• 대한응급의학회지
• /
• 제29권5호
• /
• pp.455-464
• /
• 2018

Objective: Machine learning is not yet widely used in the medical field. Therefore, this study was conducted to compare the performance of preexisting severity prediction models and machine learning based models (random forest [RF], gradient boosting [GB]) for mortality prediction in pneumonia patients. Methods: We retrospectively collected data from patients who visited the emergency department of a tertiary training hospital in Seoul, Korea from January to March of 2015. The Pneumonia Severity Index (PSI) and Sequential Organ Failure Assessment (SOFA) scores were calculated for both groups and the area under the curve (AUC) for mortality prediction was computed. For the RF and GB models, data were divided into a test set and a validation set by the random split method. The training set was learned in RF and GB models and the AUC was obtained from the validation set. The mean AUC was compared with the other two AUCs. Results: Of the 536 investigated patients, 395 were enrolled and 41 of them died. The AUC values of PSI and SOFA scores were 0.799 (0.737-0.862) and 0.865 (0.811-0.918), respectively. The mean AUC values obtained by the RF and GB models were 0.928 (0.899-0.957) and 0.919 (0.886-0.952), respectively. There were significant differences between preexisting severity prediction models and machine learning based models (P<0.001). Conclusion: Classification through machine learning may help predict the mortality of pneumonia patients visiting the emergency department.

• International Journal of Naval Architecture and Ocean Engineering
• /
• 제12권1호
• /
• pp.755-767
• /
• 2020

A significant development has been made on a new fatigue damage model applicable to Gaussian wide band stress response spectra using numerical approximation methods such as data processing, time simulation, and regression analysis. So far, most of the alternative approximate models provide slightly underestimated or overestimated damage results compared with the rain-flow counting distribution. A more reliable approximate model that can minimize the damage differences between exact and approximate solutions is required for the practical design of ships and offshore structures. The present paper provides a detailed description of the development process of a new fatigue damage model. Based on the principle of the Gaussian wide band model, this study aims to develop the best approximate fatigue damage model. To obtain highly accurate damage distributions, this study deals with some prominent research findings, i.e., the moment of rain-flow range distribution MRR(n), the special bandwidth parameter μk, the empirical closed form model consisting of four probability density functions, and the correction factor QC. Sequential prerequisite data processes, such as creation of various stress spectra, extraction of stress time history, and the rain-flow counting stress process, are conducted so that these research findings provide much better results. Through comparison studies, the proposed model shows more reliable and accurate damage distributions, very close to those of the rain-flow counting solution. Several significant achievements and findings obtained from this study are suggested. Further work is needed to apply the new developed model to crack growth prediction under a random stress process in view of the engineering critical assessment of offshore structures. The present developed formulation and procedure also need to be extended to non-Gaussian wide band processes.

• 보건행정학회지
• /
• 제3권2호
• /
• pp.159-176
• /
• 1993

The utilization of outpatient care services involves two steps of sequential decisions. The first step decision is about whether to initiate the utilization and the second one is about how many more visits to make after the initiation. Presumably, the initiation decision is largely made by the patient and his or her family, while the number of additional visits is decided under a strong influence of the physician. Implication is that the analysis of the outpatient care utilization requires to specify each of the two decisions underlying the utilization as a distinct stochastic process. This paper is concerned with the number of physician visits, which is, by definition, a discrete variable that can take only non-negative integer values. Since the initial visit is considered in the analysis of whether or not having made any physician visit, the focus on the number of visits made in addition to the initial one must be enough. The number of additional visits, being a kind of count data, could be assumed to exhibit a Poisson distribution. However, it is likely that the distribution is over dispersed since the number of physician visits tends to cluster around a few values but still vary widely. A recently reported study of outpatient care utilization employed an analysis based upon the assumption of a negative binomial distribution which is a type of overdispersed Poisson distribution. But there is an indication that the use of Poisson distribution making adjustments for over-dispersion results in less loss of efficiency in parameter estimation compared to the use of a certain type of distribution like a negative binomial distribution. An analysis of the data for outpatient care utilization was performed focusing on an assessment of appropriateness of available techniques. The data used in the analysis were collected by a community survey in Hwachon Gun, Kangwon Do in 1990. It was observed that a Poisson regression with adjustments for over-dispersion is superior to either an ordinary regression or a Poisson regression without adjustments oor over-dispersion. In conclusion, it seems the most approprite to assume that the number of physician visits made in addition to the initial visist exhibits an overdispersed Poisson distribution when outpatient care utilization is studied based upon a model which embodies the two-part character of the decision process uderlying the utilization.

• 한국지진공학회논문집
• /
• 제27권5호
• /
• pp.221-230
• /
• 2023

This study proposes a methodology for assessing seismic liquefaction hazard by implementing high-resolution three-dimensional (3D) ground models with high-density/high-precision site investigation data acquired in an area of interest, which would be linked to geotechnical numerical analysis tools. It is possible to estimate the vulnerability of earthquake-induced geotechnical phenomena (ground motion amplification, liquefaction, landslide, etc.) and their triggering complex disasters across an area for urban development with several stages of high-density datasets. In this study, the spatial-ground models for city development were built with a 3D high-precision grid of 5 m × 5 m × 1 m by applying geostatistic methods. Finally, after comparing each prediction error, the geotechnical model from the Gaussian sequential simulation is selected to assess earthquake-induced geotechnical hazards. In particular, with seven independent input earthquake motions, liquefaction analysis with finite element analyses and hazard mappings with LPI and LSN are performed reliably based on the spatial geotechnical models in the study area. Furthermore, various phenomena and parameters, including settlement in the city planning area, are assessed in terms of geotechnical vulnerability also based on the high-resolution spatial-ground modeling. This case study on the high-precision 3D ground model-based zonations in the area of interest verifies the usefulness in assessing spatially earthquake-induced hazards and geotechnical vulnerability and their decision-making support.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제17권10호
• /
• pp.2627-2642
• /
• 2023

Parkinson's disease (PD) is a typical, chronic neurodegenerative disease involving the concentration of dopamine, which can disrupt motor activity and cause different degrees of gait disturbance relevant to PD severity in patients. As current clinical PD diagnosis is a complex, time-consuming, and challenging task that relays on physicians' subjective evaluation of visual observations, gait disturbance has been extensively explored to make automatic detection of PD diagnosis and severity rating and provides auxiliary information for physicians' decisions using gait data from various acquisition devices. Among them, wearable sensors have the advantage of flexibility since they do not limit the wearers' activity sphere in this application scenario. In this paper, an attention-based temporal network (ATN) is designed for the time series structure of gait data (vertical ground reaction force signals) from foot sensor systems, to learn the discriminative differences related to PD severity levels hidden in sequential data. The structure of the proposed method is illuminated by Transformer Network for its success in excavating temporal information, containing three modules: a preprocessing module to map intra-moment features, a feature extractor computing complicated gait characteristic of the whole signal sequence in the temporal dimension, and a classifier for the final decision-making about PD severity assessment. The experiment is conducted on the public dataset PDgait of VGRF signals to verify the proposed model's validity and show promising classification performance compared with several existing methods.

• 대한치과교정학회지
• /
• 제49권2호
• /
• pp.124-136
• /
• 2019

Orthodontic treatment is more complicated when both soft and hard tissues must be considered because an impacted maxillary canine has important effects on function and esthetics. Compared with extraction of impacted maxillary canines, exposure followed by orthodontic traction can improve esthetics and better protect the patient's teeth and alveolar bone. Therefore, in order to achieve desirable tooth movement with minimal unexpected complications, a precise diagnosis is indispensable to establish an effective and efficient force system. In this report, we describe the case of a 31-year-old patient who had a labio-palatal horizontally impacted maxillary left canine with a severe occlusal alveolar bone defect and a missing maxillary left first premolar. Herein, with the aid of three-dimensional imaging, sequential traction was performed with a three-directional force device that finally achieved acceptable occlusion by bringing the horizontally impacted maxillary left canine into alignment. The maxillary left canine had normal gingival contours and was surrounded by a substantial amount of regenerated alveolar bone. The 1-year follow-up stability assessment demonstrated that the esthetic and functional outcomes were successful.

• 한국수자원학회논문집
• /
• 제45권9호
• /
• pp.861-874
• /
• 2012

SWAT(Soil and Water Assessment Tool) 모형의 적용성 검증을 위해서는 매개변수 민감도 분석 및 검 보정, 예측 불확실성 분석을 필요로 한다. 최근 SWAT 모형의 불확실성을 분석하기 위한 다양한 기법들이 개발되었는데, 본 연구는 충주댐 유역(6,581.1 $km^2$)을대상으로유역출구점의 실측 일 유출량자료(1998~2003)를 바탕으로 SWAT 모형의 유출관련 매개변수에 대한 불확실성 분석을 실시하였다. 이때 사용된 분석 기법으로는 SUFI2 (Sequential Uncertainty FItting algorithm ver.2), GLUE(Generalized Likelihood Uncertainty Estimation), ParaSol (Parameter Solution) 등을 적용하였다. 이러한 기법은 모두 SWAT-CUP (SWAT-Calibration Uncertainty Program; Abbaspour et al., 2007) 모형에 탑재되어있으며, 모형의 결과로써 검 보정, 매개변수의 민감도 분석, 각종 목적 함수 및 불확실성의 범위 등이 자동으로 산출되므로 모형의 사용자가 불확실성 평가 기법의 분석 및 비교를 손쉽게 할 수 있다. 그 결과 대표적인 목적 함수인 결정 계수($R^2$; Legates and McCabe, 1999)와 NS (Nash and Sutcliffe, 1970) 모형효율은 모두 0.67에서 0.92 사이의 값을 나타내어 대체적으로 모의가 잘 이루어졌음을 알 수 있었다. 그러나 불확실성의 범위를 나타내는 지표인 p-factor 및 r-factor에서는 평가 기법 별로 그 차이가 확연하게 드러났다. 여기서 p-factor는 불확실성 범위에 실측치가 포함되는 비율이며, r-factor는 불확실성의 상대적인 범위로 각각 1과 0에 가까울수록 모의기법의 성능이 우수함을 의미한다. 세가지 알고리듬 중에서 SUFI2의 p-factor가 약 0.79로 가장 높게 나타났으며, ParaSol의 r-factor가 0.03으로 가장 작게 나타났다. 본 연구의 결과는 SWAT 모형을 이용한 수문 모의에서 수문분석에 따른 예측결과의 불확실성을 정량적으로 평가함으로서, 모형의 적용성 평가 및 모의결과의 신뢰성 확보에 근거자료로 활용이 가능할 것으로 판단된다.

• Tuberculosis and Respiratory Diseases
• /
• 제57권4호
• /
• pp.329-335
• /
• 2004

연구 배경 : 중환자의 예후를 예측할 수 있는 지표로 현재까지 APACHE II, III 점수체계와 SAPS 등이 임상에 응용되고 있다. 1996년 유럽에서 SOFA 점수체계가 제안되었고 1998년 다기관 전향적 연구를 통하여 장기부전을 평가하고 경과를 감시하는 간편하고 효과적인 점수체계임을 입증하였다. 이 연구는 춘천성심병원에서 내과계 중환자만을 대상으로 SOFA 점수체계의 유용성을 조사하였다. 방 법 : 2000년 5월 1일부터 6월 30일까지 춘천성심병원 내과계 중환자실에 입원한 154명을 대상으로 하였고, 이중 타과환자, 타병원으로 전원된 경우, ECG 관찰만을 위하여 입원한 경우, 처치 후 관찰을 위해 48시간 이내만 체류한 환자를 제외한 105명을 연구에 포함시켰으며, 91명에서 평가가 가능하였다. SOFA 점수와 APACHE III 점수는 입원 시부터 매일 24시간 간격으로 중환자실에서 퇴실할 때까지 하였다. 결 과 : 1) 사망률은 20%였고 중환자실 체류기간의 중앙값은 3일이었다. 2) 사망군과 생존군 간의 24시간 내 SOFA 점수는 각각 $8.17{\pm}4.05$, $3.85{\pm}3.24$로 유의한 차이가 있었다. 3) 장기부전이 없는 경우의 사망률은 6.5%, 5개의 다발성 장기부전이 있는 경우 사망률은 100%였다. 4) 48시간 이상 체류했던 74명을 분석한 결과 사망군에서는 최대 SOFA 점수가 81%에서 증가하였고 생존군에서는 21%에서 증가하여 유의한 차이를 보였다. 5) 24시간 내 SOFA 점수와 APACHE III 점수와의 관련성을 평가한 결과 r=0.85 (p<0.01)로 유의한 상관관계가 있었다. 결 론 : SOFA 점수체계는 다발성 장기부전이 있는 환자에서 장기부전을 평가하고 예후를 예측하며 임상경과를 관찰하는데 이용할 수 있는 간편하고 임상적 효용성이 있는 지표로서 임상적으로 많은 도움이 될 것으로 생각한다.

• 한국환경농학회지
• /
• 제24권2호
• /
• pp.106-116
• /
• 2005

This experiment was conducted to investigate available extraction capacity and potential mobility of heavy metal according to the distribution property and contamination level of heavy metals in soils and to suggest a reform measure of soil environment assessment methodology applied with soil quality and the official soil heavy metal test methods in domestic and foreign countries. The soils were collected from the natural forest paddy with long-term application of same type fertilizer, and paddies near metal mine and industrial complex. The post-treatment methods of soil were partial extraction, acid digestion and sequential extraction methods. For the heavy metal contents with different soil properties, it was shown that their natural forest and paddy soil were slightly low and similar to the general paddy soil, while their paddies near metal mine and industrial complex were higher than the standard level of Soil Environment Protection Act. Heavy metal concentrations in the soils with different soil properties had difference between $HNO_3\;and\;HNO_3+HCl$ extractant by US-EPA 3051a method. There were highly significant positive relationships in both two methods. It was appeared that the higher extractable concentration ratio with 0.1N-HCl to total heavy metal content with $HNO_3+HCl$ extractant the greater total heavy metal content. There were highly significant positive correlationship between total heavy metal content and extractable content with 0.1N-HCl. For extractable capacity of soil extractable solution compared to the total heavy metal content it was appeared that it extractable method with 0.1N-HCl was higher than those with EDTA and DTPA. In extractable ratio with 0.1N-HCl in the contaminated paddy soils near mine and industrial complex, it was shown that the lower soil pH, the higher total heavy metal content. The order of a potential mobility coefficient by distribution of heavy metal content with ie different typies in the soil was Cd>Ni>Zn>Cu>Pb. It could be known that contamination characteristics of heavy metals with different types of soils were affected by different heavy metal components, contamination degree and soil chemical properties, and heavy metal concentration with different extractable methods had great variations with adjacent environment. To be compared with assessment methodology of soil environment impact at domestic and foreign countries with our results, it might be considered that there was necessary to make a single analysis method based on total heavy metal content with environmental overloading concept because of various analysis methods for total heavy metal content and present analysis method with great variation according to soil environment. In spite of showing higher concentration of heavy metal with acidic digestion than the extractable method, it might be considered that there is need to be adjusted the national standard of soil heavy metal contamination.

• 환경영향평가
• /
• 제16권1호
• /
• pp.69-77
• /
• 2007

This study has been conducted to assess blackish-water phenomena in Dam reservoir. To searching for the reason, we survey physical changes in reservoir and analyze metal and organic content in particulate materials and water. The blackish-water phenomena in lake A are occurred with turbidity increases in turnover season irregularly. It was reported on 6 Jan. 2005 weakly and the water column mixed with 35~40m depth and water temperature shows $7^{\circ}C$. The turbidity of AD and AM site increased up to 20NTU. Especially, AN site shows 27NTU, such a result makes that Dam manager conclude it to blackish-water phenomena. The results of sequential extraction analysis show that over 80% of Al, Cr and Fe is existed in residual form in sediment. On the other hand, the most part of Mn shows exchangeable and carbonates form, which have a good possibility of release to water column. Mn contents in pore waters of the sediment samples are also found to be ~4 times higher than Fe contents. The metal contents in pore water of different dam sites are in order of AN (Fe: 9.98, Mn: 40.6) > AD(8.33, 37.5) > DD(1.91, 2.55). According to the results of extracted organic materials from sediment, humic substances is occupied with over 85% in total organic carbon including 23~45% of humic acid (HA) and 0.9~8.5% of fulvic acid (FA). However, HA content in pore water is not detectable while FA contents, acid-soluble humic fractions is higher than that of sediment(10~15%). which indicating that FA is a main humic components affecting water color. The color unit per DOC of FA in pore waters of different dam sites are found to be higher in lake A than lake D. From the results, it could be suggested that blackish-water phenomena of lake A are mainly arise from higher concentration of Mn and water soluble organic fractions (e.g., FA) released from sediments as well as the strength of turnover in Dam reservoir.