• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.8
• /
• pp.1046-1052
• /
• 2021

Artificial intelligence is being applied in various industrial fields to the development of the fourth industry and the construction of high-performance computing environments. In the medical field, deep learning learning such as cancer, COVID-19, and bone age measurement was performed using medical images such as X-Ray, MRI, and PET and clinical data. In addition, ICT medical fusion technology is being researched by applying smart medical devices, IoT devices and deep learning algorithms. Among these techniques, medical image-based deep learning learning requires accurate finding of medical image biomarkers, minimal loss rate and high accuracy. Therefore, in this paper, we would like to compare and analyze the performance of the Cross-Entropy function used in the image classification algorithm of the loss function that derives the loss rate in the chest X-Ray image-based deep learning learning process.

• Steel and Composite Structures
• /
• v.51 no.4
• /
• pp.441-456
• /
• 2024

This paper aims to develop Machine Learning (ML) algorithms to predict the buckling resistance of cold-formed steel (CFS) channels with restrained flanges, widely used in typical CFS sheathed wall panels, and provide practical design tools for engineers. The effects of cross-sectional restraints were first evaluated on the elastic buckling behaviour of CFS channels subjected to pure axial compressive load or bending moment. Feedforward multi-layer Artificial Neural Networks (ANNs) were then trained on different datasets comprising CFS channels with various dimensions and properties, plate thicknesses, and restraining conditions on one or two flanges, while the elastic distortional buckling resistance of the elements were determined according to the Finite Strip Method (FSM). To develop less biased networks and ensure that every observation from the original dataset has the chance of appearing in the training and test set, a K-fold cross-validation technique was implemented. In addition, the hyperparameters of the ANNs were tuned using a grid search technique to provide ANNs with optimum performances. The results demonstrated that the trained ANNs were able to predict the elastic distortional buckling resistance of CFS flange-restrained elements with an average accuracy of 99% in terms of coefficient of determination. The developed models were then used to propose a simple ANN-based design formula for the prediction of the elastic distortional buckling stress of CFS flange-restrained elements. Finally, the proposed formula was further evaluated on a separate set of unseen data to ensure its accuracy for practical applications.

• The Korean Journal of Pesticide Science
• /
• v.6 no.4
• /
• pp.264-270
• /
• 2002

Studies on the resistance monitoring of green peach ahpid, Myzus persicae, its development pattern by artificial selection with imidacloprid and cross-resistance were carried out to develope resistance management strategy. Resistance ratios of M. persicae collected at Hwachon and Dunnae among 5 locations in alpine cultivation area appeared to be high as 37.2 and 16.5, respectively. Resistance of aphid to imidacloprid developed slowly up to 20 time selection, and after that it grew quickly. Imidacloprid-resistant aphid strain showed low cross-resistance ratios(<10) to most of organophosphates, carbamates, and mixed insecticides except pirimicarb(487.8), but high ratios to acetamiprid(143.0) which is one of the neonicotinoids like imidacloprid, and pyrethroids such as deltamethrin(14.9), flucythrinate(12.9) and halothrin(15.9).

• Korean Journal of Medicinal Crop Science
• /
• v.16 no.5
• /
• pp.287-293
• /
• 2008

This study was conducted to determine the characteristics of fertilization process and embryo development of Astragalus membranaceus Bunge (Astragali Radix) to provide basic data needed in its breeding. A. membranaceus showed poor seed setting when self-pollination was induced. When artificial pollination was induced, it showed less than 5% bearing in late August, but more than 13% bearing from the beginning of September 4th. The flower size was about $17.0\;mm{\times}4.0\;mm$ and pistils and stamens had the same length of 15.0mm at flowering stage. When self-pollination or cross-pollination was induced, pollen tubes extended to an ovule. While pollen tube was extending to the ovule, reproductive cell split and formed two male generative nuclei and a vegetative nucleus. In the case of self-pollination, fertilized embryo was not observed, but was formed in the case of cross-pollination. A. membranaceus is noted to have zygote self-incompatibility. In the case of cross-pollination, fertilization was observed in 6 to 8 h after pollination, where apical cell derivatives split after fertilization. A spherical pro-embryo was then formed three days after fertilization. The seed attained full shape with a seed coat showing its distinctive contour 15 days after fertilization. Thus, A. membranaceus in Leguminosae family is found to have zygote selfincompatibility although its flower shape is shown to match the self-compatibility plant.

• Restorative Dentistry and Endodontics
• /
• v.29 no.3
• /
• pp.267-272
• /
• 2004

The purpose of this in vitro study was to evaluate the effect of surface defects and cross-sectional configuration of NiTi rotary files on the fatigue life under cyclic loading. Three NiTi rotary files ($K3^{TM},{\;}ProFile^{\circledR},{\;}and{\;}HERO{\;}642^{\circledR}$) with ＃30/.04 taper were evaluated. Each rotary file was divided into 2 subgroups : control (no surface defects) and experimental group (artificial surface defects), A total of six groups of each 10 were tested. The NiTi rotary files were rotated at 300rpm using the apparatus which simulated curved canal (40 degree of curvature) until they fracture. The number of cycles to fracture was calculated and the fractured surfaces were observed with a scanning electron microscope. The data were analyzed statistically. The results showed that experimental groups with surface defects had lower number of cycles to fracture than control group but there was only a statistical significance between control and experimental group in the $K3^{TM}$ (p<0.05), There was no strong correlation between the cross-sectional configuration area and fracture resistance under experimental conditions. Several of fractured files demonstrated characteristic patterns of brittle fracture consistent with the propagation of pre-existing cracks. This data indicate that surface defects of NiTi rotary files may significantly decrease fatigue life and it may be one possible factor for early fracture of NiTi rotary files in clinical practice.

• The Journal of the Acoustical Society of Korea
• /
• v.41 no.2
• /
• pp.131-142
• /
• 2022

In order to examine the possibility of horizontal simulations of acoustic waves on the environments of big water depth variations, this study introduces a 3-dimensional model based on the pababolic equation. The model gives approximated solutions by separating the cross- and non cross-terms in the equation. Assuming artificial bathymetry (25 km × 4 km) with a source frequency 75 Hz, the simulations give clear horizontal refractions on the transmission loss distributions. The degree of refractions shows non-linear increase along the propagating range and proportional increase with water depth along the cross range. Another simulations with the real bathymetry (25 km × 8 km) also give clear horizontal refractions. The horizontal distributions present little difference with the depth resolution variations of the same data source because the model gives interpolations over the depth data before simulations. Meanwhile, the horizontal distributions show big difference with those of different data sources.

• Archives of Plastic Surgery
• /
• v.50 no.6
• /
• pp.578-585
• /
• 2023

An artificial sphincter implanted in the bulbous urethra to treat severe postprostatectomy urinary incontinence is effective, but embedding-associated complications can occur. We assessed the feasibility, efficacy, and safety of urethral graciloplasty cross-innervated by the pudendal nerve. A simulation surgery on three male fresh cadavers was performed. Both ends of the gracilis muscle were isolated only on its vascular pedicle with proximal end of the obturator nerve severed and transferred to the perineum. We examined whether the gracilis muscle could be wrapped around the bulbous urethra and whether the obturator nerve was long enough to suture with the pudendal nerve. In addition, surgery was performed on a 71-year-old male patient with severe urinary incontinence. The postoperative 12-month outcomes were assessed using a 24-hour pad test and urodynamic study. In all cadaveric simulations, the gracilis muscles could be wrapped around the bulbous urethra in a γ-loop configuration. The length of the obturator nerve was sufficient for neurorrhaphy with the pudendal nerve. In the clinical case, the postoperative course was uneventful. The mean maximum urethral closure pressure and functional profile length increased from 40.7 to 70 cm H₂O and from 40.1 to 45.3 mm, respectively. Although urinary incontinence was not completely cured, the patient was able to maintain urinary continence at night. Urethral graciloplasty cross-innervated by the pudendal nerve is effective in raising the urethral pressure and reducing urinary incontinence.

• Knowledge Management Research
• /
• v.22 no.4
• /
• pp.71-85
• /
• 2021

The purpose of this study is to find out which artificial intelligence methodology is most suitable for creating a foreign exchange rate prediction model using the indicators of bond market and interest rate market. KTBs and MSBs, which are representative products of the Korea bond market, are sold on a large scale when a risk aversion occurs, and in such cases, the USD/KRW exchange rate often rises. When USD liquidity problems occur in the onshore Korean market, the KRW Cross-Currency Swap price in the interest rate market falls, then it plays as a signal to buy USD/KRW in the foreign exchange market. Considering that the price and movement of products traded in the bond market and interest rate market directly or indirectly affect the foreign exchange market, it may be regarded that there is a close and complementary relationship among the three markets. There have been studies that reveal the relationship and correlation between the bond market, interest rate market, and foreign exchange market, but many exchange rate prediction studies in the past have mainly focused on studies based on macroeconomic indicators such as GDP, current account surplus/deficit, and inflation while active research to predict the exchange rate of the foreign exchange market using artificial intelligence based on the bond market and interest rate market indicators has not been conducted yet. This study uses the bond market and interest rate market indicator, runs artificial neural network suitable for nonlinear data analysis, logistic regression suitable for linear data analysis, and decision tree suitable for nonlinear & linear data analysis, and proves that the artificial neural network is the most suitable methodology for predicting the foreign exchange rates which are nonlinear and times series data. Beyond revealing the simple correlation between the bond market, interest rate market, and foreign exchange market, capturing the trading signals between the three markets to reveal the active correlation and prove the mutual organic movement is not only to provide foreign exchange market traders with a new trading model but also to be expected to contribute to increasing the efficiency and the knowledge management of the entire financial market.

• Proceeding of Spring/Autumn Annual Conference of KHA
• /
• 2006.11a
• /
• pp.248-252
• /
• 2006

This study is intended to understand the intrinsic attributes of the view from the apartment unit in consideration of the diverse and complex elements of the view. To this end, the Questionnaire survey was conducted to identify the tendency in the recognition by apartment dwellers. The Questionnaire survey was conducted for the apartment residents to identify their interest in and the general trend in their recognition of the view from the living rooms of their housing unit, where Questionnaire items regarding landscape elements, the distances to and location of the landscape elements, and floor locations were compiled on the basis of the results from the field survey in the previous study. Consequently, the following results have been derived. 1) Apartment residents recognize not only natural landscape elements but also artificial elements, and prefer natural elements to artificial ones. 2) It is also indicated that they recognize the distances to and locations of landscape elements and that the satisfaction for the distance and location varies depending on the type of the landscape elements. 3) Furthermore, the floor of each unit is shown to result in certain differences in the recognized landscape elements. The cross-analysis between the floor and satisfaction indicates that the higher the floor, the more satisfied the residents are with the view.

• Smart Structures and Systems
• /
• v.4 no.5
• /
• pp.583-603
• /
• 2008

In this study, a real-time damage detection method using output-only acceleration signals and artificial neural networks (ANN) is developed to monitor the occurrence of damage and the location of damage in structures. A theoretical approach of an ANN algorithm that uses acceleration signals to detect changes in structural parameters in real-time is newly designed. Cross-covariance functions of two acceleration responses measured before and after damage at two different sensor locations are selected as the features representing the structural conditions. By means of the acceleration features, multiple neural networks are trained for a series of potential loading patterns and damage scenarios of the target structure for which its actual loading history and structural conditions are unknown. The feasibility of the proposed method is evaluated using a numerical beam model under the effect of model uncertainty due to the variability of impulse excitation patterns used for training neural networks. The practicality of the method is also evaluated from laboratory-model tests on free-free beams for which acceleration responses were measured for several damage cases.