• Asian Pacific Journal of Cancer Prevention
• /
• v.14 no.10
• /
• pp.6019-6023
• /
• 2013

Lung cancer, one of the leading causes of cancer-related deaths, usually appears as solitary pulmonary nodules (SPNs) which are hard to diagnose using the naked eye. In this paper, curvelet-based textural features and clinical parameters are used with three prediction models [a multilevel model, a least absolute shrinkage and selection operator (LASSO) regression method, and a support vector machine (SVM)] to improve the diagnosis of benign and malignant SPNs. Dimensionality reduction of the original curvelet-based textural features was achieved using principal component analysis. In addition, non-conditional logistical regression was used to find clinical predictors among demographic parameters and morphological features. The results showed that, combined with 11 clinical predictors, the accuracy rates using 12 principal components were higher than those using the original curvelet-based textural features. To evaluate the models, 10-fold cross validation and back substitution were applied. The results obtained, respectively, were 0.8549 and 0.9221 for the LASSO method, 0.9443 and 0.9831 for SVM, and 0.8722 and 0.9722 for the multilevel model. All in all, it was found that using curvelet-based textural features after dimensionality reduction and using clinical predictors, the highest accuracy rate was achieved with SVM. The method may be used as an auxiliary tool to differentiate between benign and malignant SPNs in CT images.

• Korean Journal of Agricultural Science
• /
• v.27 no.1
• /
• pp.39-47
• /
• 2000

A large scale field test of prefabricated vertical drains is performed to analyze the effect of parameters of the very soft clay at a test site. Compression index and the coefficient of horizontal consolidation obtained by back-analysis from the settlement data were compared with those obtained by means of laboratory tests. The Hyperbolic, Asaoka's and The Curve fitting methods are used to estimate final settlements and coefficients of consolidation. 1. Final settlement predicted with the Hyperbolic method was the largest, and the settlements predicted with the Asaoka's and the Curve fitting methods were nearly the same range, and it was concluded that smear effect has to be considered on design in the case that spacing of drains is small 2. The relationships of the measured consolidation ratio (Urn) and the designed consolidation ratio($U_t$) were showed as $U_m$ = (1.13~1.17)$U_t$, $U_m$ = (1.07~1.20)$U_t$, $U_m$ = (1.13~1.17)$U_t$ on the Hyperbolic, Asaoka's and the Curve fitting methods, respectively. The relations on the Asaoka's and the Curve fitting methods were nearly the same range. 3. The relationships of the field compression index($C_{cfield}$) and virgin compression index($V_{cclab}$) were showed as $C_{cfield}$ = (1.26~1.45)$V_{cclab}$, $C_{cfield}$ = (1.08~1.15) $V_{cclab}$, $C_{cfield}$ = (1.04~1.21)$V_{cclab}$, on the Hyperbolic, Asaoka's and the Curve fitting methods, respectively. 4. The ratio ($C_h/C_v$) of the coefficient of vertical consolidation and the coefficient of horizontal consolidation that is obtained by back-analysis from the settlement data was $C_h$=(0.7~0.9)$C_v$, $C_h$=(0.9~1.5)$C_v$, $C_h$=(2.4~3.0)$C_v$ on the Hyperbolic, Asaoka's and the Curve fitting methods, respectively. 5. It was concluded that the exact consolidation coefficient must be determined after the final settlement is predicted again when the consolidation is finished, because the field consolidation coefficient is decreased as the time allowed to be alone is increased.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.15 no.3
• /
• pp.319-325
• /
• 2004

FHMA(frequency hopping multiple access) communication system has good performance for the LPI and AJ(low probability of intercept and anti-jamming) application. However, high PAPR(peak to average power ratio) happens in the base-station or the repeater system because of a large number of users. In general, predistorter is used to complete the HPA(high power amplifier) nonlinear characteristics. This paper analyzes BER performance when magnitude of IBO(input back oft) and the number of user are considered as the system parameters. In case of the SSPA(solid state power amplifier), the predistorter does not always work as a complete nonlinear compensator. We find that there is a minimum value of IBO for the predistorter to compensate for the nonlinear SSP A, which is changed as the number of user. If IBO is lower than 6 ㏈ at the user number of 16 and p=1, the system with predistorter is poorer than the one without predistorter. Only when the IBO is over 6 ㏈, predistorter does work as a nonlinear compensator. We call it as cross-over IBO value. TWTA improves the more compensation performance than SSPA because characteristic AM/AM of TWTA has more nonlinear than SSPA. At the BER=10$\^$-3/, there are SNR power gains of about 2.5 ㏈ and 3 ㏈ due to the predistorter when the numbers of users are 16 and 32, respectively.

• Journal of Korean Neurosurgical Society
• /
• v.60 no.2
• /
• pp.225-231
• /
• 2017

Objective : Solid bone fusion is an essential process in spinal stabilization surgery. Recently, as several minimally invasive spinal surgeries have developed, a need of artificial bone substitutes such as demineralized bone matrix (DBM), has arisen. We investigated the in vivo bone growth rate of DBM as a bone void filler compared to a local autologous bone grafts. Methods : From April 2014 to August 2015, 20 patients with a one or two-level spinal stenosis were included. A posterior lumbar interbody fusion using two cages and pedicle screw fixation was performed for every patient, and each cage was packed with autologous local bone and DBM. Clinical outcomes were assessed using the Numeric Rating Scale (NRS) of leg pain and back pain and the Korean Oswestry Disability Index (K-ODI). Clinical outcome parameters and range of motion (ROM) of the operated level were collected preoperatively and at 3 months, 6 months, and 1 year postoperatively. Computed tomography was performed 1 year after fusion surgery and bone growth of the autologous bone grafts and DBM were analyzed by ImageJ software. Results : Eighteen patients completed 1 year of follow-up, including 10 men and 8 women, and the mean age was 56.4 (32-71). The operated level ranged from L3/4 to L5/S1. Eleven patients had single level and 7 patients had two-level repairs. The mean back pain NRS improved from 4.61 to 2.78 (p=0.003) and the leg pain NRS improved from 6.89 to 2.39 (p<0.001). The mean K-ODI score also improved from 27.33 to 13.83 (p<0.001). The ROM decreased below 2.0 degrees at the 3-month assessment, and remained less than 2 degrees through the 1 year postoperative assessment. Every local autologous bone graft and DBM packed cage showed bone bridge formation. On the quantitative analysis of bone growth, the autologous bone grafts showed significantly higher bone growth compared to DBM on both coronal and sagittal images (p<0.001 and p=0.028, respectively). Osteoporotic patients showed less bone growth on sagittal images. Conclusion : Though DBM alone can induce favorable bone bridging in lumbar interbody fusion, it is still inferior to autologous bone grafts. Therefore, DBM is recommended as a bone graft extender rather than bone void filler, particularly in patients with osteoporosis.

• Journal of the Korean Geotechnical Society
• /
• v.18 no.5
• /
• pp.37-42
• /
• 2002

Artificial neural networks are efficient computing techniques that are widely used to solve complex problems in many fields. In this paper a liquefaction potential was estimated by using a back propagation neural network model applicated to cyclic triaxial test data, soil parameters and site investigation data. Training and testing of the network were based on a database of 43 cyclic triaxial test data from 00 sites. The neural networks are trained by modifying the weights of the neurons in response to the errors between the actual output values and the target output value. Training was done iteratively until the average sum squared errors over all the training patterns were minimized. This generally occurred after about 15,000 cycles of training. The accuracy from 72% to 98% was shown for the model equipped with two hidden layers and ten input variables. Important effective input variables have been identified as the NOC,$D_10$ and (N$_1$)$_60$. The study showed that the neural network model predicted a CSR(Cyclic shear stress Ratio) of silty-sand reasonably well. Analyzed results indicate that the neural-network model is more reliable than simplified method using N value of SPT.

• Journal of the Korean Geotechnical Society
• /
• v.19 no.6
• /
• pp.151-160
• /
• 2003

The stability of the designed rock slope is analysed based on two kinds of shear strength model. Besides the deterministic analysis, a probabilistic approach on Monte Carlo simulation is proposed to deal with the uncertain characteristics of the discontinuity and the results obtained from two models are compared to each other. To carry out the research of characteristics of the discontinuity, BIPS, DOM Scanline survey data and direct shear test data are used, and chi-square test is used for determining the probability distribution function. The rock slope is evaluated to be stable in the deterministic analysis, but in the probabilistic analysis, the probability of failure is more than 5%, so, it is considered that the rock slope is unstable. In the shear strength models, the probability of the failure based on the Mohr-Coulomb model(linear model) is higher than that of the Barton model. It is supported by the fact that the Mohr-Coulomb model is more sensitive to block size than the Barton model. In fact, there is no reliable way to estimate the unit cohesion of the Mohr-Coulomb model except f3r back analysis and in the case of small block failure in the slope, Mohr-Coulomb model may excessively evaluate the factor of the safety. So, the Barton model of which parameters are easily acquired using the geological survey is more reasonable for the stability of the studied slope. Also, the selection of the proper shear strength model is an important factor for slope failure analysis.

• Interaction and multiscale mechanics
• /
• v.5 no.3
• /
• pp.229-265
• /
• 2012

Hygroexpansion of wood is a known and undesired characteristic in civil engineering. When wood is exposed to changing environmental humidity, it adsorbs or desorbs moisture and warps. The resulting distortions or - at restrained conditions - cracks are a major concern in timber engineering. We herein present a multiscale model for prediction of the macroscopic hygroexpansion behavior of individual pieces of softwood from their microstructure, demonstrated for spruce. By applying poromicromechanics, we establish a link between the swelling pressure, driving the hygroexpansion of wood at the nanoscale, and the resulting macroscopic dimensional changes. The model comprises six homogenization steps, which are performed by means of continuum micromechanics, the unit cell method and laminate theory, all formulated in a poromechanical framework. Model predictions for elastic properties of wood as functions of the moisture content closely approach corresponding experimental data. As for the hygroexpansion behavior, the swelling pressure has to be back-calculated from macroscopic hygroexpansion data. The good reproduction of the anisotropy of wood hygroexpansion, based on only a single scalar calibration parameter, underlines the suitability of the model. The multiscale model constitutes a valuable tool for studying the effect of microstructural features on the macroscopic behavior and for assessing the hygroexpansion behavior at smaller length scales, which are inaccessible to experiments. The model predictions deliver input parameters for the analysis of timber at the structural scale, therewith enabling to optimize the use of timber and to prevent moisture-induced damage or failure.

• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.25 no.5
• /
• pp.276-284
• /
• 2013

A stochastic expected cost model has been suggested by combining the nonlinear cumulative damage model with the expected cost model together which can be useful for doing the preventive optimal-maintenance of the armor units of rubble-mound breakwaters. The suggested model has been satisfactorily calibrated by comparison of the results from others models, also the sensitivity analysis has been carried out in detail under the variation of the associated parameters with the model. The optimal repair times can be directly evaluated by minimizing the expected cost rates that depend on the social importances, damage intensity functions and resistance limits. Finally, the present cost model has been straightforwardly applied to the armor units of rubble-mound breakwaters. Based on the assumption of turning the damaged structure back to the state as good as new after repairs, the required optimal repair times and magnitudes can be determined quantitatively in terms of the optimum balance between the costs and benefits on the preventive maintenance.

• Korean Journal of Optics and Photonics
• /
• v.21 no.5
• /
• pp.214-223
• /
• 2010

Optimization of a 30 cm lightweight mirror was proposed with the best optical performance under various loads of gravity and thermal loads with proper boundary conditions. A pattern for a lightweight mirror was generated based on the best optical performance combined with ease of manufacturing for proper design parameters of physical properties of face sheet, back sheet, rib, and web. Evaluation of the optical performances of a telescope mirror was obtained by using the finite element analysis program, NX I-DEAS. Surface errors, individual aberration terms, such as piston, tilts, focus and other aberrations were calculated by using Zernike polynomials. The proposed telescope mirror meets well the opto-mechanical design consideration of RMS surface error less than 16 nm.

• The Korean Journal of Applied Statistics
• /
• v.22 no.5
• /
• pp.995-1005
• /
• 2009

Conditional correlation between financial time series plays an important role in risk management, asset allocation and portfolio selection and therefore diverse efforts for modeling conditional correlations in multivariate-GARCH processes have been made in last two decades. In particular, CCC (cf. Bollerslev, 1990) and DCC(dynamic conditional correlation, cf. Engle, 2002) models have been commonly used since they are relatively parsimonious in the number of parameters involved. This article is concerned with DCC modeling for multivariate GARCH processes in comparison with CCC specification. Various multivariate financial time series are analysed to illustrate possible advantages of DCC over CCC modeling.