• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.5
• /
• pp.662-667
• /
• 2022

Semantic role labeling task(SRL) is to extract predicate and arguments such as agent, patient, place, time. In the previously SRL task studies, a pipeline method extracting linguistic features of sentence has been proposed, but in this method, errors of each extraction work in the pipeline affect semantic role labeling performance. Therefore, methods using End-to-End neural network model have recently been proposed. In this paper, we propose a neural network model using the Biaffine Average Attention model for SRL task. The proposed model consists of a structure that can focus on the entire sentence information regardless of the distance between the predicate in the sentence and the arguments, instead of LSTM model that uses the surrounding information for prediction of a specific token proposed in the previous studies. For evaluation, we used F1 scores to compare two models based BERT model that proposed in existing studies using F1 scores, and found that 76.21% performance was higher than comparison models.

• The Journal of the Korea Contents Association
• /
• v.22 no.3
• /
• pp.354-360
• /
• 2022

Metabolic disease is a disease with a prevalence of 26% in Korean, and has three of the five states of abdominal obesity, hypertension, hunger glycemic disorder, high neutral fat, and low HDL cholesterol at the same time. This paper links the consumer panel data of the Rural Development Agency(RDA) and the medical care data of the National Health Insurance Service(NHIS) to generate a classification model that can be divided into a metabolic disease group and a control group through food consumption characteristics, and attempts to compare the differences. Many existing domestic and foreign studies related to metabolic diseases and food consumption characteristics are disease correlation studies of specific food groups and specific ingredients, and this paper is logistic considering all food groups included in the general diet. We created a classification model using regression, a decision tree-based classification model, and a classification model using XGBoost. Of the three models, the high-precision model is the XGBoost classification model, but the accuracy was not high at less than 0.7. As a future study, it is necessary to extend the observation period for food consumption in the patient group to more than 5 years and to study the metabolic disease classification model after converting the food consumed into nutritional characteristics.

• Progress in Medical Physics
• /
• v.29 no.1
• /
• pp.8-15
• /
• 2018

This work reports the acceptance testing and commissioning experience of the Robotic Intensity-Modulated Radiation Therapy (IMRT) M6 system with a newly released $InCise^{TM}2$ Multileaf Collimator (MLC) installed at the Yonsei Cancer Center. Acceptance testing included a mechanical interdigitation test, leaf positional accuracy, leakage check, and End-to-End (E2E) tests. Beam data measurements included tissue-phantom ratios (TPRs), off-center ratios (OCRs), output factors collected at 11 field sizes (the smallest field size was $7.6mm{\times}7.7mm$ and largest field size was $115.0mm{\times}100.1mm$ at 800 mm source-to-axis distance), and open beam profiles. The beam model was verified by checking patient-specific quality assurance (QA) in four fiducial-inserted phantoms, using 10 intracranial and extracranial patient plans. All measurements for acceptance testing satisfied manufacturing specifications. Mean leaf position offsets using the Garden Fence test were found to be $0.01{\pm}0.06mm$ and $0.07{\pm}0.05mm$ for X1 and X2 leaf banks, respectively. Maximum and average leaf leakages were 0.20% and 0.18%, respectively. E2E tests for five tracking modes showed 0.26 mm (6D Skull), 0.3 mm (Fiducial), 0.26 mm (Xsight Spine), 0.62 mm (Xsight Lung), and 0.6 mm (Synchrony). TPRs, OCRs, output factors, and open beams measured under various conditions agreed with composite data provided from the manufacturer to within 2%. Patient-specific QA results were evaluated in two ways. Point dose measurements with an ion chamber were all within the 5% absolute-dose agreement, and relative-dose measurements using an array ion chamber detector all satisfied the 3%/3 mm gamma criterion for more than 90% of the measurement points. The Robotic IMRT M6 system equipped with the $InCise^{TM}2$ MLC was proven to be accurate and reliable.

• Journal of the Korean Orthopaedic Association
• /
• v.56 no.2
• /
• pp.103-116
• /
• 2021

The use of 3-dimensional (3D) printing is becoming more common, and its use is increasing in the orthopedic surgery. Currently, there are four major methods of using 3D printing technology in orthopedic surgery. First, surgical planning simulation using 3D printing model; second, patient-specific surgical instruments; third, production of customized prosthesis using 3D printing technique; fourth, patient-specific prosthesis produced by 3D printing. The areas of orthopedic surgery where 3D printing technology can be used are shoulder joint, spine, hip and pelvis, knee joints, ankle joint, and tumors. Since the diseases and characteristics handled by each area are different, the method of using 3D printing technology is also slightly different in each area. However, using 3D printing technology in all areas can increase the efficiency of surgery, shorten the surgery time, and reduce radiation exposure intraoperatively. 3D printing technology can be of great help in treating patients with particularly complex and difficult orthopedic diseases or fractures. Therefore, the orthopedic surgeon should make the most of the benefits of the 3D printing technology so that patient can be treated effectively.

• Korean Journal of Computational Design and Engineering
• /
• v.18 no.6
• /
• pp.385-398
• /
• 2013

Opening wedge high tibial osteotomy (OWHTO) is widely used to treat unicompartmental osteoarthritis of the knee caused by degenerative deformations of the anatomical axes of the leg. However, since it is difficult to accurately plan the surgical degrees of adjustment such as coronal correction angle and tibial posterior slope angle to align the axes before the actual procedure, a number of studies have proposed analytical models to solve this problem. While previous analytical models for OWHTO were limited to specific cases, this study proposes an analytical osteotomy model (AOM) and a surgical planning system (SPS) that are suitable for a wide range of tibial morphologies and tibia conditions. The validity and generality of the model were verified in a total of 60 OWHTO cases. Results of the test showed that, as predicted, surgical degrees are affected quite significantly by tibia shape and slope of the resected surface. Comparison of the required surgical degrees and the degrees estimated from virtual surgery simulations using AOM showed a very small average difference of $0.118^{\circ}$. SPS, based on AOM, allows the operating surgeon to easily calculate surgical parameters needed to treat a patient.

• Journal of the Korean Data and Information Science Society
• /
• v.28 no.2
• /
• pp.349-359
• /
• 2017

In this paper, we conducted survival analyses by fitting the Cox proportional hazards model to stage III proximal colon cancer data obtained from the Surveillance, Epidemiology, and End Results program of the National Cancer Institute. We investigated the effect of covariates on the hazard function for death from proximal colon cancer in stage III with surgery performed and estimated the survival probability for a patient with specific covariates. We showed that the proportional hazards assumption is satisfied for covariates that were used to analyses, using a test based on the Schoenfeld residuals and plots of the Schoenfeld residuals and $log[-log\{{\hat{S}}(t)\}]$. We evaluated the model calibration and discriminatory accuracy by calibration plot and time-dependent area under the ROC curve, which were calculated using 10-fold cross validation.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.10
• /
• pp.1296-1301
• /
• 2021

With the recent increase in diabetes incidence worldwide, research has been conducted to predict diabetes through various machine learning and deep learning technologies. In this work, we present a model for predicting diabetes using machine learning techniques with German Frankfurt Hospital data. We apply outlier handling using Interquartile Range (IQR) techniques and Pearson correlation and compare model-specific diabetes prediction performance with Decision Tree, Random Forest, Knn (k-nearest neighbor), SVM (support vector machine), Bayesian Network, ensemble techniques XGBoost, Voting, and Stacking. As a result of the study, the XGBoost technique showed the best performance with 97% accuracy on top of the various scenarios. Therefore, this study is meaningful in that the model can be used to accurately predict and prevent diabetes prevalent in modern society.

• International Journal of Computer Science & Network Security
• /
• v.23 no.8
• /
• pp.17-25
• /
• 2023

The alarming global prevalence of Type 2 Diabetes Mellitus (T2DM) has catalyzed an urgent need for robust, early diagnostic methodologies. This study unveils a pioneering approach to predicting T2DM, employing the Extreme Gradient Boosting (XGBoost) algorithm, renowned for its predictive accuracy and computational efficiency. The investigation harnesses a meticulously curated dataset of 4303 samples, extracted from a comprehensive Chinese research study, scrupulously aligned with the World Health Organization's indicators and standards. The dataset encapsulates a multifaceted spectrum of clinical, demographic, and lifestyle attributes. Through an intricate process of hyperparameter optimization, the XGBoost model exhibited an unparalleled best score, elucidating a distinctive combination of parameters such as a learning rate of 0.1, max depth of 3, 150 estimators, and specific colsample strategies. The model's validation accuracy of 0.957, coupled with a sensitivity of 0.9898 and specificity of 0.8897, underlines its robustness in classifying T2DM. A detailed analysis of the confusion matrix further substantiated the model's diagnostic prowess, with an F1-score of 0.9308, illustrating its balanced performance in true positive and negative classifications. The precision and recall metrics provided nuanced insights into the model's ability to minimize false predictions, thereby enhancing its clinical applicability. The research findings not only underline the remarkable efficacy of XGBoost in T2DM prediction but also contribute to the burgeoning field of machine learning applications in personalized healthcare. By elucidating a novel paradigm that accentuates the synergistic integration of multifaceted clinical parameters, this study fosters a promising avenue for precise early detection, risk stratification, and patient-centric intervention in diabetes care. The research serves as a beacon, inspiring further exploration and innovation in leveraging advanced analytical techniques for transformative impacts on predictive diagnostics and chronic disease management.

• Quality Improvement in Health Care
• /
• v.21 no.1
• /
• pp.36-49
• /
• 2015

In Intensive Care Units (ICUs), where severely ill patients are treated, importance of reducing Hospital Acquired Infection (HAI) cannot be overstated. One of the simplest and most effective actions against HAI is proper hand hygiene (HH) behavior of Health Care Workers (HCWs). However, compliance varies across different cultures and different job types of HCWs (physicians, residents and nurses). This study aims to understand determinants of HH behavior by HCWs' job types in Korea. Qualitative analysis was performed based on Reasoned Action Approach style interviews with staff physicians, residents and nurses across 7 teaching hospitals. We found that all HCWs strongly believe HH is important in reducing HAI. There were, however, job type-specific HH behavior modifying factors; staff physicians stated feeling pressure to be HH behavior role model. Residents identified Quality Improvement team that measured compliance as a facilitator; a notable barrier for residents was senior physicians not washing their hands, because they were afraid of appearing impudent to their seniors. Nurses designated their chief nurse as a key referent. All participants mentioned heavy workload and lack of access to alcohol-based sanitizer as situational barriers, and sore and dry hand as deterrents to HH compliance.

• Journal of Digital Convergence
• /
• v.15 no.7
• /
• pp.367-373
• /
• 2017

The study aims to suggests mobile application service that helps to reduce physical and mental suffering after surgery for breast cancer patients constantly increasing. At first, I analyze literature research to understand the breast cancer and similar services for achieving study purposes. After I finally design a mobile application that offers to customed one-to-one specialist coaching service focused on post-treatment management by using a double diamond design process model. In detail, first, I analyze the current program and comprehend user needs through interviews with the person such as breast patient, protector, nurse, salesperson in cancer center. Second, I specify the entire scenario, concept and keyword based on the persona and the customer journey map representing user. Third, I propose the mobile application for the ultimate goal by producing prototype to effectively communicate with specific functional descriptions. I hope this service will improve the quality of their lives.