• Journal of Physiology & Pathology in Korean Medicine
• /
• v.38 no.2
• /
• pp.49-58
• /
• 2024

The rapid advancement of generative large language models has revolutionized various real-life domains, emphasizing the importance of exploring their applications in healthcare. This study aims to examine how generative large language models are implemented in the medical domain, with the specific objective of searching for the possibility and potential of integration between generative large language models and East Asian medicine. Through a comprehensive current state analysis, we identified limitations in the deployment of generative large language models within East Asian medicine and proposed directions for future research. Our findings highlight the essential need for accumulating and generating structured data to improve the capabilities of generative large language models in East Asian medicine. Additionally, we tackle the issue of hallucination and the necessity for a robust model evaluation framework. Despite these challenges, the application of generative large language models in East Asian medicine has demonstrated promising results. Techniques such as model augmentation, multimodal structures, and knowledge distillation have the potential to significantly enhance accuracy, efficiency, and accessibility. In conclusion, we expect generative large language models to play a pivotal role in facilitating precise diagnostics, personalized treatment in clinical fields, and fostering innovation in education and research within East Asian medicine.

• Journal of Intelligence and Information Systems
• /
• v.28 no.2
• /
• pp.101-125
• /
• 2022

Recently, with the development of computing technology and the improvement of the cloud environment, deep learning technology has developed, and attempts to apply deep learning to various fields are increasing. A typical example is anomaly detection, which is a technique for identifying values or patterns that deviate from normal data. Among the representative types of anomaly detection, it is very difficult to detect a contextual anomaly that requires understanding of the overall situation. In general, detection of anomalies in image data is performed using a pre-trained model trained on large data. However, since this pre-trained model was created by focusing on object classification of images, there is a limit to be applied to anomaly detection that needs to understand complex situations created by various objects. Therefore, in this study, we newly propose a two-step pre-trained model for detecting abnormal situation. Our methodology performs additional learning from image captioning to understand not only mere objects but also the complicated situation created by them. Specifically, the proposed methodology transfers knowledge of the pre-trained model that has learned object classification with ImageNet data to the image captioning model, and uses the caption that describes the situation represented by the image. Afterwards, the weight obtained by learning the situational characteristics through images and captions is extracted and fine-tuning is performed to generate an anomaly detection model. To evaluate the performance of the proposed methodology, an anomaly detection experiment was performed on 400 situational images and the experimental results showed that the proposed methodology was superior in terms of anomaly detection accuracy and F1-score compared to the existing traditional pre-trained model.

• KIISE Transactions on Computing Practices
• /
• v.22 no.11
• /
• pp.619-624
• /
• 2016

Social-aware video displays not only the relationships between characters but also diverse information on topics such as economics, politics and culture as a story unfolds. Particularly, the speaking habits and behavioral patterns of people in different situations are very important for the analysis of social relationships. However, when dealing with this dynamic multi-modal data, it is difficult for a computer to analyze the drama data effectively. To solve this problem, previous studies employed the deep concept hierarchy (DCH) model to automatically construct and analyze social networks in a TV drama. Nevertheless, since location knowledge was not included, they can only analyze the social network as a whole in stories. In this research, we include location knowledge and analyze the social relations in different locations. We adopt data from approximately 4400 minutes of a TV drama Friends as our dataset. We process face recognition on the characters by using a convolutional- recursive neural networks model and utilize a bag of features model to classify scenes. Then, in different scenes, we establish the social network between the characters by using a deep concept hierarchy model and analyze the change in the social network while the stories unfold.

• The Journal of Society for e-Business Studies
• /
• v.27 no.1
• /
• pp.15-28
• /
• 2022

Due to recent expansion of online market such as clothing, utilizing customer review has become a major marketing measure. User review has been used as a tool of analyzing sentiment of customers. Sentiment analysis can be largely classified with machine learning-based and lexicon-based method. Machine learning-based method is a learning classification model referring review and labels. As research of sentiment analysis has been developed, multi-modal models learned by images and video data in reviews has been studied. Characteristics of words in reviews are differentiated depending on products' and customers' categories. In this paper, sentiment is analyzed via considering review data and metadata of products and users. Gated Recurrent Unit (GRU), Long Short-Term Memory (LSTM), Self Attention-based Multi-head Attention models and Bidirectional Encoder Representation from Transformer (BERT) are used in this study. Same Multi-Layer Perceptron (MLP) model is used upon every products information. This paper suggests a multi-modal sentiment analysis model that simultaneously considers user reviews and product meta-information.

• Journal of Gastric Cancer
• /
• v.18 no.4
• /
• pp.379-391
• /
• 2018

Purpose: Gastric cancer (GC) patients with peritoneal metastasis (PM) have poor prognosis. Pressurized intraperitoneal aerosol chemotherapy (PIPAC) in combination with systemic chemotherapy is a novel treatment option for patients in stage IV of the disease. Materials and Methods: Between November 2015 and June 2018, prospective data collection was performed in 24 patients with GC and PM (median age, 57; range, 44-75 years). These patients underwent 46 PIPAC procedures with a median number of 2 interventions per patient (range, 1-6). A laparoscopic access was used and a combined therapy of cisplatin and doxorubicin aerosol was administered. Results: The median peritoneal carcinomatosis index before the 1st PIPAC was 14 (range, 2-36), and the median ascites volume in patients before the 1st PIPAC was 100 mL (range, 0-6 mL, 300 mL). Eleven patients, who received 2 or more PIPAC procedures, had decreased and stable volumes of ascites, while only 3 patients displayed increasing volume of ascites. The median overall survival was 121 days (range, 66-625 days) after the 1st PIPAC procedure, while 8 patients who received more than 3 PIPAC procedures had a median survival of 450 days (range, 206-481 days) (P=0.0376). Conclusions: Our data show that PIPAC is safe and well tolerated, and that the production of ascites can be controlled by PIPAC in GC patients. Patients, who received 2 or more PIPAC procedures, reported a stable overall quality of life. Further studies are required to document the significance of PIPAC as a palliative multimodal therapy.

• Journal of Navigation and Port Research
• /
• v.36 no.10
• /
• pp.905-910
• /
• 2012

In early 1990s, the Sea & Air Transport Cargoes (SATC) was increased annually with more than 50% rate due to the rising trade between Korea and China. However, after that, the increasing rate of the SATC was slowdown from the late 1990s, furthermore, recently it became sluggish and declined. This phenomenon is totally different compared to the skyrocketing trade volumes between two countries. In this respect, to forecast the SATC, draw out the factors for activation, and calculate the weight of priority of these factors are urgently needed. To achieve the research objectives, the ARIMA and Fuzzy-AHP were used as research methodology. The estimated volume of SATC using the data from year 2007 to 2012 on the ARIMA model, will be reached approximately 33,000 tons in year 2015. In the mean time, For drawing out and weighing the activation factors for SATC, the Fuzzy-AHP was adopted. As a result, 'Sea & Air transportation-related information system policies' is the most important factor among the principle criteria, and 'the construction of consolidation logistics center' is the most important factor among the 12 sub-principle criteria.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.2 no.1 s.2
• /
• pp.1-13
• /
• 2003

This International Standard establishes an AVI/AEI System based on radio frequency technologies. This system is intended for general application in ITS. Specially, It allows the transfer of the identification codes and further information about equipment and vehicles used in intermodal transport into such CVO and information systems related to Intermodal Transport processes. The aim of this standard is to define, describe and specify Architecture, System Parameters, Numbering/ Data structures and interface related to an AVI/AEI system to provide an enabling Standard, which, whilst allowing the system specifier to determine the performance levels and operating conditions, provides a framework for nominal interoperability. The Standard is to establish a common framework to achieve unambiguous identification in AVI/AEI applications. Thes is AVI/AEI is designed to be an 'enabling' structure to allow interoperability between different commercial systems, and not prescriptive in determining any one system. The ISO TC204 WG4 has eight active work items. A new WI on ERI is progressing quickly; three WIs for the road environment and four multimodal WIs are under development. All Work Items are joint between CEN TC278 and ISO TC204 according to the Vienna Agreement, with CEN in the lead. The work is progressing with some delay. For all the work items, the countries who have appointed experts we: Australia, Austria, Belgium, Canada, Czech, Denmark France, Germany, Japan, Korea, the Netherlands, Norway, Spain, Sweden, UK and USA. There are 30 registered experts. The study focus on the AVI/AEcl standardization and developing of the Korea standard

• Asian Pacific Journal of Cancer Prevention
• /
• v.13 no.5
• /
• pp.1809-1817
• /
• 2012

Background: Gastric cancer is frequently lethal despite aggressive multimodal therapies, and new treatment approaches are therefore needed. Retinoids are potential candidate drugs: they prevent cell differentiation, proliferation and malignant transformation in gastric cancer cell lines. They interact with nuclear retinoid receptors (the retinoic acid receptors [RARs] and retinoid X receptors [RXRs]), which function as transcription factors, each with three subclasses, ${\alpha}$, ${\beta}$ and ${\gamma}$. At present, little is known about retinoid expression and influence on prognosis in gastric cancers. Patients and Methods: We retrospectively analyzed the expression of the subtypes RARa, $RAR{\beta}$, $RAR{\gamma}$, RXRa, $RXR{\beta}$, $RXR{\gamma}$ by immunohistochemistry in 147 gastric cancers and 51 normal gastric epithelium tissues for whom clinical follow-up data were available and correlated the results with clinical characteristics. In addition, we quantified the expression of retinoid receptor mRNA using real-time PCR (RT-PCR) in another 6 gastric adenocarcinoma and 3 normal gastric tissues. From 2008 to 2010, 80 patients with gastric cancers were enrolled onto therapy with all-trans-retinoic acid (ATRA). Results: RARa, $RAR{\beta}$, $RAR{\gamma}$ and $RXR{\gamma}$ positively correlated with each other (p < 0.001) and demonstrated significantly lower levels in the carcinoma tissue sections (p < 0.01), with lower $RAR{\beta}$, $RAR{\gamma}$ and RXRa expression significantly related to advanced stages (p < =0.01). Tumors with poor histopathologic grade had lower levels of RARa and $RAR{\beta}$ in different histological types of gastric carcinoma (p < 0.01). Patients whose tumors exhibited low levels of RARa expression had significantly lower overall survival compared with patients who had higher expression levels of this receptor (p < 0.001, HR=0.42, 95.0% CI 0.24-0.73), and patients undergoing ATRA treatment had significantly longer median survival times (p = 0.007, HR=0.41, 95.0% CI 0.21-0.80). Conclusions: Retinoic acid receptors are frequently expressed in epithelial gastric cancer with a decreased tendency of expression and RARa may be an indicator of a positive prognosis. This study provides a molecular basis for the therapeutic use of retinoids against gastric cancer.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.38 no.2
• /
• pp.249-259
• /
• 2018

More than half of annual rainfall occurs in summer season in Korea due to its climate condition and geographical location. A frequency analysis is mostly adopted for designing hydraulic structure under the such concentrated rainfall condition. Among the various distributions, univariate Gumbel distribution has been routinely used for rainfall frequency analysis in Korea. However, the distributional changes in extreme rainfall have been globally observed including Korea. More specifically, the univariate Gumbel distribution based rainfall frequency analysis is often fail to describe multimodal behaviors which are mainly influenced by distinct climate conditions during the wet season. In this context, we purposed a Gumbel mixture distribution based rainfall frequency analysis with a Bayesian framework, and further the results were compared to that of the univariate. It was found that the proposed model showed better performance in describing underlying distributions, leading to the lower Bayesian information criterion (BIC) values. The mixed Gumbel distribution was more robust for describing the upper tail of the distribution which playes a crucial role in estimating more reliable estimates of design rainfall uncertainty occurred by peak of upper tail than single Gumbel distribution. Therefore, it can be concluded that the mixed Gumbel distribution is more compatible for extreme frequency analysis rainfall data with two or more peaks on its distribution.

• Journal of the Korean Geographical Society
• /
• v.51 no.6
• /
• pp.893-914
• /
• 2016

This paper describes the development of an inquiry-based fieldwork program based on Yangdong village where students explore the ways that it can develop in a sustainable way. Important considerations in an inquiry-based fieldwork design include what the key inquiry questions should be, the geographical issues of fieldwork location, the potential roles of mobile technologies, design of learning activities and a final product, and the roles of a teacher. Student fieldwork activities, including mapping land-use changes at the building level, detecting what should be changed or remain the same, and conducting interview with residents to examine their perceptions of overall tourism impacts, are supported by mobile technologies (i.e., the Collector for ArcGIS and the Google Forms). Twenty one high school students participated in a field test of the program in February 2016, which allowed authors to evaluate the program. Students' pre-, in-, and post-fieldwork activities were observed and the data and final products which they gathered and producted were analyzed. The post-program survey indicated that the students deepened and expanded their understanding of Yangdong village and expressed their satisfaction with the program in general. Incorporating mobile technologies into inquiry-based geographical fieldwork can help students involved in collaborative problem solving and creative activities in real world settings and create a shareable multimodal product combining maps, photo, and text.