• 보건행정학회지
• /
• 제9권2호
• /
• pp.1-20
• /
• 1999

This study was conducted to investigate the predictors of hospital bankruptcy in Korea and to examine the predictive power for 3 types of statistical models of hospital bankruptcy. Data on 17 financial and 4 non-financial indicators of 30 bankrupt and 30 profitable hospitals in 1. 2, and 3 years before bankruptcy were obtained from the hospital performance databank of Korea Institute of Health Services Management. Significant variables were identified through mean comparison of each indicator between bankrupt and profitable hospitals, and the predictive power of statistical models of hospital bankruptcy were compared. The major findings are as follows. 1. Nine out of 21 indicators - fixed ratio, quick ratio, operating profit to total assets, operating profit to gross revenue, normal profit to total assets,normal profit to gross revenue, net profit to gross revenue, inventories turnrounds, and added value per adjusted patient - were found to be significantly predictitive variables in Logit and Probit models. 2. The predicdtive power of discriminant model of hospital bankruptcy in 1. 2, and 3 years before bankruptcy were 85.4, 79.0, and 83.8% respectively. With regard to the predictive power of the Logit model of hospital bankruptcy, they were 82.3, 75.8, and 80.6% respectively, and of the Probit model. 87.1. 80.6, and 88.7% respectively. 3. The predictive power of the Probit model of hospital bankruptcy is better than the other two predictive models.

• 대한간호학회지
• /
• 제49권5호
• /
• pp.575-585
• /
• 2019

Purpose: The purpose of this study was to develop predictive models for pressure ulcer incidence using electronic health record (EHR) data and to compare their predictive validity performance indicators with that of the Braden Scale used in the study hospital. Methods: A retrospective case-control study was conducted in a tertiary teaching hospital in Korea. Data of 202 pressure ulcer patients and 14,705 non-pressure ulcer patients admitted between January 2015 and May 2016 were extracted from the EHRs. Three predictive models for pressure ulcer incidence were developed using logistic regression, Cox proportional hazards regression, and decision tree modeling. The predictive validity performance indicators of the three models were compared with those of the Braden Scale. Results: The logistic regression model was most efficient with a high area under the receiver operating characteristics curve (AUC) estimate of 0.97, followed by the decision tree model (AUC 0.95), Cox proportional hazards regression model (AUC 0.95), and the Braden Scale (AUC 0.82). Decreased mobility was the most significant factor in the logistic regression and Cox proportional hazards models, and the endotracheal tube was the most important factor in the decision tree model. Conclusion: Predictive validity performance indicators of the Braden Scale were lower than those of the logistic regression, Cox proportional hazards regression, and decision tree models. The models developed in this study can be used to develop a clinical decision support system that automatically assesses risk for pressure ulcers to aid nurses.

• 인터넷정보학회논문지
• /
• 제23권3호
• /
• pp.125-135
• /
• 2022

본 연구는 데이터, 네트워크, 인공지능을 기반으로 산업 생태계 조성을 위해 구축된 정부통계지표가 약국 매출액에 영향을 미치는지 머신러닝을 이용하여 변인을 탐색하고 약국 매출액 예측에 적합한 분석 기법을 제공하고자 한다. 이에, 본 연구는 28개 정부통계지표와 소매업종인 약국을 대상으로 2016년 1월부터 2021년 12월까지의 분석 데이터를 활용하여 머신러닝 기법인 랜덤 포레스트, XGBoost, LightGBM, CatBoost을 통해 예측 변인 및 성능을 탐색하였다. 분석결과 경기관련 지표인 경제심리지수, 경기동행지수순환변동치, 소비자심리지수는 약국 매출액에 영향을 미치는 중요한 변인으로 나타났고, 회귀성능은 지표 MAE, MSE, RMSE를 살펴본 결과 랜덤 포레스트가 XGBoost, LightGBM, CatBoost 보다 성능이 가장 우수하게 나타났다. 이에, 본 연구는 머신러닝 결과를 토대로 약국 매출액에 영향을 미치는 변인과 최적의 머신러닝 기법을 제시하였으며, 여러 시사점과 후속연구를 제안하였다.

• 보건행정학회지
• /
• 제12권3호
• /
• pp.54-76
• /
• 2002

This study investigated how much EVA which evaluate firm's value can explain hospital bankruptcy prediction as a explanatory variable including financial indicators in Korea. In this study, artificial neural network and logit regression which are traditional statistical were used as the model for bankruptcy prediction. Data used in this study were financial and economic value added indicators of 34 bankrupt and -:4 non-bankrupt hospitals from the Database of Korean Health Industry Development Institute. The main results of this study were as follows: First, there was a significant difference between the financial variable model including EVA and the financial variable model excluding EVA in pre-bankruptcy analysis. Second, EVA could forecast bankruptcy hospitals up to 83% by the logistic analysis. Third, the EVA model outperformed the financial model in terms of the predictive power of hospital bankruptcy. Fourth, The predictive power of neural network model of hospital bankruptcy was more powerful than the legit model. After all the result of this study will be useful to future study on EVA to evaluate bankruptcy hospitals forecast.

• 진공이야기
• /
• 제2권1호
• /
• pp.31-34
• /
• 2015

This article introduces fundamentals of self-diagnosis and predictive (or preventive) maintenance technologies for dry vacuum pumps. The state variables of dry pumps are addressed, such as the pump and motor body temperatures, consumption currents of main and booster pumps, mechanical vibration, and exhaust pressure, etc. The adaptive parametric models of the state variables of the dry pump are exploited to provide dramatic reduction of data size and computation time for self-diagnosis. Two indicators, the Hotelling's $T^2$ and the sum of squares residuals (Q), are illustrated to be quite effective and successful in diagnosing dry pumps used in the semiconductor processes.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제17권8호
• /
• pp.2016-2029
• /
• 2023

Accurate prediction of critical illness is significant for ensuring the lives and health of patients. The selection of indicators affects the real-time capability and accuracy of the prediction for critical illness. However, the diversity and complexity of these indicators make it difficult to find potential connections between them and critical illnesses. For the first time, this study proposes an indicator analysis model to extract key indicators from the preoperative and intraoperative clinical indicators and laboratory results of critical illnesses. In this study, preoperative and intraoperative data of heart failure and respiratory failure are used to verify the model. The proposed model processes the datum and extracts key indicators through four parts. To test the effectiveness of the proposed model, the key indicators are used to predict the two critical illnesses. The classifiers used in the prediction are light gradient boosting machine (LightGBM) and eXtreme Gradient Boosting (XGBoost). The predictive performance using key indicators is better than that using all indicators. In the prediction of heart failure, LightGBM and XGBoost have sensitivities of 0.889 and 0.892, and specificities of 0.939 and 0.937, respectively. For respiratory failure, LightGBM and XGBoost have sensitivities of 0.709 and 0.689, and specificity of 0.936 and 0.940, respectively. The proposed model can effectively analyze the correlation between indicators and postoperative critical illness. The analytical results make it possible to find the key indicators for postoperative critical illnesses. This model is meaningful to assist doctors in extracting key indicators in time and improving the reliability and efficiency of prediction.

• Asian-Australasian Journal of Animal Sciences
• /
• 제17권4호
• /
• pp.554-559
• /
• 2004

The objective of the present study was to investigate the accuracy of mastitis diagnostic indicators for different dairy goat breeds. Biweekly milk samples were collected from individual half mammary gland of seven Saanen and seven Alpine dairy goats in the period of 40 to 120 days in milk. With threshold value set at 2.8 and 3.1 for Alpine and Saanen dairy goats, respectively, log (SPC) offered good sensitivity (0.89, 0.93), specificity (0.88, 0.95), positive predictive value (0.75, 0.85) and negative predictive value (0.95, 0.98) as a mastitis diagnostic tool. The correlations of log (SPC) with milk yield, log (SCC), ALP, LDH, $Na^{+}$, $K^{+}$ and EC were significant in Saanen dairy goats (p<0.05), with the highest correlation coefficient (0.653) existing between log (SPC) and log (SCC). The correlations of log (SPC) with milk yield, milk fat, milk protein, log (SCC), $Na^{+}$, $K^{+}$, EC were significant in Alpine dairy goats (p<0.05), with the highest correlation coefficient (0.416) existing between log (SPC) and log (SCC). There were different best-fit regression equations with different multiple diagnostic indicators for Saanen and Alpine dairy goats. In conclusion, different breeds of dairy goats may have to adapt different mastitis diagnostic parameters for a better diagnosis.

• 대한의용생체공학회:의공학회지
• /
• 제43권1호
• /
• pp.19-26
• /
• 2022

The Comprehensive in vitro Proarrhythmic Assay(CiPA) project was launched for solving the hERG assay problem of being classified as high-risk groups even though they are low-risk drugs due to their high sensitivity. CiPA presented a protocol to predict drug toxicity using physiological data calculated based on the in-silico model. in this study, features calculated through the in-silico model are analyzed for correlation of changing action potential in the near future, and features are verified through predictive performance according to drug datasets. Using the O'Hara Rudy model modified by Dutta et al., Pearson correlation analysis was performed between 13 features(dVm/dtmax, APpeak, APresting, APD90, APD50, APDtri, Ca_peak, Ca_resting, CaD90, CaD50, CaDtri, qNet, qInward) calculated at 100 pacing, and between dVm/d_{tmax_repol} calculated at 1,000 pacing, and linear regression analysis was performed on each of the 12 training drugs, 16 verification drugs, and 28 drugs. Indicators showing high coefficient of determination(R²) in the training drug dataset were qNet 0.93, AP resting 0.83, APDtri 0.78, Ca resting 0.76, dVm/dt_max 0.63, and APD90 0.61. The indicators showing high determinants in the validated drug dataset were APDtri 0.94, APD90 0.92, APD50 0.85, CaD50 0.84, qNet 0.76, and CaD90 0.64. Indicators with high coefficients of determination for all 28 drugs are qNet 0.78, APD90 0.74, and qInward 0.59. The indicators vary in predictive performance depending on the drug dataset, and qNet showed the same high performance of 0.7 or more on the training drug dataset, the verified drug dataset, and the entire drug dataset.

• 응용통계연구
• /
• 제37권2호
• /
• pp.119-138
• /
• 2024

본 연구는 노이즈 필터링과 차원축소 등의 방법을 이용하여 텍스트 지표의 정상화에 대해 검토하고 실증 분석을 통해 동 지표의 활용가능성을 제고할 수 있는 후처리 과정을 탐색하고자 하였다. 실증분석에 대한 예측 목표 변수로 월별 선행지수 순환 변동치, BSI 전산업 매출실적, BSI 전산업 매출전망 그리고 분기별 실질 GDP SA전기비와 실질 GDP 원계열 전년동기비를 상정하고 계량경제학에서 널리 활용되는 Hodrick and Prescott 필터와 비모수 차원축소 방법론인 충분차원축소를 비정형 텍스트 데이터와 결합하여 분석하였다. 분석 결과 월별과 분기별 변수 모두에서 자료의 수가 많은 경우 텍스트 지표의 노이즈 필터링이 예측 정확도를 높이고, 차원 축소를 적용함에 따라 보다 높은 예측력을 확보할 수 있음을 확인하였다. 분석 결과가 시사하는 바는 텍스트 지표의 활용도 제고를 위해서는 노이즈 필터링과 차원 축소 등의 후처리 과정이 중요하며 이를 통해 경기 예측의 정도를 높일 수 있다는 것이다.

• Safety and Health at Work
• /
• 제13권4호
• /
• pp.394-400
• /
• 2022

Background: Impacts of exposure are generally monitored and recorded after injuries or illness occur. Yet, absence of conventional after-the-effect impacts (i.e., lagging indicators), tend to focus on physical health and injuries, and fail to inform if workers are not exposed to safety and health hazards. In contrast to lagging indicators, leading indicators are proactive, preventive, and predictive indexes that offer insights how effective safety and health. The present study is to validate an extended Voluntary Protection Programs (VPP) that consists of six leading indicators. Methods: Questionnaires were distributed to 13 organizations (response rate = 93.1%, 1,439 responses) in Taiwan. Cronbach α, multiple linear regression and canonical correlation were used to test the reliability of the extended Voluntary Protection Programs (VPP) which consists of six leading indicators (safe climate, transformational leadership, organizational justice, organizational support, hazard prevention and control, and training). Criteria-related validation strategy was applied to examine relationships of six leading indicators with six criteria (perceived health, burnout, depression, job satisfaction, job performance, and life satisfaction). Results: The results showed that the Cronbach's α of six leading indicators ranged from 0.87 to 0.92. The canonical correlation analysis indicated a positive correlation between the six leading indicators and criteria (1st canonical function: correlation = 0.647, square correlation = 0.419, p < 0.001). Conclusions: The present study validates the extended VPP framework that focuses on promoting safety and physical and mental health. Results further provides applications of the extended VPP framework to promote workers' safety and health.