• Journal of the Korean Magnetic Resonance Society
• /
• v.26 no.1
• /
• pp.1-9
• /
• 2022

There are two distinct approaches to improving the quality of protein NMR structures during refinement: all-atom force fields and accumulated knowledge-assisted methods that include Rosetta. Mao et al. reported that, for 40 proteins, Rosetta increased the accuracies of their NMR-determined structures with respect to the X-ray crystal structures (Mao et al., J. Am. Chem. Soc. 136, 1893 (2014)). In this study, we calculated 32 structures of those studied by Mao et al. using all-atom force field and implicit solvent model, and we compared the results with those obtained from Rosetta. For a single protein, using only the experimental NOE-derived distances and backbone torsion angle restraints, 20 of the lowest energy structures were extracted as an ensemble from 100 generated structures. Restrained simulated annealing by molecular dynamics simulation searched conformational spaces with a total time step of 1-ns. The use of GPU-accelerated AMBER code allowed the calculations to be completed in hours using a single GPU computer-even for proteins larger than 20 kDa. Remarkably, statistical analyses indicated that the structures determined in this way showed overall higher accuracies to their X-ray structures compared to those refined by Rosetta (p-value < 0.01). Our data demonstrate the capability of sophisticated atomistic force fields in refining NMR structures, particularly when they are coupled with the latest GPU-based calculations. The straightforwardness of the protocol allows its use to be extended to all NMR structures.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.12
• /
• pp.3364-3382
• /
• 2023

Remote sensing image segmentation plays an important role in realizing intelligent city construction. The current mainstream segmentation networks effectively improve the segmentation effect of remote sensing images by deeply mining the rich texture and semantic features of images. But there are still some problems such as rough results of small target region segmentation and poor edge contour segmentation. To overcome these three challenges, we propose an improved semantic segmentation model, referred to as MRU-Net, which adopts the U-Net architecture as its backbone. Firstly, the convolutional layer is replaced by BasicBlock structure in U-Net network to extract features, then the activation function is replaced to reduce the computational load of model in the network. Secondly, a hybrid multi-scale recognition module is added in the encoder to improve the accuracy of image segmentation of small targets and edge parts. Finally, test on Massachusetts Buildings Dataset and WHU Dataset the experimental results show that compared with the original network the ACC, mIoU and F1 value are improved, and the imposed network shows good robustness and portability in different datasets.

• Journal of information and communication convergence engineering
• /
• v.22 no.2
• /
• pp.165-171
• /
• 2024

In this study, we present a novel approach for enhancing chest X-ray image classification (normal, Covid-19, edema, mass nodules, and pneumothorax) by combining contrastive learning and machine learning algorithms. A vast amount of unlabeled data was leveraged to learn representations so that data efficiency is improved as a means of addressing the limited availability of labeled data in X-ray images. Our approach involves training classification algorithms using the extracted features from a linear fine-tuned Momentum Contrast (MoCo) model. The MoCo architecture with a Resnet34, Resnet50, or Resnet101 backbone is trained to learn features from unlabeled data. Instead of only fine-tuning the linear classifier layer on the MoCopretrained model, we propose training nonlinear classifiers as substitutes for softmax in deep networks. The empirical results show that while the linear fine-tuned ImageNet-pretrained models achieved the highest accuracy of only 82.9% and the linear fine-tuned MoCo-pretrained models an increased highest accuracy of 84.8%, our proposed method offered a significant improvement and achieved the highest accuracy of 87.9%.

• Journal of Animal Science and Technology
• /
• v.66 no.1
• /
• pp.167-177
• /
• 2024

Iris segmentation is an initial step for identifying the biometrics of animals when establishing a traceability system for livestock. In this study, we propose a deep learning framework for pixel-wise segmentation of bovine iris with a minimized use of annotation labels utilizing the BovineAAEyes80 public dataset. The proposed image segmentation framework encompasses data collection, data preparation, data augmentation selection, training of 15 deep neural network (DNN) models with varying encoder backbones and segmentation decoder DNNs, and evaluation of the models using multiple metrics and graphical segmentation results. This framework aims to provide comprehensive and in-depth information on each model's training and testing outcomes to optimize bovine iris segmentation performance. In the experiment, U-Net with a VGG16 backbone was identified as the optimal combination of encoder and decoder models for the dataset, achieving an accuracy and dice coefficient score of 99.50% and 98.35%, respectively. Notably, the selected model accurately segmented even corrupted images without proper annotation data. This study contributes to the advancement of iris segmentation and the establishment of a reliable DNN training framework.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.30 no.4B
• /
• pp.235-245
• /
• 2005

In this paper, a simplified QoS model of MPLS-based backbone network. Conventional scheme proposed by IETF(IETF schem) is to embed a DiffServ model in MPLS network. However, this approach results in overall upgrade of MPLS system and so it is difficult to deploy this approach. Our proposed model, however, uses a Vidual Link which is a set of Label Switched Path(LSP) connected from an Ingress Label Edge Router(LER) to an Egress LER. In this model, Per-Hop-Behavior(PHB) is implemented only at each LSP in ingress LER and Core Label Switch Routers(LSRs) just guarantee each LSP's bandwidth, not service. This bandwidth guarantee service is fully provided by legacy MPLS model. Also we propose flow allocation mechanism and the flow distribution among LSPs of the virtual link by the flow according to the network status. To evaluate the simplified approach, the characteristics of the approach are compared logically with these of IETF's one through simulations.

• Membrane Journal
• /
• v.30 no.1
• /
• pp.57-65
• /
• 2020

The purpose of this study was to compare the water channel morphology and the proton conductivity by changing the number of repeating units of the polymer backbone of PEMs, and to present a criterion for selecting an appropriate polymer model for MD simulation. In the model with the shortest polymer main chain, the movement of the main chain and the sulfonic acid group was observed to be large, but no change in the water channel morphology was found. In addition, due to the nature of the proton transport ability that is most affected by the water channel morphology, the proton conductivity did not show a significant correlation with the length of the polymer backbone. These results provide important information, particularly for the preparation of ionomers for binders. In general, a low molecular weight polymer electrolyte material is used for a binder ionomer. Since the movement of the main chain/sulfonic acid group is improved, it can play a role of enclosing the catalyst layer well. However, there is no change in its proton conducting performance. In conclusion, the preparation of ionomers for binders will require molecular weight and structure design with a focus on physical properties rather than proton transfer performance.

• KDI Journal of Economic Policy
• /
• v.31 no.2
• /
• pp.1-13
• /
• 2009

The monetary model suggests that nominal exchange rates between two countries will be determined by important macroeconomic variables. The existence of a cointegrating relationship among these fundamental variables is the backbone of the monetary model. In a recent paper, Rapach and Wohar (2002, Journal of International Economics) advance the literature by testing for linear cointegration in the monetary model using a century of data to increase power. They find evidence of cointegration in five or six of ten countries. We extend their work to the nonlinear framework by performing threshold cointegration tests that allow for asymmetric adjustments in two regimes. Asymmetric adjustments in exchange rates can occur, for example, if transactions costs are present or if policy makers react asymmetrically to changing fundamentals. Moreover, whereas Rapach and Wohar (2002) found it necessary to exclude the relative output variable in some cases to maintain the validity of their cointegration tests, we can include this variable as a stationary covariate to increase power. Overall, using their same long-span data, we find more support for cointegration in a nonlinear framework.

• Journal of the Korea Concrete Institute
• /
• v.21 no.5
• /
• pp.599-609
• /
• 2009

This paper discusses generalized modeling schemes for both reinforced concrete (RC) and post-tensioned (PT) flat plate buildings. In this modeling approach, nonlinear behavior due to slab flexure, moment and shear transfer at slab-column connections, and punching shear was included along with linear secant stiffness of each member or connection that accounts for concrete cracking. This generalized model was capable of simulating all different scenarios of slab-column connection failures such as brittle punching, flexure-shear interactive failure, and flexural failure followed by drift-induced punching. Furthermore, automatic detection of drift-induced punching shear and subsequent backbone curve modifications were realistically modelled by incorporating the limit state model, in which gravity shear versus drift capacity relations were adopted. The validation of the model was conducted using one-third scale two-story by two-bay RC and PT flat plate frames. The comparisons revealed that the model was robust and effective.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.28 no.10B
• /
• pp.947-956
• /
• 2003

The current IP Routing Protocolspacket networks also need to provide the network QoS based of DiffServ, RSVP, MPLStraffic model which is standardized as IETF reference model for NGN. The first topic of this paper is to propose Traffic-Balanced Routing Protocol(TBRP) to process existing best effort traffic. TBRP will process low priority interactive data and background data which is not sensitive to dealy. Secondly Hierarchical Traffic-Traffic-Scheduling Routing Protocol(HTSRP) is also proposed. HTSRP is the hierarchical routing algorithm for backbone and access networkin case of fixed-wireless convergence network. Finally, HTSRP_Q is proposed to meet the QoS requirement when user want interactive or streaming packet service. This protocol will maximize the usage of resources of access layer based on the QoS parameters and process delay-sensitive traffic. Service classes are categorized into 5 types by the user request, such as conversational, streaming, high priority interactive, low priority interactive, and background class. It could be processed efficiently by the routing protocolstraffic model proposed in this paper. The proposed routing protocolstraffic model provides the increase of efficiency and stability of the next generation network thanks to the routing according to the characteristic of the specialized service categories.

• Steel and Composite Structures
• /
• v.30 no.3
• /
• pp.201-215
• /
• 2019

In this research, the behavior of tube-in-tube BRBs (TiTBRBs) has been investigated. In a typical TiTBRB, the yielding core tube is located inside the outer restraining one to dissipate energy through extensive plastic deformation, while the outer restraining tube remains essentially elastic. With the aid of FE analyses, the monotonic and cyclic behavior of the proposed TiTBRBs have been studied as individual brace elements. Subsequently, a detailed finite element model of a representative single span-single story frame equipped with such a TiTBRB has been constructed and both monotonic and cyclic behavior of the proposed TiTBRBs have been explored under the application of the AISC loading protocol at the braced frame level. With the aid of backbone curves derived from the FE analyses, a simplified frame model has been developed and verified through comparison with the results of the detailed FE model. It has been shown that, the simplified model is capable of predicting closely the cyclic behavior of the TiTBRB frame and hence can be used for design purposes. Considering type of connection detail used in a frame, the TiTBRB member which behave satisfactorily at the brace element level under cyclic loading conditions, may suffer global buckling due to the flexural demand exerted from the frame to the brace member at its ends. The proposed TiTBRB suit tubular members of offshore structures and the application of such TiTBRB in a typical offshore platform has been introduced and studied in a single frame level using detailed FE model.