• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.20 no.4
• /
• pp.435-447
• /
• 2019

In order to prevent unnecessary re-admission, it is necessary to intensively manage the groups with high probability of re-admission. For this, it is necessary to develop a re-admission prediction model. Two - year discharge summary data of one university hospital were collected from 2016 to 2017 to develop a predictive model of re-admission. In this case, the re-admitted patients were defined as those who were discharged more than once during the study period. We conducted descriptive statistics and crosstab analysis to identify the characteristics of rehospitalized patients. The re-admission prediction model was developed using logistic regression, neural network, and decision tree. AUC (Area Under Curve) was used for model evaluation. The logistic regression model was selected as the final re-admission predictive model because the AUC was the best at 0.81. The main variables affecting the selected rehospitalization in the logistic regression model were Residental regions, Age, CCS, Charlson Index Score, Discharge Dept., Via ER, LOS, Operation, Sex, Total payment, and Insurance. The model developed in this study was limited to generalization because it was two years data of one hospital. It is necessary to develop a model that can collect and generalize long-term data from various hospitals in the future. Furthermore, it is necessary to develop a model that can predict the re-admission that was not planned.

• The Journal of Bigdata
• /
• v.5 no.2
• /
• pp.53-67
• /
• 2020

This study employees a supervised learning prediction model to detect nonconformity in advance of processed food manufacturing and processing businesses. The study was conducted according to the standard procedure of machine learning, such as definition of objective function, data preprocessing and feature engineering and model selection and evaluation. The dependent variable was set as the number of supervised inspection detections over the past five years from 2014 to 2018, and the objective function was to maximize the probability of detecting the nonconforming companies. The data was preprocessed by reflecting not only basic attributes such as revenues, operating duration, number of employees, but also the inspections track records and extraneous climate data. After applying the feature variable extraction method, the machine learning algorithm was applied to the data by deriving the company's risk, item risk, environmental risk, and past violation history as feature variables that affect the determination of nonconformity. The f1-score of the decision tree, one of ensemble models, was much higher than those of other models. Based on the results of this study, it is expected that the official food control for food safety management will be enhanced and geared into the data-evidence based management as well as scientific administrative system.

• Journal of the Korean earth science society
• /
• v.43 no.1
• /
• pp.199-209
• /
• 2022

In this study, we analyzed the preparedness of Korean citizens for earthquake hazards. For this purpose, we developed a questionnaire on the preparedness of citizens for earthquake hazards, consisting of three constructs (knowledge, awareness, and management). A total of 1,256 citizen responses were collected through probability proportionate-to-size sampling and then subjected to Rasch analysis, inferential statistical analysis, and cluster analysis. The findings are as follows. First, questionnaire analysis showed that overall, the earthquake preparedness of citizens was 'normal', with a lower management score than knowledge and awareness scores. Second, analysis of variables related to preparedness for earthquake hazards, gender, education level, actual distance, and safety awareness were found to influence preparedness for earthquake hazards. Third, correlation analysis revealed a high correlation between the three constructs of preparedness for earthquake hazards, namely knowledge, awareness, and management, indicating a structurally close relationship with each other. In addition, even if gender and education level differed, these structural correlations were similar. Through cluster analysis, the citizens were further divided into five groups; the group with moderate levels of the three constructs accounted for the majority of citizens. Considering these findings, we propose an educational orientation that fosters the preparedness of citizens for earthquake hazards.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.11
• /
• pp.147-155
• /
• 2022

In this study, we developed a system to dynamically balance a daily stock portfolio and performed trading simulations using gradient boosting and genetic algorithms. We collected various stock market data from stocks listed on the KOSPI and KOSDAQ markets, including investor-specific transaction data. Subsequently, we indexed the data as a preprocessing step, and used feature engineering to modify and generate variables for training. First, we experimentally compared the performance of three popular gradient boosting algorithms in terms of accuracy, precision, recall, and F1-score, including XGBoost, LightGBM, and CatBoost. Based on the results, in a second experiment, we used a LightGBM model trained on the collected data along with genetic algorithms to predict and select stocks with a high daily probability of profit. We also conducted simulations of trading during the period of the testing data to analyze the performance of the proposed approach compared with the KOSPI and KOSDAQ indices in terms of the CAGR (Compound Annual Growth Rate), MDD (Maximum Draw Down), Sharpe ratio, and volatility. The results showed that the proposed strategies outperformed those employed by the Korean stock market in terms of all performance metrics. Moreover, our proposed LightGBM model with a genetic algorithm exhibited competitive performance in predicting stock price movements.

• Journal of the Korean Society of Radiology
• /
• v.81 no.3
• /
• pp.610-619
• /
• 2020

Purpose The study aimed to investigate the role of Prostate Imaging Reporting and Data System version 2 (PI-RADS v2) in predicting incidental prostate cancer (PCa) or urothelial carcinoma (UCa) extension in urinary bladder (UB) cancer patients. Materials and Methods A total of 72 UB cancer patients who underwent radical cystoprostatectomy and 3 Tesla multiparametric MRI before surgery were enrolled. PI-RADS v2 ratings were assigned by two independent radiologists. All prostate specimens were examined by a single pathologist. We compared the multiparametric MRI findings rated using PI-RADS v2 with the pathologic data. Results Of the 72 UB cancer patients, 29 had incidental PCa (40.3%) and 20 showed UCa extension (27.8%), with an overlap for 3 patients. With a score of 4 as the cut-off value for predicting incidental PCa, the diagnostic accuracy was 65.3%, specificity was 90.7%, and positive predictive value (PPV) was 66.7%. The diagnostic accuracy for incidental UCa extension was 47.2%, specificity was 92.3%, and PPV was 83.3%. Conclusion Despite the low diagnostic accuracy, the PPV and specificity were relatively high. Therefore, PI-RADS v2 scores of 1, 2, or 3 may help exclude the probability of incidental PCa or UCa extension.

• Korean Journal of Radiology
• /
• v.24 no.12
• /
• pp.1249-1259
• /
• 2023

Objective: This study aimed to evaluate the clinical outcomes and cost-effectiveness of dual-energy X-ray absorptiometry (DXA) for osteoporosis screening. Materials and Methods: Eligible patients who had and had not undergone DXA screening were identified from among those aged 50 years or older at Kaohsiung Veterans General Hospital, Taiwan. Age, sex, screening year (index year), and Charlson comorbidity index of the DXA and non-DXA groups were matched using inverse probability of treatment weighting (IPTW) for propensity score analysis. For cost-effectiveness analysis, a societal perspective, 1-year cycle length, 20-year time horizon, and discount rate of 2% per year for both effectiveness and costs were adopted in the incremental cost-effectiveness (ICER) model. Results: The outcome analysis included 10337 patients (female:male, 63.8%:36.2%) who were screened for osteoporosis in southern Taiwan between January 1, 2012, and December 31, 2021. The DXA group had significantly better outcomes than the non-DXA group in terms of fragility fractures (7.6% vs. 12.5%, P < 0.001) and mortality (0.6% vs. 4.3%, P < 0.001). The DXA screening strategy gained an ICER of US$ -2794 per quality-adjusted life year (QALY) relative to the non-DXA at the willingness-to-pay threshold of US$ 33004 (Taiwan's per capita gross domestic product). The ICER after stratifying by ages of 50-59, 60-69, 70-79, and ≥ 80 years were US$ -17815, US$ -26862, US$ -28981, and US$ -34816 per QALY, respectively. Conclusion: Using DXA to screen adults aged 50 years or older for osteoporosis resulted in a reduced incidence of fragility fractures, lower mortality rate, and reduced total costs. Screening for osteoporosis is a cost-saving strategy and its effectiveness increases with age. However, caution is needed when generalizing these cost-effectiveness results to all older populations because the study population consisted mainly of women.

• Korean Journal of Radiology
• /
• v.24 no.12
• /
• pp.1284-1292
• /
• 2023

Objective: We investigated the impacts of computed tomography (CT) added to ultrasound (US) for preoperative evaluation of patients with papillary thyroid carcinoma (PTC) on staging, surgical extent, and postsurgical survival. Materials and Methods: Consecutive patients who underwent surgery for PTC between January 2015 and December 2015 were retrospectively identified. Of them, 584 had undergone preoperative additional thyroid CT imaging (CT + US group), and 859 had not (US group). Inverse probability of treatment weighting (IPTW) and propensity score matching (PSM) were used to adjust for 14 variables and balance the two groups. Changes in nodal staging and surgical extent caused by CT were recorded. The recurrence-free survival and distant metastasis-free survival after surgery were compared between the two groups. Results: In the CT + US group, discordant nodal staging results between CT and US were observed in 94 of 584 patients (16.1%). Of them, CT accurately diagnosed nodal staging in 54 patients (57.4%), while the US provided incorrect nodal staging. Ten patients (1.7%) had a change in the extent of surgery based on CT findings. Postsurgical recurrence developed in 3.6% (31 of 859) of the CT + US group and 2.9% (17 of 584) of the US group during the median follow-up of 59 months. After adjustment using IPTW (580 vs. 861 patients), the CT + US group showed significantly higher recurrence-free survival rates than the US group (hazard ratio [HR], 0.52 [95% confidence interval {CI}, 0.29-0.96]; P = 0.037). PSM analysis (535 patients in each group) showed similar HR without statistical significance (HR, 0.60 [95% CI, 0.31-1.17]; P = 0.134). For distant metastasis-free survival, HRs after IPTW and PSM were 0.75 (95% CI, 0.17-3.36; P = 0.71) and 0.87 (95% CI, 0.20-3.80; P = 0.851), respectively. Conclusion: The addition of CT imaging for preoperative evaluation changed nodal staging and surgical extent and might improve recurrence-free survival in patients with PTC.

• Journal of Chest Surgery
• /
• v.57 no.3
• /
• pp.242-251
• /
• 2024

Background: This study compared the outcomes of surgical aortic valve replacement (AVR) in patients aged 50 to 70 years based on the type of prosthetic valve used. Methods: We compared patients who underwent mechanical AVR to those who underwent bioprosthetic AVR at our institution between January 2000 and March 2019. Competing risk analysis and the inverse probability of treatment weighting (IPTW) method based on propensity score were employed for comparisons. Results: A total of 1,580 patients (984 patients with mechanical AVR; 596 patients with bioprosthetic AVR) were enrolled. There was no significant difference in early mortality between the mechanical AVR and bioprosthetic AVR groups (0.9% vs. 1.7%, p=0.177). After IPTW adjustment, the risk of all-cause mortality was significantly higher in the bioprosthetic AVR group than in the mechanical AVR group (hazard ratio [HR], 1.39; 95% confidence interval [CI], 1.07-1.80; p=0.014). Competing risk analysis revealed lower risks of stroke (sub-distributional hazard ratio [sHR], 0.44; 95% CI, 0.28-0.67; p<0.001) and anticoagulation-related bleeding (sHR, 0.35; 95% CI, 0.23-0.53; p<0.001) in the bioprosthetic AVR group. Conversely, the risk of aortic valve (AV) reintervention was higher in the bioprosthetic AVR group (sHR, 6.14; 95% CI, 3.17-11.93; p<0.001). Conclusion: Among patients aged 50 to 70 years who underwent surgical AVR, those receiving mechanical valves showed better survival than those with bioprosthetic valves. The mechanical AVR group exhibited a higher risk of stroke and anticoagulation-related bleeding, while the bioprosthetic AVR group showed a higher risk of AV reintervention.

• Clinical and Experimental Pediatrics
• /
• v.52 no.11
• /
• pp.1228-1233
• /
• 2009

Purpose : This study aimed to evaluate whether a shorter time from the arrival at a hospital to delivery is related to the occurrence of cerebral palsy in premature infants of less than 34 weeks of gestational age. Methods : We studied 142 newborns of less than 34 weeks of gestational age. The time from the arrival at the hospital to delivery was measured. The correlation between the time required for delivery and the occurrence of cerebral palsy was elucidated by diagnosing cerebral palsy in neonates using the Korean Infant Development Screening Test and neurological examination. Results : Preliminary result suggested that a shorter time from hospital arrival to delivery was related to a lower development score for gross motor activity and to a higher frequency of cerebral palsy occurrence. Moreover, it was responsible for a tendency of obtaining lower Apgar scores at 1 and 5 minutes. The shorter delivery time was associated with a higher probability of respiratory distress syndrome (RDS) occurrence when the length of delivery time was less than 6 hours and there was a higher probability of a shorter gestation period. However, the multifactor analysis revealed that there was little impact of delivery time on the occurrence of cerebral palsy. Conclusions : The length of hospital arrival time to delivery did not significantly influence the occurrence of cerebral palsy in premature infants of less than 34 weeks of gestational age.

• Tuberculosis and Respiratory Diseases
• /
• v.59 no.5
• /
• pp.473-479
• /
• 2005

Background : A pulmonary embolism often presents with nonspecific symptoms and signs. However, a delayed diagnosis can result in catastrophic outcome. The majority of preventable deaths associated with a pulmonary embolism can be ascribed to a missed diagnosis rather than to the failure of existing treatments. Therefore, accurate and rapid diagnostic methods are essential for the management of a pulmonary embolism. The recent generation of multidetector-row spiral CT scanners appears to outperform other imaging modalities in detecting a central and peripheral pulmonary embolism. However, there are some variations in the interpretations of the findings between observers. This study examined the inter-observer differences of the diagnoses in patients with a pulmonary embolism. Method : 64 patients who were diagnosed with a pulmonary embolism either clinically or with spiral chest CT from 2002 to 2004, were included. Two thoracic radiologists interpreted the multidetector-row spiral CT in terms of the diagnosis of a pulmonary embolism and the location of the thrombus independently. Among 64 patients, 14 patients were excluded because there was no evidence of a pulmonary embolism or there was different interpretation of the pulmonary embolism between radiologists. A clinical diagnosis was based on "Rules for predicting the probability of embolism". Results : The mean score of the patients according to the Wells method was $3.91{\pm}0.30$ (0-9). The accordance of the radiologists was 95% in the main, 85% in the lobar, 91.2% in the segmental, and 96% in the sub-segmental pulmonary arteries. After excluding the negative interpretation from both radiologists, their agreement was 76.2%(${\kappa}.$ 0.83) in the main, 57.6%(${\kappa}.$ 0.63) in the lobar, 51.5%(${\kappa}.$ 0.63) in the segmental, and 34.6%(${\kappa}.$ 0.49) in the sub-segmental pulmonary arteries. Conclusion : Chest CT has been recently applied to patients suspected of having a pulmonary embolism. It was found that spiral CT is a rapid test for diagnosing a thrombus, and there was reliable accordance between the observers from the area of the large pulmonary arteries. However, there was a lack of agreement between the observers in diagnosing thrombi located distal to the sub-segmental arteries.