• Journal of Digital Convergence
• /
• v.12 no.2
• /
• pp.555-561
• /
• 2014

For five diagnostic X-ray generators (DR), four units turned out to be appropriate in tests on the reproducibility of radiation output suggested in the IEC 60601-2-54 standard, but in one unit of the X-ray equipment, an item measured in a combination of 50% of the highest tube voltage of the diagnostic X-ray equipment, the test setting of Group C with authorized output doses between $1{\mu}Gy$ and $5{\mu}Gy$ of mAs turned out to be inappropriate. As a result, the radiation dose to the IEC 60601-2-54 standard for quantification standards proposed by the radiation output from diagnostic X-ray imaging device reproducibility of performance management should be aware that an important evaluation factor.

• Progress in Medical Physics
• /
• v.27 no.1
• /
• pp.37-45
• /
• 2016

In proton therapy, in vivo proton beam range verification is very important to deliver conformal dose to the target volume and minimize unnecessary dose to normal tissue. For this purpose, a multi-slit prompt-gamma camera module made of 24 scintillation detectors and 24-channel signal processing system is under development. In the present study, we have developed and tested a dual-mode signal processing system, which can operate in the energy calibration mode and the fast data acquisition mode, to process the signals from the 24 scintillation detectors. As a result of performance test, using the energy calibration mode, we were able to perform energy calibration for the 24 scintillation detectors at the same time and determine the discrimination levels for the detector channels. Further, using the fast data acquisition mode, we were able to measure a prompt-gamma distribution induced by a 45 MeV proton beam. The measured prompt gamma distribution was found similar to the proton dose distribution at the distal fall-off region, and the estimated beam range was $17.13{\pm}0.76mm$, which is close to the proton beam range of 16.15 mm measured by an EBT film.

• Journal of Intelligence and Information Systems
• /
• v.9 no.1
• /
• pp.227-249
• /
• 2003

Prediction of corporate failure using past financial data is a well-documented topic. Early studies of bankruptcy prediction used statistical techniques such as multiple discriminant analysis, logit and probit. Recently, however, numerous studies have demonstrated that artificial intelligence such as neural networks can be an alternative methodology for classification problems to which traditional statistical methods have long been applied. In building neural network model, the selection of independent and dependent variables should be approached with great care and should be treated as model construction process. Irrespective of the efficiency of a teaming procedure in terms of convergence, generalization and stability, the ultimate performance of the estimator will depend on the relevance of the selected input variables and the quality of the data used. Approaches developed in statistical methods such as correlation analysis and stepwise selection method are often very useful. These methods, however, may not be the optimal ones for the development of neural network model. In this paper, we propose a genetic algorithms approach to find an optimal or near optimal input variables fur neural network modeling. The proposed approach is demonstrated by applications to bankruptcy prediction modeling. Our experimental results show that this approach increases overall classification accuracy rate significantly.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.29 no.7
• /
• pp.550-559
• /
• 2018

The effectiveness of deep neural networks (DNNs) for detection and classification of micro-Doppler signals generated by human walking and background noise sources is investigated. Previous research included a complex process for extracting meaningful features that directly affect classifier performance, and this feature extraction is based on experiences and statistical analysis. However, because a DNN gradually reconstructs and generates features through a process of passing layers in a network, the preprocess for feature extraction is not required. Therefore, binary classifiers and multiclass classifiers were designed and analyzed in which multilayer perceptrons (MLPs) and DNNs were applied, and the effectiveness of DNNs for recognizing micro-Doppler signals was demonstrated. Experimental results showed that, in the case of MLPs, the classification accuracies of the binary classifier and the multiclass classifier were 90.3% and 86.1%, respectively, for the test dataset. In the case of DNNs, the classification accuracies of the binary classifier and the multiclass classifier were 97.3% and 96.1%, respectively, for the test dataset.

• Nutrition Research and Practice
• /
• v.14 no.2
• /
• pp.102-108
• /
• 2020

BACKGROUND/OBJECTIVES: In Oriental medicine, certain foods may be beneficial or detrimental based on an individual's constitution; however, the scientific basis for this theory is insufficient. The purpose of this study was to investigate the effect of body constitution, based on the Sasang type of Korean traditional medical classification system, on the bioavailability of soy isoflavones of Cheonggukjang, a quick-fermented soybean paste. SUBJECTS/METHODS: A pilot study was conducted on 48 healthy Korean men to evaluate the bioavailability of isoflavone after ingestion of food based on constitution types classified by the Sasang typology. The participants were classified into the Taeeumin (TE; n = 15), Soyangin (SY; n = 15), and Soeumin (SE; n = 18) groups. Each participant ingested 50 g of Cheonggukjang per 60 kg body weight. Thereafter, blood was collected, and the soy isoflavone metabolites were analyzed by ultra-performance liquid chromatography/quadrupole time-of-flight mass spectrometry. Ntrikinetic analysis of individual isoflavone-derived metabolites was performed. RESULTS: Our nutrikinetic analysis identified 21 metabolites derived from isoflavones in the blood samples from 48 healthy Korean men (age range, 21-29 years). Significant differences were observed in the time to maximum concentration (T_max) and elimination half-life (t_1/2) for nine metabolites among the three groups. The T_max and t_1/2 of the nine metabolites were higher in the SE group than in the other groups. Moreover, the absorption rates, as determined by the area under the plasma-level curve (AUC) values of intact isoflavone, were 5.3 and 9.4 times higher in the TE group than in the SY and SE groups, respectively. Additionally, the highest AUC values for phase I and II metabolites were observed in the TE group. CONCLUSIONS: These findings indicate that isoflavone bioavailability, following Cheonggukjang insgestion, is high in individuals with the TE constitution, and relatively lower in those with the SE and SY constitutions.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.21 no.11
• /
• pp.180-186
• /
• 2020

ICT convergence using smart livestock is that in the first-generation dairy smart farm model, each device made by several manufacturers uses its own communication method, limiting the mutual operation of each device. This study uses a model based on open control technology to secure interoperability of existing ICT devices and to manage data efficiently. The open integrated control derived from this process is the software interface structure of Open API. It is an observer that serves as real-time data collection according to the communication method of ICT devices and sensors located at each end. It consists of a broker that connects and transmits to the upper integrated management server. As a result of the performance analysis through verification of two first-generation dairy smart farm model sites, the average daily milk production increased compared to the previous year (farm A 5.13%, farm B 1.33%, p<0.05). Cow days open (DO) was reduced by 17.5% on farm A and 13.3% for farm B(p<0.05). Cows require an adaptation period after the introduction of the ICT device, but if continuous effects are observed, the effect of production can be expected to increase gradually.

• Journal of radiological science and technology
• /
• v.41 no.5
• /
• pp.493-504
• /
• 2018

The diagnostic radiation equipment is managed in accordance with the "Rules for Safety Management of Diagnostic Radiation Equipment" enacted in 1995. The equipments should be inspected before use and every three years after use in accordance with the [Appendix 1] of the same rule. The inspection standard has been maintained without particular revision since enacted. But, over the past two decades new types of equipments have been manufactured and used. So, it is necessary to revise [Appendix 1] by making inspection items and inspection standards. In this study, we revised the classification system of equipments and reviewed international standards of IEC 60601 series, IEC 61223 series and AAPM TG 18 On-line Report No.03. And identified the problem of current inspection standards. Through this, we revised, deleted and added the inspection items and inspection standard of each equipment to meet the domestic circumstances. As a result of the study, we reorganized the classification system of equipment which are current classified as 5 classes into 22 classes as X-ray system etc. (7 classes), CT system etc. (5 classes) and Dental X-ray system etc. (10 classes). And then, we developed 70 inspection items for 6 types of equipments according to the reorganized classification system of equipments. The inspection items and inspection standards derived from this study have been proposed to the KCDC and will be applied to the revision of the Rule's [Appendix 1]. Therefore, we expect to be used as reference materials for domestic medical center, inspection institutions, and equipment manufacturing import companies.

• Textile Coloration and Finishing
• /
• v.28 no.1
• /
• pp.48-56
• /
• 2016

Improvement of heat sink technology related to the continuous implementation performance and extension of device-life in circumstance of easy heating and more compact space has been becoming more important issue as multi-functional integration and miniaturization trend of electronic gadgets and products has been generalized. In this study, it purposed to minimize of decline of the heat diffusivity by gluing polymer through compounding of inorganic particles which have thermal conductive properties. We used NH-9300 as base resin and used inorganic fillers such as silicon carbide(SiC), aluminum nitride(AlN), and boron nitride(BN) to improve heat diffusivity. After making film which was made from 100 part of acrylic resin mixed hardener(1.0 part more or less) with inorganic particles. The film was matured at $80^{\circ}C$ for 24h. Diffusivity were tested according to sorts of particles and density of particles as well as size and structure of particle to improve the effect of heat sink in view of morphology assessing diffusivity by LFA(Netzsch/LFA 447 Nano Flash) and adhesion strength by UTM(Universal Testing Machine). The correlation between diffusivity of pure inorganic particles and composite as well as the relation between density and morphology of inorganic particles has been studied. The study related morphology showed that globular type had superior diffusivity at low density of 25% but on the contarary globular type was inferior to non-globular type at high density of 80%.

• Journal of KIISE:Software and Applications
• /
• v.34 no.12
• /
• pp.1045-1055
• /
• 2007

Latent topic models are statistical models which automatically captures salient patterns or correlation among features underlying a data collection in a probabilistic way. They are gaining an increased popularity as an effective tool in the application of automatic semantic feature extraction from text corpus, multimedia data analysis including image data, and bioinformatics. Among the important issues for the effectiveness in the application of latent topic models to the massive data set is the efficient learning of the model. The paper proposes an accelerated learning technique for PLSA model, one of the popular latent topic models, by an incremental EM algorithm instead of conventional EM algorithm. The incremental EM algorithm can be characterized by the employment of a series of partial E-steps that are performed on the corresponding subsets of the entire data collection, unlike in the conventional EM algorithm where one batch E-step is done for the whole data set. By the replacement of a single batch E-M step with a series of partial E-steps and M-steps, the inference result for the previous data subset can be directly reflected to the next inference process, which can enhance the learning speed for the entire data set. The algorithm is advantageous also in that it is guaranteed to converge to a local maximum solution and can be easily implemented just with slight modification of the existing algorithm based on the conventional EM. We present the basic application of the incremental EM algorithm to the learning of PLSA and empirically evaluate the acceleration performance with several possible data partitioning methods for the practical application. The experimental results on a real-world news data set show that the proposed approach can accomplish a meaningful enhancement of the convergence rate in the learning of latent topic model. Additionally, we present an interesting result which supports a possible synergistic effect of the combination of incremental EM algorithm with parallel computing.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.19 no.5
• /
• pp.45-53
• /
• 2018

Among the various slope stabilization methods, the green soil method based on the growth of plants is advantageous to the environment, but the durability and slope stability are insufficient when the green soil method is applied to a steep slope and rock slope sites. Therefore, in this study, green soil, which improved the adhesion performance and the vegetation environment, was developed using cementitious materials and ECG, and the durability and slope stability as well as the possibility of its use as a rock vegetation base material were assessed. From the results, the adhesive force and internal friction angle were higher than that of the existing green soil so that it could be used for in situ construction. The soil hardness value was 26 mm, which was slightly higher than that of the best growth condition of the plant, 18~23 mm, and the drying shrinkage strain was approximately 3%; hence, it is not expected to affect the durability of green soil. The results of a rainfall intensity simulation for evaluating the slope adhesion force showed that slope failure did not occur under all conditions. The damage decreased with increasing slope angle. Therefore, the green soils developed in this study have excellent durability and slope stability and can be used for rock slope sites.