• 농업과학연구
• /
• 제47권2호
• /
• pp.205-217
• /
• 2020

In this study, a deep learning-based tillage boundary detection method for autonomous tillage by a tractor was developed, which consisted of image cropping, object classification, area segmentation, and boundary detection methods. Full HD (1920 × 1080) images were obtained using a RGB camera installed on the hood of a tractor and were cropped to 112 × 112 size images to generate a dataset for training the classification model. The classification model was constructed based on convolutional neural networks, and the path boundary was detected using a probability map, which was generated by the integration of softmax outputs. The results show that the F1-score of the classification was approximately 0.91, and it had a similar performance as the deep learning-based classification task in the agriculture field. The path boundary was determined with edge detection and the Hough transform, and it was compared to the actual path boundary. The average lateral error was approximately 11.4 cm, and the average angle error was approximately 8.9°. The proposed technique can perform as well as other approaches; however, it only needs low cost memory to execute the process unlike other deep learning-based approaches. It is possible that an autonomous farm robot can be easily developed with this proposed technique using a simple hardware configuration.

• The Korean Journal of Pain
• /
• 제27권2호
• /
• pp.139-144
• /
• 2014

Background: To manage intractable cancer pain, an alternative to systemic analgesics is neuraxial analgesia. In long-term treatment, intrathecal administration could provide a more satisfactory pain relief with lower doses of analgesics and fewer side-effects than that of epidural administration. However, implantable drug delivery systems using intrathecal pumps in Korea are very expensive. Considering cost-effectiveness, we performed epidural analgesia as an alternative to intrathecal analgesia. Methods: We retrospectively investigated the efficacy, side effects, and complications of epidural morphine and local anesthetic administration through epidural catheters connected to a subcutaneous injection port in 29 Korean terminal cancer patients. Patient demographic data, the duration of epidural administration, preoperative numerical pain rating scales (NRS), side effects and complications related to the epidural catheterization and the drugs, and the numerical pain rating scales on the 1st, 3rd, 7th and 30th postoperative days were determined from the medical records. Results: The average score for the numerical pain rating scales for the 29 patients decreased from $7{\pm}1.0$ at baseline to $3.6{\pm}1.4$ on postoperative day 1 (P < 0.001). A similar decrease in pain intensity was maintained for 30 days (P < 0.001). Nausea and vomiting were the most frequently reported side effects of the epidural analgesia and two patients (6.9%) experienced paresthesia. Conclusions: Epidural morphine and local anesthetic infusion with a subcutaneous pump seems to have an acceptable risk-benefit ratio and allows a high degree of autonomy to patients with cancer pain.

• 대한지역사회영양학회지
• /
• 제11권6호
• /
• pp.779-784
• /
• 2006

Several studies about hospital malnutrition have been reported that about more than 40% of hospitalized patients are having nutritional risk factors and hospital malnutrition presents a high prevalence. People in a more severe nutritional status ended up with a longer length of hospital stay and higher hospital cost. Nutrition screening tools identify individuals who are malnourished or at risk of becoming malnourished and who may benefit from nutritional support. For the early detection and treatment of malnourished hospital patients , few valid screening instruments fur Koreans exist. Therefore, the aim of this study was to develop a simple, reliable and valid malnutrition screening tool that could be used at hospital admission to identify adult patients at risk of malnutrition using medical electrical record data. Two hundred and one patients of the university affiliated medical center were assessed on nutritional status and classified as well nourished, moderately or severely malnourished by a Patient-Generated subjective global assessment (PG-SGA) being chosen as the 'gold standard' for defining malnutrition. The combination of nutrition screening questions with the highest sensitivity and specificity at prediction PG-SGA was termed the nutrition screening index (NSI). Odd ratio, and binary logistic regression were used to predict the best nutritional status predictors. Based on regression coefficient score, albumin less than 3.5 g/dl, body mass index (BMI) less than $18.5kg/m^2$, total lymphocyte count less than 900 and age over 65 were determined as the best set of NSI. By using best nutritional predictors receiver operating characteristic curve with the area under the curve, sensitivity and 1-specificity were analyzed to determine the best optimal cut-off point to decide normal or abnormal in nutritional status. Therefore simple and beneficial NSI was developed for identifying patients with severe malnutrition. Using NSI, nutritional information of the severe malnutrition patient should be shared with physicians and they should be cared for by clinical dietitians to improve their nutritional status.

• 한국정보통신학회논문지
• /
• 제13권10호
• /
• pp.2011-2018
• /
• 2009

인터넷의 등장으로 국가간의 경계는 사라지고 글로벌 자유경제 체제로 전환되는 현 시점에서 사물을 지능화하고 서비스하는 단계인 유비쿼터스 사회는 융합의 산업혁명으로 전개될 것이다. 따라서 향후 도시의 경쟁력은 정치, 경제, 문화관광, 교통, 정보통신, 환경, 방재 등 사회 전반에 활용되는 유비쿼터스 단위별 정보의 질과 양에 따라 좌우될 것으로 예상된다. 이러한 이유로 현재 각 국가와 도시는 유비쿼터스 선도를 주창하며 서비스를 개발하고 네트워크도 광대역 통신망으로 구축하고 있다. 본 논문에서는 우리나라에서 추진하고 있는 도시별 U-City 추진현황과 향후 전망을 분석하고 비용절감과 활용측면을 고려하여 광대역 자가통신 네트워크를 확보하고 구축하는 방안을 연구하였다. 특히, U-City와 자가통신 네트워크를 활용한 융합 모델 개발로 도시 경쟁력을 높이기 위한 정책 방안을 제시한다.

• 상하수도학회지
• /
• 제34권1호
• /
• pp.35-43
• /
• 2020

Turbidity has various effects on the water quality and ecosystem of a river. High turbidity during floods increases the operation cost of a drinking water supply system. Thus, the management of turbidity is essential for providing safe water to the public. There have been various efforts to estimate turbidity in river systems for proper management and early warning of high turbidity in the water supply process. Advanced data analysis technology using machine learning has been increasingly used in water quality management processes. Artificial neural networks(ANNs) is one of the first algorithms applied, where the overfitting of a model to observed data and vanishing gradient in the backpropagation process limit the wide application of ANNs in practice. In recent years, deep learning, which overcomes the limitations of ANNs, has been applied in water quality management. LSTM(Long-Short Term Memory) is one of novel deep learning algorithms that is widely used in the analysis of time series data. In this study, LSTM is used for the prediction of high turbidity(>30 NTU) in a river from the relationship of turbidity to discharge, which enables early warning of high turbidity in a drinking water supply system. The model showed 0.98, 0.99, 0.98 and 0.99 for precision, recall, F1-score and accuracy respectively, for the prediction of high turbidity in a river with 2 hour frequency data. The sensitivity of the model to the observation intervals of data is also compared with time periods of 2 hour, 8 hour, 1 day and 2 days. The model shows higher precision with shorter observation intervals, which underscores the importance of collecting high frequency data for better management of water resources in the future.

• 한국정보통신학회:학술대회논문집
• /
• 한국해양정보통신학회 2009년도 추계학술대회
• /
• pp.657-662
• /
• 2009

인터넷의 등장으로 국가간의 경계는 사라지고 글로벌 자유경제 체제로 전환되는 현 시점에서 사물을 지능화하고 서비스하는 단계인 유비쿼터스 사회는 융합의 산업혁명으로 전개될 것이다. 따라서 향후 도시의 경쟁력은 정치, 경제, 문화관광, 교통, 정보통신, 환경, 방재 등 사회 전반에 활용되는 유비쿼터스 단위별 정보의 질과 양에 따라 좌우될 것으로 예상된다. 이러한 이유로 현재 각 국가와 도시는 유비쿼터스 선도를 주창하며 서비스를 개발하고 네트워크도 광대역 통신망으로 구축하고 있다. 본 논문에서는 우리나라에서 추진하고 있는 도시별 U-City 추진현황과 향후 전망을 분석하고 비용절감과 활용측면을 고려하여 광대역 자가통신 네트워크를 확보하고 구축하는 방안을 연구하였다. 특히, U-City와 자가통신 네트워크를 활용한 융합모델 개발로 도시 경쟁력을 높이기 위한 정책 방안을 제시한다.

• 대한간호학회지
• /
• 제49권2호
• /
• pp.113-125
• /
• 2019

Purpose: The purpose of this study was to examine the effects of a daily life-based physical activity enhancement program performed by middle-aged women at risk for cardiovascular disease. Methods: This study used a randomized control group pretest-posttest design. Middle-aged women aged 45 to 64 were recruited from two outpatient cardiology departments, and randomly assigned to an experimental group (n=28) and a control group (n=30). For the experimental group, after providing one-on-one counseling and education, we provided customized text messages to motivate them in daily life. To monitor the practice of physical activity, they also used an exercise diary and mobile pedometer for 12 weeks. Subjects' physical activities (MET-min/week) were measured using the International Physical Activity Questionnaire (IPAQ). Their physiological data were obtained by blood tests using a portable analyzer, and the data were analyzed using the SPSS 21.0/WIN program. Results: There were significant differences in exercise self-efficacy, health behavior, IPAQ score, body fat, body muscle, and fasting blood sugar between the two groups. However, there were no significant differences in total cholesterol, hemoglobin A1c, high-density lipoprotein cholesterol, and waist-to-hip ratio. Conclusion: Strengthening physical activity in daily life without being limited by cost burden and time and space constraints. Therefore, it is essential to motivate middle-aged women at risk for cardiovascular disease to practice activities that are easily performed in their daily lives.

• Safety and Health at Work
• /
• 제10권2호
• /
• pp.219-223
• /
• 2019

Background: Ice cooling vests can cause tissue damage and have no flexibility. Therefore, these two undesirable properties of ice cooling vest were optimized, and the present study was aimed to compare the impact of the optimized ice cooling vest and a commercial paraffin cooling vest on physiological and perceptual strain under controlled conditions. Methods: For optimizing, hydrogel was used to increase the flexibility and a layer of the ethylene vinyl acetate foam was placed into the inside layer of packs to prevent tissue damage. Then, 15 men with an optimized ice cooling vest, with a commercial paraffin cooling vest, and without a cooling vest performed tests including exercise on a treadmill (speed of 2.8 km/hr and slope of %0) under hot ($40^{\circ}C$) and dry (40 %) condition for 60 min. The physiological strain index and skin temperature were measured every 5 and 15 minutes, respectively. The heat strain score index and perceptual strain index were also assessed every 15 minutes. Results: The mean values of the physiological and perceptual indices differed significantly between exercise with and without cooling vests (P < 0.05). However, the difference of the mean values of the indices except the value of the skin temperature during the exercises with the commercial paraffin cooling vest and the optimized ice cooling vest was not significant (P > 0.05). Conclusions: The optimized ice cooling vest was as effective as the commercial paraffin cooling vest to control the thermal strain. However, ice has a greater latent heat and less production cost.

• 한국치위생학회지
• /
• 제19권4호
• /
• pp.493-502
• /
• 2019

Objectives: This study aimed to investigate the current status of oral health applications developed for smartphones because they can be used as a new educational medium to manage and improve oral health. Methods: This study examined 60 basic oral health applications provided by Google Play Store and Apple App Store as of May 2019 and examined delivery contents, delivery methods, application types, and other information. Results: Apple included 65.4% of oral apps in the game category whereas Android included 64.3% in the education category (p>0.05). All Apple's apps and 71.4% of Android apps were developed overseas (p<0.01). The delivery contents were 61.5% for Brushing + tooth decay in Apple, and 78.6% for others (oral care products and gum diseases) in Android (p>0.05). For the delivery method, game + video was 65.4% in Apple, and game and other methods (text, image, augmented reality) was 42.9% in Android (p>0.05). In the case of application type, play type was the most common with 88.5% in Apple, and 46.4% play type and 39.3% other type (text, appreciation, problem-solving types) in Android (p<0.01). In addition, play type was high in both education (53.8%) and game (90.0%) categories (p>0.05). The average review score was 4.30 in the education category, 4.34 in the case of brushing and care (delivery contents), 4.37 in the case of using game + video (delivery methods), and 4.57 in the case of Play + other types (application type) (p>0.05). Conclusions: The use of healthcare apps is expected to increase owing to improved lifestyles, an increase in the elderly population, cost-effectiveness, and convenience that is not affected by time and place. Effective use of oral health apps will require the participation of dental professionals in the development process to identify the exact status, expand subjects, and provide appropriate information.

• 정보처리학회논문지:소프트웨어 및 데이터공학
• /
• 제11권10호
• /
• pp.427-436
• /
• 2022

산업통상자원부에서 제공하는 KOTRA 무역 데이터는 해당 품목과 해당 국가에 대하여 GDP, 관세율, 비즈니스 점수, 과/차년도 수출금액 등을 제공한다. 그러나 무역 수출품목은 수없이 많을뿐더러 그에 따른 대량의 데이터를 매년 수작업 기반 분석을 통해 유의미한 결과를 이끌어내는 것은 상당히 큰 시간과 비용을 요구한다. 따라서 이번 연구에선 대량의 데이터를 학습하여 단기간에 저비용으로 결과 예측이 가능한 다층 퍼셉트론 모델을 구현하고 성능을 평가하였다. 먼저 딥러닝 기반 무역 수출 가격 예측 모델을 일반적 다변량 회귀 모델과 비교하였을 때, 예측 오류와 학습 시간 측면에서 통계적으로 우수한 성능을 보였다. 수출 가격 데이터는 시계열 속성이 있을 것으로 예상하는 바, 은닉 노드들이 모두 연결된 다층 퍼셉트론과 순환 신경망을 이용하여 수출 가격 데이터를 예측하였다. 그 결과 새로운 데이터에 대해 수출 가격 예측을 위한 일반화 능력은 순환 신경망이 우수한 성능을 보였으나, 다층 퍼셉트론이 무역 수출 가격 예측에서 더 뛰어난 성능을 보였다. 추후 장기간 데이터를 확보한다면, 순환 신경망 혹은 트랜스포머 기반 딥러닝 모델을 이용하여 더 뛰어난 수출 가격 예측이 가능할 것으로 사료된다.