• Journal of Broadcast Engineering
• /
• v.28 no.3
• /
• pp.293-301
• /
• 2023

Unstandardized medical data collection and management are still being conducted manually, and studies are being conducted to classify CT data using deep learning to solve this problem. However, most studies are developing models based only on the axial plane, which is a basic CT slice. Because CT images depict only human structures unlike general images, reconstructing CT scans alone can provide richer physical features. This study seeks to find ways to achieve higher performance through various methods of converting CT scan to 2D as well as axial planes. The training used 1042 CT scans from five body parts and collected 179 test sets and 448 with external datasets for model evaluation. To develop a deep learning model, we used InceptionResNetV2 pre-trained with ImageNet as a backbone and re-trained the entire layer of the model. As a result of the experiment, the reconstruction data model achieved 99.33% in body part classification, 1.12% higher than the axial model, and the axial model was higher only in brain and neck in contrast classification. In conclusion, it was possible to achieve more accurate performance when learning with data that shows better anatomical features than when trained with axial slice alone.

• International Journal of Computer Science & Network Security
• /
• v.23 no.1
• /
• pp.112-119
• /
• 2023

Traditionally used for networking computers and communications, the Internet has been evolving from the beginning. Internet is the backbone for many things on the web including social media. The concept of social networking which started in the early 1990s has also been growing with the internet. Social Networking Sites (SNSs) sprung and stayed back to an important element of internet usage mainly due to the services or provisions they allow on the web. Twitter and Facebook have become the primary means by which most individuals keep in touch with others and carry on substantive conversations. These sites allow the posting of photos, videos and support audio and video storage on the sites which can be shared amongst users. Although an attractive option, these provisions have also culminated in issues for these sites like posting offensive material. Though not always, users of SNSs have their share in promoting hate by their words or speeches which is difficult to be curtailed after being uploaded in the media. Hence, this article outlines a process for extracting user reviews from the Twitter corpus in order to identify instances of hate speech. Through the use of MPCA (Modified Principal Component Analysis) and ECNN, we are able to identify instances of hate speech in the text (Enhanced Convolutional Neural Network). With the use of NLP, a fully autonomous system for assessing syntax and meaning can be established (NLP). There is a strong emphasis on pre-processing, feature extraction, and classification. Cleansing the text by removing extra spaces, punctuation, and stop words is what normalization is all about. In the process of extracting features, these features that have already been processed are used. During the feature extraction process, the MPCA algorithm is used. It takes a set of related features and pulls out the ones that tell us the most about the dataset we give itThe proposed categorization method is then put forth as a means of detecting instances of hate speech or abusive language. It is argued that ECNN is superior to other methods for identifying hateful content online. It can take in massive amounts of data and quickly return accurate results, especially for larger datasets. As a result, the proposed MPCA+ECNN algorithm improves not only the F-measure values, but also the accuracy, precision, and recall.

• Journal of the Korean Geotechnical Society
• /
• v.26 no.1
• /
• pp.45-54
• /
• 2010

It is important to calculate the natural frequency of a piled structure in the design stage in order to prevent resonance-induced damage to the pile foundation and analyze the dynamic behavior of the piled structure during an earthquake. In this paper, a simple but relatively accurate method employing a mass-spring model is presented for the evaluation of the natural frequency of a pile-soil system. Greatly influencing the calculation of the natural frequency of a piled structure, the spring stiffness between a pile and soil was evaluated by using the coefficient of subgrade reaction, the p-y curve, and the subsoil elastic modulus. The resulting natural frequencies were compared with those of 1-g shaking table tests. The comparison showed that the natural frequency of the pile-soil system could be most accurately calculated by constructing a stiffness matrix with the spring stiffness of the Reese (1974) method, which utilizes the coefficient of the subgrade reaction modulus, and Yang's (2009) dynamic p-y backbone curve method. The calculated natural frequencies were within 5% error compared with those of the shaking table tests for the pile system in dry dense sand deposits and 5% to 40% error for the pile system in saturated sand deposits depending on the occurrence of excess pore water pressure in the soil.

• Journal of Navigation and Port Research
• /
• v.47 no.3
• /
• pp.167-173
• /
• 2023

The manufacturing industry is the backbone of the Korean economy. Among them, the petrochemical industry is a strategic growth industry, which makes a profit through reexports based on eminent technology in South Korea which imports all of its crude oil. South Korea imports whole amount of crude oil, which is the raw material for many manufacturing industries, by sea transportation. Therefore, it must respond swiftly to a highly volatile tanker freight market. This study aimed to make an early warning model of crude oil shipping market using a signal approach. The crisis of crude oil shipping market is defined by BDTI. The overall leading index is made of 38 factors from macro economy, financial data, and shipping market data. Only leading correlation factors were chosen to be used for the overall leading index. The overall leading index had the highest correlation coefficient factor of 0.499 two months ago. It showed a significant correlation coefficient five months ago. The QPS value was 0.13, which was found to have high accuracy for crisis prediction. Furthermore, unlike other previous time series forecasting model studies, this study quantitatively approached the time lag between economic crisis and the crisis of the tanker ship market, providing workers and policy makers in the shipping industry with an framework for strategies that could effectively deal with the crisis.

• Journal of Internet Computing and Services
• /
• v.24 no.6
• /
• pp.91-98
• /
• 2023

Recently, according to development of artificial intelligence, a wide range of industry being automatic and optimized. Also we can find out some research of using supervised learning for deteceting defect of railway in domestic rail industry. However, there are structures other than rails on the track, and the fastener is a device that binds the rail to other structures, and periodic inspections are required to prevent safety accidents. In this paper, we present a method of reducing cost for labeling using semi-supervised and transfer model trained on rail fastener data. We use Resnet50 as the backbone network pretrained on ImageNet. At first we randomly take training data from unlabeled data and then labeled that data to train model. After predict unlabeled data by trained model, we adopted a method of adding the data with the highest probability for each class to the training data by a predetermined size. Futhermore, we also conducted some experiments to investigate the influence of the number of initially labeled data. As a result of the experiment, model reaches 92% accuracy which has a performance difference of around 5% compared to supervised learning. This is expected to improve the performance of the classifier by using relatively few labels without additional labeling processes through the proposed method.

• Applied Chemistry for Engineering
• /
• v.35 no.4
• /
• pp.296-302
• /
• 2024

An ionic conjugated polymer-iron (III) chloride composite was prepared via in-situ quaternization polymerization of 2-ethynylpyridine (2EP) using iron (III) chloride. Various instrumental methods revealed that the chemical structure of the resulting conjugated polymer (P2EP)-iron (III) chloride composite has the conjugated backbone system having the designed pyridinium ferric chloride complexes. The polymerization mechanism was assumed to be that the activated triple bond of 2-ethynylpyridinium salt, formed at the first reaction step, is easily susceptible to the step-wise polymerization, followed by the same propagation step that contains the propagating macroanion and monomeric 2-ethynylpyridinium salts. The electro-optical and electrochemical properties of the P2EP-FeCl₃ composite were studied. In the UV-visible spectra of P2EP-FeCl₃ composite, the absorption maximum values were 480 nm and 533 nm, and the PL maximum value was 598 nm. The cyclic voltammograms of the P2EP-FeCl₃ composite exhibited irreversible electrochemical behavior between the oxidation and reduction peaks. The kinetics of the redox process of composites were found to be very close to a diffusion-controlled process from the plot of the oxidation current density versus the scan rate.

• Journal of the Korean Society of Radiology
• /
• v.18 no.5
• /
• pp.541-549
• /
• 2024

The various structures of artificial neural networks, such as convolutional neural networks (CNNs) and recurrent neural networks (RNNs), have been extensively studied and served as the backbone of numerous models. Among these, a transformer architecture has demonstrated its potential for natural language processing and become a subject of in-depth research. Currently, the techniques can be adapted for image processing through the modifications of its internal structure, leading to the development of Vision transformer (ViT) models. The ViTs have shown high accuracy and performance with large data-sets. This study aims to develop a ViT-based model for detecting pneumonia using chest X-ray images and quantitatively evaluate its performance. The various architectures of the ViT-based model were constructed by varying the number of encoder blocks, and different patch sizes were applied for network training. Also, the performance of the ViT-based model was compared to the CNN-based models, such as VGGNet, GoogLeNet, and ResNet. The results showed that the traninig efficiency and accuracy of the ViT-based model depended on the number of encoder blocks and the patch size, and the F1 scores of the ViT-based model ranged from 0.875 to 0.919. The training effeciency of the ViT-based model with a large patch size was superior to the CNN-based models, and the pneumonia detection accuracy of the ViT-based model was higher than that of the VGGNet. In conclusion, the ViT-based model can be potentially used for pneumonia detection using chest X-ray images, and the clinical availability of the ViT-based model would be improved by this study.

• Journal of the Korea Society of Computer and Information
• /
• v.29 no.10
• /
• pp.35-45
• /
• 2024

In this paper, we propose Multi-Scale Feature Knowledge Distillation for 3D Object Detection (M3KD), which extracting knowledge from the teacher model, and transfer to the student model consider with multi-scale feature map. To achieve this, we minimize L2 loss between feature maps at each pyramid level of the student model with the correspond teacher model so student model can mimic the teacher model backbone information which improves the overall accuracy of the student model. We apply the class logits knowledge distillation used in the image classification task, by allowing student model mimic the classification logits of the teacher model, to guide the student model to improve the detection accuracy. In KITTI (Karlsruhe Institute of Technology and Toyota Technological Institute) dataset, our M3KD (Multi-Scale Feature Knowledge Distillation for 3D Object Detection) student model achieves 30% inference speed improvement compared to the teacher model. Additionally, our method achieved an average improvement of 1.08% in 3D mean Average Precision (mAP) across all classes and difficulty levels compared to the baseline student model. Furthermore, when integrated with the latest knowledge distillation methods such as PKD and SemCKD, our approach achieved an additional 0.42% and 0.52% improvement in 3D mAP, respectively, further enhancing performance.

• Polymer(Korea)
• /
• v.24 no.4
• /
• pp.505-512
• /
• 2000

The mixture density data for poly(lactide-co-glycolide) [PLGA] with supercritical $CO_2$, CHF$_3$ and CHClF$_2$ were obtained in the temperature range of 27 to 10$0^{\circ}C$ and at pressures as high as 3000 bar (PLGA$_{x}$, Where the molar concentration of glycolide in the backbone, x, range from 0 to 50 mol%). The PLA-$CO_2$, PLA-CHF$_3$, and PLA-CHClF$_2$ systems dissolve in the pressure less than 1430 below 700, and below 100 bar, respectively. The mixture density shows from 1.084 to 1.334 g/cm$^3$ at temperatures from 27 to 93$^{\circ}C$. The PLGA$_{15}$ -$CO_2$ mixture dissolves at pressures of below 1900 bar and the mixture density is in the range of 1.158 to 1.247 g/cm$^3$ at temperatures between 37 and 92$^{\circ}C$. The solubilities of the PLGA$_{25}$ for $CO_2$, CHF$_3$, and CHClF$_2$ are shown to pressure as high as 2390, 1470, and 118 bar, respectively, and the mixture density exhibits iron 1.154 to 1.535 g/cm$^3$ at temperatures from 29 to 81$^{\circ}C$. The PLGA$_{50}$-$CO_2$ system does not dissolve at 24$0^{\circ}C$ and 3000 bar while the PLGA$_{50}$-CHCIF$_2$ does easily at 5$0^{\circ}C$ and 100 bar. The mixture density for the PLGA-CHClF$_2$ system increases even at low pressures as the glycolide molar concentration increases.es.es.

• Journal of Ginseng Research
• /
• v.24 no.1
• /
• pp.8-17
• /
• 2000

Estrogen can influence on the expression of behaviors not associated directly with reproduction, including learning and memory. Recently estrogen has received considerable attention for its effects on neuroprotection and neural circuits in brain areas associated with cognition. Although estrogen replacement therapy may be helpful to postmenopausal women, it also results in a number of harmful side effects. Ginseng also has steroidal qualities and contains several ginsenoside components which have similar backbone structure to estrogen. The objectives of this experiment were 1) to examine the effects of estrogen and 2) to investigate the effects of ginsenosides as estrogenic agent on learning and memory using the Morris water maze, a traditional experimental task for spatial memory. In the experiments designed here, ovariectomized mice were implanted subcutaneously with Sila, itic capsules containing 17${\beta}$-estradiol (100∼250 $\mu\textrm{g}$/$m\ell$), panaxadiol (PD) and panaxatriol (PT) saponins (15∼100 $\mu\textrm{g}$/$m\ell$) diluted with sesame oil. In the first set of experiment, the effects of estradiol on learning and memory during the Morris water maze was examined. When estradiol was delivered via Silastic capsules following training improved spatial memory performance in ovariectomized female mice. In the second set of experiment, three different PD and PT saponin concentrations were delivered via Silastic implants to ovariectomized female mice and their effects were compared with estrogenic effects. Results of three separate experiments demonstrated that estradiol, PD and PT administrated by Silastic implants for 2 weeks prior to water maze training significantly improved spatial memory performance compared to ovariectomized (OVX) mice, as indicated by lower escape latency over trial. The positive effect of estradiol suggests that estrogen can affect performance on learning and memory. In addition, the positive effect of PD and PT saponins suggest that ginsenosides have an estrogen-like effects in mediating learning and memory related behavior action.