• Journal of Oral Medicine and Pain
• /
• v.14 no.1
• /
• pp.19-24
• /
• 1989

The purpose of this study was to investigate the reliability of electromyographic examination in a day and detect the objective measurement timing of electromyography. 18 normal students who were in dental collage of C national university were selected for this study. They had no symptoms on temporomandibular joint area, no restorations and missing tooth on dentition. Their molar relationship was normal or class I of Angle's classification. Electromyography had been taken every one hour from 9:00 A.M. to 5:00 P.M. in the state of clenching and rest by using Bioelectric processor EM2 interfaced with mandibular kinesiograph K-6 diagnostic system. We compared and analyzed the variances of electromyography values of each subjects. The obtained results were as follows : There were no significant differences on variability of electromyography values in each state clenching and rest. In the aspect of this circumstances, the use of electromyography could be a good objective procedure to diagnose the temporomandibular disorders and to detect the effect to the treatment of temporomandibular disorders at any time in a day from 9:00 A.M. to 5:00 P.M.

• Journal of Korean Institute of Industrial Engineers
• /
• v.28 no.2
• /
• pp.147-161
• /
• 2002

In novelty detection, one attempts to discriminate abnormal patterns from normal ones. Novelty detection is quite difficult since, unlike usual two class classification problems, only normal patterns are available for training. Auto-Associative Multi-Layer Perceptron (AAMLP) has been shown to provide a good performance based upon the property that novel patterns usually have larger auto-associative errors. In this paper, we give a mathematical analysis of 2-layer AAMLP's output characteristics and empirical results of 2-layer and 4-layer AAMLPs. Various activation functions such as linear, saturated linear and sigmoid are compared. The 2-layer AAMLPs cannot identify non-linear boundaries while the 4-layer ones can. When the data distribution is multi-modal, then an ensemble of AAMLPs, each of which is trained with pre-clustered data is required. This paper contributes to understanding of AAMLP networks and leads to practical recommendations regarding its use.

• The Transactions of the Korea Information Processing Society
• /
• v.7 no.6
• /
• pp.1834-1843
• /
• 2000

In this paper, we propose a modular neural network based classification method for handwritten numerals utilizing rotated and translated images of an input image. The whole numeral pattern space is divided into smaller spaces which overlap each other and form multiple clusters. On these multiple clusters, multiple multilayer perceptrons (MLP) neural networks, specialized in those clusters, are constructed. Thus, each MLP acts as an expert network on the corresponding cluster. An MLP is also used as a gating network functioning as a mediator among the multiple MLPs. In the learning phase, an input numeral image is dithered by tow geometric operations of translation and rotation so that new numeral images similar to original one are generated. In the recognition phase, we utilize not only input numeral image, but also nearly generated images through the rotation and the translation of the original image. Thus, multiple output values for those generated images were combined to make class decision by various combination methods. The experimental results confirm the validity of the proposed method.

• Proceedings of the KSRS Conference
• /
• v.2
• /
• pp.623-626
• /
• 2006

Group of genes controls the functioning of a cell by complex interactions. These interacting gene groups are called Gene Regulatory Networks (GRNs). Two previous data mining approaches, clustering and classification have been used to analyze gene expression data. While these mining tools are useful for determining membership of genes by homology, they don't identify the regulatory relationships among genes found in the same class of molecular actions. Furthermore, we need to understand the mechanism of how genes relate and how they regulate one another. In order to detect regulatory relationships among genes from time-series Microarray data, we propose a novel approach using frequent pattern mining and chain rule. In this approach, we propose a method for transforming gene expression data to make suitable for frequent pattern mining, and detect gene expression patterns applying FP-growth algorithm. And then, we construct gene regulatory network from frequent gene patterns using chain rule. Finally, we validated our proposed method by showing that our experimental results are consistent with published results.

• Journal of the korean academy of Pediatric Dentistry
• /
• v.23 no.3
• /
• pp.680-687
• /
• 1996

Surgical removal of impacted mesiodens can be performed easily when exact position of mesiodens is identified. This case report is argued about methodological approach of exact antero-posterior postioning of mesiodens using conventional cross-sectional occlusal film and periapical film. The author concludes, 1. Among various methods of positioning mesiodens, exact position of mesiodens can be determined with occlusal film and periapical film. 2. On operation, exact antero-posterior position of mesiodens can be determined with comparing occlusal images of adjacent teeth and anatomic structure to real ones. 3. It is important that exact removal course of mesiodens has to be determined in addition to exact determination of one's position, and that it has to be determined in regard to position, morphological basis, direction of impacted pattern of mesiodens and adjacent anatomic structure. 4. In 2 cases presented, both are mesiodens of inverted conical type, and impacted direction are class I and III respectively according to classification author suggested, and surgery can be perfomed with ease by different approach directions.

• Journal of Korean Society for Atmospheric Environment
• /
• v.6 no.1
• /
• pp.11-21
• /
• 1990

To quantitatively estimate the effect of yellow sand(loess) fromt he Northern China, various soil sources having similar chemical compositions to yellow sands should be separated and identified. After that, mass contribution for yellow sand can be calculated. The study showed that it was impossible to solve this problem by the traditional bulk analyses. However, particle-by-particle analysis by a CCSEM (computer controlled scanning electron microscope) gave enormous potentials to solve it. To perform this study, seven soil source data analyzed by CCSEM were obtained from Texas, U.S.A. Initially, each soil date was classified into two groups, coarse and fine particle groups since the particle number distribution showed a minimum occurring at 5.2$\mu$m of aerodynamic diameter. Particles in each group were then classified into one of the 283 homogeneous particle classes by the universal classification rule which had been built by an expert system in the early study. Further, mass fractions and their uncertainties for each class in each source were calculated by the Jackknife method, and then source profile matrix for the 7 soil sources was created. To use the profile matrix in the study of source contribution, it is necessary to test the degree of collinearity among sources. The profiles were tested by the singular value decomposition method. As a result, each soil source characterized by artificially created variables was totally independent each other and is ready to use in source contribution studies as a receptor model.

• Journal of Chest Surgery
• /
• v.28 no.12
• /
• pp.1122-1131
• /
• 1995

One hundred eighty-eight patients[August.23,1988,through July.30,1994 underwent aortic[AVR , mitral[MVR , or double [DVR valve replacement with the St.Jude Medical prosthesis. The author analyzed 100 patients with valvular heart disease,who underwent valve replacement with the St.Jude Medical prothesis from 1990 to 1994, at Hanyang University hospital Cardiovascular department. Information on volume and functional change of the heart chamber can be obtained by cardiac echocardiography and cartheterization. Out of 100 patients, 40 patients were male[40% and 60 patients were female [60% . Age ranged from 13 years to 68 years, with mean age of 42.6 years. Mean height was 160.3cm and mean body weight was 54.9Kg. According to NYHA functional classification, class III is most frequent and 60 patients could be classfied under it. MVR [involved Redo MVR was performed in 40 patients, AVR [involved Redo AVR was performed in 18 patients, and DVR [involved Redo DVR was performed in 42 patients. Warfarin [Coumadin anticoagulation was recommended for all patients. Life long warfarin anticoagulation was necessary to all patients who underwent valve replacement with St.Jude Medical prosthesis. Ideal prothrombin time was maintained about 30% during warfarinization. There were no case of mechanical failure. It followed a comparison of echocardiography before and after valve replacement at Hanyang University hospital [30 patients and a preoperative evaluation of cardiac catheterization and angiography [64 patients . The St.Jude Medical cardiac valve is a viable alternative in the surgical therapy of valvular heart disease.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.12
• /
• pp.5643-5656
• /
• 2016

The detection accuracy of steganalysis depends on many factors, including the embedding algorithm, the payload size, the steganalysis feature space and the properties of the cover source. In practice, the cover source mismatch (CSM) problem has been recognized as the single most important factor negatively affecting the performance. To address this problem, we propose a new framework for blind, universal steganalysis which uses traditional steganalyst features. Firstly, cover images with the same statistical properties are searched from a reference image database as aided samples. The test image and its aided samples form a whole test set. Then, by assuming that most of the aided samples are innocent, we conduct outlier detection on the test set to judge the test image as cover or stego. In this way, the framework has removed the need for training. Hence, it does not suffer from cover source mismatch. Because it performs anomaly detection rather than classification, this method is totally unsupervised. The results in our study show that this framework works superior than one-class support vector machine and the outlier detector without considering the image retrieval process.

• Journal of the Korean Wood Science and Technology
• /
• v.47 no.3
• /
• pp.324-335
• /
• 2019

Timber from plantation forests has inferior physical and mechanical properties compared to timber from natural forest because it is mostly from fast-growing tree species that are cut at a young age. Filling cell voids with methyl methacrylate (MMA) can improve the wood properties. The purpose of this study was to determine the physical and mechanical properties of MMA-impregnated wood from three fast-growing wood species, namely jabon (Anthocephalus cadamba (Roxb.) Miq.), mangium (Acacia mangium Willd) and pine (Pinus merkusii Jungh. & de Vriese). Wood samples were either immersed in MMA monomer or impregnated with it and then heated to induce the polymerization process. Jabon, which was the lowest density wood, had the highest polymer loading, followed by pine and mangium. The physical and mechanical properties of samples were affected by wood species and the presence of MMA, with higher-density wood having better properties than wood with a lower density. Physical and mechanical properties of MMA wood were enhanced compared to untreated wood. Furthermore, the impregnation process was better than immersion process resulting the physical and mechanical properties. Based on MOR values, the MMA woods were one strength class higher compared to untreated wood with regard to Strength Classification of Indonesian Wood.

• International Journal of Computer Science & Network Security
• /
• v.21 no.12spc
• /
• pp.526-538
• /
• 2021

Machine and deep learning-based models are emerging techniques that are being used to address prediction problems in biomedical data analysis. DNA sequence prediction is a critical problem that has attracted a great deal of attention in the biomedical domain. Machine and deep learning-based models have been shown to provide more accurate results when compared to conventional regression-based models. The prediction of the gene sequence that leads to cancerous diseases, such as prostate cancer, is crucial. Identifying the most important features in a gene sequence is a challenging task. Extracting the components of the gene sequence that can provide an insight into the types of mutation in the gene is of great importance as it will lead to effective drug design and the promotion of the new concept of personalised medicine. In this work, we extracted the exons in the prostate gene sequences that were used in the experiment. We built a Deep Neural Network (DNN) and Bi-directional Long-Short Term Memory (Bi-LSTM) model using a k-mer encoding for the DNA sequence and one-hot encoding for the class label. The models were evaluated using different classification metrics. Our experimental results show that DNN model prediction offers a training accuracy of 99 percent and validation accuracy of 96 percent. The bi-LSTM model also has a training accuracy of 95 percent and validation accuracy of 91 percent.