Evidence-based approaches for establishing the 2015 Dietary Reference Intakes for Koreans

Sangah Shin,Subeen Kim,Hyojee Joung 한국영양학회 2018 Nutrition Research and Practice Vol.12 No.6

BACKGROUND/OBJECTIVES: The Dietary Reference Intakes for Koreans (KDRIs), a set of reference intake values, have served as a basis for guiding a balanced diet that promotes health and prevents disease in the general Korean population. In the process of developing DRIs, a systematic review has played an important role in helping the DRI committees make evidence-based and transparent decisions for updating the next DRIs. Thus, the 2015 KDRI steering committee applied the systematic review framework to the revision process of the KDRIs. The purpose of this article is to summarize the revision process for the 2015 KDRIs by focusing on the systematic review framework. MATERIALS/METHODS: The methods used to develop the systematic review framework for 2015 KDRIs followed the Agency for Healthcare Research and Quality and the Tufts Evidence-based Practice Center. The framework for systematic review of the 2015 KDRIs comprised of the 3 following steps: (1) development of an analytic framework and refinement of key questions and search terms; (2) literature search and data extraction; and, (3) appraisal of the literature and summarizing the results. RESULTS: A total of 203,237 studies were retrieved through the above procedure, with 2,324 of these studies included in the analysis. General information, main results, comments of reviewers, and results of quality assessment were extracted and organized by study design. The average points of quality appraisals were 3.0 (range, 0-5) points for intervention, 6.1 (0-9) points for cohort, 6.0 (3-9) points for nested case-control, 5.4 (1-8) points for case-control, 14.6 (0-22) points for cross-sectional studies, and 7.0 (0-11) points for reviews. CONCLUSIONS: Systematic review helped to establish the 2015 KDRIs as a useful tool for evidence-based approach. Collaborative efforts to improve the framework for systematic review should be continued for future KDRIs.

Dietary patterns and prostate cancer risk in Japanese: the Japan Public Health Center-based Prospective Study (JPHC Study)

Shin, Sangah,Saito, Eiko,Sawada, Norie,Ishihara, Junko,Takachi, Ribeka,Nanri, Akiko,Shimazu, Taichi,Yamaji, Taiki,Iwasaki, Motoki,Sasazuki, Shizuka,Inoue, Manami,Tsugane, Shoichiro Springer-Verlag 2018 Cancer causes & control Vol.29 No.6

Dietary patterns and colorectal cancer risk in middle-aged adults: A large population-based prospective cohort study

Shin, Sangah,Saito, Eiko,Sawada, Norie,Ishihara, Junko,Takachi, Ribeka,Nanri, Akiko,Shimazu, Taichi,Yamaji, Taiki,Iwasaki, Motoki,Sasazuki, Shizuka,Inoue, Manami,Tsugane, Shoichiro,Tsugane, S.,Sawada, Elsevier 2018 Clinical nutrition Vol.37 No.3

<P><B>Summary</B></P> <P><B>Background & aims</B></P> <P>A finding between dietary pattern and cancer may provide visions beyond the assessment of individual foods or nutrients. We examined the influence of dietary pattern with colorectal cancer (CRC) among a Japanese population.</P> <P><B>Methods</B></P> <P>A total of 93,062 subjects (43,591 men, 49,471 women) who participated in the Japan Public Health Center-based Prospective Study were followed from 1995–1998 to the end of 2012, during which 2482 cases of CRC (1514 men, 968 women) were newly identified. Dietary data was obtained from a validated food-frequency questionnaire between 1995 and 1998.</P> <P><B>Results</B></P> <P>Three dietary pattern was derived from principal components factor: prudent, westernized, and traditional pattern. After controlled for potential confounders, the prudent pattern showed a decreased association of CRC risk in men (HR for highest quintile vs lowest: 0.85; 95% CI: 0.72–1.00; <I>P</I> trend <0.05), slightly more strongly with distal colon cancer (<I>P</I> trend <0.05); but an increased risk of rectal cancer in women (<I>P</I> trend <0.05). The westernized pattern showed a significant positive linear trend for colon (<I>P</I> trend <0.05) and distal cancer (<I>P</I> trend <0.05) in women. There was no apparent association of traditional Japanese dietary pattern on the overall or any specific sites risk of CRC.</P> <P><B>Conclusions</B></P> <P>A prudent dietary pattern showed an inverse association with CRC risk in men, and a westernized pattern was related with a higher risk of colon and distal cancer in women.</P>

여고생의 불안 및 우울상태와 식행태와의 관련성 연구

신상아(Sangah Shin),권경남(Kyoung Nam Kwon),이은지(Eunji Lee),정효지(Hyojee Joung) 서울대학교 보건환경연구소 2015 보건학논집 Vol.52 No.1

식품으로 인한 유해물질 노출평가를 위한 설문지 개발

신상아 (Shin,Sangah ),최슬기 (Choi,Seul Ki ),김혜미 (Kim,Hyemi ),이경윤 (Kyung Youn ),신상희 (Shin,Sanghee ),이정원 (Lee,Jung Won ),유수현 (Yu,Soo Hyun ),남혜선 (Nam,Hye-Soen ),윤혜정 (Yoon,Hae-Jung ),정효지 (Joung,Hyojee ) 韓國營養學會 2011 Journal of Nutrition and Health Vol.44 No.1

The aim of this study was to develop a reliable dietary questionnaire to assess human exposure to food-borne hazards. Eleven food-borne hazards were chosen as a priority control list through a literature review and advisory committees. The 11 food-borne hazards were phthalate, aflatoxin, bisphenol A, polycyclic aromatic hydrocarbons, dioxin, polychlorinated biphenyls, mercury, lead, cadmium, arsenic, and acrylamide. The characteristics, exposure level, and paths of these hazards were reviewed, and questionnaire items were identified to assess human exposure from the literature. A questionnaire was developed for each selected food based on its characteristics. Based on the items in the individual questionnaires, a comprehensive questionnaire, which contained demographic characteristics, job information, socioeconomic factors, health related lifestyles, and dietary behaviors, was developed. A 99-item food frequency questionnaire (FFQ) to assess food-borne hazard exposure was also developed. The FFQ included frequency of food intake during the previous year, container type for purchasing and storing food, and cooking method. The questionnaire developed in this study could be applied to assess dietary factors during an exposure assessment of food-borne hazards in a large population. A validation study for the questionnaire is needed before applying it to surveys. (Korean J Nutr 2011 44(1): 61 ~ 73)

서울시 일부 청소년의 골질량 관련요인

신상아(Shin Sangah),홍경의(Hong Kyung Eui),최현정(Choi Hyun Jeong),노지현(Roh Ji Hyun),정효지(Joung Hyojee) 韓國營養學會 2008 Journal of Nutrition and Health Vol.41 No.2

청소년기의 골건강에 영향을 줄 수 있는 신체계측치와 사회경제적 요인, 출생 당시의 모체의 건강상태 및 모유수유, 대상자의 건강상태, 생활습관 및 식습관 요인을 분석한 본 연구의 결과를 요약하면 다음과 같다. 1) 조사 대상자의 평균 연령은 14.2세, 신장 162.7 ㎝, 체중 57.3 ㎏이었고, 비만 및 과체중 유병률은 27.3%였다. 우측 종골의 골밀도는 0.518 g/㎠, 골질량은 2.241 g이었다. 2) 월 3만원 이상 용돈을 사용하는 남학생의 골질량이 3만원 이하 용돈을 사용하는 남학생에 비해서 유의하게 높았고, 형제가 없는 남학생의 경우 골질량이 유의하게 높았다. 출생시 어머니의 나이가 30세 이전인 학생들의 골질량이 유의하지는 않지만 높은 경향이 있었다. 3) 골절 경험이 있는 남학생의 골절량이 골절 경험이 없는 남학생에 비해 유의하게 높았고, 2차 성징이 늦은 남학생의 골질량이 높았다. 영양제를 섭취 하지 않은 남학생의 골질량이 높았고, 활동적인 선체 활동과 규직적인 운동, 야외활동을 많이 하는 학생들의 골질량이 유의하게 높았다. 4) 다양한 음식을 골고루 섭취하는 남학생의 골질량이 유의하게 높았고, 충분한 물을 섭취할 때 남학생과 여학생의 골질량이 유의하게 높았다. 밥 위주의 전통식을 하루 두끼 이상 섭취하는 여학생의 골질량이 유의하게 낮았다. 치즈, 생선류, 시금치 및 무청나물의 섭취가 대상자들의 골질량에 관련성이 있었다. 5) 다중회귀분석 모델에서 선장, BMI, 운동 횟수, 시금치 또는 무청나물의 섭취횟수의 골질량 설명력은 64.9%였다. 이상의 연구 결과에서 청소년기 골질량 향상을 위해서는 규칙적인 운동과 선체활동으로 체지방률을 낮추고, 칼슘 급원 식품과 더불어 다양한 음식을 골고루 섭취하도록 해야 할 것이다. 따라서 본 연구의 결과를 토대로 청소년기 학생들에게 골질량 향상을 위해 신체 활동 증가를 통한 생활습관 교육 프로그램과 올바른 식습관 형성을 위한 식사 교육 프로그램을 실시한다면 이들이 성인이 되었을 때 발생 할 수 있는 골감소증 및 골다공증을 예방할 수 있을 것으로 사료된다. The purpose of this study was to investigate factors related to bone mineral contents (BMC) at os calcis of adolescents. The subjects were 604 students (327 boys and 277 girls) between 7th and 9th grade in Seoul, Korea. The mean age was 14.2±0.9 years. General characteristics were collected by a questionnaire, bone mineral contents (BMC) were measured in os calcis by Duel Energy X-ray Absorptiometry (PIXI, General Electronics, USA) and height and weight were measured by bioelectrical impedance analysis method (Inbody 4.0, Biospace Co. Ltd, Seoul, Korea). Mean BMC of os cal sis were 2.241 g. Height (p<.0001), weight (p<.0001), BMC (p<.0001) of boys were higher than those of girls. Percent body fat of girls, however, was higher than that of boys (p<.0001). Anthropometric measurement, pocket money, sibling, experience of fracture, pubertal stage, intake of supplements, physical or outdoors activity, and eating habit had significant influence on BMC of boys. Anthropometric measurement, physical or outdoors activity, and eating habit had significant influence on BMC of girls. Multivariate regression with adjustment for sex and age showed that BMC was associated positively with height, BMI, frequency of regular activity, and intake of spinach or radish leaves (all p<0.05), and negatively with percent body fat (p<.0000 and Korean traditional diet pattern (p=0.01). On the basis of these results, it is recommended to develop nutrition education and physical activity program for adolescents to improve BMC and prevent osteoporosis. (Korean J Nutr 2008; 41 (2): 156~164)

서울대학교 교내식당 메뉴의 영양가분석 및 평가

정효지(Hyojee Joung),신상아(Sangah Shin),박은영(Eunyoung Park),권경남(Kyoungnam Kwon),이경희(Kyunghee Lee),정선심(Sunsim Jeong),주정정(Tingting Zhu),신상희(Sanghee Shin),신아영(Ah Young Shin) 서울대학교 보건환경연구소 2009 보건학논집 Vol.46 No.1

순차 모형과 언어 자질 벡터를 이용한 한국어 토론 데이터의 선형 논증 구조 분석

이상아(Sangah Lee),신효필(Hyopil Shin) Korean Institute of Information Scientists and Eng 2018 정보과학회논문지 Vol.45 No.12

Current studies on argument mining provide tree-structured argumentation structures based on relational nuclearities and discourse relations between sentences in each document. In this case, inconsistencies between related sentences may occur, constructing a full argumentation structure for a document by the bottom-up method. This paper introduces relations between the topic of texts and sentences to provide a frame of argumentation structure. Automatic analysis of argumentation structure uses contextual information from documents, as argument types defined for each sentence are applied to the sequential model. In this paper, we vectorized sentences using bag-of-words of morphemes, word embedding of morphemes, and some linguistic features extracted from the sentence respectively, and used those vectors as inputs of models to predict argument types in the document. As a result, the combination of linguistic features and the sequential model revealed the best result in the experiment, showing 0.68 as the f1-score.

소규모 데이터 기반 한국어 버트 모델

이상아(Sangah Lee),장한솔(Hansol Jang),백연미(Yunmee Baik),박수지(Suzi Park),신효필(Hyopil Shin) Korean Institute of Information Scientists and Eng 2020 정보과학회논문지 Vol.47 No.7

Recent models for the sentence embedding use huge corpus and parameters. They have massive data and large hardware and it incurs extensive time to pre-train. This tendency raises the need for a model with comparable performance while economically using training data. In this study, we proposed a Korean-specific model KR-BERT, using sub-character level to character-level Korean dictionaries and BidirectionalWordPiece Tokenizer. As a result, our KR-BERT model performs comparably and even better than other existing pre-trained models using one-tenth the size of training data from the existing models. It demonstrates that in a morphologically complex and resourceless language, using sub-character level and BidirectionalWordPiece Tokenizer captures language-specific linguistic phenomena that the Multilingual BERT model missed.

감정 분석을 위한 BERT 사전학습모델과 추가 자질 모델의 결합

이상아(Sangah Lee),신효필(Hyopil Shin) Korean Institute of Information Scientists and Eng 2021 정보과학회논문지 Vol.48 No.7

It is known that BERT can capture various linguistic knowledge from raw text via language modeling without using any additional hand-crafted features. However, some studies have shown that BERT-based models with an additional use of specific language knowledge have higher performance for natural language processing problems associated with that knowledge. Based on such finding, we trained a sentiment-combined model by adding sentiment features to the BERT structure. We constructed sentiment feature embeddings using sentiment polarity and intensity values annotated in a Korean sentiment lexicon and proposed two methods (external fusing and knowledge distillation) to combine sentiment-combined model with a general-purpose BERT pre-trained model. The external fusing method resulted in higher performances in Korean sentiment analysis tasks with movie reviews and hate speech datasets than baselines from other pre-trained models not fused with sentiment-combined models. We also observed that adding sentiment features to the BERT structure improved the model’s language modeling and sentiment analysis performance. Furthermore, when implementing sentiment-combined models, training time and cost could be decreased by using a small-scale BERT model with a small number of layers, dimensions, and steps.