• 한국인공지능학회지
• /
• 제11권3호
• /
• pp.29-34
• /
• 2023

Recently, the rapid development of artificial intelligence technology, many studies are being conducted to predict the risk of heart disease in order to lower the mortality rate of cardiovascular diseases worldwide. This study presents exercise or dietary improvement contents in the form of a software app or web to patients with cardiovascular disease, and cardiovascular disease through digital devices such as mobile phones and PCs. LR, LDA, SVM, XGBoost for the purpose of developing "Life style Improvement Contents (Digital Therapy)" for cardiovascular disease care to help with management or treatment We compared and analyzed cardiovascular disease prediction models using machine learning algorithms. Research Results XGBoost. The algorithm model showed the best predictive model performance with overall accuracy of 80% before and after. Overall, accuracy was 80.0%, F1 Score was 0.77~0.79, and ROC-AUC was 80%~84%, resulting in predictive model performance. Therefore, it was found that the algorithm used in this study can be used as a reference model necessary to verify the validity and accuracy of cardiovascular disease prediction. A cardiovascular disease prediction analysis algorithm that can enter accurate biometric data collected in future clinical trials, add lifestyle management (exercise, eating habits, etc.) elements, and verify the effect and efficacy on cardiovascular-related bio-signals and disease risk. development, ultimately suggesting that it is possible to develop lifestyle improvement contents (Digital Therapy).

• Hip & pelvis
• /
• 제36권2호
• /
• pp.135-143
• /
• 2024

Purpose: Hip fractures are associated with increased mortality. The identification of risk factors of mortality could improve patient care. The aim of the study was to identify risk factors of mortality after surgery for a hip fracture and construct a mortality model. Materials and Methods: A cohort study was conducted on patients with hip fractures at two institutions. Five hundred and ninety-seven patients with hip fractures that were treated in the tertiary hospital, and another 147 patients that were treated in a secondary hospital. The perioperative data were collected from medical charts and interviews. Functional Assessment Measure score, Short Form-12 and mortality were recorded at 12 months. Patients and surgery variables that were associated with increased mortality were used to develop a mortality model. Results: Mortality for the whole cohort was 19.4% at one year. From the variables tested only age >80 years, American Society of Anesthesiologists category, time to surgery (>48 hours), Charlson comorbidity index, sex, use of anti-coagulants, and body mass index <25 kg/m² were associated with increased mortality and used to construct the mortality model. The area under the curve for the prediction model was 0.814. Functional outcome at one year was similar to preoperative status, even though their level of physical function dropped after the hip surgery and slowly recovered. Conclusion: The mortality prediction model that was developed in this study calculates the risk of death at one year for patients with hip fractures, is simple, and could detect high risk patients that need special management.

• Earthquakes and Structures
• /
• 제25권6호
• /
• pp.469-479
• /
• 2023

A probabilistic seismic damage analysis is an essential procedure to identify seismically vulnerable structures, prioritize the seismic retrofit, and ultimately minimize the overall seismic risk. To assess the seismic risk of multiple structures within a region, a large number of nonlinear time-history structural analyses must be conducted and studied. As a result, each assessment requires high computing resources. To overcome this limitation, we explore a deep learning-based metamodel to enable the prediction of the mean and the standard deviation of the seismic damage distribution of track-on steel-plate girder railway bridges in Korea considering the geometric variation. For machine learning training, nonlinear dynamic time-history analyses are performed to generate 800 high-fidelity datasets on the seismic response. Through intensive trial and error, the study is concentrated on developing an optimal machine learning architecture with the pre-identified variables of the physical configuration of the bridge. Additionally, the prediction performance of the proposed method is compared with a previous, well-defined, response surface model. Finally, the statistical testing results indicate that the overall performance of the deep-learning model is improved compared to the response surface model, as its errors are reduced by as much as 61%. In conclusion, the model proposed in this study can be effectively deployed for the seismic fragility and risk assessment of a region with a large number of structures.

• Architectural research
• /
• 제21권2호
• /
• pp.49-57
• /
• 2019

Risk assessment during pre-construction phase is important due to the uncertainty of the risks that may exist in projects. Risk checklist is a method to systematically classify and organize the risks that have been experienced in the past, and to identify the risk factors that may be present in the future projects. In addition, risk value assessment based on checklists plays a key role in risk management, and various risk assessment researches have been conducted to carry out this systematically. However, previous approaches have limitations in common, this is because risk values are evaluated individually in risk checklists, which ignore interdependencies among risk factors and neglect the emergence of co-occurrence of risks. Hence, when multiple risk factors cooccur, they cannot be far off from the conventional method of summing the total risk value to establish the risk response strategy. Most of risk factors are interdependent and may have multiple effects if occurred than expected. In particular, specific cause can be overlapped if multiple risks co-occur, and this may result in overestimation of the risk response for the future project. Thus, the objective of this research is to propose a model to help decision makers to quantify the risk value reflecting the interdependency during the identification phase using existing risk checklist that is currently being practiced in actual construction projects. The proposed model will provide the guideline to support the prediction and identification of the interdependency of risks in practice. In addition, the better understanding and prediction of the exceeding risk response by co-occurring risks during the risk identification phase for decision makers.

• 한국안전학회지
• /
• 제33권4호
• /
• pp.98-104
• /
• 2018

Macroscopic accident analyses have been conducted to incorporate transportation safety into long-term transportation planning. In macro-level accident prediction model, exposure variable(e.g. a settled population) have been used as fundamental explanatory variable under the concept that each trip will be subjected to a probable risk of accident. However, a settled population may be embedded error by exclusion of active population concept. The objective of this research study is to develop macro-level accident prediction model using floating population variable(concept of including a settled population and active population) collected from mobile phone data. The concept of accident prediction models is introduced utilizing exposure variable as explanatory variable in a generalized linear regression with assumption of a negative binomial error structure. The goodness of fit of model using floating population variable is compared with that of the each models using population and the number of household variables. Also, log transformation models are additionally developed to improve the goodness of fit. The results show that the log transformation model using floating population variable is useful for capturing the relationships between accident and exposure variable and generally perform better than the models using other existing exposure variables. The developed model using floating population variable can be used to guide transportation safety policy decision makers to allocate resources more efficiently for the regions(or zones) with higher risk and improve urban transportation safety in transportation planning step.

• 한국재난정보학회 논문집
• /
• 제16권4호
• /
• pp.842-848
• /
• 2020

연구목적: 노후화된 발전기의 지속 가능한 운영을 위하여 효율적이며, 안전한 운영이 중요하다. 효율적 운영이란 경제적 관점이며, 안전한 운영은 발전 설비의 치명적 사고 발생에 대한 발생 이전의 사전 조치를 말한다. 그러므로 발전기의 지속가능 운영 모니터링을 위하여 관련된 센서 설치와 이를 기반으로 지속 가능에 대한 예측할 수 있는 모델에 대한 연구가 필요하다. 연구방법: 전기와 열에 대한 수요 예측, 엔진의 성능과 이상을 탐지하는 예측, 그리고 재 난 안전에 대한 예측 모델을 제시하였다. 이를 위하여 필요한 센서를 정의하였으며, 이를 기반으로 예측 모델을 각각 개발하여 수행하였다. 연구결과: 수요 예측 모델은 기존의 79%에서 90% 이상으로 예측 정확도를 향상시켰으며, 다른 2개 모델도 시스템의 지속가능한 안정적 운영을 지원하였다. 결론: 노후화된 발전설비의 지속가능 운영을 지원하기 위한 3가지 종류의 예측 모델을 개발하고 이를 제이비주식회사의 발전 설비에 실제 적용하여 운영하고 있다.

• 문화기술의 융합
• /
• 제6권4호
• /
• pp.289-294
• /
• 2020

기능성 위장장애는 Helicobacter pylori 감염 및 비 스테로이드성 항염증제의 사용 등의 원인으로 발생하는 소화기 계통 질환이다. 그동안 기능성 위장장애의 위험요인에 대한 많은 연구들이 수행되어졌으나, 한국인에 대한 기능성 위장장애 예측 모델 제시에 대한 연구는 없는 실정이다. 따라서 본 연구의 목적은 중년 및 노년층을 대상으로 인구학적정보, 비만정보, 혈액정보, 영양성분 정보를 바탕으로 머신러닝을 이용하여 기능성위장장애 예측 모델을 구현하고 평가하는 것이다. 모델생성을 위해 wrapper-based variable selection 메소드와 naive Bayes 알고리즘이 사용되었다. 여성 예측 모델의 분류 정확도는 0.712의 the area under the receiver operating characteristics curve(AUC) 값을 나타냈고, 남성에서는 여성보다 낮은 0.674의 AUC값이 나타났다. 이러한 연구결과는 향후 중년 및 노년층의 위장장애 질환의 예측과 예방에 활용될 수 있다.

• Journal of the Korean Data and Information Science Society
• /
• 제18권4호
• /
• pp.915-925
• /
• 2007

Multivariate GARCH has been useful to model dynamic relationships between volatilities arising from each component series of multivariate time series. Methodologies including EWMA(Exponentially weighted moving-average model), DVEC(Diagonal VEC model), BEKK and CCC(Constant conditional correlation model) models are comparatively reviewed for bivariate time series. In addition, these models are applied to evaluate VaR(Value at Risk) and to construct joint prediction region. To illustrate, bivariate stock prices data consisting of Samsung Electronics and LG Electronics are analysed.

• 한국대기환경학회지
• /
• 제18권4호
• /
• pp.327-334
• /
• 2002

The work presents sensitivity and uncertainty analysis of 2-compartment model for the evaluation of indoor radon pollution in a house. Effort on the development of such model is directed towards the prediction of the generation and transfer of radon in indoor air released from groundwater. The model is used to estimate a quantitative daily human exposure through inhalation of such radon based on exposure scenarios. However, prediction from the model has uncertainty propagated from uncertainties in model parameters. In order to assess how model predictions are affected by the uncertainties of model inputs, the study performs a quantitative uncertainty analysis in conjunction with the developed model. An importance analysis is performed to rank input parameters with respect to their contribution to model prediction based on the uncertainty analysis. The results obtained from this study would be used to the evaluation of human risk by inhalation associated with the indoor pollution by radon released from groundwater.

• 임상간호연구
• /
• 제19권2호
• /
• pp.265-274
• /
• 2013

Purpose: The aim of this study was to determine the risk factors contributed to unplanned readmission to intensive care unit (ICU) and to investigate the prediction model of unplanned readmission. Methods: We retrospectively reviewed the electronic medical records which included the data of 3,903 patients who had discharged from ICUs in a university hospital in Seoul from January 2011 to April 2012. Results: The unplanned readmission rate was 4.8% (n=186). The nine variables were significantly different between the unplanned readmission and no readmission groups: age, clinical department, length of stay at 1st ICU, operation, use of ventilator during 24 hours a day, APACHE II score at ICU admission and discharge, direct nursing care hours and Glasgow coma scale total score at 1st ICU discharge. The clinical department, length of stay at 1st ICU, operation and APACHE II score at ICU admission were the significant predictors of unplanned ICU readmission. The predictive model's area under the curve was .802 (p<.001). Conclusion: We identified the risk factors and the prediction model associated with unplanned ICU readmission. Better patient assessment tools and knowledge about risk factors could contribute to reduce unplanned ICU readmission rate and mortality.