• Proceedings of the Korean Society of Near Infrared Spectroscopy Conference
• /
• 2001.06a
• /
• pp.1245-1245
• /
• 2001

Food adulteration is a serious consumer fraud and a matter of concern to food processors and regulatory agencies. A range of analytical methods have been investigated to facilitate the detection of adulterated or mis-labelled foods & food ingredients but most of these require sophisticated equipment, highly-qualified staff and are time-consuming. Regulatory authorities and the food industry require a screening technique which will facilitate fast and relatively inexpensive monitoring of food products with a high level of accuracy. Near infrared spectroscopy has been investigated for its potential in a number of authenticity issues including meat speciation (McElhinney, Downey & Fearn (1999) JNIRS, 7(3), 145-154; Downey, McElhinney & Fearn (2000). Appl. Spectrosc. 54(6), 894-899). This report describes further analysis of these spectral sets using a hierarchical approach and binary decisions solved using logistic regression. The sample set comprised 230 homogenized meat samples i. e. chicken (55), turkey (54), pork (55), beef (32) and lamb (34) purchased locally as whole cuts of meat over a 10-12 week period. NIR reflectance spectra were recorded over the wavelength range 400-2498nm at 2nm intervals on a NIR Systems 6500 scanning monochromator. The problem was defined as a series of binary decisions i. e. is the meat red or white\ulcorner is the red meat beef or lamb\ulcorner, is the white meat pork or poultry\ulcorner etc. Each of these decisions was made using an individual binary logistic model based on scores derived from principal component or partial least squares (PLS1 and PLS2) analysis. The results obtained were equal to or better than previous reports using factorial discriminant analysis, K-nearest neighbours and PLS2 regression. This new approach using a combination of exploratory and logistic analyses also appears to have advantages of transparency and the use of inherent structure in the spectral data. Additionally, it allows for the use of different data transforms and multivariate regression techniques at each decision step.

• Proceedings of the Korean Society of Near Infrared Spectroscopy Conference
• /
• 2001.06a
• /
• pp.1152-1152
• /
• 2001

Food adulteration is a serious consumer fraud and a matter of concern to food processors and regulatory agencies. A range of analytical methods have been investigated to facilitate the detection of adulterated or mis-labelled foods & food ingredients but most of these require sophisticated equipment, highly-qualified staff and are time-consuming. Regulatory authorities and the food industry require a screening technique which will facilitate fast and relatively inexpensive monitoring of food products with a high level of accuracy. Near infrared spectroscopy has been investigated for its potential in a number of authenticity issues including meat speciation (McElhinney, Downey & Fearn (1999) JNIRS, 7(3), 145 154; Downey, McElhinney & Fearn (2000). Appl. Spectrosc. 54(6), 894-899). This report describes further analysis of these spectral sets using a hierarchical approach and binary decisions solved using logistic regression. The sample set comprised 230 homogenized meat samples i. e. chicken (55), turkey (54), pork (55), beef (32) and lamb (34) purchased locally as whole cuts of meat over a 10-12 week period. NIR reflectance spectra were recorded over the wavelength range 400-2498nm at 2nm intervals on a NIR Systems 6500 scanning monochromator. The problem was defined as a series of binary decisions i. e. is the meat red or white\ulcorner is the red meat beef or lamb\ulcorner, is the white meat pork or poultry\ulcorner etc. Each of these decisions was made using an individual binary logistic model based on scores derived from principal component or partial least squares (PLS1 and PLS2) analysis. The results obtained were equal to or better than previous reports using factorial discriminant analysis, K-nearest neighbours and PLS2 regression. This new approach using a combination of exploratory and logistic analyses also appears to have advantages of transparency and the use of inherent structure in the spectral data. Additionally, it allows for the use of different data transforms and multivariate regression techniques at each decision step.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.29 no.2B
• /
• pp.155-162
• /
• 2009

Univariate frequency analyses are widely used in practical hydrologic design. However, a storm event is usually characterized by amount, intensity, and duration of the storm. To fully understand these characteristics and to use them appropriately in hydrologic design, a multivariate statistical approach is necessary. This study applied a Gumbel mixed model to a bivariate storm frequency analysis using hourly rainfall data collected for 46 years at the Seoul rainfall gauge station in Korea. This study estimated bivariate return periods of a storm such as joint return periods and conditional return periods based on the estimation of joint cumulative distribution functions of storm characteristics. These information on statistical behaviors of a storm can be of great usefulness in the analysis and assessment of the risk associated with hydrologic design problems.

• Asian Pacific Journal of Cancer Prevention
• /
• v.14 no.12
• /
• pp.7309-7314
• /
• 2013

Aim: To investigate factors that affect the place of death (POD) of terminal cancer patients. Materials and Methods: We recruited 702 consecutive patients (${\geq}18$ years) from 12 centers during July 2005 to October 2006, and 481 completed the questionnaire. In April 2011, we linked the data for 96.0% (n=462) of the deceased patients to the POD using the 2005-2009 death certificate data of Korea's National Statistical Office. The primary outcome variable was POD, and the predictive value of variables pertaining to patients and caregivers was evaluated using univariate and multivariate analyses. Results: Most patients died in a hospital (91.5%, n=441) and age, education, preference for place of terminal care, wish to use hospice/palliative care services, terminal cancer awareness, time between diagnosis and death, and global quality-of-life subscale of the EORTC QLQ-C30 of patients, and education and preference for place of terminal care of caregivers were significant predictors in univariate analyses. On multivariate analysis, patients and caregivers who preferred hospital/palliative care as the terminal care option over home care [adjusted odds ratio (aOR), 2.68; 95% confidential interval (CI), 1.18-7.04 and aOR: 2.65; 95%CI: 1.15-6.09 for patient and caregiver preferences, respectively] and caregivers who were highly educated (aOR, 3.19; 95%CI, 1.44-7.06) were predictors of POD. Conclusions: Most of the terminal cancer patients died in a hospital. Our findings indicate that major predictors of hospital deaths are preference of both the patient and caregiver for hospital/palliative care as the terminal care option and higher education of the caregiver.

• Journal of Gastric Cancer
• /
• v.2 no.1
• /
• pp.5-11
• /
• 2002

Purpose: The purpose of this study is to identify immunohistochemical evidence of lymph-node micrometastasis in histologic node-negative gastric cancer patients and to evaluate the prognostic significance of lymph-node micrometastasis.Materials and Methods: A retrospective study of 50 gastric cancer patients who underwent curative resections from October 1990 to November 1994 was performed. Two consecutive sections were prepared: one for ordinary hematoxylin and eosin staining, and the other for immunohistochemical staining with Pan cytokeratin antibody (Novocastra, UK). In the univariate analysis, the survival rate was calculated using the Life Table Method, and the multivariate analysis was determined using a Cox Proportional HazardsModel. The statistical analyses of the relationships between the clinicopathologic factors and micrometastases were performed by using a Chi-square test. Results: Of 2522 harvested lymph nodes, 81 ($4.1\%$) nodes and 19 ($38\%$) of 50 patients were identified as having lymphnode micrometastases by using immunohistochemical staining for cytokeratin. The incidence of lymph-node micrometastases was significantly higher in diffuse type carcinomas ($54\%$, P=0.024) and in patients with serosal invasion ($52.2\%$, P=0.05). For patients with lymph-node micrometastases (n=19), the 5-year survival rate was significantly decreased ($73.7\%$, P=0.015). The Lauren's classirication (P=0.021) and the depth of invasion (P=0.035) were shown by multivariate analysis to have a significant relationship with the presence of micrometastases. Multivariate analysis revealed that lymph-node micrometastasis was independently correlated with survival in histologic node-negative gastic cancer patients. Conclusion: The presence of cytokeratin detected lymphnode micrometastases correlates with the worse prognosis for patients with histologic node-negative gastric cancer.

• Journal of Environmental Science International
• /
• v.28 no.3
• /
• pp.303-320
• /
• 2019

Water quality is characterized by various complex factors. Therefore, a systematic understanding of water quality trends is required to carry out a proper evaluation. In this study, we analyzed the spatio-temporal water quality characteristics of the Nakdong River using five-year data from 2012 to 2016. Data was collected on the pH, DO, BOD, COD, SS, TN, TP, TOC, WT, EC, $NH_3-N$, $NO_3-N$, $PO_4-P$, Chl-a, rainfall, and total and fecal coliforms. A total of 38 water quality measurement stations, from Andong1 to Gupo, were considered. Statistical analyses including trend, cluster, and factor analyses were conducted to identify the dominant water quality components affecting the Nakdong River. The Nakdong River was spatially classified into three groups for up-stream (Andong1 to Sangju1), mid/up-stream (Donam to Dalseong), and mid/down-stream (Hwawonnaru to Gupo) data collection, and temporally into two groups for summer/fall (7~10), and the rest of the season (11~6) data. The water quality of the entire Nakdong River showed trends similar to the mid/down-stream section, which indicates the importance of water quality management in this section. Suspended solids, phosphorus, and coliform groups were established as important factors to be considered in the summer/fall season across the river, especially in the mid/down-stream section. Nitrogen and organic matter were identified as important factors to be considered in the rest of the season, especially in the mid/up-stream section. This study could help determine the water quality components that should be intensively monitored in the Nakdong River.

• Journal of Hospice and Palliative Care
• /
• v.8 no.1
• /
• pp.45-51
• /
• 2005

Purpose: The purpose of this study is to examine the factors surrounding burnout of nurses caring for cancer patients. Methods: The sample of this study was conveniently selected among nurses who had hospice care experiences working in General Hospitals located in Seoul. This study was conducted by a self-administered questionnaire. Two hundred forty four questionnaires were retrieved and the response rate was 81.3%. The period of data collection was from February 25th to March 5th in 1994. Mean, standard deviation, T-test ANOVA, and multiple regression analysis were performed for statistical analysis. Results: The data showed that respondents reported to have burnout as many as 2.71 out of a 5.0 score. Bivariate analyses indicated that those who had hospice education reported to have a lower burnout than those without hospice education. Multivariate regression analyses revealed factors associated with burnout the nurses have had. They include being a Christian, higher job satisfaction, and experiences of hospice education. Hospice education reducing burnout for the nurses was observed by hierarchial multiple regression analyses, after controlling out the effect of coping methods, sociodemographic characteristics, job satisfaction, and job-related stresses on experience of burnout. This observation was not hue for physical and psychological burnout but for burnout in general and emotional one. But this was not confirmed among the nurses with type A personality. Conclusion: The findings of this study have a weakness in generalizability due to the sampling methodology used in this study. However, for the better hospice care further research with a probability sampling method are necessary.

• Asian Pacific Journal of Cancer Prevention
• /
• v.16 no.5
• /
• pp.1839-1843
• /
• 2015

Background: MicroRNAs are a class of noncoding RNAs which regulate multiple cellular processes during tumor development. The purpose of this report is to investigate the clinicopathological and prognostic significance of miR-218 in human gliomas. Materials and Methods: Quantitative RT-PCR (qRT-PCR) was conducted to detect the expression of miR-218 in primary normal human astrocytes, three glioma cell lines and 98 paired glioma and adjacent normal brain tissues.Associations of miR-218 with clinicopathological variables of glioma patients were statistically analyzed. Finally, a survival analysis was performed using the Kaplan-Meier method and Cox's proportional hazards model. Results: The expression level of miR-218 in primary normal human astrocytes was significantly higher than that in glioma cell lines (p<0.01). Also, the expression level of miR-218 in glioma tissues was significantly downregulated in comparison with that in the adjacent normal brain tissues (p<0.001). Statistical analyses demonstrated that low miR-218 expression was closely associated with advanced WHO grade (p=0.002) and low Karnofsky performance score (p=0.010) of glioma patients. Kaplan-Meier analysis with the log-rank test showed that patients with low-miR-218 expression had poorer disease-free survival and overall survival (p=0.0045 and 0.0124, respectively). Multivariate analysis revealed that miR-218 expression was independently associated with the disease-free survival (p=0.009) and overall survival (p=0.004) of glioma patients. Conclusions: Our results indicate that miR-218 is downregulated in gliomas and that its status might be a potential valuable biomarker for glioma patients.

• The Korean Journal of Applied Statistics
• /
• v.22 no.4
• /
• pp.745-758
• /
• 2009

Cluster analysis is the automated search for groups of related observations in a data set. To group the observations into clusters many techniques has been proposed, and a variety measures aimed at validating the results of a cluster analysis have been suggested. In this paper, we compare complete linkage, Ward's method, K-means and model-based clustering and compute validity measures such as connectivity, Dunn Index and silhouette with simulated data from multivariate distributions. We also select a clustering algorithm and determine the number of clusters of Korean consumers based on Korean consumers' palatability scores for Hanwoo bull in BBQ cooking method.

• Journal of Life Science
• /
• v.17 no.9 s.89
• /
• pp.1217-1223
• /
• 2007

I investigated the reproduction system of nine natural populations of P. discolor in Korea and two Mongolian P. conferta populations. The measurements of 19 quantitative or qualitative morphological characters were taken on each of total individuals directly from their natural habitats. Multivariate principal component analyses (PCA) were conducted to detect differences among populations consid-ering several characters simultaneously of variances using the statistical analysis system. 19 morpho-logical characteristics between Korean Potentilla species and Mongolian Potentilla species showed a slight heterogeneity of variance. The length of internodes (LFL and LSI) and characteristics of root (LLR and NOR) were shown a significant difference between two species (P<0.05). The number of ra-mets in P. conferta decreased with increasing geographic distance from viviparity. However, P. discolor has most ramets at distance intervals $60{\sim}80$ cm. In light conditions, P. discolor was significantly less resilience than P. conferta. In drought conditions, although there was not shown significant difference, P. conferta was less resilience than P. discolor. The core analysis indicates that P. conferta is the more resistant species than P. discolor and usually propagates by clonal growth during several strong envi-ronmental disadvantages such as drought events.