• Journal of Institute of Convergence Technology
• /
• v.11 no.1
• /
• pp.1-5
• /
• 2021

Deep learning based methods achieve state-of-the-art accuracy, however, they typically rely on supervised training with large labeled datasets. It is known in many medical applications that labeling medical images requires significant expertise and much time, and typical hand-tuned approaches for data augmentation fail to capture the complex variations in such images. This paper proposes a 3D image augmentation method to overcome these difficulties. It allows us to enrich diversity of training data samples that is essential in medical image segmentation tasks, thus reducing the data overfitting problem caused by the fact the scale of medical image dataset is typically smaller. Our numerical experiments demonstrate that the proposed approach provides significant improvements over state-of-the-art methods for 3D medical image segmentation.

• Journal of Korea Multimedia Society
• /
• v.25 no.8
• /
• pp.1203-1211
• /
• 2022

Melanoma is a skin cancer that starts in pigment-producing cells (melanocytes). The death rates of skin cancer like melanoma can be reduced by early detection and diagnosis of diseases. It is common for doctors to spend a lot of time trying to distinguish between skin lesions and healthy cells because of their striking similarities. The detection of melanoma lesions can be made easier for doctors with the help of an automated classification system that uses deep learning. This study presents a new approach for melanoma classification based on an ensemble of deep convolution neural networks and a Log-Gabor filter. First, we create the Log-Gabor representation of the original image. Then, we input the Log-Gabor representation into a new ensemble of deep convolution neural networks. We evaluated the proposed method on the melanoma dataset collected at Yonsei University and Dongsan Clinic. Based on our numerical results, the proposed framework achieves more accuracy than other approaches.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2022.05a
• /
• pp.627-630
• /
• 2022

Over the past few years, fake news has become one of the most significant problems. Since it is impossible to prevent people from spreading misinformation, people should analyze the news themselves. However, this process takes some time and effort, so the routine part of this analysis should be automated. There are many different approaches to this problem, but they only analyze the text and messages, ignoring the images. The fake news problem should be solved using a complex analysis tool to reach better performance. In this paper, we propose the approach of training an Artificial Intelligence using an unsupervised learning algorithm, combined with online data parsing tools, providing independence from subjective data set. Therefore it will be more difficult to spread fake news since people could quickly check if the news or article is trustworthy.

• Geomechanics and Engineering
• /
• v.32 no.4
• /
• pp.397-409
• /
• 2023

This study examines two traditional approaches (non-linear elastic and elasto-plastic) in association with 2D and 3D FEM analyses of a box-section pile embedded in sand. A particular emphasis is placed on stress singularities concerning both reentrant corners of the pile section and the resulting tension zones. From the experience gained in this study, non-linear elastic soil models are less restrictive when one considers stress singularities and their possible effects on convergence of the solution. At least for monotonic loading, when compared with field tests, non-linear elastic models yield better results than the plasticity ones. On the other hand, although elasto-plastic models are not limited to monotonic loading, they are much more sensitive to stress singularities. For this reason, a spherical elastic region is necessary at the pile tip to ensure convergence. Without this region, one must artificially impose an apparent cohesion to limit the tension stresses within a sand medium.

• Biomolecules & Therapeutics
• /
• v.32 no.2
• /
• pp.171-182
• /
• 2024

All cells are equipped with intricate signaling networks to meet the energy demands and respond to the nutrient availability in the body. AMP-activated protein kinase (AMPK) is among the most potent regulators of cellular energy balance. Under ATP -deprived conditions, AMPK phosphorylates substrates and affects various biological processes, such as lipid/glucose metabolism and protein synthesis. These actions further affect the cell growth, death, and functions, altering the cellular outcomes in energy-restricted environments. AMPK plays vital roles in maintaining good health. AMPK dysfunction is observed in various chronic diseases, making it a promising target for preventing and alleviating such diseases. Herein, we highlight the different AMPK functions, especially in allergy, aging, and cancer, to facilitate the development of new therapeutic approaches in the future.

• Nonlinear Functional Analysis and Applications
• /
• v.29 no.2
• /
• pp.307-332
• /
• 2024

Two inertial extragradient-type algorithms are introduced for solving convex pseudomonotone variational inequalities with fixed point problems, where the associated mapping for the fixed point is a 𝜌-demicontractive mapping. The algorithm employs variable step sizes that are updated at each iteration, based on certain previous iterates. One notable advantage of these algorithms is their ability to operate without prior knowledge of Lipschitz-type constants and without necessitating any line search procedures. The iterative sequence constructed demonstrates strong convergence to the common solution of the variational inequality and fixed point problem under standard assumptions. In-depth numerical applications are conducted to illustrate theoretical findings and to compare the proposed algorithms with existing approaches.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.40 no.5
• /
• pp.410-416
• /
• 2003

In the affine projection algorithm(APA), the projection dimension depends on a number of projection basis and of elements of input vector used for updating of coefficients of the adaptive filter. The projection dimension is closely related to a convergence speed of the APA, and it determines computational complexity. As the adaptive filter approaches to steady state, convergence speed is decreased. Therefore it is possible to reduce projection dimension that determines convergence speed. In this paper, we proposed the variable dimension affine projection algorithm (VDAPA) that controls the projection dimension and uses the relation between variations of coefficients of the adaptive filter and convergence speed of the APA. The proposed method reduces computational complexity of the APA by modifying the number of projection basis on convergence state. For demonstrating the good performances of the proposed method, simulation results are compared with the APA and normalized LMS algorithm in convergence speed and computational quantity.

• Journal of the Korea Convergence Society
• /
• v.10 no.10
• /
• pp.249-257
• /
• 2019

The purpose of this study was to find ways to promote residents' participation as an important key for urban regeneration and enhancing urban competition. The Korea Convergence Society. The case of the design and operation of the community academy as part of the urban regeneration project (the New Ddeul Village Project) in the S district of Busan Metropolitan City was analyzed based on the strength model. Based on the job, this study presented some empirical discussions for a successful community engagement program. First, move away from the pathological model based on regional problems and approach it as a positive and potential-oriented strength model. Second, it is co-prosperity by utilizing the resources of local universities around the village. Third, to meet the diverse needs of local residents, they should seek regional support and exchanges through the convergence of college majors.

• Journal of Digital Convergence
• /
• v.19 no.8
• /
• pp.327-335
• /
• 2021

The tendency of the current times require education to focus on convergence, and the same applies to the essence of ceramics and design base imagination and creativity. For effective integration, a wide range of experimentations is required both in terms of academic and methodic approaches. This study analyzes the [TeamTeam Class] curriculum, converging ceramics with design initiated in the second semester (autumn semester) of 2020. Through reference materials on ceramics and design convergence education, it was possible to classify the following 5 categories: Subject, Method, Management, space and communication. The aim of the study is to find resolutions to overcome existing issues and problems in search of more effective methods. Although this study is based on convergence education, [TeamTeam Class] at Kookmin University, I hope to extend it further to also consider education after COVID-19.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.05a
• /
• pp.357-359
• /
• 2023

This paper proposes a hybrid recommendation system (RS) model that overcomes the limitations of traditional approaches such as data sparsity, cold start, and scalability by combining collaborative filtering and context-aware techniques. The objective of this model is to enhance the accuracy of recommendations and provide personalized suggestions by leveraging the strengths of collaborative filtering and incorporating user context features to capture their preferences and behavior more effectively. The approach utilizes a novel method that combines contextual attributes with the original user-item rating matrix of CF-based algorithms. Furthermore, we integrate k-mean++ clustering to group users with similar preferences and finally recommend items that have highly rated by other users in the same cluster. The process of partitioning is the use of the rating matrix into clusters based on contextual information offers several advantages. First, it bypasses of the computations over the entire data, reducing runtime and improving scalability. Second, the partitioned clusters hold similar ratings, which can produce greater impacts on each other, leading to more accurate recommendations and providing flexibility in the clustering process. keywords: Context-aware Recommendation, Collaborative Filtering, Kmean++ Clustering.