• Proceedings of the Korea Information Processing Society Conference
• /
• 2020.11a
• /
• pp.784-787
• /
• 2020

폐암 환자의 생존율을 예측할 때 미국암연합회(AJCC)의 TNM병기 분류체계에 의해 진단되는 최종병기를 많이 사용한다. 최종병기는 폐암환자의 임상데이터 중 하나로 종양의 위치, 크기, 전이정도를 고려하여 환자의 폐암 상태를 판별하는 정보이다. 최종병기는 개략적인 환자의 상황을 설명하는 데 효과적이지만, 보다 구체적인 생존분석을 위해서는 임상데이터 뿐만 아니라 PET/CT와 같은 영상 데이터를 함께 분석해야 한다. 이 논문에서는 데이터 과학적 접근을 통해 폐암환자의 임상데이터, CT영상과 PET영상 등 다양한 종류의 데이터를 함께 활용하는 생존분석기법을 검토한다. 실험을 통해 다중 모드 데이터를 활용하는 생존분석을 위해 비선형모델 개발과 Feature임베딩 기법 고도화가 필요함을 확인하였다.

• Phonetics and Speech Sciences
• /
• v.15 no.4
• /
• pp.109-118
• /
• 2023

Cardiac arrest is a critical medical emergency where immediate response is essential for patient survival. This is especially true for Out-of-Hospital Cardiac Arrest (OHCA), for which the actions of emergency medical services in the early stages significantly impact outcomes. However, in Korea, a challenge arises due to a shortage of dispatcher who handle a large volume of emergency calls. In such situations, the implementation of a machine learning-based OHCA detection program can assist responders and improve patient survival rates. In this study, we address this challenge by developing a machine learning-based OHCA detection program. This program analyzes transcripts of conversations between responders and callers to identify instances of cardiac arrest. The proposed model includes an automatic transcription module for these conversations, a text-based cardiac arrest detection model, and the necessary server and client components for program deployment. Importantly, The experimental results demonstrate the model's effectiveness, achieving a performance score of 79.49% based on the F1 metric and reducing the time needed for cardiac arrest detection by 15 seconds compared to dispatcher. Despite working with a limited dataset, this research highlights the potential of a cardiac arrest detection program as a valuable tool for responders, ultimately enhancing cardiac arrest survival rates.

• Proceedings of the Korean Society of Applied Pharmacology
• /
• 1993.04a
• /
• pp.170-170
• /
• 1993

유전독성물절외 검출과 평가에 용이하게 사용할 수 있는 모델 시스템의 개발 및 DNA 회복기작을 규명할 목적으로 수종의 돌연변이, 발암원을 이용하여 배양 포유동물세포 및 어류세포에서 세포생존률, DNA 합성 및 복제억제의 양상 등을 비교 분석하였다. MMS및 MNNG 와 같은 알칼라제는 CHO 세포에서 유의한 DNA 합성저해, DNA 복재억제, DNA 단사절단 및 비주기성 DNA 합성률의 증가를 유발하였다. Benzo(a)pyrene과 3-methylcholanthrene좌 같은 DNA 상해 전구물질의 경우 유전독성 여부의 판정에는 반드시 S-9/15과 같은 대사활성계 또는 mouse embryonic fibroblast와 같은 대사 활성능이 있는 세포와의 co-culture system들이 필요함을 확인하였으며, 이들에 의한 DNA 상해와 복재억제 유도의 작용양상은 자외선의 작용양상과 유사하였다. 배양 어류세포에서 자외선에 의한 세포생존율의 측정, 광재활성능의 분석 및 자외선에 의해 유발된 피리미딘 이량체 절제능 검토 및 DNA 합성 저해능의 결과를 분석함으로써 유전독성 평가를 위한 모델 시스템 구축의 기초결과를 얻었다.

• Proceedings of the Korea Society for Simulation Conference
• /
• 1998.10a
• /
• pp.182-186
• /
• 1998

최근 산업체, 공공기관 등 거의 모든 분야에서 인터넷 기술을 이용한 클라이언트/서버환경으로 시스템을 구축하고 있다. 본 논문에서는 이러한 환경에서 DIS와 ADS의 개념을 이용한 전투 시뮬레이션을 수행하여 기존 모델의 비현실적인 가정과 문제의 난이도로 인한 여러 제약을 해결할 수 있으며 시간 및 시스템의 활용면에서 효율의 증가에 대한 모델을 구현하였다. 본 전투 시뮬레이션 모델은 인터넷 프로토콜의 표준으로 자리하고 있는 TCP/IP를 이용한 것이기 때문에 이동통신을 이용한 무선 데이터는 물론, 인터넷 접속이 가능한 곳이라면 어떠한 환경에서도 수행이 가능하다. 노드에서 시뮬레이션이 종료되면 여기서 발생한 생존 대수, 발사시간 분포 등의 정보는 서버에서 적절한 의사결정을 수행한 후 다른 노드로 전달하게 된다.

• Journal of the Korean Institute of Gas
• /
• v.25 no.3
• /
• pp.53-58
• /
• 2021

The number of patients with cardiovascular diseases who experience an out-of-hospital cardiac arrest (OHCA) are increasing among young adults as well as the aged population. An automated external defibrillator (AED) is vital in improving survival rates of OHCA victims. Survival rates of OHCA were shown to decline exponentially in time to defibrillation, yet studies in Korea are uncommon that captures the properties of their survival rates in examining optimal locations of AEDs. In this study, we worked on the maximal gradual coverage location problem (MGCLP) with exponential decay coverage function to decide on their optimal locations. The exponential decay coverage function mitigates the drawback of over-estimating survival rates of OHCA patients. It is expected that a more sophisticated facility location problem will be developed to identify the "emergent" characteristics of pedestrians who responds to the OHCA occurrence by incorporating random pedestrian locations and movement through simulation.

• Proceedings of the Korean Information Science Society Conference
• /
• 2004.10c
• /
• pp.10-12
• /
• 2004

무선 센서네트워크는 센싱 지역에 분산되어 있는 초소형 센서 노드들이 감지/처리한 데이터를 수집노드로 전송하여 원격의 사용자가 센싱 지역의 상황인지를 가능하게 하는 유비쿼터스 컹퓨팅의 기반 네트워크이다. 전력 사용이 극히 제한된 센서 노드를 이용하여 무선 센서 망을 구성, 유지하며 데이터를 수집하기 위해서는 효율적인 망의 형태와 이에 따른 네트워크 배치 전략을 필요로 한다. 클러스터 기반 네트워크의 형태는 밀집도가 높은 센서네트워크에서 데이터병합을 수행하고 노드간 에너지 소비 균형을 이루기 위한 효과적인 구조이다. 본 논문에서는 클러스터 기반의 단일 홉 전송 구조에서 데이터 수집률을 높이고 노드간 에너지 사용의 균형을 이루게 하여 네트워크의 생존시간을 최대화할 수 있는 네트워크 배치 방법을 제안하였다. 클러스터링 기법에 따른 에너지 소비 모델을 분석하고 이를 통해 노드의 적절한 밀집도를 산출하였다. 싱크로부터 멀어질수록, 센서필드의 중앙보다는 외곽에 노드의 배치를 조밀하게 배치하여 네트워크의 생존시간을 늘렸다. 시뮬레이션을 통해 효율적인 네트워크의 배치가 노드간 에너지 소비의 균형을 이루도록 하여 네트워크의 생존시간을 늘일 수 있음을 확인하였다.

• The Korean Journal of Applied Statistics
• /
• v.25 no.2
• /
• pp.279-291
• /
• 2012

When fitting a Cox proportional hazards model with missing covariates, it is inefficient to exclude observations with missing values in the analysis. Furthermore, if the missing-data mechanism is not Missing Completely At Random(MCAR), it may lead to biased parameter estimation. Many approaches have been suggested to handle the Cox proportional hazards model when covariates are sometimes missing, but they are based on the selection model. This paper suggest an approach to handle Cox proportional hazards model with missing covariates by using the pattern-mixture model (Little, 1993). The pattern-mixture model is expressed by the joint distribution of survival time and the missing-data mechanism. In the pattern-mixture model, many models can be considered by setting up various restrictions, and different results under various restrictions indicate the sensitivity of the model due to missing covariates. A simulation study was conducted to show the sensitivity of parameter estimation under different restrictions in a pattern-mixture model. The proposed approach was also applied to mouse leukemia data.

• Journal of the Korean Society of Marine Environment & Safety
• /
• v.23 no.5
• /
• pp.473-481
• /
• 2017

Seafarer sea survival training, such as basic safety refresher training and advanced safety refresher training, in accordance with the STCW Convention, is an indispensable program that can increase the crew survival rate during emergency situations at sea. It is important for crew members to carry out theoretical and practical training with various safety equipment in order to effectively train according to IMO model courses. Therefore, this study suggests the following measures to improve safety training facilities for seafarers by reviewing survival training requirements based on the IMO model course and comparing and analyzing related facilities based on operating cases from domestic and overseas training institutes. First, it is necessary to establish a training environment where seafarers can practice utilizing various, updated safety equipment such as marine evacuation equipment (slides, chutes, etc.). Second, it is necessary to construct an educational environment in which learners can directly or indirectly experience realistic emergency situations by installing marine environment simulation facilities with such equipment as a wave generator, rain fall device, wind generating device, etc. Third, it is also necessary to develop and expand customized training using virtual reality equipment in addition to experiential training, audiovisual training and simulation training.

• Korean Journal of Social Welfare
• /
• v.37
• /
• pp.1-31
• /
• 1999

In Korea, little is known about unemployment duration and exit rate from unemployment. This paper empirically examines the duration of unemployment using data for the years 1996 and 1997 on unemployed individuals who are eligible for unemployment insurance benefits in Korea. A parametric survival model (log-normal model) is adopted to identify factors predicting transitions to reemployment. Factors that affect unemployment duration are sex, age, employment duration (year), prior salary, region, prior employment industry, cause of unemployment, officially determined unemployment benefit duration, degree of benefit exhaustion, and amount of benefits for early reemployment. However, education is not statistically significant In degree of benefit exhaustion, the exit rate from unemployment decreases as benefit exhaustion is approached. In amount of benefits for early reemployment, the exit rate from unemployment increases as amount of benefits increases. Hazards for reemployment gradually increase until 80 days after unemployment and gradually decrease in the following period. Thus, we find that distribution of hazards for reemployment has log-normal shapes between inverted U and inverted L This paper takes advantage of a unique analysis about unemployment duration and exit rate from unemployment in the Korean Unemployment Insurance system which functions as the most valuable social safely-net mechanism in the recent national economic crisis. Indeed, this paper provides a basic knowledge about realities of unemployed individuals in the Unemployment Insurance system and identifies research areas that require further study.

• Smart Media Journal
• /
• v.12 no.8
• /
• pp.18-25
• /
• 2023

The COVID-19 outbreak in 2020 caused continuous damage worldwode, especially the smart tourism industry was hit directly by the blockade of sky roads and restriction of going out. At a time when overseas travel and domestic travel have decreased significantly, the number of tourist hotels that are colsed and closed due to the continued deficit is increasing. Therefore, in this study, licensing data from the Ministry of Public Administraion and Security were collected and visualized to understand the operation status of the tourism and lodging industry. The machine learning classification algorithm was applied to implement the business status prediction model of the tourist hotel, the performance of the prediction model was optimized using the ensemble algorithm, and the performance of the model was evaluated through 5-Fold cross-validation. It was predicted that the survival rate of tourist hotels would decrease somewhat, but the actual survival rate was analyzed to be no different from before COVID-19. Through the prediction of the business status of the hotel industry in this paper, it can be used as a basis for grasping the operability and development trends of the entire tourism and lodging industry.