• Journal of the Korean Society of Radiology
• /
• v.11 no.3
• /
• pp.153-159
• /
• 2017

Chronic kidney disease can be treated if it is detected early, but as the disease progresses, it becomes impossible to recover. Finally, renal replacement therapy such as transplantation or dialysis should be used. Ultrasonography is used to diagnose kidney cancer, inflammatory disease, nodular disease, and chronic kidney disease. It is used to identify information about degree of inflammation using information such as kidney size, internal echo characteristics. Currently, the degree of disease in the clinic uses the value of glomerular filtration rate. However, even in ultrasound, changes in the degree of inflammation and disease can be observed. In this study, we used ultrasound images to quantify the changes in brightness, size, cortex, and subclinical changes of the kidney with progression of the disease, and compared them with the glomerular filtration rate used in clinical practice. In 105 cases, we performed 35 cases of normal kidney, 35 cases of early kidney disease, and 35 cases of terminal kidney. The brightness of the cortex of the image was obtained and the difference in brightness between the cortex and the proximal portion was obtained by the slope. The graph of the portion which was not smooth due to the ultrasonic characteristics was used as the function regrass. The size reduction was obtained from the original data. The results were as follows: It was proportional to the glomerular filtration rate. It is considered that the algorithm can be applied to the disease if the algorithm study continues.

• Journal of KIISE:Computing Practices and Letters
• /
• v.8 no.3
• /
• pp.344-355
• /
• 2002

Having a variety of good characteristics against other pattern recognition techniques, Multilayer Perceptron (MLP) has been used in wide applications. But, it is known that Error Backpropagation (EBP) algorithm which MLP uses in learning has a defect that requires relatively long leaning time. Because learning data in pattern recognition contain abundant redundancies, in order to increase learning speed it is very effective to use online-based teaming methods, which update parameters of MLP pattern by pattern. Typical online EBP algorithm applies fixed learning rate for each update of parameters. Though a large amount of speedup with online EBP can be obtained by choosing an appropriate fixed rate, fixing the rate leads to the problem that the algorithm cannot respond effectively to different leaning phases as the phases change and the learning pattern areas vary. To solve this problem, this paper defines learning as three phases and proposes a Instant Learning by Varying Rate and Skipping (ILVRS) method to reflect only necessary patterns when learning phases change. The basic concept of ILVRS is as follows. To discriminate and use necessary patterns which change as learning proceeds, (1) ILVRS uses a variable learning rate which is an error calculated from each pattern and is suppressed within a proper range, and (2) ILVRS bypasses unnecessary patterns in loaming phases. In this paper, an experimentation is conducted for speaker verification as an application of pattern recognition, and the results are presented to verify the performance of ILVRS.

• Journal of KIISE
• /
• v.42 no.5
• /
• pp.609-620
• /
• 2015

Viewers of videos are likely to absorb more information from the part of the screen that attracts visual attention. This fact has led to the visual attention models that are being used in producing and evaluating videos. In this paper, we investigate the factors that are significant to visual attention and the mathematical form of the visual attention model. We then estimated the visual attention probability using the statistical design of experiments. The analysis of variance (ANOVA) verifies that the motion velocity, distance from the screen, and amount of defocus blur affect human visual attention significantly. Using the response surface modeling (RSM), we created a visual attention score model that concerns the three factors, from which we calculate the visual attention probabilities (VAPs) of image pixels. The VAPs are directly applied to existing gradient based 3D effect perception measurement. By giving weights according to our VAPs, our algorithm achieves more accurate measurement than the existing method. The performance of the proposed measurement is assessed by comparing them with subjective evaluation as well as with existing methods. The comparison verifies that the proposed measurement outperforms the existing ones.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.40 no.4
• /
• pp.268-281
• /
• 2004

A study was carried out in order to estimate the deformation of the pound net according to the current by the model test in the circulating water channel. The tension of the frame rope and the variation of net shape were measured to investigate the deforming of the model pound net in the flow. The results are obtained as follows; 1. The experimental equation between tensions (R) of the frame rope and velocity (ν)was found to be R=$19.58v^{1.98}$($r^2$=0.98) in case of the upperward flow with fish court net and R=$26.90v^{1.72}$($r^2$=0.95)at the upperward flow with bag net according to the velocity from 0.0m/s to 0.6m/s, respectively. 2. As the variation of flow speed inside of the model net was gradually decreased according as which is passed through netting panels, in case of the upperward flow with fish court net, the flow speed was about 70% of initial flow speed at 0.1m/s, 60% at 0.2m/s, 50% at 0.3m/s and 40% 0.4~0.6m/s at the measurement point(h) inside of the first bag net, respectively. In case of the upperward flow with bag net, as the flow speed was steeply decreased according as which if passed through the second bag net, it was 30~60% of the initial flow speed and was 20~30% inside of the first bag net and was about 10~20% inside of the inclined passage net. 3. In case of the upperward flow with fish court net, the variation of deformed angle of fish court net was from 0$^{\circ}$ to 70$^{\circ}$and that of inclined passage net was from 0$^{\circ}$ to 63$^{\circ}$and that of the second bag net was from 0$^{\circ}$ to 47$^{\circ}$ . 4. In case of the upperward flow with fish court net, the variation of deformed angle of the second bag net was changed from 0$^{\circ}$ to 70$^{\circ}$and that of the inclined passage net was from 0$^{\circ}$ to 55$^{\circ}$ and that of the fish court net was from 0$^{\circ}$ to 50$^{\circ}$. The depth ratio of the first bag net was changed from 0% to 35% and that of the second bag net was from 0% to 20% and that of the inclined passage net was from 0% to 35%. In the flow speed 0.5m/s, the inclined passage net was raised up to the entry of the bag net and then prevented it more over 90%. 5. To be increased the opening volume of pound net, it needs to attach the added weight outside of the fish court net, inclined passage net and bag net. At the same time, it needs to adjust the tension of the twine for maintenance of the shape.

• Journal of Korean Society of Forest Science
• /
• v.106 no.3
• /
• pp.340-352
• /
• 2017

In This study, we examined the gap between supply and demand of forest policy by year through big data analysis for macroscopic evaluation of the 5th Basic Forest Plan. We collected unstructured data based on keywords related to the projects mentioned in the news, SNS and so on in the relevant year for the policy demand side; and based on the documents published by the Korea Forest Service for the policy supply side. based on the collected data, we specified the network structure through the social network analysis technique, and identified the gap between supply and demand of the Korea Forest Service's policies by comparing the network of the demand side and that of the supply side. The results of big data analysis indicated that the network of the supply side is less radial than that of the demand side, implying that various keywords other than forest could considerably influence on the network. Also we compared the trends of supply and demand for 33 keywords related to 27 major projects. The results showed that 7 keywords shows increasing demand but decreasing supply: sustainable, forest management, forest biota, forest protection, forest disease and pest, urban forest, and North Korea. Since the supply-demand gap is confirmed for the 7 keywords, it is necessary to strengthen the forest policy regarding the 7 keywords in the 6th Basic Plan.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2006.05a
• /
• pp.180-188
• /
• 2006

수리 전도도는 hydraulic gradient에 대한 flux의 비율 또는, flux-hydraulic gradient 직선의 기울기를 나타내며, 포화된 토양에서의 물의 이동이 포화수리 전도도이고 불포화된 토양에서의 이동이 불포화수리전도도이다. 일반적인 밭 상태에서의 토양수분 조건은 불포화수리 전도도로 표시하는 것이 적절하나 그 상태를 표현하기가 쉽지 않다. 토양의 불포화 상태를 나타내는데 가장 많이 쓰이고 있는 VGM(van Genuchten Mualem) 모형은 토양수분 포텐셜과 수분함량의 함수로 구성된 모형이며 몇 가지 매개변수가 필요하다. VGM 모형의 매개변수를 얻기 위해 본 연구에서는 VGM 모형의 매개변수를 계산해주는 프로그램인 Rosetta를 사용하였다. Rosetta 모형은 신경그물 얼개(neural network)를 이용하여 토양의 물리적 자료들인 토성이나 모래, 미사, 점토 함량 또는 용적밀도나 33kPa, 1500kPa에서의 토양수분 함량 자료를 가지고 VGM의 매개변수인 Ko(effective saturated hydraulic conductivity), ${\theta}r$(residual soil water content), ${\theta}s$(saturated soil water content), L, n, m(=1-1/n)을 예측하는 모형으로 미국 농무성(USDA-ARS)에서 개발한 프로그램이다. Rosetta를 이용하여 10kPa에서의 불포화수리 전도도를 예측하였다. 또한 Gardner(1958)와 Wooding (1968)의 모형을 기반으로 하여 만들어진 tension infiltrometer의 포화수리 전도도 값을 Gardner 식에 적용하여 1, 3, 5, 7kPa에서의 불포화수리 전도도 값을 17개 토양통을 대상으로 하여 구했다. 토양수분 potential이 3kPa에서는 물의 이동이 거의 없는 토양들이 있었는데 반해 남계통을 비롯한 학곡통, 회곡통, 백산통, 상주통, 석천통, 예산통 등 7개의 토양은 3kPa에서도 약간의 물의 이동이 있었다. 이는 모암이 화강 편마암인 관계로 토양 내에 물의 이동에 영향을 미치는 자갈의 함량이 높았기 때문일 것으로 생각되고 추후의 연구에서는 이 부분에 대한 내용도 검토되어야 할 것이다. 또한, 1kPa에서 물의 이동은 삼각통에서 35.21 cm/day로 이동 속도가 가장 컸으며 그 뒤로 예산통, 화봉통, 학곡통, 백산통 등이 토양에서 빠른 속도로 이동하였다. 가천통이나 석천통 및 우곡통은 1kPa에서의 이동 속도가 아주 느린 토양으로 판단되었다. 또한, 포화되지 않은 상태인 1kPa에서 물의 이동 속도를 VGM 모형에 의해 예측된 값과 측정된 값으로 비교하였을 때 불포화 수리 전도도가 예측되지 않은 토양(석천통, 지곡통, 풍천통)이 존재하여 불포화 수리 전도도 특성평가에 대한 VGM 모형의 적용성에 문제를 보였다. 이는 결과적으로 논이라는 영농형태가 존재하는 우리나라에서 토양의 수리적 특성해석을 위한 VGM 모형의 적용성에 한계가 있을 것으로 판단되었다.

• KDI Journal of Economic Policy
• /
• v.34 no.4
• /
• pp.117-156
• /
• 2012

This paper decomposes and estimates trend/cyclical components of some key macro variables-GDP, inflation, and interest rate, using a simple DSGE model along with flexible trend specification. The extracted cyclical components of output and interest rate are similar to HP-filtered counterparts, despite some differences in persistence and volatility, while inflation resembles that from BK filtering. This implies that the usual practice of applying a single filtering method to the data of interest may be problematic. When the baseline model is extended to incorporate consumption habit and price indexation, habit turns out to be important in explaining the persistence of business cycles. Comparison of several alternative models shows that the usual practice of estimation of DSGE model using filtered data leads to biased results. Finally, various sensitivity analyses illustrate that (1) allowing for correlation between structural cyclical shocks and trend shocks and (2) including irregular components (in inflation rate) may deliver interesting/important implication for gap estimates.

• The KIPS Transactions:PartB
• /
• v.11B no.3
• /
• pp.281-288
• /
• 2004

In order to extract and analyze complex features of the behavior of animals in response to external stimuli such as toxic chemicals, we implemented an adaptive computational method to characterize changes in the behavior of chironomids in response to treatment with the insecticide, diazinon. In this paper, we propose an energy minimization model to extract the features of response behavior of chironomids under toxic treatment, which is applied on the image of velocity vectors. It is based on the improved active contour model and the variations of the energy functional, which are produced by the evolving active contour. The movement tracks of individual chironomid larvae were continuously measured in 0.25 second intervals during the survey period of 4 days before and after the treatment. Velocity on each sample track at 0.25 second intervals was collected in 15-20 minute periods and was subsequently checked to effectively reveal behavioral states of the specimens tested. Active contour was formed around each collection of velocities to gradually evolve to find the optimal boundaries of velocity collections through processes of energy minimization. The active contour which is improved by T. Chan and L. Vese is used in this paper. The energy minimization model effectively revealed characteristic patterns of behavior for the treatment versus no treatment, and identified changes in behavioral states .is the time progressed.

• The Korean Journal of Nuclear Medicine Technology
• /
• v.16 no.2
• /
• pp.115-119
• /
• 2012

Purpose : The initial Breast-Specific Gamma Imaging (BSGI) protocol included bilateral breast imaging with 2 views of each breast-craniocaudal (CC) and mediolateral oblique (MLO). Furthermore, Axillary lymph nodes view can be acquired easily. The most meaningful prognosis factor for prediction of breast cancer is whether or not the breast cancer has metastasized to the lymph nodes. However, axillary view doesn't conduct in clinical. This article collates a diverse data of BSGI and describes technical details to acquire optimal imaging. Materials and Methods : A retrospective review was performed on 343 patients who had undergone BSGI between May 2011 and March 2012. Patients who had undergone BSGI received intravenous injection of 740 MBq (20 mCi) $^{99m}Tc$-sestamibi. Results : The following contents are the technical details for optimal axillary imaging. $^{99m}Tc$-sestamibi should be administered using an indwelling venous catheter or scalp needle followed by 10 cc of saline to flush to reduce extravasation and vascular trapping. After administration, patients raise their arm over their head and exercise with stress ball for 1 full minute. A lead shield attached to the gamma camera is removed and patients axilla is placed as close as possible to the camera at a $90^{\circ}$ angle. A lead apron is placed across the shoulder to reduce background from other organs. Acquisition time is enough for 120 sec~180 sec. Conclusion : If patients undergo bilateral axillary imaging as a standard with CC, MLO views, it could improve cancer treatment. Result of this study could maximize efficiency axillary imaging of breast cancer patients.

• Journal of the Korean Society of Radiology
• /
• v.5 no.6
• /
• pp.369-375
• /
• 2011

In dental radiography, panoramic views cause distortion and thus may bring about inaccurate results in the process of quantitative analysis. In this connection, there has recently been an increasing use of cone beam computed tomography (CBCT) that is known to provide high-resolution images and positional information. In this study, a dental computed tomography unit, 'DCT-90-P IMPLAGRAPHY (Vatech, Korea)', was applied to 20 patients for 24 seconds respectively, with a tube voltage of 85kVp and a tube current of 7mA. The data of CBCT were three-dimensionally reconstructed by use of a computer program, and were histomorphometrically analyzed. The results showed that the diameter of mandibular canal is less distorted at a certain inclination of the mandibular body. The image tends to seem more distended in proportion to the distance between the subject and film. Also, the image tends to be affected according as it is out of focus. In conclusion, it requires that the image should be reconstructed in light of anatomic position and structure.