• Investigative Magnetic Resonance Imaging
• /
• v.25 no.4
• /
• pp.313-322
• /
• 2021

Purpose: To find diagnostic image features, to compare diagnostic performance of multiphase CT versus gadoxetic acid disodium-enhanced MRI (GAD-MRI), and to evaluate the impact of analyzing Liver Imaging Reporting and Data System (LI-RADS) imaging features, for distinguishing combined hepatocellular-cholangiocarcinoma (CHC) from hepatocellular carcinoma (HCC). Materials and Methods: Ninety-six patients with pathologically proven CHC (n = 48) or HCC (n = 48), diagnosed June 2008 to May 2018 were retrospectively analyzed in random order by three radiologists with different experience levels. In the first analysis, the readers independently determined the probability of CHC based on their own knowledge and experiences. In the second analysis, they evaluated imaging features defined in LI-RADS 2018. Area under the curve (AUC) values for CHC diagnosis were compared between CT and MRI, and between the first and second analyses. Interobserver agreement was assessed using Cohen's weighted κ values. Results: Targetoid LR-M image features showed better specificities and positive predictive values (PPV) than the others. Among them, rim arterial phase hyperenhancement had the highest specificity and PPV. Average sensitivity, specificity, and AUC values were higher for MRI than for CT in both the first (P = 0.008, 0.005, 0.002, respectively) and second (P = 0.017, 0.026, 0.036) analyses. Interobserver agreements were higher for MRI in both analyses (κ = 0.307 for CT, κ = 0.332 for MRI in the first analysis; κ = 0.467 for CT, κ = 0.531 for MRI in the second analysis), with greater agreement in the second analysis for both CT (P = 0.001) and MRI (P < 0.001). Conclusion: Rim arterial phase hyperenhancement on GAD-MRI can be a good indicator suggesting CHC more than HCC. GAD-MRI may provide greater accuracy than CT for distinguishing CHC from HCC. Interobserver agreement can be improved for both CT and MRI by analyzing LI-RADS imaging features.

• Journal of Intelligence and Information Systems
• /
• v.29 no.3
• /
• pp.419-437
• /
• 2023

As the number of thin filers in Korea surpasses 12 million, there is a growing interest in enhancing the accuracy of assessing their credit default risk to generate additional revenue. Specifically, researchers are actively pursuing the development of default prediction models using machine learning and deep learning algorithms, in contrast to traditional statistical default prediction methods, which struggle to capture nonlinearity. Among these efforts, Graph Neural Network (GNN) architecture is noteworthy for predicting default in situations with limited data on thin filers. This is due to their ability to incorporate network information between borrowers alongside conventional credit-related data. However, prior research employing graph neural networks has faced limitations in effectively handling diverse categorical variables present in credit information. In this study, we introduce the Transformer embedded Graph Convolutional Network (TeGCN), which aims to address these limitations and enable effective default prediction for thin filers. TeGCN combines the TabTransformer, capable of extracting contextual information from categorical variables, with the Graph Convolutional Network, which captures network information between borrowers. Our TeGCN model surpasses the baseline model's performance across both the general borrower dataset and the thin filer dataset. Specially, our model performs outstanding results in thin filer default prediction. This study achieves high default prediction accuracy by a model structure tailored to characteristics of credit information containing numerous categorical variables, especially in the context of thin filers with limited data. Our study can contribute to resolving the financial exclusion issues faced by thin filers and facilitate additional revenue within the financial industry.

• Atmosphere
• /
• v.34 no.1
• /
• pp.55-67
• /
• 2024

This paper aims to provide a detailed introduction to the concept of the Ratio of Predictable Component (RPC) and the Signal-to-Noise Paradox. Then, we derive insights from them by exploring the paradoxical features by conducting a seasonal and regional analysis through ensemble expansion in KMA's climate prediction system (GloSea). We also provide an explanation of the ensemble generation method, with a specific focus on stochastic physics. Through this study, we can provide the predictability limits of our forecasting system, and find way to enhance it. On a global scale, RPC reaches a value of 1 when the ensemble is expanded to a maximum of 56 members, underlining the significance of ensemble expansion in the climate prediction system. The feature indicating RPC paradoxically exceeding 1 becomes particularly evident in the winter North Atlantic and the summer North Pacific. In the Siberian Continent, predictability is notably low, persisting even as the ensemble size increases. This region, characterized by a low RPC, is considered challenging for making reliable predictions, highlighting the need for further improvement in the model and initialization processes related to land processes. In contrast, the tropical ocean demonstrates robust predictability while maintaining an RPC of 1. Through this study, we have brought to attention the limitations of potential predictability within the climate prediction system, emphasizing the necessity of leveraging predictable signals with high RPC values. We also underscore the importance of continuous efforts aimed at improving models and initializations to overcome these limitations.

• Journal of the Korean Society of Radiology
• /
• v.83 no.5
• /
• pp.1032-1045
• /
• 2022

Peripheral arterial disease (PAD) is common in elderly patients. Lower-extremity CT angiography (LE-CTA) can be useful for detecting PAD and planning its treatment. PAD can also be accurately evaluated on reconstructed monoenergetic images (MEIs) from low kiloelectron volt (keV) to high keV images using dual-energy CT. Low keV images generally provide higher contrast than high keV images but also feature more severe image noise. The noise-reduced virtual MEI reconstruction algorithm, called the Mono+ technique, was recently introduced to overcome such image noise. Therefore, this pictorial review aimed to present the imaging findings of PAD on LE-CTA and compare low and high keV images with those subjected to the Mono+ technique. We found that, in many cases, the overall and segmental image qualities were better and metal artifacts and venous contamination were decreased in the high keV images.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.54 no.1
• /
• pp.96-110
• /
• 2017

Most of vehicle detection studies using the existing general lens or wide-angle lens have a blind spot in the rear detection situation, the image is vulnerable to noise and a variety of external environments. In this paper, we propose a method that is detection in harsh external environment with noise, blind spots, etc. First, using a fish-eye lens will help minimize blind spots compared to the wide-angle lens. When angle of the lens is growing because nonlinear radial distortion also increase, calibration was used after initializing and optimizing the distortion constant in order to ensure accuracy. In addition, the original image was analyzed along with calibration to remove fog and calibrate brightness and thereby enable detection even when visibility is obstructed due to light and dark adaptations from foggy situations or sudden changes in illumination. Fog removal generally takes a considerably significant amount of time to calculate. Thus in order to reduce the calculation time, remove the fog used the major fog removal algorithm Dark Channel Prior. While Gamma Correction was used to calibrate brightness, a brightness and contrast evaluation was conducted on the image in order to determine the Gamma Value needed for correction. The evaluation used only a part instead of the entirety of the image in order to reduce the time allotted to calculation. When the brightness and contrast values were calculated, those values were used to decided Gamma value and to correct the entire image. The brightness correction and fog removal were processed in parallel, and the images were registered as a single image to minimize the calculation time needed for all the processes. Then the feature extraction method HOG was used to detect the vehicle in the corrected image. As a result, it took 0.064 seconds per frame to detect the vehicle using image correction as proposed herein, which showed a 7.5% improvement in detection rate compared to the existing vehicle detection method.

• Journal of the Korean Geographical Society
• /
• v.31 no.1
• /
• pp.49-67
• /
• 1996

On of Revolutionary changes of the commercial facilities is appearance of supermarket since the 1970's. It means modernization of retail establishment size. Simultaneously, and then path of distribution resulted in change. Namely, function of wholesale and retail is mixed, path of distribution is shortened, and then distribution function become more influential. By these changes, producer, manager and consumer can be benefit. That is, a change of commercial part is accurance to change of society and economy of region. The aim of the study is to examine commercial properties of supermarket in Kwangju city, and then, it is concerned with life of residents. Factor analysis is applied to the study. The results are as follows: Four factors are abstracted. The first factor of large explanation on properties of supermarket is accounted for years of management, land lent, number of establishments of wholesale${\cdot>$retail${\cdot>$service, age of manager, distance of neighbour residence, and so on. Namely, these may be regarded as commercial environmental feature. Moreover, regional distribution on the ground of factor score is concentrated on CBD and it's neighbour area, which are the areas of highest land lent in Kwangju city. The second factor is accounted for number of supermarket and number of population in each 'dong' (smallest factor-scored area is new housing complex in an outer ring of Kwangju. The third factor is accounted for number of workers, and size of establishment. Regional distribution is concentrated on neighbour of CBD and the outer ring area. These are scattered in contrast to the first and second factors which are continuous. The fourth factor is noted on a sale price of non-foodstuffs, that is, a feature of formation of goods. The highest factor score areas are the CBD, neighbour area of the CBD and a part of an outer ring area. On the other hand, the property of supermarket is preferred to 'the near distance' according to consumer's viewpoint. Furthermore, the location of supermarket, which is actually used by residents, is almost near residental area. It means that condition of location of supermarket is important in the distance of neighbour residence. Besides, supermarkets in Kwangju city are actually situated at residental area, that is, those locations in Kwangju city correspond to preference of consumer's viewpoint. Moreover, this result corresponds to the first factor of the analysis. Namely, the distance of property of supermarket from neighbour residence has very important effect on both of consumer and manager. In the end, supermarket is strongly related to life of residents. It means that supermarket is also responsible to quality of life. On the other hand, commercial facilities have been changed according to consumer's needs and social environments. For these reasons, commercial properties of supermarket must be reexamined in according to diverse commercial facilities and change of commercial environments.

• Journal of Internet Computing and Services
• /
• v.15 no.3
• /
• pp.101-107
• /
• 2014

With the development of online service, recent forms of databases have been changed from static database structures to dynamic stream database structures. Previous data mining techniques have been used as tools of decision making such as establishment of marketing strategies and DNA analyses. However, the capability to analyze real-time data more quickly is necessary in the recent interesting areas such as sensor network, robotics, and artificial intelligence. Landmark window-based frequent pattern mining, one of the stream mining approaches, performs mining operations with respect to parts of databases or each transaction of them, instead of all the data. In this paper, we analyze and evaluate the techniques of the well-known landmark window-based frequent pattern mining algorithms, called Lossy counting and hMiner. When Lossy counting mines frequent patterns from a set of new transactions, it performs union operations between the previous and current mining results. hMiner, which is a state-of-the-art algorithm based on the landmark window model, conducts mining operations whenever a new transaction occurs. Since hMiner extracts frequent patterns as soon as a new transaction is entered, we can obtain the latest mining results reflecting real-time information. For this reason, such algorithms are also called online mining approaches. We evaluate and compare the performance of the primitive algorithm, Lossy counting and the latest one, hMiner. As the criteria of our performance analysis, we first consider algorithms' total runtime and average processing time per transaction. In addition, to compare the efficiency of storage structures between them, their maximum memory usage is also evaluated. Lastly, we show how stably the two algorithms conduct their mining works with respect to the databases that feature gradually increasing items. With respect to the evaluation results of mining time and transaction processing, hMiner has higher speed than that of Lossy counting. Since hMiner stores candidate frequent patterns in a hash method, it can directly access candidate frequent patterns. Meanwhile, Lossy counting stores them in a lattice manner; thus, it has to search for multiple nodes in order to access the candidate frequent patterns. On the other hand, hMiner shows worse performance than that of Lossy counting in terms of maximum memory usage. hMiner should have all of the information for candidate frequent patterns to store them to hash's buckets, while Lossy counting stores them, reducing their information by using the lattice method. Since the storage of Lossy counting can share items concurrently included in multiple patterns, its memory usage is more efficient than that of hMiner. However, hMiner presents better efficiency than that of Lossy counting with respect to scalability evaluation due to the following reasons. If the number of items is increased, shared items are decreased in contrast; thereby, Lossy counting's memory efficiency is weakened. Furthermore, if the number of transactions becomes higher, its pruning effect becomes worse. From the experimental results, we can determine that the landmark window-based frequent pattern mining algorithms are suitable for real-time systems although they require a significant amount of memory. Hence, we need to improve their data structures more efficiently in order to utilize them additionally in resource-constrained environments such as WSN(Wireless sensor network).

• The korean journal of orthodontics
• /
• v.32 no.4 s.93
• /
• pp.235-240
• /
• 2002

Surgical-orthodontic treatment is an increasingly more common dental procedure whose unique psycho-social or psychological feature has not been fairly quantified objectively. Since the treatment of a surgical-orthodontic (or orthodontic) patient is part science and part art, a subjective recognition of a patient about his or her own treatment may be more important than technical success during and after treatment. Therefore, the knowledge of the patient's underlying psychological status could be useful in the prediction of patient's response to surgical-orthodontic (or orthodontic) treatment. The purpose of this study was to investigate and evaluate the psychological difference between conventional orthodontic patient and surgical-orthodontic patient by using locus of control (LOC) examination. Locus of control scale has been proven to be extremely useful in the prediction of a variety of human behaviors. Two types of locus of control data (I-score, Internal locus of control score : I-score, External locus of control score) were obtained for 42 surgical-orthodontic patients and randomly selected 42 conventional orthodontic patients (as a control group) matched for age and gender. No statistically significant difference was observed in the scale of internal and external locus of control between the groups of surgical-orthodontic patients and conventional orthodontic patients. However, in the group of surgical-orthodontic patients, males showed lower E-score (external locus of control score, higher internal locus of control tendency) than females. The results of this study suggested that the psychological background of surgical-orthodontic patients, in contrast with that of cosmetic surgery patients, has a similar disposition with that of orthodontic patients.

• Korean Journal of Ichthyology
• /
• v.13 no.2
• /
• pp.117-122
• /
• 2001

Three species, Gymnogobius urotaenia, G. sp. 1 and G. sp. 2, collected at Hosan-river, Hosan-ri, Samcheock-si, Gangwon-do, Korea, were examined for the study of sexual dimorphism. The three species have peculiar nuptial pigmentation during the breeding season: all females of the three species showed an intense black coloration on the ventral part of the head, the pelvic fin, the anal fin and the margin of the first dorsal fin. There was a difference between G. sp. 1 and the other two species in 1/3 from margin of first dorsal fin. Namely, G. sp. 1 showed an intense black coloration, whereas the other two species had an intense yellow coloration. All females of the three species showed the same, intensely yellow coloration in the ventral part of the body. Males of G. urotaenia and G. sp. 2 had almost the same nuptial pigmentation, however the former showed diffuse black coloration in the anal fin, whereas the latter had an intensely black anal fin. Males of G. sp. 1 differed from those of the other two species in their yellow color on the ventral part of the head, the pelvic f ins, the anal fin, in contrast to the diffuse black coloration in those parts in the other species. The genital papilla of males was triangle-shaped, but that of females was heart-shaped. Significant differences between sexes include: G. urotaenia, 7 characters (body depth, head length, snout length, upper jaw length, head depth, pectoral fin ray length, eye diameter); G. sp. 1, 11 characters (body depth, head length, caudal peduncle length, caudal peduncle depth, snout length, upper jaw length, head width, head depth, anal fin ray length, pectoral fin ray length and caudal fin ray length); G. sp. 2, 10 characters (body depth, head length, caudal peduncle depth, snout length, upper jaw length, head width, anal fin ray length, pectoral f in ray length, caudal fin ray length and eye diameter). Of the 5 characters which showed differences in all three species (body depth, head length, snout length, upper jaw length and pectoral fin ray length), females had higher values than males only in body depth, whereas males values in all other characters. The upper jaw length was the most distinctive sexually dimorphic feature, the upper jaw extended beyond the posterior margin of the eyes (p<0.001) in males but not in females.

• Journal of the Mineralogical Society of Korea
• /
• v.16 no.1
• /
• pp.33-47
• /
• 2003

Various applied-mineralogical characterization including measurements of surface area, size distribution, swelling index, and viscosity were done for some domestic bentonites in order to decipher the rheological properties and their controlling factors. The bentonites, which are Ca-type and relatively low-grade (rnontmorillonite contents: 30 ∼ 75 wt%), occur mostly as subhedral lamellas with the size range of 2 ∼ 4 $\mu\textrm{m}$. The size distribution of mineral fractions in bentonite suspension is dominant in the range of 10 ∼ 100 $\mu\textrm{m}$, and though rather complicated, exhibits roughly bimodal patterns. The feature is more conspicuous in the case of zeolitic bentonite. The bentonites have surface areas ranging 269 ∼ 735 $\m^2$/g, which are measured by EGME adsorption method. The EGME surface areas are nearly proportional to the rnontmorillonite contents, moisture contents, or total CEC. In the surface area measurements, zeolitic bentonites have slightly higher values than those zeolite- free types. The measured swelling index and viscosity of domestic bentonites are comparatively low in values. The swelling values of bentonites were measured to be 250∼500% at maximum by progressively mixing amounts of 2 ∼ 5 wt% Na$_2$CO$_3$, which varies depending on the contents of rnontmorillonite and other impurities, especially zeolite. Much amount of sodium carbonate is required for optimum swelling property of zeolitic bentonited which has usually strong Na- exchanged capacity. The bentonites, which are comparatively feldspar-rich and low in size and crystallinity, tend to be higher in viscosity values. Tn addition, the viscosity is largely higher in case of the bentonites with higher pH in suspension. However, the rheological properties of bentonites such as swelling index and viscosity do not show any obvious relationships with rnontmorillonite contents and mean particle size in suspension. In contrast, roughly speaking, the swelling index of bentonites is reversely proportional to the values of surface area which can be regarded as a collective physico-chemical parameter encompassing all the effects caused by mineral composition, surface charge, particle size, morphological farm, and etc. in bentonites. Thus, the rheological properties in bentonite suspension appear to be rather complicated characteristics which mainly depend on the flocculation of clay particles and the mode of particle association, i.e. quasicrystals, controlled by surface charge, morphology, size, and texture of rnon-tmorillonite, and which partly affected by the finer impurities such as zeolite.