• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.05a
• /
• pp.206-208
• /
• 2022

As the big data industry has recently developed significantly, interest in privacy violations caused by personal information leakage has increased. There have been attempts to automate this through named entity recognition in natural language processing. In this paper, named entity recognition data is constructed semi-automatically by identifying sentences with de-identification information from de-identification information in Korean Wikipedia. This can reduce the cost of learning about information that is not subject to de-identification compared to using general named entity recognition data. In addition, it has the advantage of minimizing additional systems based on rules and statistics to classify de-identification information in the output. The named entity recognition data proposed in this paper is classified into twelve categories. There are included de-identification information, such as medical records and family relationships. In the experiment using the generated dataset, KoELECTRA showed performance of 0.87796 and RoBERTa of 0.88.

• Journal of the Institute of Convergence Signal Processing
• /
• v.23 no.3
• /
• pp.173-178
• /
• 2022

Recently, various computing technologies such as artificial intelligence, big data, and IoT are developing. In particular, artificial intelligence-based deepfake technology is being used in various fields such as the content and medical industry. Deepfake technology is a combination of deep learning and fake, and is a technology that synthesizes a person's face or body through deep learning, which is a core technology of AI, to imitate accents and voices. This paper uses deepfake technology to study the creation of virtual characters through the synthesis of animation models and real person photos. Through this, it is possible to minimize various cost losses occurring in the animation production process and support writers' work. In addition, as deepfake open source spreads on the Internet, many problems emerge, and crimes that abuse deepfake technology are prevalent. Through this study, we propose a new perspective on this technology by applying the deepfake technology to children's material rather than adult material.

• Journal of Appropriate Technology
• /
• v.5 no.1
• /
• pp.46-53
• /
• 2019

It should be supported by the ease and economic production in terms of appropriate technology to be widely used in the real applying fields. In particular, the diagnosis of diseases is one of the most difficult areas to apply to the developing countries compared to other fields because of the high costs of professional manpower, equipment, and medial reagents. In recent years, paper, which is one of the most common materials around, has been used to fabricate the disease diagnostic chips, particularly without any special facilities or the equipment. These new technologies lead the possibility of using point-of-care testing devices that can be manufactured quickly in the field by lowering the production cost. These paperbased technologies will contribute to solve many of the deadliest medical problems in developing countries and ultimately improve the quality of human life.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2023.07a
• /
• pp.179-181
• /
• 2023

본 논문에서는 GAN 기반의 영상 생성 방법론을 이용해 delayed PET 영상을 생성하는 연구를 수행하였다. PET은 양전자를 방출하는 방사성 동위원소를 표지한 방사성의약품의 체내 분포를 시각화함으로서 암 세포 진단에 이용되는 의료영상 기법이다. 하지만 PET의 스캔 과정에서 방사성의약품이 체내에 분포하는 데에 걸리는 시간이 오래 걸린다는 문제점이 존재한다. 따라서 본 연구에서는 방사성의약품이 충분히 분포되지 않은 상태에서 얻은 PET 영상을 통해 목표로 하는 충분히 시간이 지난 후에 얻은 PET 영상을 생성하는 모델을 GAN (generative adversarial network)에 기반한 image-to-image translation(I2I)를 통해 수행했다. 특히, 생성 전후의 영상 간의 영상 쌍을 고려한 paired I2I인 Pix2pix와 이를 고려하지 않은 unpaired I2I인 CycleGAN 두 가지의 방법론을 비교하였다. 연구 결과, Pix2pix에 기반해 생성한 delayed PET 영상이 CycleGAN을 통해 생성한 영상에 비해 영상 품질이 좋음을 확인했으며, 또한 실제 획득한 ground-truth delayed PET 영상과의 유사도 또한 더 높음을 확인할 수 있었다. 결과적으로, 딥러닝에 기반해 early PET을 통해 delayed PET을 생성할 수 있었으며, paired I2I를 적용할 경우 보다 높은 성능을 기대할 수 있었다. 이를 통해 PET 영상 획득 과정에서 방사성의약품의 체내 분포에 소요되는 시간을 딥러닝 모델을 통해 줄여 PET 이미징 과정의 시간적 비용을 절감하는 데에 크게 기여할 수 있을 것으로 기대된다.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.17 no.2
• /
• pp.99-107
• /
• 2024

The recent digital transformation in many industries has led to an explosion of data, which has exponentially increased the amount of data that companies need to generate and process. As a result, enterprises are leveraging ERP systems to manage and analyze large amounts of data in real time. However, due to cost and time issues in processing large amounts of data in existing ERP systems, it is essential to apply data archiving technology that can compress and store data in real time in existing systems. Therefore, this paper aims to identify the trends of the target technology by utilizing patent data on ERP-based real-time data archiving technology, analyze the core patents, and create new core patents based on them.

• Journal of the Korea Society of Computer and Information
• /
• v.29 no.3
• /
• pp.43-54
• /
• 2024

Recently, there has been a lot of research on utilizing Language Models, and Large Language Models have achieved innovative results in various tasks. However, the practical application faces limitations due to the constrained resources and costs required to utilize Large Language Models. Consequently, there has been recent attention towards methods to effectively utilize models within given resources. Curriculum Learning, a methodology that categorizes training data according to difficulty and learns sequentially, has been attracting attention, but it has the limitation that the method of measuring difficulty is complex or not universal. Therefore, in this study, we propose a methodology based on data heterogeneity-based Curriculum Learning that measures the difficulty of data using reliable prior information and facilitates easy utilization across various tasks. To evaluate the performance of the proposed methodology, experiments were conducted using 5,000 specialized documents in the field of information communication technology and 4,917 documents in the field of healthcare. The results confirm that the proposed methodology outperforms traditional fine-tuning in terms of classification accuracy in both LoRA fine-tuning and full fine-tuning.

• Clinical and Experimental Pediatrics
• /
• v.48 no.4
• /
• pp.401-405
• /
• 2005

Purpose : The use of mechanically-assisted ventilators at home reduces morbidity and improves the quality of life in children with chronic respiratory failure. But in Korea there is no clinical data of children with home mechanical ventilation. We investigated ventilator types, duration, the causes of failure or death, and the cost needed for care. Methods : We retrospectively analyzed the medical records of 21 children who were admitted and who applied for home mechanical ventilation at the Pediatric Intensive Care Unit in Asan Medical Center. Phone interviews took place after discharge. and interviewed by phone after discharge. Results : The median age was 31 months; the median duration with ventilator was 25 months. Underlying diseases were 16 neuromuscular diseases, one metabolic disease and four chronic respiratory diseases. The types of ventilator were pressure and volume type(16 and five patients, respectively). The frequency of ventilation failure was once per 19 months. Weaning could be performed in three cases. Frequencies of admission after receiving ventilators were 1.7 times per year; the most common cause was pneumonia. Nine patients(43%) died; four of them died because of endotracheal tube obstruction. The costs for medical care were about 1,110,000 won per month. Conclusion : There is an increment in the numbers of individuals who need mechanical ventilation support. The most common cause of death was endotracheal tube obstruction. The most important problem for the patients was medical cost. There needs to be more interest in patients with ventilator and social welfare systems to support their families need to be prepared.

• Journal of radiological science and technology
• /
• v.30 no.1
• /
• pp.53-59
• /
• 2007

A safety management rule of the diagnosis radiation system which opened a court 2006 February 10th was promulgated for safety of the radiation worker, patients and patients' family members. The purpose of this study is to minimize injury by radiation that can happen to patients and people around a sick ward when managing mobile X-ray system. This study analyzed sickroom environment of mobile X-ray examination and the statistical data of the Konkuk medical Information System(KIS) and the Picture Archiving Communication System(PACS). This study also investigated patient conditions, infection, relation information and related data, when the sickroom mobile X-ray examination is used. Through data analysis, many problems were expected such as restriction of space side, manpower and expense of business side, satisfaction degree decline of patient and protector of operation side. Therefore, we tried to restrict examination of multi bed sickroom, and to use treatment room in each ward to solve problem mentioned. As a result, the whole sickroom mobile X-ray examination rate decreased to near 50%, and mobile X-ray examination rate for inpatients decreased to more than 85%. This study shows that several attempts we did should be helpful for manpower, patients satisfaction and expenses. Also, they should protect patients in sickroom from unnecessary radiation exposure and could minimize inconvenience of patients and their family members from x-ray examination.

• Journal of Korean Academy of Nursing
• /
• v.24 no.4
• /
• pp.545-556
• /
• 1994

Medical insurance and health care delivery system enabled Korean people to get the necessary medical service, but it caused increased needs for medical service, and resulted in the occurence of some problems such as a lack of manpower and medical facilities. In order to solve these problems, many countries, which already had medical insurance system had developed home care system and it has been regarded effective both in reducing costs and in increasing the rates of turnover of bed. Recently, Korea has included home nursing care in its health care delivery system, and some models of the hospital based home nursing care had been tried and its effects had been evaluated. So, author tried to run a home nursing care for the Cesarean section mothers and evaluate Its effects both in the mother's health and costs. This study was designed as a Quasi-experimental study. Subjects were thirty mothers who got Cesarean section operation in hospital in Seoul. Experimental group consisted of 15 volunteers, and control group were selected by means of matching technique. Data were gathered from February 1st to March 26th by two assistants who were trained by author. Experimental group were discharged on the 4th day after their operation, and got nursing care and assessment about their home three times on the 5th, 6th, and 7th day. Control group stayed in the hospital until 7th day as usual and were checked on the same day as above mentioned To evaluate the state of physiological recovery, vital signs, H.O.F, presence of edema in the legs, bathing, appetite, sleep, presence of pain or discomfort in the breasts, amount of lochia, color of lochia, defecation urination. To compare incidence of complication in experimental group with that in control group, specific assessment was done such variables as smell of lochia, presence of inflammation of operation wound, dizziness, and presence of immobilization in the extremities. The activities of daily living were checked Satisfaction of nursing were checked To calculate costs, author asked subjects to specify expenditure including hospital charge, traffic enpenses, and food expenses. The results were as fellows. 1. On effectiveness of home nursing careThere were n significant differences between experimental and control group in incidence of abnormal symptoms and any complication. The number of taking a bath [POD #5 P=0.001, #6 P=0.0003, #7 P=0.001] and the degree of appetite [POD #5 P=0.03, #6 P=0.02, #7 P=0.013] were significantly higher in experimental group than in control group. Contrary to author's expectation, the degree of the activities of daily living in experimental group was not higher than that of control group. All of the experimental group said they were satisfied with the home nursing care. 2. Cost analysis 1) Hospital charge of experimental group was lower than that of control group. ［P=0.009］ By taking home nursing care, average period of hospitalization was shortened to 3.1 days, and family members could save 22.8 hours. Total amount of money saved by early discharge was 3,443,093 Won. It is estimated that total amount of money saved by early discharge in a year will be 40,398,956 Won. 2) Home nursing care charge of 15 mothers was 1,781,633 Won. It is estimated that total amount of money Saved by it in a year Will be 20,904,493 Won. It was lower altogether than hospital charge of the three days which is 5th, 6th, 7th day of operation. The average cost of single home visit was calculated 10,940 Won. It took 87 minutes per round and it costed 1,017.3 Won. The average hour of home care was 39.0 minutes. 3) It is expected that early discharge can bring forth the increase of hospital income. On the condition that the rate of running bed is 100%, the expected increase of hospital income will be 202,374, 026 Won in a year. Suggestions for further study and nursing practice are as follows : 1. For the welfare of patients and the increased rates of running bed, home nursing care system should be included in the hospital nursing care system. 2. Studies to test effect of home nursing care on the patients with other diseases are needed. 3. Establishment of law on the practice of home nursing care is strongly recommended.

• Korean Business Review
• /
• v.17 no.2
• /
• pp.63-87
• /
• 2004

To create and maintain comparative supremacy as a strategic tool of business, many organizations have introduced informational technology and system. By using this system, Some companies got a beneficial value for achieving organizational goals but others could not obtain their effectiveness and efficiency. In particular, a lot of organizations that tried to make strategic supremacy with e-commercial trade are under hard condition because of poor profit. It implies that it is essential to identify and analyse the consumer who uses e-commercial trade. This paper, therefore, focusing on internet shopping malls between business and consumer as one of areas of e-commercial trades, shows the difference between consumer expectation and performance. The results of this study are as follows: First, as for the significant difference of influencing factors to consumer satisfactions according to the types of internet shopping malls, there is a meaningful difference in consumer anxiety and internet usefulness, but not in consumer service. Prior to verify the differences in detail on consumer's anxiety and internet usefulness, we examined that there is any difference between expectation and performance. T-test was used for the variants of consumer anxiety and internet usefulness, and its meaningful probability was 0.000, which means that both showed statistically significant difference. Based on the results, we also found that regardless of the types of internet shopping malls, consumer expectation was greater than performance. although the difference between expectation and performance was not equal according to the internet shopping malls. Second, a regression analysis was performed to understand the relation between consumer service, internet usefulness, consumer anxiety, and consumer satisfaction, it was found that consumer service, internet usefulness, consumer anxiety had significantly effected on consumer satisfaction. Third, To verify the relation between consumer satisfaction and repurchase-intentions, intentions to spread out, Pearson correlation analysis was used. it was found that consumer satisfaction had positive effect on both intentions. This study has some limitations because of the shorts of money and time. since the sample of this study was consumers who have ever bought one or more products via internet shopping mall, this sample was appropriate. but the major parts of sample were college students, and the sample size was so small. therefore this results should carefully be generalized. For further study, it is required to select more precise samples and to include more variables.