• Journal of radiological science and technology
• /
• v.47 no.1
• /
• pp.29-37
• /
• 2024

This study aims to evaluate the performance of the U-Net based learning model that may vary depending on the histogram equalization algorithm. The subject of the experiment were 17 radiology students of this college, and 1,727 data sets in which the region of interest was set in the thyroid after acquiring ultrasound image data were used. The training set consisted of 1,383 images, the validation set consisted of 172 and the test data set consisted of 172. The equalization algorithm was divided into Histogram Equalization(HE) and Contrast Limited Adaptive Histogram Equalization(CLAHE), and according to the clip limit, it was divided into CLAHE8-1, CLAHE8-2. CLAHE8-3. Deep Learning was learned through size control, histogram equalization, Z-score normalization, and data augmentation. As a result of the experiment, the Attention U-Net showed the highest performance from CLAHE8-2 to 0.8355, and the U-Net and BSU-Net showed the highest performance from CLAHE8-3 to 0.8303 and 0.8277. In the case of mIoU, the Attention U-Net was 0.7175 in CLAHE8-2, the U-Net was 0.7098 and the BSU-Net was 0.7060 in CLAHE8-3. This study attempted to confirm the effects of U-Net, Attention U-Net, and BSU-Net models when histogram equalization is performed on ultrasound images. The increase in Clip Limit can be expected to increase the ROI match with the prediction mask by clarifying the boundaries, which affects the improvement of the contrast of the thyroid area in deep learning model learning, and consequently affects the performance improvement.

• Journal of Korea Multimedia Society
• /
• v.24 no.7
• /
• pp.849-859
• /
• 2021

Prostate cancer (PCa) is a fatal disease that occurs in men. In general, PCa cells are found in the prostate gland. Early diagnosis is the key to prevent the spreading of cancers to other parts of the body. In this case, deep learning-based systems can detect and distinguish histological patterns in microscopy images. The histological grades used for the analysis were benign, grade 3, grade 4, and grade 5. In this study, we attempt to use transfer learning and fine-tuning methods as well as different model architectures to develop and compare the models. We implemented MobileNet, ResNet50, and DenseNet121 models and used three different strategies of freezing layers techniques of fine-tuning, to get various pre-trained weights to improve accuracy. Finally, transfer learning using MobileNet with the half-layer frozen showed the best results among the nine models, and 90% accuracy was obtained on the test data set.

• Hwankyungkyoyuk
• /
• v.22 no.4
• /
• pp.95-110
• /
• 2009

The purpose of the study was to develop The Stream Environmental education u-learning contents for elementary school students. For the development of content, the researchers commissioned detailed examination to experts to confirm validity, did a literature review and hosted expert forums. In addition, to enhance accessibility, they asked fairytale writers to develop easier and more valid scenarios and narrations of u-learning contents for elementary school students. The development content is for 18 hours of education and has three sections: i) Preparation, ii) Exploration, and iii) Arrangement. Since the content has been developed based on SCROM, it is expected to have re-usability, accessibility, compatibility and durability. Based on evaluation criteria of u-learning contents suggested in the research methods, the research group commissioned evaluation to ten experts in environmental education of each school level. Recommendations for applying the content developed in this study and further research are as follows: First, the developed content should be actively promoted and provided both online and offline so that elementary school students can fully utilize them. To this end, the website of the Ministry of Environment and u-learning training centers of universities of education should be used. Since content requires interaction not only between learners of the content but also between learners and operators, additional administrative and financial support should be provided. Second, this study focuses on the development of u-learning contents for elementary school students. Further studies are needed to develop content for secondary school students.

• 제어로봇시스템학회:학술대회논문집
• /
• 1995.10a
• /
• pp.123-126
• /
• 1995

Characteristics of control system design using Universal Learning Network (U.L.N.) are that a system to be controlled and a controller are both constructed by U.L.N. and that the controller is best tuned through learning. U.L.N has the same generalization ability as N.N.. So the controller constructed by U.L.N. is able to control the system in a favorable way under the condition different from the condition of the control system in learning stage. But stability can not be realized sufficiently. In this paper, we propose a robust control method using U.L.N. and second order derivatives of U.L.N.. The proposed method can realize better performance and robustness than the commonly used Neural Network. Robust control considered here is defined as follows. Even though initial values of node outputs change from those in learning, the control system is able to reduce its influence to other node outputs and can control the system in a preferable way as in the case of no variation. In order to realize such robust control, a new term concerning the variation is added to a usual criterion function. And parameter variables are adjusted so as to minimize the above mentioned criterion function using the second order derivatives of criterion function with respect to the parameters. Finally it is shown that the controller constricted by the proposed method works in an effective way through a simulation study of a nonlinear crane system.

• 한국정보교육학회:학술대회논문집
• /
• 2006.08a
• /
• pp.101-107
• /
• 2006

컴퓨터와 통신기술이 발전함에 따라 최근의 교육환경은 학습자 스스로 학습내용, 학습 시간 및 학습 순서를 선택하고 조직하는 유비쿼터스 학습방향으로 나아가고 있다. 이러한 유비쿼터스 학습에서는 학습의 효과를 높이기 위하여 전문가의 도움이나 조언 등이 절실히 필요하다. 이를 위해 유비쿼터스 학습에 스케폴딩과 멘토링의 개념을 접목시키면 학습효과를 극대화 시킬 수 있다. 따라서 본 연구에서는 스캐폴딩과 e-Mentoring을 활용한 u-Learning 교수학습모형을 설계하고 이를 구현하여 학생들의 학업성취도와 학습태도에 미치는 영향을 알아보고 u-Learning 환경에서 다양한 교수학습 방법의 적용 가능성을 검증해보았다.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.10 no.1
• /
• pp.49-53
• /
• 2010

e-learning and m-learning have some problems that data transmission frequently discontinuously, communication cost increases, the computation speed of mass data drops, battery limitation in the mobile learning environments. In this paper, we propose the PULIMS for u-learning systems. The proposed system intellectualize the education environment using intelligent mobile agent, supports the customized education service, and helps that learners feasible access to the education information through mobile phone. We can see the fact that the efficience of proposed method is outperformed that of the conventional methods. The PULIMS is new technology that can be used to learn whenever and wherever learners want in Ubiquitous education environment.

• International Journal of Internet, Broadcasting and Communication
• /
• v.4 no.1
• /
• pp.8-12
• /
• 2012

The purpose of this study is to review the possible effect of the learning paradigm shift from traditional method to ubiquitous learning. What are the societal issues that need to be address in order to design a new pedagogical platform trending from e-learning to m-learning and now the u-learning? That without the proper study of how learning environment may affect the learning process of an individual will lead to poor quality of education. This new era of learning environment offer a big opportunity for "anytime, anywhere" learning. Thus, Lifelong learning is at hand of everyone. Maximizing the benefit of new trend will be a great help and addressing the limitations will lead to quality education.

• Asia-pacific Journal of Multimedia Services Convergent with Art, Humanities, and Sociology
• /
• v.8 no.3
• /
• pp.279-288
• /
• 2018

In this paper, we present the results of educational innovation case study using U-Learning Box and Ubiquitous-based Test(UBT) system for 6 sample primary schools in Nepal. As Nepal is considered to be a developing country with electricity problem to the school, the U-Learning Box, consisting of a small and easy-to-use tablet PC for teacher and a small smart beam with its own battery was evaluated as the optimum solution to support continuous basic English and hygiene education for these schools. And UBT technology using tablet PC was used to evaluate and analyze basic English learning ability of the students, which helped us realized that it is necessary to improve the educational environment and develop suitable educational contents. We hope that the global educational innovation using U-Learning Box and UBT technology will become a successful model for global equality of educational opportunity project for developing countries including Nepal.

• Korean Institute of Interior Design Journal
• /
• v.19 no.3
• /
• pp.214-223
• /
• 2010

Information & Communication Technology(ICT), called ubiquitous computing, changes our environment in many ways. It also acts on educational environment in concert with new pedagogy. Korean Government gives an efforts to set up the u-School as the educational environment supported by ICT. This study aims to find efficient uses of ICT to cope with the changing education environment and emerging needs of users. The survey research follows the pilot study on the model of future school. The students and teachers in six schools located in Busan answered the questionnaire and 752 individuals were analyzed by the statistical method including frequencies, crosstabs, ANOVA, and factor analysis. The conclusions are as follows: (1) Highly demanded were the u-School facilities to provide pleasant environment as the teachers and students have to spend most of time in classroom, or school premises. In addition, the infrastructures were so desperately required that students are able to control their textbooks and belongings, and efficiently utilize a variety of data. (2) The classroom volume should be planned a bit smaller than the current number of students so that teachers are relieved from the burden of too many student. The needs of students to interact with more students can be bolstered by utilization of ICT. (3) The respondents feel the high needs of computer for student, electronic blackboard, electronic lecture desk, microphone and computer for teacher as the teaching and learning facilities. This showed high needs for teaching and learning facilities enabling the active and individual learning through the use of ICT at education environment. In conclusion, the study proposes the fundamental data to combine u-School elements proper to the current learning and school environment, from the analysis on the use of school facilities, with the subjects, the users of education environment.

• Journal of the Korean Society of Radiology
• /
• v.18 no.2
• /
• pp.93-99
• /
• 2024

The digital medical imaging, especially, computed tomography (CT), should necessarily be considered in terms of noise distribution caused by converting to X-ray photon to digital imaging signal. Recently, the denoising technique based on deep learning architecture is increasingly used in the medical imaging field. Here, we evaluated noise reduction effect according to various noise types based on the U-net deep learning model in the lung CT images. The input data for deep learning was generated by applying Gaussian noise, Poisson noise, salt and pepper noise and speckle noise from the ground truth (GT) image. In particular, two types of Gaussian noise input data were applied with standard deviation values of 30 and 50. There are applied hyper-parameters, which were Adam as optimizer function, 100 as epochs, and 0.0001 as learning rate, respectively. To analyze the quantitative values, the mean square error (MSE), the peak signal to noise ratio (PSNR) and coefficient of variation (COV) were calculated. According to the results, it was confirmed that the U-net model was effective for noise reduction all of the set conditions in this study. Especially, it showed the best performance in Gaussian noise.