• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.5
• /
• pp.1396-1412
• /
• 2023

Conversation modeling is an important and challenging task in the field of natural language processing because it is a key component promoting the development of automated humanmachine conversation. Most recent research concerning conversation modeling focuses only on the current utterance (considered as the current question) to generate a response, and thus fails to capture the conversation's logic from its beginning. Some studies concatenate the current question with previous conversation sentences and use it as input for response generation. Another approach is to use an encoder to store all previous utterances. Each time a new question is encountered, the encoder is updated and used to generate the response. Our approach in this paper differs from previous studies in that we explicitly separate the encoding of the question from the encoding of its context. This results in different encoding models for the question and the context, capturing the specificity of each. In this way, we have access to the entire context when generating the response. To this end, we propose a deep neural network-based model, called the Context Model, to encode previous utterances' information and combine it with the current question. This approach satisfies the need for context information while keeping the different roles of the current question and its context separate while generating a response. We investigate two approaches for representing the context: Long short-term memory and Convolutional neural network. Experiments show that our Context Model outperforms a baseline model on both ConvAI2 Dataset and a collected dataset of conversational English.

• Nuclear Medicine and Molecular Imaging
• /
• v.43 no.2
• /
• pp.120-128
• /
• 2009

Purpose: Early detection of recurrence is an important factor for long term survival of patients with colorectal cancer. Measurement of serum levels of CEA, CA 19-9, CT and PET/CT has been commonly used in the postoperative surveillance of colorectal cancer. The purpose of this study was to compare the diagnostic ability of PET/CT, tumor marker and CT for recurrence in colorectal cancer patients after treatment. Materials and Methods: F-18 FDG PET/CT imaging was performed in 189 colorectal cancer patients who underwent curative surgical resection and/or chemotherapy. Measurement of serum levels of CEA, CA 19-9 and CT imaging were performed within 2 months of PET/CT examination. Final diagnosis of recurrence was made by biopsy, radiologic studies or clinical follow-up for 6 months after each study. Results: Overall sensitivity, specificity of PET/CT was 94.7%, 91.1%, while those of serum CEA were 44.7% and 97.3%, respectively. Sensitivity and specificity were 94.2%, 90.4% for PET/CT and better than those of combined CEA and CA 19-9 measurement(52.1%, 88.5%) in 174 patients measured available both CEA and CA 19-9 data. In 115 patients with both tumor markers and CT images available, PET/CT showed similar sensitivity but higher specificity(92.9%, 91.3%) compared to combination of tumor markers and CT images(92.9%, 74.1%). Conclusion: PET/CT was superior for detection of recurred colorectal cancer patients compared with both CEA, CA 19-9, and even with combination of both tumor markers and CT. Therefore PET/CT could be used as a routine surveillance examination to detect recurrence or metastasis of colorectal cancer.

• The Korean Journal of Nuclear Medicine
• /
• v.32 no.4
• /
• pp.344-353
• /
• 1998

Purpose: The purpose of this study was to evaluate the diagnostic accuracy of Tc-99m labeled antigranulocyte antibody immunoscintigrapy in the diagnosis of osteomyelitis and compare with the results of triphasic bone scan. Materials and Methods: The study population was 39 patients (22 male, 17 female) who had uncertain diagnoses of osteomyelitis. Fifteen patients had history of orthopedic surgery, and 5 had previous fracture. One milligram of monoclonal antibody against NCA-95 was labeled with 370 MBq of Tc-99m, injected intravenously, and 4 hour images were obtained. Triphasic bone scan images were obtained in 30 patients. The final diagnosis was confirmed by bacteriologic culture, biopsy or long term clinical follow up. Results: Twenty one patients were confirmed to have osteomyelitis (1 acute, 20 chronic). Eighteen patients were without osteomyelitis. Antigranulocyte antibody immunoscintigraphy had a sensitivity of 71% (15/21), and a specificity of 89% (16/18), while the sensitivity and specificity of triphasic bone scan was 93% (13/14) and 38% (6/16), respectively. Antigranulocyte antibody scan showed higher specificity of 100% (11/11) in comparison with 33% (3/9) of triphasic bone scan in patients with history of orthopedic surgery or fracture. Conclusion: Antigranulocyte antibody immunoscintigraphy is more specific than that of triphasic bone scan and may be helpful in patients with history of surgery or fracture. However, sensitivity is lower than triphasic bone scan in the detection of chronic osteomyelitis.

• THE INTERNATIONAL COMMERCE & LAW REVIEW
• /
• v.57
• /
• pp.283-303
• /
• 2013

When a trade conflict arises related to an officially supported export credit programme, The World Trade Organization(WTO), decides on whether the programme is a forbidden subsidy stipulated in the Agreement on Subsidies and Countervailing Measures(the ASCM Agreement). Korea was taken to the WTO panel two times for the export credit programme. One is the semiconductor case in 2002 and the other was the shipbuilding disputes in 2004. And, In 2012, the U.S. Commerce Department ruled K-SURE's export insurance for Korean refrigerator manufacturers as a forbidden subsidy even if the case was not taken to the WTO. This paper examines the significance of export credit programmes on the WTO ASCM Agreement and discusses how to operate these programmes so they would not infringe upon the Agreement by analyzing the actual cases of WTO subsidy conflicts that involved Korean enterprises in relation to export credit programmes for the purpose of determining the related issues and impacts. From this research the results were as follows: First, on whether export credit is a prohibited subsidy, the deciding factor was whether a benefit has been conferred to the beneficiary. On the presence of a benefit, the WTO panel used market benchmarks as the main criteria. Thus, official export credit agencies(ECAs) should be careful not to provide export credit support which had been granted to the beneficiary at better than market terms. Second, in the case of export credit, the special status of ECA as a public body receiving government support itself does not constitute a subsidy. However, caution must be taken not to provide export credit that may lead to WTO ASCM subsidy conflicts involving a certain exporter or industry by setting up clear and valid regulations and fair work processes in the operation of export credit programmes. Third, item (j) of Annex I cannot be interpreted reversely as this item is for interpreting the presence of a prohibited subsidy, not the presence of a benefit. Thus, an export credit program that confers a financial contribution, a benefit and specificity, could qualify as a prohibited subsidy. Fourth, ECAs not only have to maintain long-term account balance but also introduce additional measures to meet this long-term balance such as a clear and systematic premium system. Finally, export credit programmes that are not defined in item (j) of Annex I of the ASCM Agreement would not deemed as an prohibited export subsidy as long as the continued support of the programmes are not being forced.

• Korean Security Journal
• /
• no.61
• /
• pp.9-38
• /
• 2019

As the scales & density of the Korean national infrastructures have been increased, they will be identified as rich and attractive potential targets for intensified North Korea's attack in the rear region and terrorism attack. In addition, due to changes in security environment such as drone threats and lack of security forces under the 52-hour workweek law, I think that it is the proper time point to reevaluate the effectiveness and appropriateness of the current physical protection system and its shift to a new system. In this study, the direction and improvement of the perimeter physical protection systems of the national infrastructures are to be studied from the viewpoints of its concepts of operations and design methodology, focusing on the nuclear power plant. The reason why we focus on nuclear power plants is because they cause wide-range and long-term damages caused by radioactive materials disperal and pollution, along with short-term damage caused by the interruption of electricity generation in the event of damage to nuclear power plants. With the aim of extracting improvement directions, as we will comprehensively review domestic research trends and domestic·overseas related laws, and consider Korea's specificity, we try to reframe the concept of operation - systematization, mobilization and flexibility -, and establish criteria on system change. In order to improve the technical performance of the new perimeter physical protection system, we study on high-fidelity·multi-methodology based integrated design methodology, breaking from individual silo-type design methods, and I suggest improvement of government procurement, its expansion to export business and other national infrastructure.

• Clinical and Experimental Pediatrics
• /
• v.50 no.11
• /
• pp.1061-1066
• /
• 2007

Purpose : Early identification of developmental disabilities allows intervention at the earliest possible point to improve the developmental potential. The Ages and Stages Questionnaire (ASQ), a parent- completed questionnaire, can be used as a substitute for formal screening tests. The purpose of this study was to evaluate the validity of the Korean version of the ASQ (K-ASQ) as a screening tool for detecting developmental delay of young Korean children in the setting of a busy pediatric outpatient clinic. Methods : Parents completed the K-ASQ in the waiting room of the pediatric outpatient clinic of St. Mary's Hospital, Catholic University Medical College. Out of 150 completed the ASQ, 67 who were born term and had no previous diagnosis of developmental delay, congenital anomalies, or neurological abnormalities were enrolled. The cut-off values of less than 2 standard deviations (SD) below the mean for the ASQ were used to define a "fail", and children who failed in one or more domains tested were classified as "screen-positive". Diagnosis of developmental delay was made when the developmental indices fell below -1 SD of the Bayley Scales of Infant Development-II. Results : (1) The mean age of children was $16.4{\pm}7.4$ months. Ten children (14.9%) were small-for- gestational age infants. The mean birth weight and gestational age were $3.1{\pm}0.6kg$ and $38.8{\pm}1.4$ weeks. Nine children (13.4%) were twins and 33 (49.0%) were male. The mean maternal education in years was $13.6{\pm}2.4$, and 31.3% had full-time jobs. The time for completing the ASQ was $10.2{\pm}3.0$ minutes. (2) Seventeen children (25.4%) were classified as screen-positive, four of them were delayed in development. Among eight children diagnosed with developmental delay, four were screen-positive and the other four were screen-negative by the ASQ. (3) The test characteristics of the ASQ were as follows: sensitivity (50.0%); specificity (78.0%); positive predictive value (23.5%); negative predictive value (92.0%). Conclusion : The high negative predictive value of the K-ASQ supports its use as a screening tool for developmental delay in the setting of a pediatric outpatient clinic.

• Clinical and Experimental Pediatrics
• /
• v.49 no.3
• /
• pp.298-304
• /
• 2006

Purpose : We compared the asthma predictive index(API) and the modified asthma predictive index (mAPI) of the Tuscon Children's Respiratory Study Group in Korean children with recurrent wheezing. We investigated the atopic profiles and presence of allergen sensitization of each risk group, and ascertained the significant clinical risk factors. Methods : Two hundred and sixty two children, who visited for recurrent wheezing from 1998 to 2005, were enrolled and divided into groups by API and mAPI. We investigated the history of the patients and their families, atopic profiles, and sensitization to aeroallergen and food allergens. Twenty nine children were followed up to 6 years of age and we evaluated the sensitivity, specificity and positive and negative predictive value of both indices. Results : The high risk group of API were of older age, were more likely to be sensitized to aeroallergen(P=0.001) and food allergen(P=0.034) and had higher levels of total eosinophil count, eosinophil percent, serum ECP, total IgE, and D.p-, D.f-specific IgE. High risk group of mAPI showed higher levels of atopic markers such as egg-, milk-, D.p- and D.f-specific IgE. Even though API did not include allergen sensitization, the high risk group was more significantly sensitized to common allergens than the low risk group. Twenty nine children were followed up until 6 years of age; therefore 15 children were diagnosed as asthma, clinically. The sensitivity, specificity, positive and negative predictive values of mAPI were higher than API. Conclusion : Both high risk groups of API and mAPI had higher levels of atopic markers and were more sensitized to common allergens. These findings suggest that sensitization to aeroallergens and food allergens are more objective markers as asthma predictive indices. In addition, mAPI is a more reliable index in predicting asthma in Korean children with recurrent wheezing than is API. But only 29 patients were followed until the age of 6, so we need to include more children with long term follow up for future study.

• Investigative Magnetic Resonance Imaging
• /
• v.12 no.2
• /
• pp.153-160
• /
• 2008

Purpose : To evaluate if at the time of diagnosis of Legg-Calve-Perthes disease, MR imaging is superior to radiographs for identifying patients with a favorable long term prognosis as determined by maintenance of a spherical configuration of the femoral head epiphysis on follow-up radiographs. Materials and Methods : We retrospectively analyzed plain radiographs and MR images of 48 femoral heads in 41 patients with Legg-Calve-Perthes disease at time of diagnosis. Femoral head involvement was estimated on initial plain radiographs and MR images using the Catterall classification. Prognosis as determined by maximum femoral head deformity was determined on follow-up radiographs evaluated according to Stulberg classification. Results: Catterall classification of greater than 1 at MR imaging had 100% sensitivity (14/14) for detecting patients with a poor prognosis as indicated by deformity of the femoral head on follow-up radiographs (Stulberg class > 2). Initial radiographs had a lower sensitivity of 57% (8/14) for detecting patients with poor prognosis. Specificity for MR imaging was 21% (7/34) which was not significantly different from 32% (11/34) specificity for initial radiographs. Conclusion: Catterall group 1 at initial MR imaging indicates favorable prognosis for Legg-Calve-Perthes disease. Patients with more extensive involvement of the femoral head can have a good outcome, however they are at risk for loss of spherical configuration of the femoral head and subsequent osteoarthritis in adulthood. MR imaging may be superior to radiographs for identifying a subgroup of patients with favorable prognosis.

• Tuberculosis and Respiratory Diseases
• /
• v.43 no.3
• /
• pp.377-387
• /
• 1996

Background : In 1980, WHO made a definition in which the term "impairment" as applied to the respiratory system is used to describe loss of lung function, "disability" the resulting diminution in exercise capacity. The measurement of pulmonary function during exercise would give us information about overall functional capacity and respiratory performance that would be lacking in tests performed at rest. We conducted this study to investigate the role of resting pulmonary function test and exercise test for assessing impairment/disability in patients with chronic airflow obstruction(CAO). Method : We studied 19 patients with CAO. The spirometry and body plethysmograph were performed in stable condition. And then patients performed a progressive incremental exercise test to a symptom-limited maximum using cycle ergometer. Patients were divided in two groups, severe and non-severe impairment, according to the resting PFTs and compaired each other. A patient was considered to be severely impaired if FVC < 50 %, FEV1 < 40 % or FEV1/FVC < 40 %. Results : 1) The airway obstruction and hypoxemia of severe impairment group were more severe and exercise performance was markedly reduced compairing to non-severe impairment group. 2) The severe impairment group showed ventilatory limitation during exercise test and the limiting symptomes ware dyspnea in 9/10 patients. 3) The impairment and disability of the patients with tuberculous destructed lung were most marked in patients with CAO. 4) The FEV1 was the most prevalent criterion for the determination of severe impairment based on resting PFTs and was the valuable best correlated to V02max(r=0.81, p < 0.001). 5) The sensitivity of exercise limits for predicting severe disability according to resting PFTs was 80 % and specificity 89 %. Conclusion : In patients with severe CAO, FEV1 is a good predictive of exercise performance and impairment measured by resting PFTs can predict a disability by exercise test.

• Journal of the Korean Geographical Society
• /
• v.42 no.4
• /
• pp.560-581
• /
• 2007

The purpose of this paper is to present a conceptual framework and a stage of development of trust building and to study the factors affecting on the trust building in economics space. Conceptual framework on trust building in economics space is combined of in the three approaches. The macro(structural and institutional) approach includes normative and regulative factors(laws, norms), and positionality in social and economic systems(beliefs, political ideologies, institutions). The meso(intersubjective) approach contains the personal fronts(expressive factors, social cues, significant symbols) and settings(physical space, intermediary such as technologies & knowledges). The micro(subjective) approach comprises the willingness(internalization of value) and calculation(risk and uncertainties analysis) of economic actors. According to sustainable cooperation among economic actors, trust building to the macro(structural and institutional) level, the meso(intersubjective) level, and the micro(subjective) level develop by stages. The factors such as long-term and repeated interaction, information sharing and reciprocity, interdependence and asset specificity, uncertainty, proximity, and culture & norm of corporate and formal institution are determinants on the trust building across economic actors in economic space.