease-free survival (DFS) by comparing with traditional absolute positive lymph node number (pN classification) for prediction of breast cancer (BC) progrnosis. Methods and Patients: We retrospectively reviewed patients who received comprehensive therapy in Department of Breast Surgery, Hubei Cancer Hospital, China from Jan 2002 to Dec 2006 (Group A), and Department of Breast and Thyroid Surgery, Renmin Hospital of Wuhan University, China from Jun 2008 to May 2012 (Group B). Patients were allocated to low-risk (${\leq}0.20$), intermediate-risk (> 0.20 but ${\leq}0.65$), high-risk (>0.65) groups by LNR. The primary endpoint was 5-DFS. Results: A total of 294 patients were included in our study. LNR was verified as a negative prognostic factor for DFS (P=0.002 in Group A, P<0.0001 in Group B). Then we found the effects of pN and LNR delamination on disease-free survival (DFS) had statistical significance (P=0.012 for pN and P=0.031 for LNR stratification in Group A, both of them P<0.001 in Group B). Compared to pN staging, LNR staging displayed superior performance in prognosis, the adjusted hazard ratio of recurrence being 2.07 (95%CI, 1.07 to 4.0) for intermediate risk group (P=0.030) and 2.44 (95%CI, 1.21 to 4.92) for high risk group (P=0.013) in Group A. Conclusions: LNR stratification proved an adverse prognostic factor of DFS in lymph nodes positive invasive BC using cut-off values 0.20 and 0.65, and was more predictive than traditional pN classification for 5-DFS.