• 한국융합학회논문지
• /
• 제12권10호
• /
• pp.365-372
• /
• 2021

본 연구의 목적은 자연어처리에 의해 생성된 욕창간호진술문의 특성을 파악하고, 욕창 단계판별 예측정확도를 평가하기 위함이다. 욕창관련 간호기록은 서술통계를 이용하여 분석하였고, 워드클라우드 생성기를 활용하여 욕창예방 간호기록에서 단어의 특성을 파악하였다. 딥러닝을 이용하여 욕창단계판별 정확도(accuracy ratio) 를 구하였다. 연구결과, 욕창의 단계에 대한 기록 중 2단계와 심부조직손상의심단계가 각각 23.1% 와 23.0 % 로 가장 많았고, 빈도수가 높은 핵심단어는 홍반, 수포, 가피, 부위, 크기 등으로 나타났다. 예측의 정확도가 높은 단계는 0단계, 심부조직손상의심단계, 2단계 순으로 나타났다. 따라서, 이를 활용하여 임상적 의사결정지지 시스템으로 개발된다면, 임상간호사의 욕창관리역량 향상 전략 개발에 기초가 될 수 있을 것이다.

• 한국응급구조학회지
• /
• 제25권3호
• /
• pp.145-162
• /
• 2021

Purpose: The 2018 General survey of emergency assistance was conducted to examine the working conditions and welfare, including educational direction, interests, and awareness of work, of the fire department emergency medical technicians (EMT). This would be used as basic data for future policy directions. Methods: Among the fire-fighting officers in 16 cities nationwide, emergency rescue workers engaged in first-aid activities were targeted. With prior consent, a survey was conducted through electronic documents. Of the total 1,227 people, responses from 1,151 were finally analyzed, excluding 76 who did not respond appropriately. Results: The working conditions and welfare of 119 firefighters were moderate, but in the fields of education and interest, the learning according to the regulations was high. In particular, satisfaction with the scope of work was found to be below average. However, it was positive that it will play a role as a social safety net in the future and will converge with cutting-edge science. Conclusion: Although this study was a total investigation of the EMT survey, conducting an EMT survey on all fire fighters in Korea is difficult. Further research is needed, particularly on first-class emergency medical personnel who play a major role in 119 paramedics.

• 한국융합학회논문지
• /
• 제10권1호
• /
• pp.7-12
• /
• 2019

신경회로망은 적합한 수학적 모델에 대한 가정 없이 데이터로부터 유용한 정보를 추출해서 예측에 필요한 입출력 관계를 정의할 수 있어서 주가 예측에 널리 사용되어 왔다. 본 논문에서는 신경회로망 모델을 사용하여 일별 KOrea composite Stock Price Index (KOSPI) 종가를 예측한다. 예측된 종가를 기반으로 KOSPI에 연동해 변동하는 Exchange Traded Funds (ETFs)의 거래를 위한 알파 매매를 제안한다. 본 논문에 제안된 방법으로 KOSPI 예측 신경회로망 모델들을 구현하고 예측 정확도를 평가한다. 구현된 신경회로망 모델(NN1)의 학습 오차(MAPE)는 0.427, 평가 오차는 0.627이다. 평가용 데이터를 사용해 알파 매매를 시뮬레이션하면 수익률은 7.16 ~ 15.29 %를 보인다. 이는 125 거래일 데이터로 거둔 수익률로 제안된 알파 매매가 효과적임을 보인다.

• 한국정보통신학회논문지
• /
• 제23권5호
• /
• pp.547-554
• /
• 2019

2015년 교육과정 개편으로 중학교 정보교과가 필수로 지정됨에 따라 초중고등학교는 물론 대학에서도 프로그래밍 교육을 강화하고 있다. 아두이노는 프로그래밍 교육을 위해 사용되는 대표적인 도구 중 하나로 그 유용성은 다양한 사례 연구를 통해 입증되었다. 하지만 기존 아두이노 기반 키트는 연결 방법이 복잡하거나, 확장성이 떨어지는 등 하드웨어 의존적인 단점을 가지고 있다. 이러한 문제점을 보완하기 위한 방안으로 모듈 기반의 구조를 가지고, 동일한 인터페이스를 통해 기능 확장이 가능하며, 다양한 수준에서 학습에 사용할 수 있는 아두이도 호환 키트 설계를 제안하였다. 이 논문에서는 제안한 키트의 요구 조건을 만족시키는 FRUTO 키트와 이를 사용하기 위한 소프트웨어 구현 방법을 설명한다. FRUTO 키트는 몇 번의 디자인 변경을 통해 현재의 형태로 결정되었으며, 현재 출시 전 테스트 진행을 준비 중에 있다.

• International Journal of Advanced Culture Technology
• /
• 제8권4호
• /
• pp.82-88
• /
• 2020

With the development of medical care in the 21st century and the rapid development of the 4th industry, electronic devices and household goods taking into account the physical and mental aging of the silver generation have been developed, and apps related to health and health are generally developed and operated. The apps currently used by the silver generation are a form that provides information on diseases by focusing on prevention rather than treatment, such as safety management apps for the elderly living alone and methods for preventing diseases. There are not many apps that provide information on foods that have a direct effect and nutrients in that food, and research on apps that can obtain information about individual foods is insufficient. In this paper, we propose an app that analyzes food factors and provides self-medication for health promotion of the silver generation. This app allows the silver generation to conveniently and easily obtain information such as nutrients, calories, and efficacy of food they need. In addition, this app collects/categorizes healthy food information through a textom solution-based crawling agent, and stores highly relevant words in a data resource. In addition, wide deep learning was applied to enable self-medication recommendations for food. When this technique is applied, the most appropriate healthy food is suggested to people with similar eating patterns and tastes in the same age group, and users can receive recommendations on customized healthy foods that they need before eating. This made it possible to obtain convenient healthy food information through a customized interface for the elderly through a smartphone.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제15권5호
• /
• pp.1814-1828
• /
• 2021

Low-light image enhancement is a key technique to overcome the quality degradation of photos taken under scotopic vision illumination conditions. The degradation includes low brightness, low contrast, and outstanding noise, which would seriously affect the vision of the human eye recognition ability and subsequent image processing. In this paper, we propose an approach based on deep learning and Retinex theory to enhance the low-light image, which includes image decomposition, illumination prediction, image reconstruction, and image optimization. The first three parts can reconstruct the enhanced image that suffers from low-resolution. To reduce the noise of the enhanced image and improve the image quality, a super-resolution algorithm based on the Laplacian pyramid network is introduced to optimize the image. The Laplacian pyramid network can improve the resolution of the enhanced image through multiple feature extraction and deconvolution operations. Furthermore, a combination loss function is explored in the network training stage to improve the efficiency of the algorithm. Extensive experiments and comprehensive evaluations demonstrate the strength of the proposed method, the result is closer to the real-world scene in lightness, color, and details. Besides, experiments also demonstrate that the proposed method with the single low-light image can achieve the same effect as multi-exposure image fusion algorithm and no ghost is introduced.

• 한국정보전자통신기술학회논문지
• /
• 제14권4호
• /
• pp.328-337
• /
• 2021

석면, 벤젠과 같이 발암물질 1급인 미세먼지는 각종 질병에 원인이 되고 있다. 초 미세먼지 확산은 코로나 바이러스 확산의 중요한 원인중 하나이다. 본 논문은 2015년부터 2019년까지 서울시 평균 기온, 강수량, 평균 풍속등의 기상 데이터와 SO2, NO2, O3,등의 대기 환경 데이터를 기반으로 미세먼지와 초 미세먼지를 분석하고 예측한다. 계절별과 월별로 미세먼지와 초미세먼지 현황을 파악·분석하며 미세먼지를 예측하기 위해 기계학습 모델 중 선형회귀, SVM, 앙상블 모델을 이용하여 비교 분석하였다. 또한 미세먼지와 초 미세먼지 발생에 영향을 미치는 중요한 피쳐(속성)를 파악한다. 본 논문이 파악한 결과 3월에 가장 (초)미세먼지가 높았고, 8월에서 9월까지 (초)미세먼지가 가장 낮았다. 기상 데이터일 경우 (초)미세먼지에 가장 영향을 미치는 데이터가 평균 기온이며, 기상 데이터와 대기 환경 데이터일 경우 NO2가 (초)미세먼지 발생에 가장 크게 작용하였다.

• 대한의용생체공학회:의공학회지
• /
• 제42권6호
• /
• pp.268-276
• /
• 2021

Functional near-infrared spectroscopy-based brain-computer interface (fNIRS-based BCI) has been receiving much attention. However, we are practically constrained to obtain a lot of fNIRS data by inherent hemodynamic delay. For this reason, when employing machine learning techniques, a problem due to the high-dimensional feature vector may be encountered, such as deteriorated classification accuracy. In this study, we employ an elastic net-based feature selection which is one of the embedded methods and demonstrate the utility of which by analyzing the results. Using the fNIRS dataset obtained from 18 participants for classifying brain activation induced by mental arithmetic and idle state, we calculated classification accuracies after performing feature selection while changing the parameter α (weight of lasso vs. ridge regularization). Grand averages of classification accuracy are 80.0 ± 9.4%, 79.3 ± 9.6%, 79.0 ± 9.2%, 79.7 ± 10.1%, 77.6 ± 10.3%, 79.2 ± 8.9%, and 80.0 ± 7.8% for the various values of α = 0.001, 0.005, 0.01, 0.05, 0.1, 0.2, and 0.5, respectively, and are not statistically different from the grand average of classification accuracy estimated with all features (80.1 ± 9.5%). As a result, no difference in classification accuracy is revealed for all considered parameter α values. Especially for α = 0.5, we are able to achieve the statistically same level of classification accuracy with even 16.4% features of the total features. Since elastic net-based feature selection can be easily applied to other cases without complicated initialization and parameter fine-tuning, we can be looking forward to seeing that the elastic-based feature selection can be actively applied to fNIRS data.

• 여성건강간호학회지
• /
• 제28권3호
• /
• pp.174-186
• /
• 2022

Purpose: This study aimed to investigate the effects of virtual reality used in maternal-child nursing clinical practicums on nursing students' competencies through a systematic review. Methods: The inclusion criteria were peer-reviewed papers in English or Korean presenting analytic studies of maternal-child nursing practicums using virtual reality. An electronic literature search of the Cochrane Library, CINAHL, EMBASE, ERIC, PubMed, and Research Information Sharing System databases was performed using combinations of the keywords "nursing student," "virtual reality," "augmented reality," "mixed reality," and "virtual simulation" from February 4 to 15, 2022. Quality appraisal was performed using the RoB 2 and ROBINS-I tools for randomized controlled trials (RCTs) and non-RCTs, respectively. Results: Of the seven articles identified, the RCT study (n=1) was deemed to have a high risk of bias, with some items indeterminable due to a lack of reported details. Most of the non-RCT studies (n=6) had a moderate or serious risk of bias related to selection and measurement issues. Clinical education using virtual reality had positive effects on knowledge, skills, satisfaction, self-efficacy, and needs improvement; however, it did not affect critical thinking or self-directed learning. Conclusion: This study demonstrated that using virtual reality for maternal-child nursing clinical practicums had educational effects on a variety of students' competencies. Considering the challenges of providing direct care in clinical practicums, virtual reality can be a viable tool that supplements maternal-child nursing experience. Greater rigor and fuller reporting of study details are required for future research.

• 한국기록관리학회지
• /
• 제21권4호
• /
• pp.65-80
• /
• 2021

기록관리전문가의 역할은 전통적인 기록관리 뿐 아니라 정보기술을 기반으로 전자기록관리까지 주도적으로 실행하는 역할로 확장되고 있다. 본 연구에서는 기록관리전문가 양성과정인 기록관리학 대학원 교육지침서의 구성요소, 지식범주, 지식내용의 적절성과 타당성을 검증하기 위하여 전문가 델파이 평가를 실시하였다. 전문가 평가 결과 교육지침서의 구성요소는 서문, 교육 내용, 교원 요건, 교육 과정, 행정적 지원, 결론이고 교육 내용은 8개 지식범주와 44개 지식내용으로 구성되었다. 본 연구는 교육지침서 모형 개발을 통해 기록관리학 대학원에서 준수해야 하는 최소한의 기준인 구성요소를 제시하고, 이론과 실무 역량을 갖춘 기록관리전문가 양성을 위한 교육내용을 제시하였다. 본 연구 모형은 향후 교육지침서 제정을 위한 기초자료로 활용될 수 있을 것이다.