• Genomics & Informatics
• /
• v.14 no.4
• /
• pp.138-148
• /
• 2016

The success of genome-wide association studies (GWASs) has enabled us to improve risk assessment and provide novel genetic variants for diagnosis, prevention, and treatment. However, most variants discovered by GWASs have been reported to have very small effect sizes on complex human diseases, which has been a big hurdle in building risk prediction models. Recently, many statistical approaches based on penalized regression have been developed to solve the "large p and small n" problem. In this report, we evaluated the performance of several statistical methods for predicting a binary trait: stepwise logistic regression (SLR), least absolute shrinkage and selection operator (LASSO), and Elastic-Net (EN). We first built a prediction model by combining variable selection and prediction methods for type 2 diabetes using Affymetrix Genome-Wide Human SNP Array 5.0 from the Korean Association Resource project. We assessed the risk prediction performance using area under the receiver operating characteristic curve (AUC) for the internal and external validation datasets. In the internal validation, SLR-LASSO and SLR-EN tended to yield more accurate predictions than other combinations. During the external validation, the SLR-SLR and SLR-EN combinations achieved the highest AUC of 0.726. We propose these combinations as a potentially powerful risk prediction model for type 2 diabetes.

• International Journal of Internet, Broadcasting and Communication
• /
• v.16 no.3
• /
• pp.290-297
• /
• 2024

We investigate the efficacy of ensemble learning methods, specifically the soft voting technique, for enhancing heart disease prediction accuracy. Our study uniquely combines Logistic Regression, SVM with RBF Kernel, and Random Forest models in a soft voting ensemble to improve predictive performance. We demonstrate that this approach outperforms individual models in diagnosing heart disease. Our research contributes to the field by applying a well-curated dataset with normalization and optimization techniques, conducting a comprehensive comparative analysis of different machine learning models, and showcasing the superior performance of the soft voting ensemble in medical diagnosis. This multifaceted approach allows us to provide a thorough evaluation of the soft voting ensemble's effectiveness in the context of heart disease prediction. We evaluate our models based on accuracy, precision, recall, F1 score, and Area Under the ROC Curve (AUC). Our results indicate that the soft voting ensemble technique achieves higher accuracy and robustness in heart disease prediction compared to individual classifiers. This study advances the application of machine learning in medical diagnostics, offering a novel approach to improve heart disease prediction. Our findings have significant implications for early detection and management of heart disease, potentially contributing to better patient outcomes and more efficient healthcare resource allocation.

• Journal of Sasang Constitutional Medicine
• /
• v.32 no.2
• /
• pp.10-21
• /
• 2020

Objectives Genome-wide association studies(GWAS) is a useful method to identify genetic associations for various phenotypes. The purpose of this study was to develop predictive models for Sasang constitution types using genetic factors. Methods The genotypes of the 1,999 subjects was performed using Axiom Precision Medicine Research Array (PMRA) by Life Technologies. All participants were prescribed Sasang Constitution-specific herbal remedies for the treatment, and showed improvement of original symptoms as confirmed by Korean medicine doctor. The genotypes were imputed by using the IMPUTE program. Association analysis was conducted using a logistic regression model to discover Single Nucleotide Polymorphism (SNP), adjusting for age, sex, and BMI. Results & Conclusions We developed models to predict Korean medicine constitution types using identified genectic factors and sex, age, BMI using Random Forest (RF), Support Vector Machine (SVM), and Neural Network (NN). Each maximum Area Under the Curve (AUC) of Teaeum, Soeum, Soyang is 0.894, 0.868, 0.767, respectively. Each AUC of the models increased by 6~17% more than that of models except for genetic factors. By developing the predictive models, we confirmed usefulness of genetic factors related with types. It demonstrates a mechanism for more accurate prediction through genetic factors related with type.

• Journal of Fisheries and Marine Sciences Education
• /
• v.29 no.3
• /
• pp.757-766
• /
• 2017

This study investigated not only the existence of differently functioned item due to gender but also domain. In this study, the randomly selected data of TIMSS 2007, which consist of 681 male and 646 women, were analyzed. To detect differently functioned items, this study employed Raju method. For Raju method, three-parameter logistic model was selected. Signed and unsigned area between two item characteristic curve were measured within the real ability range. An item which was detected commonly SA and UA area in Raju method was defined as a differently functioned item. As a result of this study, six items among twenty seven items of mathematics in the TIMSS 2007 were differently functioned item. Five items among those six items, were in favor of boys and one item was in favor of girls. Number, Geometric Shapes and Measures, and Applying were in favor of boys. but Data Display, Reasoning were in favor of girls. The conclusion of this study was summarized as existing differently functioned items in TIMSS 2007 and difference between favorable domain based gender. Finally, it is desirable to consider the differently functioned items by relating those item content for improving the test reliability of TIMSS 2007.

• International Journal of Reliability and Applications
• /
• v.10 no.2
• /
• pp.89-99
• /
• 2009

Oriented Strand Board (OSB) is an important engineered wood product used in housing construction which has a lower environmental impact or "carbon footprint." In this paper, reliability and statistical tools are applied to gain insights on the strand thickness of OSB panels. An OSB panel consists of several hundred wood strands that are resinated and pressed. The variability of OSB strand thickness for six manufacturers in the Eastern United States is examined as a whole, as well as individually. Little research exists on OSB strand thickness across mills even though strand thickness variability has been documented in laboratory experiments to greatly influence the dimensional stability of OSB panels. Our aims are to quantify and characterize strand thickness, plus apply reliability techniques, such as Kaplan-Meier curves, to characterize the probability of strand thickness. We further explore graphically and statistically the thickness of the strands.

• IE interfaces
• /
• v.16 no.1
• /
• pp.103-110
• /
• 2003

Growth curves are widely used in forecasting the market demand. When there are only a few data points available, the estimated model parameters have a low confidence. In this case, if some expert opinions are available, it would be better for predicting future demand to adjust the model parameters using these information. This paper proposes the methodology for re-estimation of model parameters in growth curves when adjusting market potential and/or time of maximum sales. We also provide the detailed procedures for five growth curves including Bass, Logistic, Gompertz, Weibull and Cumulative Lognormal models. Applications to real data are also included.

• Asian Pacific Journal of Cancer Prevention
• /
• v.17 no.4
• /
• pp.2089-2092
• /
• 2016

Background: The impact of mean platelet volume (MPV) on prognosis, diagnosis and response to therapy in cancer patients has been widely investigated. In the present study, we evaluated whether MPV at diagnosis has predictive value for pathologic complete response (pCR) after neoadjuvant chemotherapy in patients with locally advanced breast cancer (LABC). Materials and Methods: A total of 109 patients with LABC from Akdeniz University and Antalya Research and Training Hospital were evaluated retrospectively. Results: ROC curve analysis suggested that the optimum MPV cut-off point for LABC patients with pCR (+) was 8.15 (AUC:0.378, 95%CI [0.256-0.499], p=0.077). The patients with MPV <8.15 had higher pCR rates (29.2% vs. 13.1%, p=0.038). After binary logistic regression analysis, MPV and estrogen receptor absence were independent predictors for pCR. Conclusions: MPV has an independent predictive value for pCR after neoadjuvant chemotherapy in patients with LABC.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.30 no.3
• /
• pp.172-181
• /
• 1994

The mesh selectivity of diamond and suare mesh cod-ends at the Southern Korean Sea and the East China Sea were compared for Pampus argenteus, Trachurus japonicus, Trichiurus lepturus. Selection trials were carried out using diamond and square mesh cod-end by trouser type cod-end with cover net. of which the mesh cod-end has four types : A(51.2mm), B(70.2mm), C(77.6mm), D(88.0mm). Selection curves and selection parameters were calculated using a logistic model. The results obained are summarized as follows : 1. Harvest fish : In B. C and D type selection range and fifty percent selection length of the square mesh were about 21mm, 11mm : 12mm, 18mm and 34mm, 5mm higher than those of the diamond mesh, respectively. Selection factor of master curve for the diamond mesh was 1.54 and for the square mesh was 1.68. The optimum mesh size for the diamond mesh was 97.4mm and for the square mesh was 89.3mm, the difference was 8.1mm. 2. Horse mackerel : In A type, selection range was nearly the same for the diamond and the square mesh, but fifty percent selection length of the square mesh was 43mm higher than the diamond mesh. In B. C and D type, selection range and fifty percent selection length of the square mesh were about 6mm, 3mm : 24mm, 21mm and 11mm, 42mm higher than those of the diamond mesh, respectively. Selection factor of master curve for the diamond mesh was 2.37, for the square mesh was 2.77. The optimum mesh size for the diamond mesh was 78.1mm and for the square mesh was 66.8mm, the difference was 11.3mm. 3. Hair tail : In A, B and C type, selection range of the square mesh was about 34mm, 8mm, 60mm higher than those of the diamond mesh. Fifty percent selection length for the diamond mesh was about 5mm, 7mm, 8mm higher than that of the square mesh. Selection factor of master curve for the diamond mesh was 3.11, for the square mesh was 3.48. The optimum mesh size for the diamond mesh was 64.3mm and for the square mesh was 57.5mm, the difference was 6.8mm.

• Journal of the Korea Convergence Society
• /
• v.13 no.1
• /
• pp.399-408
• /
• 2022

The purpose of this study was to compare the usefulness of the lipid ratio indicators for the diagnosis of metabolic syndrome in the elderly aged 65 years or older. From January 2018 to December 2020, 1,464 people aged 65 years or older who underwent a health checkup at a general hospital in Seoul were included. Lipid ratio indicators were measured through blood tests. The prevalence of metabolic syndrome according to the quartiles of the lipid ratio index was confirmed by logistic regression analysis. In addition, the metabolic syndrome predictive ability and cutoff value of the lipid ratio indices were estimated with the receiver operating characteristic(ROC) curve. The correlation between atherogenic index of plasma(AIP) and waist circumference was the highest in both men and women(r=0.278, p<0.001 vs r=0.252, p<0.001). As for the lipid ratio indices, the incidence of metabolic syndrome was higher in the fourth quartile than in the first quartile. The area under the ROC curve(AUC) value of AIP was higher at 0.826(95% CI=0.799-0.850) and 0.852(95% CI=0.820-0.881) for men and women, respectively, compared to other lipid ratio indicators, and the optimal cutoff values for both men and women was 0.44(p<0.001). Therefore, the AIP among the lipid ratio indicators was found to be the most useful index for diagnosing metabolic syndrome in the elderly aged 65 years or older.

• Korean Journal of Community Nutrition
• /
• v.19 no.5
• /
• pp.490-498
• /
• 2014

Objectives: This study was conducted to propose the need of re-establishing the criteria of the body weight classification in the elderly. We compared the Asia-Pacific Region Criteria (APR-C) with Entropy Model Criteria (ENT-C) using Morbidity rate of chronic diseases which correlates significantly with Body Mass Index (BMI). Methods: Subjects were 886 elderly female participating in the 2007-2009 Korea National Health and Nutrition Examination Survey (KNHANES). We compared APR-C with those of ENT-C using Receiver Operating Characteristics (ROC) curve and logistic regression analysis. Results: In the case of the morbidity of hypertension, the results were as follows: Where it was in the T-off point of APR-C, sensitivity was 67.5%, specificity was 43.1%, and Youden's index was 10.6. While in the cut-off point of ENT-C, it was 56.7%, 56.6%, and 13.3 respectively. In the case of the morbidity of diabetes, the results were as follows: In the cut-off point of APR-C, Youden's index was 14.2. While in the cut-off point of ENT-C, it was 17.2 respectively. The Area Under the ROC Curve (AUC) of the subjects who had more than 2 diseases among hypertension, diabetes, and dyslipidemia was 0.615 (95% CI: 0.578-0.652). Compared to the normal group, the odds ratio of the hypertension group which will belong to the overweight or obesity was 1.79 (95% CI: 1.30-2.47) in the APR-C, and 2.04 (95% CI: 1.49-2.80) in the ENT-C (p < 0.001). Conclusions: We conclude that the optimal cut-off point of BMI to distinguish between normal weight and overweight was $24kg/m^2$ (ENT-C) rather than $23kg/m^2$ (APR-C).