• Korean Journal of Radiology
• /
• v.22 no.3
• /
• pp.476-488
• /
• 2021

Objective: We aimed to develop a deep neural network for segmenting lung parenchyma with extensive pathological conditions on non-contrast chest computed tomography (CT) images. Materials and Methods: Thin-section non-contrast chest CT images from 203 patients (115 males, 88 females; age range, 31-89 years) between January 2017 and May 2017 were included in the study, of which 150 cases had extensive lung parenchymal disease involving more than 40% of the parenchymal area. Parenchymal diseases included interstitial lung disease (ILD), emphysema, nontuberculous mycobacterial lung disease, tuberculous destroyed lung, pneumonia, lung cancer, and other diseases. Five experienced radiologists manually drew the margin of the lungs, slice by slice, on CT images. The dataset used to develop the network consisted of 157 cases for training, 20 cases for development, and 26 cases for internal validation. Two-dimensional (2D) U-Net and three-dimensional (3D) U-Net models were used for the task. The network was trained to segment the lung parenchyma as a whole and segment the right and left lung separately. The University Hospitals of Geneva ILD dataset, which contained high-resolution CT images of ILD, was used for external validation. Results: The Dice similarity coefficients for internal validation were 99.6 ± 0.3% (2D U-Net whole lung model), 99.5 ± 0.3% (2D U-Net separate lung model), 99.4 ± 0.5% (3D U-Net whole lung model), and 99.4 ± 0.5% (3D U-Net separate lung model). The Dice similarity coefficients for the external validation dataset were 98.4 ± 1.0% (2D U-Net whole lung model) and 98.4 ± 1.0% (2D U-Net separate lung model). In 31 cases, where the extent of ILD was larger than 75% of the lung parenchymal area, the Dice similarity coefficients were 97.9 ± 1.3% (2D U-Net whole lung model) and 98.0 ± 1.2% (2D U-Net separate lung model). Conclusion: The deep neural network achieved excellent performance in automatically delineating the boundaries of lung parenchyma with extensive pathological conditions on non-contrast chest CT images.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.26 no.1
• /
• pp.27-38
• /
• 2002

The effects of Reynolds number on the non-nulling calibrations of a cone-type ave-type probe in low-speed flows have been investigated at Reynolds numbers of 2.04$\times$10$^3$, 4.09$\times$10$^3$and 6.13$\times$10$^3$. The calibration is conducted at the pitch and yaw angles in ranges between -35 degrees and 35 degrees with an angle interval of 5 degrees. In addition to the calibration coefficients, reduced pitch and yaw angles, static and total pressures, and velocity magnitude are obtained through a typical non-nulling reduction procedure. The result shows that each calibration coefficient, in general, is a function of both the pitch and yaw angles, so that the pre-existing calibration data in a nulling mode are not enough in accounting far the full non-nulling calibration characteristics. Due to interference of the probe stem, the calibration coefficient are more sensitive to Reynolds number at positive pitch angles than at negative ones. The calibration data reduced in this study may serve as a guide line in the estimation of uncertainty intervals resulted from the Reynolds number effects at low Reynolds numbers.

• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.30 no.3
• /
• pp.114-122
• /
• 2018

This paper introduces a non-hydrostatic wave model SWASH for simulating wave interactions with porous structures. This model calculates the flow in porous media based on volume-averaged Reynolds-averaged Navier-Stokes equations (VARANS) in ${\sigma}$-coordinate. The empirical coefficients of resistance used to account for the flow in a porous media often need to be measured or calibrated. In this study, the empirical resistance coefficients used in the model are calibrated and validated using laboratory experiments, involving dam-break flow through porous media, and solitary wave interactions with a porous structure. It is shown that the agreement between experimental and numerical results is generally satisfactory. It is also confirmed that non-hydrodynamic model, SWASH, is computationally much more efficient than the three-dimensional porous flow models based on VOF approach.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.9 no.1
• /
• pp.165-176
• /
• 2013

This paper introduces the high density discrete wavelet transform using quincunx sampling, which is a discrete wavelet transformation that combines the high density discrete transformation and non-separable processing method, each of which has its own characteristics and advantages. The high density discrete wavelet transformation is one that expands an N point signal to M transform coefficients with M > N. The high density discrete wavelet transformation is a new set of dyadic wavelet transformation with two generators. The construction provides a higher sampling in both time and frequency. This new transform is approximately shift-invariant and has intermediate scales. In two dimensions, this transform outperforms the standard discrete wavelet transformation in terms of shift-invariant. Although the transformation utilizes more wavelets, sampling rates are high costs and some lack a dominant spatial orientation, which prevents them from being able to isolate those directions. A solution to this problem is a non separable method. The quincunx lattice is a non-separable sampling method in image processing. It treats the different directions more homogeneously than the separable two dimensional schemes. Proposed wavelet transformation can generate sub-images of multiple degrees rotated versions. Therefore, This method services good performance in image processing fields.

• Journal of Ocean Engineering and Technology
• /
• v.15 no.4
• /
• pp.115-119
• /
• 2001

The simulation of motion responses of a dolphin-moored ocean structure in shallow water when it cllides with a ship, has been carried out. The equation of motion in the time domain according to Cummin's theory is employed, and solved by making use of the Newmark-${\beta}$ method. The added mass and damping coefficients involved in the equations are abtained from a three-dimensional panel method in the frequency domain. The impact forces due to ship collision are calculated using both the elastic and non-elastic modelings. The mooring forces for dolphin systems of ocean structure are regarded as linear spring forces.

• Proceedings of the Korea Committee for Ocean Resources and Engineering Conference
• /
• 2001.05a
• /
• pp.103-107
• /
• 2001

When ship claps against the ocean structure sited at shallow water, the time simulation of motion responses of dolphin-moored ocean structure is presented. The equatien of motion based on Cummin's theory of impulse responses are employed, and solved in time domain by using the Newmark $\beta$ method. The added mass and damping coefficients involved in the equations are obtained from a three-dimensional panel method in the frequency domain. The impact forces due to ship collision are modeled as two method, and those are elastic and non-elastic collisions. The mooring forces for dolphin systems of scean structure are considered as linear spring system.

• International Journal of Industrial Entomology and Biomaterials
• /
• v.30 no.1
• /
• pp.6-16
• /
• 2015

This study established the genetic characterisation of 10 microsporidian isolates infecting non-mulberry silkworms (Antheraea assamensis and Samia cynthia ricini) collected from biogeographical forest locations in the State of Assam, India, using PCR-based markers assays: inter simple sequence repeat (ISSR) and random amplified polymorphic DNA (RAPD). A Nosema type species (NIK-1s_mys) was used as control for comparison. The shape of mature microsporidian spores were observed oval to elongated, measuring 3.80 to $4.90{\mu}m$ in length and 2.60 to $3.05{\mu}m$ in width. Fourteen ISSR primers generated reproducible profiles and yielded 178 fragments, of which 175 were polymorphic (98%), while 16 RAPD primers generated reproducible profiles with 198 amplified fragments displaying 95% of polymorphism. Estimation of genetic distance coefficients based on dice coefficients method and clustering with un-weighted pair group method using arithmetic average (UPGMA) analysis was done to unravel the genetic diversity of microsporidians infecting Indian muga and eri silkworm. The similarity coefficients varied from 0.385 to 0.941 in ISSR and 0.083 to 0.938 in RAPD data. UPGMA analysis generated dendrograms with two microsporidian groups, which appear to be different from each other. Based on Euclidean distance matrix method, 2-dimensional distribution also revealed considerable variability among different identified microsporidians. Clustering of these microsporidian isolates was in accordance with their host and biogeographic origin. Both techniques represent a useful and efficient tool for taxonomical grouping as well as for phylogenetic classification of different microsporidians in general and genotyping of these pathogens in particular.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.12 no.5
• /
• pp.19-24
• /
• 2004

Entrainment of ambient gas into a transient diesel spray is a crucial factor affecting the following preparation of combustible mixture. In this study, the entrainment characteristics of ambient gas for a non-evaporating transient diesel were investigated using a common-rail injection system. The effects of ambient gas density and nozzle hole geometry were assessed with entrainment coefficient. Laser Doppler Velocimetry (LDV) technique was introduced to measure the entrainment speed of ambient gas into a spray. There appeared a region where the entrainment coefficients remained almost constant while injection rates were still changing. The effect of common-rail pressure, which altered the slope of injection rate curve, was hardly noticed at this region. Entrainment coefficient increased with ambient gas density, that is, the effect of ambient gas density was greater than that of turbulent jet whose entrainment coefficient remained constant. The non-dimensional distance was defined to reflect the effect of nozzle hole diameter and ambient gas density together. The mean value of entrainment coefficient was found to increase with non-dimensional distance from the nozzle tip, which would be suggested as the guideline for the nozzle design.

• Nuclear Engineering and Technology
• /
• v.54 no.1
• /
• pp.36-48
• /
• 2022

As a valid numerical method to obtain a high-resolution result of a flow field, computational fluid dynamics (CFD) have been widely used to study coolant flow and heat transfer characteristics in fuel rod bundles. However, the time-consuming, iterative calculation of Navier-Stokes equations makes CFD unsuitable for the scenarios that require efficient simulation such as sensitivity analysis and uncertainty quantification. To solve this problem, a reduced-order model (ROM) based on proper orthogonal decomposition (POD) and machine learning (ML) is proposed to simulate the flow field efficiently. Firstly, a validated CFD model to output the flow field data set of the rod bundle is established. Secondly, based on the POD method, the modes and corresponding coefficients of the flow field were extracted. Then, an deep feed-forward neural network, due to its efficiency in approximating arbitrary functions and its ability to handle high-dimensional and strong nonlinear problems, is selected to build a model that maps the non-linear relationship between the mode coefficients and the boundary conditions. A trained surrogate model for modes coefficients prediction is obtained after a certain number of training iterations. Finally, the flow field is reconstructed by combining the product of the POD basis and coefficients. Based on the test dataset, an evaluation of the ROM is carried out. The evaluation results show that the proposed POD-ROM accurately describe the flow status of the fluid field in rod bundles with high resolution in only a few milliseconds.

• Structural Engineering and Mechanics
• /
• v.50 no.3
• /
• pp.287-304
• /
• 2014

A computer based iterative numerical procedure has been developed to analyse reinforced high strength concrete columns subjected to horizontal wave loads and eccentric vertical load by taking the material, geometrical and wave load non-linearity into account. The behaviour of the column has been assumed, to be represented by Moment-Thrust-Curvature relationship of the column cross-section. The formulated computer program predicts horizontal load versus deflection behaviour of a column up to failure. The developed numerical model has been applied to analyse several column specimens of various slenderness, structural properties and axial load ratios, tested by other researchers. The predicted values are having a better agreement with experimental results. A simplified user friendly hydrodynamic load model has been developed based on Morison equation supplemented with a wave slap term to predict the high frequency non-linear impulsive hydrodynamic loads arising from steep waves, known as ringing loads. A computer program has been formulated based on the model to obtain the wave loads and non-dimensional wave load coefficients for all discretised nodes, along the length of column from instantaneous free water surface to bottom of the column at mud level. The columns of same size and material properties but having different slenderness ratio are analysed by the developed numerical procedure for the simulated wave loads under various vertical thrust. This paper discusses the results obtained in detail and effect of slenderness in resisting wave loads under various vertical thrust.