• Journal of Sleep Medicine
• /
• v.15 no.2
• /
• pp.48-54
• /
• 2018

Objectives: The aim of this study was to develop a predicting model for the optimal continuous positive airway pressure (CPAP) for obstructive sleep apnea (OSA) patient with obesity by using a machine learning. Methods: We retrospectively investigated the medical records of 162 OSA patients who had obesity [body mass index (BMI) ≥ 25] and undertaken successful CPAP titration study. We divided the data to a training set (90%) and a test set (10%), randomly. We made a random forest model and a least absolute shrinkage and selection operator (lasso) regression model to predict the optimal pressure by using the training set, and then applied our models and previous reported equations to the test set. To compare the fitness of each models, we used a correlation coefficient (CC) and a mean absolute error (MAE). Results: The random forest model showed the best performance {CC 0.78 [95% confidence interval (CI) 0.43-0.93], MAE 1.20}. The lasso regression model also showed the improved result [CC 0.78 (95% CI 0.42-0.93), MAE 1.26] compared to the Hoffstein equation [CC 0.68 (95% CI 0.23-0.89), MAE 1.34] and the Choi's equation [CC 0.72 (95% CI 0.30-0.90), MAE 1.40]. Conclusions: Our random forest model and lasso model ($26.213+0.084{\times}BMI+0.004{\times}$apnea-hypopnea index+$0.004{\times}oxygen$ desaturation index-$0.215{\times}mean$ oxygen saturation) showed the improved performance compared to the previous reported equations. The further study for other subgroup or phenotype of OSA is required.

• Journal of Trauma and Injury
• /
• v.34 no.1
• /
• pp.13-20
• /
• 2021

Purpose: Nutritional therapy in the intensive care unit is an essential factor for patient progress. The purpose of this study was to compare resting energy expenditure (REE) calculated by prediction equations (PEs) to the REE measured by indirect calorimetry (IC) in trauma patients. Methods: Patients admitted to the trauma intensive care unit who received mechanical ventilation between January and December 2015 were enrolled. REE was measured by IC (CCM Express, MGC Diagnostics) and calculated by the following PEs: Harris-Benedict, Fleisch, Robertson and Reid, Ireton-Jones, and the maximum value (25 kcal/kg/day) of the European Society for Clinical Nutrition and Metabolism (ESPEN). All patients were ventilated at a fraction of inspired oxygen (FiO2) below 60%. Results: Of the 31 patients included in this study, 24 (77.4%) were men and seven (22.6%) were women. The mean age of the patients was 49.7±13.2 years, their mean weight was 68.1±9.6 kg, and their mean Injury Severity Score was 26.1±11.3. The mean respiratory quotient on IC was 0.93±0.19, and their mean FiO2 was 38.72%±6.97%. The mean REE measured by IC was 2,146±444.36 kcal/day, and the mean REE values calculated by the PEs were 1,509.39±205.34 kcal/day by the Harris and Benedict equation, 1,509.39±154.33 kcal/day by the Fleisch equation, and 1,443.39±159.61 kcal/day by the Robertson and Reid equation. The Ireton-Jones equation yielded a higher value (2,278.90±202.35 kcal/day), which was not significantly different from the value measured using IC (p=0.53). The ESPEN maximum value (1,704.03±449.36 kcal/day) was lower, but this difference was likewise not significant (p=0.127). Conclusions: The REE measured by IC was somewhat higher than that calculated using PEs. Further studies are needed to determine the proper nutritional support for trauma patients.

• Steel and Composite Structures
• /
• v.23 no.5
• /
• pp.543-555
• /
• 2017

The mechanical properties of steel fiber reinforced high strength concrete (SFRHSC) made with binary and ternary blends of metakaolin (MK) and ground pumice (GP) are investigated in this study. The investigated properties are ultrasonic pulse velocity ($U_{pv}$), compressive strength ($f_c$), flexural strength ($f_f$) and splitting tensile strength ($f_{st}$) of SFRHSC. A total of 16 steel fiber reinforced concrete mixtures were produced by a total binder content of $500kg/m^3$ for determining the effects of MK and GP on the mechanical properties. The design $f_c$ was acquired from 70 to 100 MPa by using a low water-binder ratio of 0.2. The test results exhibit that high strength concrete can be obtained by replacing the cement with MK and GP. Besides, correlations between these results are executed for comprehending the relationship between mechanical properties of SFRHSC and the strong correlations are observed between these properties. Moreover, two models in the gene expression programming (GEP) for predicting the $f_c$ of SFRHSC made with binary and ternary blends of MK and GP have been developed. The results obtained from these models are compared with the experimental results. These comparisons proved that the results of equations obtained from these models seem to agree with the experimental results.

• Computers and Concrete
• /
• v.19 no.6
• /
• pp.631-639
• /
• 2017

This study experimentally investigated the flexural capacity of a concrete beam reinforced with a newly developed GFRP bar that overcomes the lower modulus of elasticity and bond strength compared to a steel bar. The GFRP bar was fabricated by thermosetting a braided pultrusion process to form the outer fiber ribs. The mechanical properties of the modulus of elasticity and bond strength were enhanced compared with those of commercial GFRP bars. In the four-point bending test results, all specimens failed according to the intended failure mode due to flexural design in compliance with ACI 440.1R-15. The effects of the reinforcement ratio and concrete compressive strength were investigated. Equations from the code were used to predict the deflection, and they overestimated the deflection compared with the experimental results. A modified model using two coefficients was developed to provide much better predictive ability, even when the effective moment of inertia was less than the theoretical $I_{cr}$. The deformability of the test beams satisfied the specified value of 4.0 in compliance with CSA S6-10. A modified effective moment of inertia with two correction factors was proposed and it could provide much better predictability in prediction even at the effective moment of inertia less than that of theoretical cracked moment of inertia.

• Journal of the Korean Institute of Gas
• /
• v.14 no.5
• /
• pp.19-25
• /
• 2010

The flash points for the systems, o-xylene+n-pentanol and m-xylene+n-hexanol, were measured by using Tag open-cup tester(ASTM D1310-86). The experimental data were compared with the values calculated by the Raoult's law and the optimization method using van Laar and Wilson equations. The calculated values based on the optimization method were found to be better than those based on the Raoult's law. The predictive curve of the flash point prediction model based on the van Laar equation described the experimentally-derived data more effectively than was the case when the prediction model was based upon the Wilson equation.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.29 no.9
• /
• pp.33-41
• /
• 2015

To improve the power quality of distributed generation (DG) systems even in the presence of distorted grid condition, dual current control scheme of a grid-connected inverter is proposed. The proposed current control scheme is achieved by decomposing the inverter state equations into the fundamental and harmonic components. The derived models are employed to design dual current controllers. The conventional PI decoupling current controller is used in the fundamental model to control the main power flow in DG systems. At the same time, the predictive control is applied in the harmonic model to suppress undesired harmonic currents to zero quickly. To decompose the voltage inputs and state variables into the fundamental and harmonic components, the fourth order band pass filter (BPF) is designed in the discrete-time domain for a digital implementation. For experimental verification, 2kVA prototype of a grid-connected inverter has been constructed using digital signal processor (DSP) TMS320F28335. The effectiveness of the proposed strategy is demonstrated through comparative simulation and experimental results.

• Earthquakes and Structures
• /
• v.14 no.4
• /
• pp.285-295
• /
• 2018

In this study, the effect of local site conditions (site class and site amplifications) and structural variability are investigated on fragility functions of typical building structures. The study area is chosen as Eastern Turkey. The fragility functions are developed using site-specific uniform hazard spectrum (UHS). The site-specific UHS is obtained based on simulated ground motions. The implementation of ground motion simulation into seismic hazard assessment has the advantage of investigating detailed local site effects. The typical residential buildings in Erzincan are represented by equivalent single degree of freedom systems (ESDOFs). Predictive equations are accomplished for structural seismic demands of ESDOFs to derive fragility functions in a straightforward manner. To study the sensitivity of fragility curves to site class, two sites on soft and stiff soil are taken into account. Two alternative site amplification functions known as generic and theoretical site amplifications are examined for these two sites. The reinforced concrete frames located on soft soil display larger fragilities than those on stiff soil. Theoretical site amplification mostly leads to larger fragilities than generic site amplification more evidently for reinforced concrete buildings. Additionally, structural variability of ESDOFs is generally observed to increase the fragility especially for rigid structural models.

• Asian-Australasian Journal of Animal Sciences
• /
• v.18 no.3
• /
• pp.358-363
• /
• 2005

Five crossbred beef cattle (Simmental${\times}$yellow cattle, Shantung Province) fitted with permanent cannulae in the rumen and T-type cannulae at the proximal duodenum and terminal ileum, were fed five different diets containing corn, cotton meal or soybean meal and ammoniated straw to determine the dry matter, crude protein and amino acid flows in duodenal and ileum digesta, and to calculate the regression equations between theoretical and experimental concentration of AA in duodenal digesta. The results showed that there was a strong correlation between experimental (g/d, y) and theoretical CP flows (g/d, x) at the proximal duodenum, the $R^2$-value regression equation of crude protein is very high (0.9636). The $R^2$-value regression equation of the limiting amino acid (such as Met or Lys) is high (0.7573 or 0.9252 respectively). This results indicated that we can formulate better diets fed to beef cattle according to the theoretical amino acid concentration. A mathematical model has been successfully constructed for predicting the supply of absorbed amino acids at the proximal duodenum in growing beef cattle.

• International Journal of Aeronautical and Space Sciences
• /
• v.15 no.2
• /
• pp.123-137
• /
• 2014

The aviation gas turbine is composed of many expensive and highly precise parts and operated in high pressure and temperature gas. When breakdown or performance deterioration occurs due to the hostile environment and component degradation, it severely influences the aircraft operation. Recently to minimize this problem the third generation of predictive maintenance known as condition based maintenance has been developed. This method not only monitors the engine condition and diagnoses the engine faults but also gives proper maintenance advice. Therefore it can maximize the availability and minimize the maintenance cost. The advanced gas turbine health monitoring method is classified into model based diagnosis (such as observers, parity equations, parameter estimation and Gas Path Analysis (GPA)) and soft computing diagnosis (such as expert system, fuzzy logic, Neural Networks (NNs) and Genetic Algorithms (GA)). The overview shows an introduction, advantages, and disadvantages of each advanced engine health monitoring method. In addition, some practical gas turbine health monitoring application examples using the GPA methods and the artificial intelligent methods including fuzzy logic, NNs and GA developed by the author are presented.

• Computers and Concrete
• /
• v.19 no.5
• /
• pp.501-507
• /
• 2017

The predictive utility of a damage model depends heavily on its particular choice of a damage variable, which serves as a macroscopic approximation in describing the underlying micromechanical processes of microdefects. In the case of spatially perfectly randomly distributed microcracks or microvoids in all directions, isotropic damage model is an appropriate choice, and scalar damage variables were widely used for isotropic or one-dimensional phenomenological damage models. The simplicity of a scalar damage representation is indeed very attractive. However, a scalar damage model is of somewhat limited use in practice. In order to entirely characterize the isotropic damage behaviors of damaged materials in multidimensional space, a system theory of isotropic double scalar damage variables, including the expressions of specific damage energy release rate, the coupled constitutive equations corresponding to damage, the conditions of admissibility for two scalar damage effective tensors within the framework of the thermodynamics of irreversible processes, was provided and analyzed in this study. Compared with the former studies, the theoretical formulations of double scalar damage variables in this study are given in the form of matrix, which has many features such as simpleness, directness, convenience and programmable characteristics. It is worth mentioning that the above-mentioned theoretical formulations are only logically reasonable. Owing to the limitations of time, conditions, funds, etc. they should be subject to multifaceted experiments before their innovative significance can be fully verified. The current level of research can be regarded as an exploratory attempt in this field.