Bayesian Test of Quasi-Independence in a Sparse Two-Way Contingency Table

Kwak, Sang-Gyu,Kim, Dal-Ho The Korean Statistical Society 2012 Communications for statistical applications and me Vol.19 No.3

We consider a Bayesian test of independence in a two-way contingency table that has some zero cells. To do this, we take a three-stage hierarchical Bayesian model under each hypothesis. For prior, we use Dirichlet density to model the marginal cell and each cell probabilities. Our method does not require complicated computation such as a Metropolis-Hastings algorithm to draw samples from each posterior density of parameters. We draw samples using a Gibbs sampler with a grid method. For complicated posterior formulas, we apply the Monte-Carlo integration and the sampling important resampling algorithm. We compare the values of the Bayes factor with the results of a chi-square test and the likelihood ratio test.

Effectiveness of pulsed radiofrequency treatment on cervical radicular pain : A meta-analysis

Kwak, Sang Gyu,Lee, Dong Gyu,Chang, Min Cheol Wolters Kluwer Health 2018 Medicine Vol.97 No.31

<P><B>Abstract</B></P><P><B>Background:</B></P><P>Cervical radicular pain is a challenging medical problem in terms of therapeutic management. Recently, pulsed radiofrequency (PRF) stimulation on the dorsal root ganglion (DRG) has been used to control several types of chronic pain. However, its effect on cervical radicular pain is still not well studied. To conduct a meta-analysis of available clinical studies on PRF treatment in patients with cervical radicular pain induced by cervical spine disease that was not responsive to other conservative treatments.</P><P><B>Methods:</B></P><P>A comprehensive database search was conducted on PubMed, Embase, Cochrane Library, and SCOPUS. We included studies published up to August 31, 2017, that fulfilled our inclusion and exclusion criteria. The pain degrees measured using visual analog scale (VAS) at pretreatment and after PRF on the DRG were collected for the meta-analysis. The Cochrane Collaboration's Handbook and Newcastle–Ottawa scale were used for the methodological quality assessments of included studies. The meta-analysis was performed using the Comprehensive Meta-analysis Version 2.</P><P><B>Results:</B></P><P>A total of 67 patients from one RCT study, 2 prospective observational studies, and one retrospective study were included in this meta-analysis. The pooled data of the 4 included studies showed that overall VAS after the PRF treatment was significantly reduced (<I>P </I>≤<I> </I>.001). In the subgroup analysis according to follow-up evaluation time points, the pain was significantly reduced at 2 weeks, 1 month, 3 months, and 6 months after the procedure (2 weeks: <I>P = </I>.02; 1, 3, and 6 months: <I>P <</I> .001).</P><P><B>Conclusion:</B></P><P>According to the results of the meta-analysis, the use of PRF on the DRG is effective for alleviating cervical radicular pain, which was unresponsive to oral medications, physical therapy, or epidural steroid injection.</P>

Two Bayesian methods for sample size determination in clinical trials

Kwak, Sang-Gyu,Kim, Dal-Ho,Shin, Im-Hee,Kim, Ho-Gak,Kim, Sang-Gyung The Korean Data and Information Science Society 2010 한국데이터정보과학회지 Vol.21 No.6

Sample size determination is very important part in clinical trials because it influences the time and the cost of the experimental studies. In this article, we consider the Bayesian methods for sample size determination based on hypothesis testing. Specifically we compare the usual Bayesian method using Bayes factor with the decision theoretic method using Bayesian reference criterion in mean difference problem for the normal case with known variances. We illustrate two procedures numerically as well as graphically.

Bayesian estimation for finite population proportions in multinomial data

Kwak, Sang-Gyu,Kim, Dal-Ho The Korean Data and Information Science Society 2012 한국데이터정보과학회지 Vol.23 No.3

We study Bayesian estimates for finite population proportions in multinomial problems. To do this, we consider a three-stage hierarchical Bayesian model. For prior, we use Dirichlet density to model each cell probability in each cluster. Our method does not require complicated computation such as Metropolis-Hastings algorithm to draw samples from each density of parameters. We draw samples using Gibbs sampler with grid method. We apply this algorithm to a couple of simulation data under three scenarios and we estimate the finite population proportions using two kinds of approaches We compare results with the point estimates of finite population proportions and their standard deviations. Finally, we check the consistency of computation using differen samples drawn from distinct iterates.

Bayesian estimation for finite population proportions in multinomial data

Sang Gyu Kwak,Dal Ho Kim 한국데이터정보과학회 2012 한국데이터정보과학회지 Vol.23 No.3

We study Bayesian estimates for finite population proportions in multinomial prob-lems. To do this, we consider a three-stage hierarchical Bayesian model. For prior, we use Dirichlet density to model each cell probability in each cluster. Our method does not require complicated computation such as Metropolis-Hastings algorithm to draw samples from each density of parameters. We draw samples using Gibbs sampler with grid method. We apply this algorithm to a couple of simulation data under three sce-narios and we estimate the finite population proportions using two kinds of approaches. We compare results with the point estimates of finite population proportions and their standard deviations. Finally, we check the consistency of computation using diffierent samples drawn from distinct iterates.

Normality Test in Clinical Research

( Sang Gyu Kwak ),( Sung-hoon Park ) 대한류마티스학회 2019 대한류마티스학회지 Vol.26 No.1

In data analysis, given that various statistical methods assume that the distribution of the population data is normal distribution, it is essential to check and test whether or not the data satisfy the normality requirement. Although the analytical methods vary depending on whether or not the normality is satisfied, inconsistent results might be obtained depending on the analysis method used. In many clinical research papers, the results are presented and interpreted without checking or testing normality. According to the central limit theorem, the distribution of the sample mean satisfies the normal distribution when the number of samples is above 30. However, in many clinical studies, due to cost and time restrictions during data collection, the number of samples is frequently lower than 30. In this case, a proper statistical analysis method is required to determine whether or not the normality is satisfied by performing a normality test. In this regard, this paper discusses the normality check, several methods of normality test, and several statistical analysis methods with or without normality checks. (J Rheum Dis 2019;26:5-11)

Two Bayesian methods for sample size determination in clinical trials

Sang Gyu Kwak,Dal Ho Kim,Im Hee Shin,Ho Gak Kim,Sang Gyung Kim 한국데이터정보과학회 2010 한국데이터정보과학회지 Vol.21 No.6

Sample size determination is very important part in clinical trials because it influ-ences the time and the cost of the experimental studies. In this article, we consider the Bayesian methods for sample size determination based on hypothesis testing. Specif-ically we compare the usual Bayesian method using Bayes factor with the decision theoretic method using Bayesian reference criterion in mean difference problem for the normal case with known variances. We illustrate two procedures numerically as well as graphically.

Evidence-based clinical recommendations for hypofractionated radiotherapy: exploring efficacy and safety - Part 1. Brain and head and neck

Soo-Yoon Sung(Soo-Yoon Sung),Jin Ho Song(Jin Ho Song),Byoung Hyuck Kim(Byoung Hyuck Kim),Yoo-Kang Kwak(Yoo-Kang Kwak),Kyung Su Kim(Kyung Su Kim),Gyu Sang Yoo(Gyu Sang Yoo),Hwa Kyung Byun(Hwa Kyung Byu 대한방사선종양학회 2024 Radiation Oncology Journal Vol.42 No.1

Advances in radiotherapy (RT) techniques, including intensity-modulated RT and image-guided RT, have allowed hypofractionation, increasing the fraction size over the conventional dose of 1.8–2.0 Gy. Hypofractionation offers advantages such as shorter treatment times, improved compliance, and under specific conditions, particularly in tumors with a low α/β ratio, higher efficacy. It was initially explored for use in RT for prostate cancer and adjuvant RT for breast cancer, and its application has been extended to various other malignancies. Hypofractionated RT (HFRT) may also be effective in patients who are unable to undergo conventional treatment owing to poor performance status, comorbidities, or old age. The treatment of brain tumors with HFRT is relatively common because brain stereotactic radiosurgery has been performed for over two decades. However, re-irradiation of recurrent lesions and treatment of elderly or frail patients are areas under investigation. HFRT for head and neck cancer has not been widely used because of concerns regarding late toxicity. Thus, we aimed to provide a comprehensive summary of the current evidence for HFRT for brain tumors and head and neck cancer and to offer practical recommendations to clinicians faced with the challenge of choosing new treatment options.

Afatinib-Induced Acute Fatal Pneumonitis in Metastatic Lung Adenocarcinoma

Sang Hoon Yoo,Jin Ah Ryu,Seo Ree Kim,Su Yun Oh,Gu Sung Jung,Dong Jae Lee,Bong Gyu Kwak,Yu Hyun Nam,Kyung Hyun Kim,Young Jun Yang 대한가정의학회 2016 Korean Journal of Family Medicine Vol.37 No.6

Afatinib is an oral tyrosine kinase inhibitor (TKI) that inhibit Endothelial Growth Factor Receptor (EGFR), Human Epidermal Growth Factor Receptor 2 (HER2), and HER4. The common side effects of EGFR TKI are rash, acne, di-arrhea, stomatitis, pruritus, nausea, and loss of appetite. Drug induced pneumonitis is the less common adverse ef-fects of EGFR TKI. Afatinib, 2nd generation EGFR TKI is anticipated to overcome drug resistance from 1st genera-tion EGFR TKI according to preclinical study, and several studies are being conducted to compare clinical efficacy between 1st and 2nd EGFR TKI. Several cases of rug induced acute fatal pneumonitis were reported after use of er-lotinib or gefitinib. However, a case of acute fatal pneumonitis associated with afatinib was note reported except drug induced pneumonitis in other clinical study. Here, we present a cases of acute severe pneumonitis related with afatinib in metastatic lung adenocarcinoma with literature review.

Flexible metal-insulator-metal (MIM) devices for plastic film AM-LCD

Sung-Jei Hong,Chan-Jae Lee,Jeong-In Han,Won-Keun Kim,Dae-Gyu Moon,Min-Gi Kwak,Sung-Kyu Park,Yong-Hoon Kim 한국물리학회 2002 Current Applied Physics Vol.2 No.3

We developed new flexible metal–insulator–metal (MIM) devices subject to plastic film substrate. The structure of the MIM device is that a Ta2O5 insulator is covered with two flexible Al electrodes on both sides. The flexible structure of the MIM device was successfully fabricated applyingour own etch-free process. 2002 Elsevier Science B.V. All rights reserved.