• Proceedings of the Korea Water Resources Association Conference
• /
• 2021.06a
• /
• pp.136-136
• /
• 2021

In this study, a deep convolutional neural network (DCNN) model is proposed for short-term precipitation forecasting using weather radar-based images. The DCNN model is a combination of convolutional neural networks, autoencoder neural networks, and U-net architecture. The weather radar-based image data used here are retrieved from competition for rainfall forecasting in Korea (AI Contest for Rainfall Prediction of Hydroelectric Dam Using Public Data), organized by Dacon under the sponsorship of the Korean Water Resources Association in October 2020. This data is collected from rainy events during the rainy season (April - October) from 2010 to 2017. These images have undergone a preprocessing step to convert from weather radar data to grayscale image data before they are exploited for the competition. Accordingly, each of these gray images covers a spatial dimension of 120×120 pixels and has a corresponding temporal resolution of 10 minutes. Here, each pixel corresponds to a grid of size 4km×4km. The DCNN model is designed in this study to provide 10-minute predictive images in advance. Then, precipitation information can be obtained from these forecast images through empirical conversion formulas. Model performance is assessed by comparing the Score index, which is defined based on the ratio of MAE (mean absolute error) to CSI (critical success index) values. The competition results have demonstrated the impressive performance of the DCNN model, where the Score value is 0.530 compared to the best value from the competition of 0.500, ranking 16^th out of 463 participating teams. This study's findings exhibit the potential of applying the DCNN model to short-term rainfall prediction using weather radar-based images. As a result, this model can be applied to other areas with different spatiotemporal resolutions.

• Journal of Cardiovascular Imaging
• /
• v.31 no.2
• /
• pp.98-104
• /
• 2023

BACKGROUND: We aimed to investigate left ventricular (LV) global longitudinal strain (GLS) in end-stage renal disease patients and its change after kidney transplantation (KT). METHODS: We retrospectively reviewed patients who underwent KT between 2007 and 2018 at two tertiary centers. We analyzed 488 patients (median age, 53 years; 58% male) who had obtained echocardiography both before and within 3 years after KT. Conventional echocardiography and LV GLS assessed by two-dimensional speckle-tracking echocardiography were comprehensively analyzed. Patients were classified into three groups according to the absolute value of pre-KT LV GLS (|LV GLS|). We compared longitudinal changes of cardiac structure and function according to pre-KT |LV GLS|. RESULTS: Correlation between pre-KT LV EF and |LV GLS| were statistically significant, but the constant was not high (r = 0.292, p < 0.001). |LV GLS| was widely distributed at corresponding LV EF, especially when the LV EF was > 50%. Patients with severely impaired pre-KT |LV GLS| had significantly larger LV dimension, LV mass index, left atrial volume index, and E/e' and lower LV EF, compared to mildly and moderately reduced pre-KT |LV GLS|. After KT, the LV EF, LV mass index, and |LV GLS| were significantly improved in three groups. Patients with severely impaired pre-KT |LV GLS| showed the most prominent improvement of LV EF and |LV GLS| after KT, compared to other groups. CONCLUSIONS: Improvements in LV structure and function after KT were observed in patients throughout the full spectrum of pre-KT |LV GLS|.

• The Journal of Korean Philosophical History
• /
• no.42
• /
• pp.259-294
• /
• 2014

The key concept of political thoughts of Confucianism is 'Tian(天)=the Heaven' and 'Ren(人)=the Human being'. In particular, the meaning of 'the Heaven(天)' is profoundly exclusive in Confucianism in comparison to the other concepts. It is because that 'the Heaven' is connected with 'Ri(理)' which explains the human being and the heaven metaphysically, and the discussions about those concepts have also been widening to the political field. And if the concept of 'the Heaven's nature(天命)' is excluded, it is difficult for us to discuss the essence of political thought of Confucianism. This paper argued how the politicalization of 'the Heaven' which was private dimension of monarch in the strict sense, changed to the public dimension of people in Japan in the early 18th century through some theories of the two famous philosopher, Ito-Jinsai(伊藤仁齋) and Ogyu-Sorai(荻生?徠) who belonged to the 'Kogakuha(古學派)'. The doctrines of Chu-tzu(朱子學) has hold absolute authority over the history of the oriental political thoughts. But the authority have gone through the stage of disorganization in a unique space called Eto(江戶) era of Japan. Therefore, the interpretations and concept establishments in a new dimension are naturally formed in that periods about 'Min(民)=People'. Based on discussions like this, the purpose of this paper is to study how those movements are emerged in what kinds of thinking and debating process. In part of Ito-Jinsai, this paper have reconsidered the concept of 'People' which Jinsai had thought and examined closely with critical mind about 'In(仁)=Mercy'. In case of Ogyu-Sorai, this paper have argued his political theory of 'the welfare of the people in the nation(安民天下)', in clarifying with the concept of 'the Roads of the Saint(先王의 道)' and the meaning about his declaration that "the Six Scripture(六經) is things(物)". The meaning of political view of this two philosopher is in the point to suggest some kinds of cure for problems of those days with the former days values. In the Jinsai's doctrine, that cure is 'the realization of the rule of right(王道)' based on new interpretation of 'Mersy'. And in case of Sorai, that is a form of independence of politics from morality based on 'the Roads of the Saint'.

• Proceedings of the Korean Vacuum Society Conference
• /
• 2016.02a
• /
• pp.93-93
• /
• 2016

Large scale integrated circuits (LSIs) has been improved by the shrinkage of the circuit dimensions. The smaller chip sizes and increase in circuit density require the miniaturization of the line-width and space between metal interconnections. Therefore, an extreme precise control of the critical dimension and pattern profile is necessary to fabricate next generation nano-electronics devices. The pattern profile control of plasma etching with an accuracy of sub-nanometer must be achieved. To realize the etching process which achieves the problem, understanding of the etching mechanism and precise control of the process based on the real-time monitoring of internal plasma parameters such as etching species density, surface temperature of substrate, etc. are very important. For instance, it is known that the etched profiles of organic low dielectric (low-k) films are sensitive to the substrate temperature and density ratio of H and N atoms in the H2/N2 plasma [1]. In this study, we introduced a feedback control of actual substrate temperature and radical density ratio monitored in real time. And then the dependence of etch rates and profiles of organic films have been evaluated based on the substrate temperatures. In this study, organic low-k films were etched by a dual frequency capacitively coupled plasma employing the mixture of H2/N2 gases. A 100-MHz power was supplied to an upper electrode for plasma generation. The Si substrate was electrostatically chucked to a lower electrode biased by supplying a 2-MHz power. To investigate the effects of H and N radical on the etching profile of organic low-k films, absolute H and N atom densities were measured by vacuum ultraviolet absorption spectroscopy [2]. Moreover, using the optical fiber-type low-coherence interferometer [3], substrate temperature has been measured in real time during etching process. From the measurement results, the temperature raised rapidly just after plasma ignition and was gradually saturated. The temporal change of substrate temperature is a crucial issue to control of surface reactions of reactive species. Therefore, by the intervals of on-off of the plasma discharge, the substrate temperature was maintained within ${\pm}1.5^{\circ}C$ from the set value. As a result, the temperatures were kept within $3^{\circ}C$ during the etching process. Then, we etched organic films with line-and-space pattern using this system. The cross-sections of the organic films etched for 50 s with the substrate temperatures at $20^{\circ}C$ and $100^{\circ}C$ were observed by SEM. From the results, they were different in the sidewall profile. It suggests that the reactions on the sidewalls changed according to the substrate temperature. The precise substrate temperature control method with real-time temperature monitoring and intermittent plasma generation was suggested to contribute on realization of fine pattern etching.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.10 no.3
• /
• pp.249-255
• /
• 2017

The finger vein recognition technology is a method to acquire a finger vein image by illuminating infrared light to the finger and to authenticate a person through processes such as feature extraction and matching. In order to recognize a finger vein, a 2D mask-based two-dimensional convolution method can be used to detect a finger edge but it takes too much computation time when it is applied to a low cost micro-processor or micro-controller. To solve this problem and improve the recognition rate, this study proposed an extraction method for the region of interest based on virtual core points and moving average filtering based on the threshold and absolute value of difference between pixels without using 2D convolution and 2D masks. To evaluate the performance of the proposed method, 600 finger vein images were used to compare the edge extraction speed and accuracy of ROI extraction between the proposed method and existing methods. The comparison result showed that a processing speed of the proposed method was at least twice faster than those of the existing methods and the accuracy of ROI extraction was 6% higher than those of the existing methods. From the results, the proposed method is expected to have high processing speed and high recognition rate when it is applied to inexpensive microprocessors.

• The Journal of the Korea Contents Association
• /
• v.10 no.11
• /
• pp.318-328
• /
• 2010

This paper as basic work on the regional economic policy of Korean government considers economic conditions and fundamentals of the cold started 5+2 mega-regional economic zones and makes a comparative study of regional economic power of national mega-regional economic zones based on the established economic indicator. Therefore, this study aims to consider henceforward policy direction of the mega-regional economic zone and the validity of the regional economic policy, which promote autonomous mega-regional economic zone in dimension to secure the global competitiveness. According to the comparison result of regional economic power index, the capital area had absolute superiority, while Gangwon, Honam and Jeju areas were relatively weak. Since the given regional conditions are dissimilar, the government must consider regional characteristics and economic fundamentals carefully, as they push henceforth regional development policy for the mega-regional economic zone. What is more, the government should promote a balance of the regional development, through to maintain demand-based policy and demand-pull policy flexibly, which are based on demand analysis, as well as unequal quota budget and policy for regions, which are relatively falling behind in terms of development and income.

• Journal of the Korea Society of Computer and Information
• /
• v.16 no.12
• /
• pp.285-296
• /
• 2011

UN, OECD, ITU and other international organizations regularly announce ISI (Information Society Index) to utilize in establishing and evaluating information policies. ISI is utilized as important data for countries to evaluate their information policy performance and select future projects. As the advancement of information systems, the importance of information security has been emerged. Accordingly, NISI (National Information Security Index) has been required. NISI number is the most clearly figure to express the characteristics of a particular group's information security. It can be utilized in determining information security policies. Currently, questionnaire method has been used to calculate NISI number. But there is an absolute lack of statistical data, and the reliability of surveyed statistical data is problematic. The objective of this paper is to show how to collect precise micro data of each company's information security index numbers, and to develop an OLAP database system which calculating NISI numbers by using those micro data. In this process of the survey, we presented the technique to collect the data more systematically, and to analyze the data without using questionnaire method. OLAP architecture performs only well on the facts that are summarizable along each dimension, where all hierarchy schemas are distributive. Therefore we transformed the non-distributive hierarchy schema into the distributive hierarchy schema to implement OLAP database system. It is thought that this approach will be useful one from an implementation and schema design point of view.

• Progress in Medical Physics
• /
• v.14 no.1
• /
• pp.15-19
• /
• 2003

High dose rate (HDR) brachytherapy for treating a cervix carcinoma has become popular, because it eliminates many of the problems associated with conventional brachytherapy. In order to improve the clinical effectiveness with HDR brachytherapy, a dose calculation algorithm, optimization procedures, and image registrations need to be verified by comparing the dose distributions from a planning computer and those from a phantom. In this study, the phantom was fabricated in order to verify the absolute doses and the relative dose distributions. The measured doses from the phantom were then compared with the treatment planning system for the dose verification. The phantom needs to be designed such that the dose distributions can be quantitatively evaluated by utilizing the dosimeters with a high spatial resolution. Therefore, the small size of the thermoluminescent dosimeter (TLD) chips with a dimension of <1/8"and film dosimetry with a spatial resolution of <1mm used to measure the radiation dosages in the phantom. The phantom called a pelvic phantom was made from water and the tissue-equivalent acrylic plates. In order to firmly hold the HDR applicators in the water phantom, the applicators were inserted into the grooves of the applicator holder. The dose distributions around the applicators, such as Point A and B, were measured by placing a series of TLD chips (TLD-to-TLD distance: 5mm) in the three TLD holders, and placing three verification films in the orthogonal planes. This study used a Nucletron Plato treatment planning system and a Microselectron Ir-192 source unit. The results showed good agreement between the treatment plan and measurement. The comparisons of the absolute dose showed agreement within $\pm$4.0 % of the dose at point A and B, and the bladder and rectum point. In addition, the relative dose distributions by film dosimetry and those calculated by the planning computer show good agreement. This pelvic phantom could be a useful to verify the dose calculation algorithm and the accuracy of the image localization algorithm in the high dose rate (HDR) planning computer. The dose verification with film dosimetry and TLD as quality assurance (QA) tools are currently being undertaken in the Catholic University, Seoul, Korea.

• Korean Journal of Applied Biomechanics
• /
• v.15 no.1
• /
• pp.29-44
• /
• 2005

The purpose of this study is to reveal the effects of two different kicks, the drop kick and the punt kick, into the kicking motion, through the kinetic comparative analysis of the kicking motion, which is conducted when one kicks a soccer goal. To grasp kinetic changing factors, which is performed by individual's each body segment, I connected kicking motions, which were analyzed by a two dimension co-ordination, into the personal computer to concrete the digits of it and smoothed by 10Hz. Using the smoothed data, I found a needed kinematical data by inputting an analytical program into the computer. The result of comparative analysis of two kicking motions can be summarized as below. 1. There was not a big difference between the time of the loading phase and the time of the swing phase, which can affect the exact impact and the angle of balls aviation direction. 2. The two kicks were not affected the timing and the velocity of the kicking leg's segment. 3. In the goal kick motion, the maximum velocity timing of the kicking leg's lower segment showed the following orders: the thigh(-0.06sec), the lower leg(-0.05sec), the foot(-0.018sec) in the drop kick, and the thigh(-0.06sec), the lower leg(-0.05sec), the foot(-0.015sec) in the punt kick. It showed that whipping motion increases the velocity of the foot at the time of impact. 4. At the time of impact, there was not a significant difference in the supporting leg's knee and ankle. When one does the punt kick, the subject spreads out his hip joint more at the time of impact. 5. When the impact performed, kicking leg's every segment was similar. Because the height of the ball is higher in the punt kick than in the drop kick, the subject has to stretch the knees more when he kicks a ball, so there is a significant affect on the angle and the distance of the ball's flying. 6. When one performs the drop kick, the stride is 0.02m shorter than the punt kick, and the ratio of height of the drop kick is 0.05 smaller than the punt kick. This difference greatly affects the center of the ball, the supporting leg's location, and the location of the center of gravity with the center of the ball at the time of impact. 7. Right before the moment of the impact, the center of gravity was located from the center of the ball, the height of the drop kick was 0.67m ratio of height was 0.37, and the height of the punt kick was 0.65m ratio of height was 0.36. The drop kick was located more to the back 0.21m ratio of height was 0.12, the punt kick was located more to the back 0.28m ratio of height was 0.16. 8. There was not a significant difference in the absolute angle of incidence and the maximum distance, but the absolute velocity of incidence showed a significant difference. This difference is caused from that whether players have the time to perform of not; the drop kick is used when the players have time to perform, and punt kick is used when the players launch a shifting attack. 9. The surface reaction force of the supporting leg had some relation with the approaching angle. Vertical reaction force (Fz) showed some differences in the two movements(p<0.05). The maximum force of the right and left surface reaction force (Fx) didn't have much differences (p<0.05), but it showed the tendency that the maximum force occurs before the peak force of the front and back surface (Fy) occurs.

• Journal of Chest Surgery
• /
• v.29 no.12
• /
• pp.1306-1315
• /
• 1996

This study reviewed the changes of hemodynamlcs and centrAl pulmonary artery dimension in 54 patients who underwent bidirectional cavopulmonary shunt(BCPS) between February 1992 and December 1995 at Seoul National University Childrell's Hospital. Ag and body weight of patients averaged 36.8 $\pm$ 37.7 months and 8.0$\pm$3.0 kg, respectively Eightynine percent of patients had more than 2 violations of the risk factors for Fontan operation, resulting overall hospital mortality of 16.6%(9154). Serial hemodynamic and anglographic examinations before and mean 16.3 $\pm$ 14.3 months iirter BCPS were compared. The arterial oxygen saturation improved from a preoperative value of 71 9: 10.1 % to 79.H $\pm$ 8. 5% (n:4), p<0.05). The values of arterial oxygen saturation were lower as the age of the patients with BCPS in place was older(n=22, R'=0.341, p=0.004). A mean pulmonary artery pressure and pulmonary vascular resistance reduced from 31 $\pm$17 to 1).5$\pm$3.SmmHg(n=22, p<0.05) and from ).2$\pm$2.1 to 2.3$\pm$2.7 unit (n=7. p>0.05), respectively. Follow-up study showed a significant Increase of absolute values of ipsilateral pulmoanry artery (n: 14, p<0.05), but no change of contralateral pulmonary artreries (n: 14. p=not significant(HSI). However, there w re significant decreases in diameters of both ipsilateral and contralateral pulmonary arteries standardized by patients' body surface areas(16.8% decrease, n: 14, p< 0. 05 for ipsilatreal, 25.1%, n=14, p<0.05 for contralateral). Pulmonary artery indices for cross sectional areas of both pulmonary arterises decreased 9.3 $\pm$ 13.8% with showing a trend of more decrease as the follow-up duration was longer, We conclude that the bidirectional cavopulmonary shunt provide an excellent.