• 한국멀티미디어학회논문지
• /
• 제12권7호
• /
• pp.1032-1042
• /
• 2009

본 연구는 태깅된 웹 이미지의 검색에서 태그들의 의미정보를 미리 추출하여 검색 시에 이용하고자 하는 것이다. 일반적으로 웹 이미지의 태그들은 사용자들에 의해 순서 구분 없이 무작위로 매겨지며 많게는 그 수가 100여개에 이른다. 본 논문에서는 이 태그들 간에 의미정보가 많이 공유된 것일수록 해당 이미지를 설명하는 중요 태그가 될 것임에 착안하여 이미지와 태그 정보가 업 로드되는 시점에 중요도에 따른 우선 태그를 결정하고 이를 검색에 활용하는 방법을 소개한다 제안된 방법은 워드넷에 기반하여 태그의 연관성점수를 계산하고 이를 이용하여 다단계 검색으로 태징된 웹 이미지를 검색한다. 평가를 위하여 제안된 방법으로 검색된 결과와 검색어와 태그의 단순 비교방식인 기존의 검색을 비교하였으며 실험 결과, 정확도와 재현율에서 본 시스템의 우수함을 확인할 수 있었다.

• 대한지역사회영양학회지
• /
• 제10권6호
• /
• pp.860-868
• /
• 2005

Anemia diagnosed early in pregnancy is associated with increased risks of low birth weight and preform delivery. The purposes of this study were to assess the maternal iron status during pregnancy and to evaluate the relationships between the iron indices of maternal-umbilical cord serum iron and ferritin levels and pregnancy outcomes. Dietary intakes of the pregnant women were estimated by 24 hour-recall (3 times). Serum iron and ferritin levels in maternal blood and umbilical cord were measured at 1st-, 2nd-, 3rd- trimester and delivery, respectively. The mean of maternal se겨m iron levels of the trimester and delivery were $124.27\;{\mu}g/dl,\;97.03\;{\mu}g/dl,\;94.32\;{\mu}g/dl,\;and\;145.53\;{\mu}g/dl$. Those maternal levels were significantly lower than that of umbilical cord blood ($222.59\;{\mu}g/dl$). Serum ferritin levels of maternal trimester and delivery were 22.68 $22.68\;{\mu}g/l,\;11.09\;{\mu}g/l,\;14.18\;{\mu}g/l,\;and\;\;24.54\;{\mu}g/l$, which were significantly lower than those of umbilical cord blood ($184.35\;{\mu}g/l$) (p < 0.0001). This prevalence of anemia of total subjects was $30.3\%$ by WHO criteria (Hb < 11.0 g/dl, Hct < $33\%$). Iron levels of 2nd-trimester was significantly higher in the normal group than in the anemia group. And ferritin levels of 3rd-trimester and delivery was significantly higher in the normal group than in the anemia group. Therefore, we suggest for successful pregnancy outcome and delivery differential iron supplementation programs will be carried out with individual Pregnant women on the basis of pre-Pregnancy nutrition. (Korean J Community Nutrition 10(6) : $860\∼868$, 2005)

• 대한지역사회영양학회지
• /
• 제16권2호
• /
• pp.278-293
• /
• 2011

The purpose of this study was to examine the research trend of nutrition for the recent 12 years from 1996 to 2007 by analyzing 734 articles published in the Korean Journal of Community Nutrition. The majority of the articles (61.4%) were classified as survey types in terms of data collection methods. Most of the subjects used in the articles were adults (28.8%), and the subject whose research has been increased at the highest rate was "patients". The most frequent keywords in the title of articles were "nutrient intake" (231times), "food service" (92times), "dietary habits" (69times), and "obesity" (69times). The keywords that have appeared more frequently with the years were "osteoporosis" (450.0%), "menopause" (350.0%) and "dietary attitudes" (208.3%). As for research interests, "nutrient intake" was dominant in the early stage of research while "disease", "dietary habits", "dietary attitudes" and "nutrition education" have increased in recent years. Some of the most common methods of nutrition assessment were "dietary intake" (41.2%), "anthropometric" (34.0%) and "biochemical test" (14.7%). The most common methods of dietary intake were "24-hours recall" (28.6%) and "dietary habits" (23.3%). The results of this study showed some biases in data collection methods, gender of the subjects, and study areas. Moreover, inconsistent terminologies, questionnaire contents, and measures were used for the researches on dietary behaviors, dietary habits, dietary attitudes, which made it difficult to compare their results for each research. Therefore, standardized research methods and terminologies need to be developed regarding dietary practices.

• Nutrition Research and Practice
• /
• 제8권3호
• /
• pp.319-326
• /
• 2014

BACKGROUND/OBJECTIVES: The immigrant population has grown considerably in South Korea since the early 1990s due to international marriages. Dietary changes in immigrants are an important issue, because they are related to health and disease patterns. This study was conducted to compare changes in dietary intake between baseline and follow-up periods. SUBJECTS/METHODS: Two hundreds thirty three Vietnamese female married immigrants. Baseline data were collected during 2006-2009, and the follow-up data were collected during 2008 and 2010. Food consumption was assessed using a 1-day 24-hour recall. RESULTS: The amount of the total food consumed (P < 0.001) including that of cereals (P = 0.004), vegetables (P = 0.003), and fruits (P = 0.002) decreased at follow-up compared to that at baseline, whereas consumption of milk and dairy products increased (P = 0.004). Accordingly, the overall energy and nutrient intake decreased at follow-up, including carbohydrates (P = 0.012), protein (P = 0.021), fiber (P = 0.008), iron (P = 0.009), zinc (P = 0.006), and folate (P = 0.002). Among various anthropometric and biochemical variables, mean skeletal muscle mass decreased (P = 0.012), plasma high density lipoprotein-cholesterol increased, (P = 0.020) and high sensitivity C-reactive protein decreased at follow-up (P < 0.001). CONCLUSIONS: A long-term follow-up study is needed to investigate the association between changes in food and nutrient intake and anthropometric and biochemical variables in these Vietnamese female marriage immigrants.

• Nutrition Research and Practice
• /
• 제9권5호
• /
• pp.547-553
• /
• 2015

BACKGROUNDS/OBJECTIVES: The effects of malnutrition on growth of toddlers and preschoolers by socioeconomic status are not well known. This study aimed to examine the effects of dietary intake on growth outcomes in toddlers and preschoolers by household income level. SUBJECTS/METHODS: The study population was a total of 1,687 children aged 1 to 5 years that participated in the KNHANES from 2009 to 2011. Growth of children was assessed by height for age (HFA) and weight for height (WFH). Children were classified into three groups according to children's HFA and WFH compared to the $10^{th}$ and $90^{th}$ percentiles of the 2007 Korean Children and Adolescent Growth Standard. Average monthly household income was divided into quartile groups. Dietary intake data were obtained by using the one day 24-hr recall method. Risks of inadequate intake of nutrients and unfavorable growth were estimated by using a multiple logistic regression model adjusted for sex, age, region, and energy intake. RESULTS: The low HFA group (<$10^{th}$ percentile) had significantly lower intakes of energy, carbohydrate, and thiamin as compared with the high group (${\geq}90^{th}$ percentile). For WFH status, vitamin C intake was lower in the low group than in the high group. Household income level was related to WFH status but not HFA. Children from lower income households were more likely to have high WFH than those from higher income households (P for trend = 0.038). Household income status was also significantly related with risk of inadequate intake of micronutrients such as thiamin (P for trend = 0.032) and vitamin C (P for trend = 0.002), showing higher odds of inadequate intakes in children from lower income households. CONCLUSIONS: Children from lower income households were prone to be overweight and to have inadequate intakes of micronutrients such as thiamin and vitamin C. To reduce nutritional and health disparities, collective action in the public sector is required from early life.

• Nutrition Research and Practice
• /
• 제15권4호
• /
• pp.504-515
• /
• 2021

BACKGROUND/OBJECTIVES: Malnutrition risk and malnutrition among the elderly is a public health concern. In combating this health-related problem, it is critically important to evaluate the risk factors in a multidimensional way and to apply appropriate nutrition intervention based on the results. SUBJECTS/METHODS: A cross-sectional study was conducted on 215 elderly patients (32.6% male, 67.4% female) in a geriatric outpatient clinic of a hospital in Turkey. Nutritional questionnaires that incorporated the 24-h recall method were applied to determine general characteristics of patients, their health status, nutritional habits, and daily energy and nutrient intakes. Mini Nutritional Assessment was used to determine nutritional status. Relevant anthropometric measurements were obtained. RESULTS: The subjects' mean age was 76.1 ± 7.0 years, and the prevalence of malnutrition (n = 7) and risk of malnutrition (n = 53) among the 215 subjects was 3.2% and 24.7%, respectively. Patients with malnutrition or risk of malnutrition were found to be single, have a depression diagnosis, in an older age group, have less appetite, more tooth loss, have more frequent swallowing/chewing difficulty, and have more frequent meal skipping. In addition, mean daily energy, carbohydrate, fat, fiber, vitamin E, vitamin B1, vitamin B2, vitamin B6, vitamin C, folates, potassium, magnesium, phosphorus, iron intake, and water consumption were found to be statistically significantly low in subjects with malnutrition or risk of malnutrition. After performing regression analysis to determine confounding factors, malnutrition risk was significantly associated with marital status, loss of teeth, appetite status, and depression. CONCLUSIONS: Routine nutritional screening and assessment of the elderly should be performed. If nutritional deficiencies cannot be diagnosed early and treated, self-sufficiency in the elderly may deteriorate, resulting in increased institutionalization.

• Journal of Animal Science and Technology
• /
• 제63권6호
• /
• pp.1453-1463
• /
• 2021

Feeding is the most important behavior that represents the health and welfare of weanling pigs. The early detection of feed refusal is crucial for the control of disease in the initial stages and the detection of empty feeders for adding feed in a timely manner. This paper proposes a real-time technique for the detection and recognition of small pigs using a deep-leaning-based method. The proposed model focuses on detecting pigs on a feeder in a feeding position. Conventional methods detect pigs and then classify them into different behavior gestures. In contrast, in the proposed method, these two tasks are combined into a single process to detect only feeding behavior to increase the speed of detection. Considering the significant differences between pig behaviors at different sizes, adaptive adjustments are introduced into a you-only-look-once (YOLO) model, including an angle optimization strategy between the head and body for detecting a head in a feeder. According to experimental results, this method can detect the feeding behavior of pigs and screen non-feeding positions with 95.66%, 94.22%, and 96.56% average precision (AP) at an intersection over union (IoU) threshold of 0.5 for YOLOv3, YOLOv4, and an additional layer and with the proposed activation function, respectively. Drinking behavior was detected with 86.86%, 89.16%, and 86.41% AP at a 0.5 IoU threshold for YOLOv3, YOLOv4, and the proposed activation function, respectively. In terms of detection and classification, the results of our study demonstrate that the proposed method yields higher precision and recall compared to conventional methods.

• International Journal of Computer Science & Network Security
• /
• 제22권6호
• /
• pp.131-138
• /
• 2022

Novel Coronavirus (COVID-19) is viewed as one of the main general wellbeing theaters on the worldwide level all over the planet. Because of the abrupt idea of the flare-up and the irresistible force of the infection, it causes individuals tension, melancholy, and other pressure responses. The avoidance and control of the novel Covid pneumonia have moved into an imperative stage. It is fundamental to early foresee and figure of infection episode during this troublesome opportunity to control of its grimness and mortality. The entire world is investing unimaginable amounts of energy to fight against the spread of this lethal infection. In this paper, we utilized machine learning and deep learning techniques for analyzing what is going on utilizing countries shared information and for detecting the climate factors that effect on spreading Covid-19, such as humidity, sunny hours, temperature and wind speed for understanding its regular dramatic way of behaving alongside the forecast of future reachability of the COVID-2019 around the world. We utilized data collected and produced by Kaggle and the Johns Hopkins Center for Systems Science. The dataset has 25 attributes and 9566 objects. Our Experiment consists of two phases. In phase one, we preprocessed dataset for DL model and features were decreased to four features humidity, sunny hours, temperature and wind speed by utilized the Pearson Correlation Coefficient technique (correlation attributes feature selection). In phase two, we utilized the traditional famous six machine learning techniques for numerical datasets, and Dense Net deep learning model to predict and detect the climatic factor that aide to disease outbreak. We validated the model by using confusion matrix (CM) and measured the performance by four different metrics: accuracy, f-measure, recall, and precision.

• Imaging Science in Dentistry
• /
• 제52권4호
• /
• pp.351-357
• /
• 2022

Purpose: Convolutional neural networks (CNNs) have rapidly emerged as one of the most promising artificial intelligence methods in the field of medical and dental research. CNNs can provide an effective diagnostic methodology allowing for the detection of early-staged diseases. Therefore, this study aimed to evaluate the performance of a deep CNN algorithm for apical lesion segmentation from panoramic radiographs. Materials and Methods: A total of 1000 panoramic images showing apical lesions were separated into training (n=800, 80%), validation (n=100, 10%), and test (n=100, 10%) datasets. The performance of identifying apical lesions was evaluated by calculating the precision, recall, and F1-score. Results: In the test group of 180 apical lesions, 147 lesions were segmented from panoramic radiographs with an intersection over union (IoU) threshold of 0.3. The F1-score values, as a measure of performance, were 0.828, 0.815, and 0.742, respectively, with IoU thresholds of 0.3, 0.4, and 0.5. Conclusion: This study showed the potential utility of a deep learning-guided approach for the segmentation of apical lesions. The deep CNN algorithm using U-Net demonstrated considerably high performance in detecting apical lesions.

• 대한조현병학회지
• /
• 제20권1호
• /
• pp.15-22
• /
• 2017

Objectives : It has been shown that early intervention is crucial for favorable outcome in patients with schizophrenia. However, development of biomarkers for predicting prognosis of psychotic disorder still requires more research. In this study, we aimed to investigate whether baseline mismatch negativity (MMN) predict prognosis in patients with first episode psychosis (FEP). Methods : Twenty-four patients with FEP and matched healthy controls (HCs) were examined with MMN at baseline, and their clinical status were re-assessed after 1 year. Repeated-measures analysis of variance was performed to compare baseline MMN between the two groups. Multiple regression analysis was used to identify factors predicting prognosis in FEP patients during the follow-up period. Results : MMN amplitudes at baseline were significantly reduced in patients with FEP compared to healthy controls. In the multiple regression analysis, baseline MMN amplitude significantly predicted later improvement of performances on digit span and delayed recall of California Verbal Learning Test. However, baseline MMN did not predicted improvement of clinical symptoms. Conclusion : These results indicate that MMN may be a possible predictor of improvement in cognitive functioning in patients with FEP. Future study with larger sample and longer follow-up period would be needed to confirm the findings of the current study.