• Korean Journal of Radiology
• /
• v.21 no.12
• /
• pp.1317-1325
• /
• 2020

Objective: The aim of this study was to prospectively evaluate whether liver stiffness (LS) assessments, obtained by two-dimensional (2D)-shear wave elastography (SWE) with a propagation map, can evaluate liver fibrosis stage using histopathology as the reference standard. Materials and Methods: We prospectively enrolled 123 patients who had undergone percutaneous liver biopsy from two tertiary referral hospitals. All patients underwent 2D-SWE examination prior to biopsy, and LS values (kilopascal [kPa]) were obtained. On histopathologic examination, fibrosis stage (F0-F4) and necroinflammatory activity grade (A0-A4) were assessed. Multivariate linear regression analysis was performed to determine the significant factors affecting the LS value. The diagnostic performance of the LS value for staging fibrosis was assessed using receiver operating characteristic (ROC) analysis, and the optimal cut-off value was determined by the Youden index. Results: Reliable measurements of LS values were obtained in 114 patients (92.7%, 114/123). LS values obtained from 2D-SWE with the propagation map positively correlated with the progression of liver fibrosis reported from histopathology (p < 0.001). According to the multivariate linear regression analysis, fibrosis stage was the only factor significantly associated with LS (p < 0.001). The area under the ROC curve of LS from 2D-SWE with the propagation map was 0.773, 0.865, 0.946, and 0.950 for detecting F ≥ 1, F ≥ 2, F ≥ 3, and F = 4, respectively. The optimal cut-off LS values were 5.4, 7.8, 9.4, and 12.2 kPa for F ≥ 1, F ≥ 2, F ≥ 3, and F = 4, respectively. The corresponding sensitivity and specificity of the LS value for detecting cirrhosis were 90.9% and 88.4%, respectively. Conclusion: The LS value obtained from 2D-SWE with a propagation map provides excellent diagnostic performance in evaluating liver fibrosis stage, determined by histopathology.

• Journal of Trauma and Injury
• /
• v.37 no.3
• /
• pp.201-208
• /
• 2024

Purpose: The number of elderly patients with trauma is increasing; therefore, precise models are necessary to estimate the mortality risk of elderly patients with trauma for informed clinical decision-making. This study aimed to develop machine learning based predictive models that predict 30-day mortality in severely injured elderly patients with trauma and to compare the predictive performance of various machine learning models. Methods: This study targeted patients aged ≥65 years with an Injury Severity Score of ≥15 who visited the regional trauma center at Chungbuk National University Hospital between 2016 and 2022. Four machine learning models-logistic regression, decision tree, random forest, and eXtreme Gradient Boosting (XGBoost)-were developed to predict 30-day mortality. The models' performance was compared using metrics such as area under the receiver operating characteristic curve (AUC), accuracy, precision, recall, specificity, F1 score, as well as Shapley Additive Explanations (SHAP) values and learning curves. Results: The performance evaluation of the machine learning models for predicting mortality in severely injured elderly patients with trauma showed AUC values for logistic regression, decision tree, random forest, and XGBoost of 0.938, 0.863, 0.919, and 0.934, respectively. Among the four models, XGBoost demonstrated superior accuracy, precision, recall, specificity, and F1 score of 0.91, 0.72, 0.86, 0.92, and 0.78, respectively. Analysis of important features of XGBoost using SHAP revealed associations such as a high Glasgow Coma Scale negatively impacting mortality probability, while higher counts of transfused red blood cells were positively correlated with mortality probability. The learning curves indicated increased generalization and robustness as training examples increased. Conclusions: We showed that machine learning models, especially XGBoost, can be used to predict 30-day mortality in severely injured elderly patients with trauma. Prognostic tools utilizing these models are helpful for physicians to evaluate the risk of mortality in elderly patients with severe trauma.

• Geotechnical Engineering
• /
• v.13 no.6
• /
• pp.107-122
• /
• 1997

This paper presents a ground surface settlement prediction method for shield tunneling in cohesive soils. In order to develop the method, a parametric study on shield tunneling was performed by using a threetimensional elasto-plastic finite element analysis, which can simulate the construction procedure. By using the results of the finite element analysis, the ground movement mechanism was investigated and a base which relates the ground surface settlement and iuluencing factors was formed. The data base was then used to formulate semi -empirical equations for both surface settlement ratio above tunnel face and imflection point by means of a regression analysis. Furthermore, a prediction method for transverse and longitudinal surface settlement profiles was suggested by using the leveloped equations in conjunction with the normal probability curve. Effectiveness of the developed method was illustrated by comparing settlement profiles obtained by using the developed method with the results of finite element analysis and measured data. Based on the comparison, it was concluded that the developed method can be effectively rosed for practical applications at least within the conditions investigated.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.32 no.6B
• /
• pp.373-378
• /
• 2012

In this study, a new empirical equation for the transverse dispersion coefficient has been developed based on the theoretical background in river bends. The nonlinear least-square method was applied to determine regression coefficients of the equation. The estimated dispersion coefficients derived by the new equation were compared with observed transverse dispersion coefficients acquired from natural rivers and coefficients calculated by the other existing empirical equations. From a comparison of the existing transverse dispersion equations and the new proposed equation, it appears that the behavior of the existing formula in a relative sense is very much dependent on the friction factor and the river geometry. However, the new proposed equation does not vary widely according to variation of friction factor. Also, it was revealed that the equation proposed in this study becomes an asymptotic curve as the curvature effect increases.

• Journal of The Korean Society of Agricultural Engineers
• /
• v.58 no.3
• /
• pp.57-69
• /
• 2016

In this study, we analyzed the extreme rainfall distribution scenarios based on probable rainfall calculation and applying various time distribution models over the landslide high risk zones in urban areas. We used observed rainfall data form total 71 ASOS (Automated Synoptic Observing System) station and AWS (Automatic Weather Station) in KMA (Korea Meteorological Administration), and we analyzed the linear trends for 1-hr and 24-hr annual maximum rainfall series using simple linear regression method, which are identified their increasing trends with slopes of 0.035 and 0.660 during 1961-2014, respectively. The Gumbel distribution was applied to obtain the return period and probability precipitation for each duration. The IDF (Intensity-Duration-Frequency) curves for landslide high risk zones were derived by applying integrated probability precipitation intensity equation. Results from IDF analysis indicate that the probability precipitation varies from 31.4~38.3 % for 1 hr duration, and 33.0~47.9 % for 24 hr duration. It also showed different results for each area. The $Huff-4^{th}$ Quartile method as well as Mononobe distribution were selected as the rainfall distribution scenarios of landslide high risk zones. The results of this study can be used to provide boundary conditions for slope collapse analysis, to analyze sediment disaster risk, and to use as input data for risk prediction of debris flow.

• Composites Research
• /
• v.26 no.4
• /
• pp.265-272
• /
• 2013

Fiber-reinforced plastic (FRP) is applied to fabricate the main structures of composite boats. Most of them are made from molds. These products deform after releasing from the mold and they also deform in high temperature environment. Therefore, experimental investigation and evaluation of deformation of laminated composite structures under various manufacturing and working conditions are necessary. The specimens of L-shape and curveshape were made from unsaturated polyester resin and fiberglass material. Input factors (independent variables) are percentage of hardener and manufacturing temperature and four levels of working temperature and output factor is the deformation which is measured on these specimens. From the results, it was observed that the higher the hardener rate and temperature, the lower the deformation. When the working temperature increased, the specimens showed great variations for the initial deformation values. Besides, the values of deformation or input factors could be predicted by regression equations.

• Journal of the Earthquake Engineering Society of Korea
• /
• v.22 no.3
• /
• pp.129-137
• /
• 2018

On November 15, 2017, Pohang earthquake occurred. Its local magnitude was announced to be $M_L=5.4$ by Korea Meteorological Administration (KMA). Ground motion data recorded at KMA stations were obtained from their data bases. From the data, horizontal and vertical response spectra, and V/H ratio were calculated. The horizontal spectra were defined as GMRotI50 spectra. From the statistical analysis of the GMRotI50 spectra, a mean plus one standard deviation spectrum in lognormal distribution is obtained. Regression analysis is performed on this curve to determine the shape of spectrum including transition periods. Applying the same procedure, the shape and transition periods of vertical spectrum were obtained. These results were compared with theKorean standard design spectra, which were developed from domestic and overseas intra-plate earthquake records, and Gyeongju earthquake response spectra. The response spectra of Pohang earthquake were found to be almost identical with the newly proposed design spectra. Even the V/H ratios showed good agreement. These results confirmed that the method adopted when developing the standard design spectra were valid and the developed design spectra were reliable.

• Asian Pacific Journal of Cancer Prevention
• /
• v.13 no.2
• /
• pp.505-509
• /
• 2012

Background: Matrix metalloproteinases (MMP) are proteolytic enzymes that are essentially involved in turnover of the extracellular matrix (ECM). The aim was to investigate the diagnostic value of MMP-7 and MMP-10 as tumor markers in pleural effusion (PE) and evaluate the value of combining MMP-7, MMP-10 and carcinoembryonic antigen (CEA) assays as diagnostic aids for malignant cells. Materials and Methods: A total of 179 patients with PE (87 malignant and 92 benign) were included in this study. The levels of MMP-7 and MMP-10 were measured using ELISA. Results: Values for MMP-7 and MMP-10 were significantly higher in malignant PE than those in benign PE (P<0.01). Among all variables evaluated, logistic regression found that MMP-7 and MMP-10 were significantly correlated with the presence of malignant disease (P<0.01). Analysis of receiver operating characteristics (ROC) curves showed that the area under the curve of MMP-10 (0.806) was significantly larger than that of MMP-7 (0.771) and CEA (0.789) (P<0.01). With parallel interpretation, the combination of MMP-10 and CEA achieved the higher sensitivity of 94.6%. The combination of MMP-7 and CEA in serial interpretation was able to boost the specificity to 95.7%. The combination of MMP-7, MMP-10 and CEA produced better sensitivity, specificity, PPV and NPV than MMP-7 and MMP-10 alone. Conclusion: MMP-7 and MMP-10 in PE may represent helpful adjuncts to conventional diagnostic tools in ruling out malignancy as a probable diagnosis, thus guiding the selection of patients who might benefit from further invasive procedures.

• Computers and Concrete
• /
• v.21 no.3
• /
• pp.249-259
• /
• 2018

The use of recycled aggregate in concrete is gaining much attention due to the growing need for sustainability in construction. In the present study, Self Compacting Concrete (SCC) is made using both natural and recycled aggregate (crushed recycled concrete aggregate from building demolished waste) and performance of recycled aggregate based SCC for the bond behaviour of reinforcement is evaluated. The major factors that influence the bond like concrete compressive strength (Mix-A, B and C), diameter of bar ($D_b=10$, 12 and 16 mm) and embedment length of bar ($L_d=2.5Db$, $5D_b$ and full depth of specimen) are the parameters considered in the present study in addition to type of aggregates (natural and recycled aggregates). The mix proportions of Natural Aggregate SCC (NASCC) are arrived based on the specifications of IS 10262. The mix proportions also satisfy the guidelines of EFNARC. In case of Recycled Aggregate SCC (RASCC), both the natural coarse and fine aggregates are replaced 100% by volume with that of recycled aggregates. These mixes are also evaluated for fresh properties as per EFNARC. The hardened properties like compressive strength, split tensile strength and flexural strength are also determined. The pull-out test is conducted as per the specifications of IS 2770 (Part-1) for determining the bond strength of reinforcement. Bond stress versus slip curves were plotted and a typical comparison of RASCC is made with NASCC. The fracture energy i.e., area under the bond stress slip curve is determined. With the use of recycled aggregates, reduction in maximum bond stress is noticed whereas, the normalised maximum bond stress is higher in case of recycled aggregates. Based on the experimental results, regression analysis is conducted and an equation is proposed to predict the maximum bond stress of RASCC. The equation is in good agreement with the experimental results. The available models in the literature are made use to predict the maximum bond stress and compare the present results.

• Tuberculosis and Respiratory Diseases
• /
• v.80 no.2
• /
• pp.201-209
• /
• 2017

Background: Methicillin-resistant Staphylococcus aureus (MRSA) infection is a severe and life-threatening disease in patients with community-onset (CO) pneumonia. However, the current guidelines lack specificity for a screening test for MRSA infection. Methods: This study was retrospectively conducted in elderly patients aged ${\geq}65years$, who had contracted CO-pneumonia during hospitalization at the Jeju National University Hospital, between January 2012 and December 2014. We analyzed the risk factors of MRSA in these patients and developed a scoring system to predict MRSA infection. Results: A total of 762 patients were enrolled in this study, including 19 (2.4%) with MRSA infection. Healthcare-associated pneumonia (HCAP) showed more frequent MRSA infection compared to community-acquired pneumonia (4.4% vs. 1.5%, respectively; p=0.016). In a multivariate logistic regression analysis, admissions during the influenza season (odds ratio [OR], 2.896; 95% confidence interval [CI], 1.022-8.202; p=0.045), chronic kidney disease (OR, 3.555; 95% CI, 1.157-10.926; p=0.027), and intensive care unit admission (OR, 3.385; 95% CI, 1.035-11.075; p=0.044) were identified as predictive factors for MRSA infection. However, the presence of HCAP was not significantly associated with MRSA infection (OR, 1.991; 95% CI, 0.720-5.505; p=0.185). The scoring system consisted of three variables based on the multivariate analysis, and showed moderately accurate diagnostic prediction (area under curve, 0.790; 95% CI, 0.680-0.899; p<0.001). Conclusion: MRSA infection would be considered in elderly CO-pneumonia patients, with three risk factors identified herein. When managing elderly patients with pneumonia, clinicians might keep in mind that these risk factors are associated with MRSA infection, which may help in selecting appropriate antibiotics.