• 한국경영과학회:학술대회논문집
• /
• 대한산업공학회/한국경영과학회 1991년도 춘계공동학술대회 발표논문 및 초록집; 전북대학교, 전주; 26-27 Apr. 1991
• /
• pp.143-143
• /
• 1991

• Korean Journal of Acupuncture
• /
• 제33권3호
• /
• pp.102-113
• /
• 2016

Objectives : This study is aimed at developing and discussing the prediction model of blood stasis pattern of traditional Korean medicine(TKM) using machine learning algorithms: multiple logistic regression and decision tree model. Methods : First, we reviewed the blood stasis(BS) questionnaires of Korean, Chinese, and Japanese version to make a integrated BS questionnaire of patient-reported outcomes. Through a human subject research, patients-reported BS symptoms data were acquired. Next, experts decisions of 5 Korean medicine doctor were also acquired, and supervised learning models were developed using multiple logistic regression and decision tree. Results : Integrated BS questionnaire with 24 items was developed. Multiple logistic regression models with accuracy of 0.92(male) and 0.95(female) validated by 10-folds cross-validation were constructed. By decision tree modeling methods, male model with 8 decision node and female model with 6 decision node were made. In the both models, symptoms of 'recent physical trauma', 'chest pain', 'numbness', and 'menstrual disorder(female only)' were considered as important factors. Conclusions : Because machine learning, especially supervised learning, can reveal and suggest important or essential factors among the very various symptoms making up a pattern identification, it can be a very useful tool in researching diagnostics of TKM. With a proper patient-reported outcomes or well-structured database, it can also be applied to a pre-screening solutions of healthcare system in Mibyoung stage.

• Journal of the Korean Statistical Society
• /
• 제31권2호
• /
• pp.237-250
• /
• 2002

Motivated by Cook's (1986) assessment of local influence by investigating the curvature of a surface associated with the overall discrepancy measure, this paper extends this idea to the linear regression model with AR(p) disturbances. Diagnostic for the linear regression models with AR(p) disturbances are discussed when simultaneous perturbations of the response vector are allowed. For the derived criterion, numerical studies demonstrate routine application of this work.

• 응용통계연구
• /
• 제25권3호
• /
• pp.503-512
• /
• 2012

Detecting outliers in a linear regression model eventually fails when similar observations are classified differently in a sequential process. In such circumstances, identifying clusters and applying certain methods to the clustered data can prevent a failure to detect outliers and is computationally efficient due to the reduction of data. In this paper, we suggest to implement a clustering procedure for this purpose and provide examples that illustrate the suggested procedure applied to the Hadi-Simonoff (1993) method, reverse Hadi-Simonoff method, and Gentleman-Wilk (1975) method.

• Communications for Statistical Applications and Methods
• /
• 제14권2호
• /
• pp.401-411
• /
• 2007

In this paper we propose an influence measure for detecting potentially influential observations using the infinitesimal perturbation and the local influence in the log-linear regression model. Also, we propose a goodness-of-fit measure for variable selection. A real data set are used for illustration.

• Journal of the Korean Statistical Society
• /
• 제22권1호
• /
• pp.27-37
• /
• 1993

Since the analytic expression for a diagnostic in the logistic regression model is not available, one-step estimation is often used by a case-deletion point of view. In this paper, infinitesimal perturbation approach is used, and it is shown that the scale transformation of infinitesimal perturbation approach is eventually equal to the weighted perturbation of local influence approach and the replacement measure. Also, multiple cases deletion for the masking effect is considered.

• Journal of Gastric Cancer
• /
• 제21권4호
• /
• pp.335-351
• /
• 2021

Purpose: An underlying factor for the failure of several clinical trials of anti-epidermal growth factor receptor (EGFR) therapies is the lack of an effective method to identify patients who overexpress EGFR protein. The quantitative dot blot method (QDB) was used to measure EGFR protein levels objectively, absolutely, and quantitatively. Its feasibility was evaluated for the prognosis of overall survival (OS) of patients with gastric cancer. Materials and Methods: Slices of 2×5 ㎛ from formalin-fixed paraffin-embedded gastric cancer specimens were used to extract total tissue lysates for QDB measurement. Absolutely quantitated EGFR protein levels were used for the Kaplan-Meier OS analysis. Results: EGFR protein levels ranged from 0 to 772.6 pmol/g (n=246) for all gastric cancer patients. A poor correlation was observed between quantitated EGFR levels and immunohistochemistry scores with ρ=0.024 and P=0.717 in Spearman's correlation analysis. EGFR was identified as an independent negative prognostic biomarker for gastric cancer patients only through absolute quantitation, with a hazard ratio of 1.92 (95% confidence interval, 1.05-3.53; P=0.034) in multivariate Cox regression OS analysis. A cutoff of 208 pmol/g was proposed to stratify patients with a 3-year survival probability of 44% for patients with EGFR levels above the cutoff versus 68% for those below the cutoff based on Kaplan-Meier OS analysis (log rank test, P=0.002). Conclusions: A QDB-based assay was developed for gastric cancer specimens to measure EGFR protein levels absolutely, quantitatively, and objectively. This assay should facilitate clinical trials aimed at evaluation of anti-EGFR therapies retrospectively and prospectively for gastric cancer.

• 대한한의진단학회지
• /
• 제17권1호
• /
• pp.63-76
• /
• 2013

The aims of this study are to investigate a logisitic regression equation of the vacuous pulse and the replete pulse for efficacy evaluation of clip-type pulsimeter by using magnetic Hall device. To evaluate the efficacy of clip-type pulsimeter by using magnetic Hall device as sensing the minute movement of a radial artery, one research clinical trial have been performed. The number of subject was 120, the clinical data of patients did treated with a normal statistical method. The systolic peak amplitude, the reflective peak amplitude and time, and the notch peak amplitude and time are analyzed major efficacy parameters to discern the vacuous pulse and the replete pulse. The equations included of five parameters such as systolic peak amplitude, the reflective peak amplitude and time, and the notch peak amplitude and notch amplitude time for determination of the vacuous pulse and the replete pulse were deducted by statistical logistic regression method. It suggests that the logistic regression equations are possible to develop the oriental algorithm for pulse diagnosis.

• Journal of the Korean Statistical Society
• /
• 제35권2호
• /
• pp.213-224
• /
• 2006

We investigate the influence of subjects or observations on regression coefficients of generalized estimating equations using the influence function and the derivative influence measures. The influence function for regression coefficients is derived and its sample versions are used for influence analysis. The derivative influence measures under certain perturbation schemes are derived. It can be seen that the influence function method and the derivative influence measures yield the same influence information. An illustrative example in longitudinal data analysis is given and we compare the results provided by the influence function method and the derivative influence measures.

• Communications for Statistical Applications and Methods
• /
• 제8권2호
• /
• pp.311-317
• /
• 2001

It is necessary to check the linearity of selected covariates in regression diagnostics. There are various graphical methods using residual plots such as partial residual plots, augmented partial residual plots and combining conditional expectation and residual plots. In this paper, we propose the modified pseudolikelihood ratio test statistics based on these residual plots to test linearity of selected covariate. These test statistics which measure the distance between the nonparametric and parametric models are derived as a ratio of quadratic forms. The approximate distribution of these statistics is calculated numerically by using three moments. The power comparison of these statistics is given.