http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
밀도 프로파일 추출 방법에 따른 염색체 분류 성능 분석
송혜정(Hye-Jeong Song),최광원(Kwang-Won Choi),권경훈(Kyoung-Hun Kwon),이용식(Yong-Sik Lee),김유섭(Yong-Sik Lee),김종대(Jong-Dae Kim) 한국정보기술학회 2011 한국정보기술학회논문지 Vol.9 No.2
This work presents the performance comparison according to the density profile extraction methods for the automated chromosome classification system based on banding pattern. Chromosome length and density profiles were chosen as the features for the classification system. They were estimated from the medial axes of the chromosomes that were detected by the conventional thinning algorithm. The extracted medial axes extended to the boundaries, because they do not cover both ends of chromosomes. Also filters applied along or perpendicular to the medial axes to reduce the noise. This paper investigated the classification performances of the above feature extraction methods. The dataset consists of 4561 chromosomes were supported by Gendix. Classification method used 3-NN. Experimental result showed that employing the extension of the medial axis and the vertical filter improved performance by 13%.
송혜정 ( Hye Jeong Song ),이덕희 ( Duk Hee Lee ) 한국치위생학회 2015 한국치위생학회지 Vol.15 No.1
Objectives: The purpose of the study is to investigate the relationship between bone mineral density and remaining teeth in Korean adults. Methods: Using National Health and Nutrition Examination Survey(KNHANES) 5th Year 1(2010), 1,985 adults over 50 years old having bone mineral density test were selected. Male adults were 883, and female adults were 1,102. The raw data consisted of general characteristics of the subjects, bone density data, and oral examination data. The questionnaire consisted of general health survey, laboratory examination study, and nutrition study. Data were analyzed using SPSS 18.0 program for multiple regression analysis, ANOVA, t-test. The variables included general characteristics of the subjects, health behavior, diet habit, and oral health behavior. General characteristics consisted of gender, age, residence area, education level, marital status, monthly income, and vocation. Health behavior consisted of smoking, alcohol drinking, body mass index(BMI), physical activity, and mental health. Diet habit consisted of frequency of meal, milk, coffee, and calcium ingestion. Oral health behavior consisted of frequency of tooth brushing, use of oral health care devices, and oral examination. Systemic diseases were measured by the questionnaire for hypertension, diabetes mellitus, and hyperlipidemia. Bone density was measured in T-score of femoral bone, neck of femur, and lumbar vertebrae. Bone density was classified into normal(T-score ≥ -1.0), osteopenia(-2.5 < T-score < -1.0), and osteoporosis (T-score ≤ -2.5). The remaining teeth were measured by the differences between normal teeth and missing teeth. Results: Smoking and age are the most influencing factors on the number of remaining teeth and bone density(p<0.01)(p<0.05). Those who had higher bone density, younger age, and nonsmoker significantly had more remaining teeth. Those who were men, diabetic, lowly educated women, and low birth weighted persons tended to have less number of teeth(p<0.01). Conclusions: The study suggested that there were significant differences between men and women according to the oral health management behavior and systemic disease control. Proper management of the oral health and good quality of bone mineral density can reserve the remaining teeth through the whole life.
박정희 ( Jeong Hee Park ),허남숙 ( Nam Suk Heo ),송혜정 ( Hye Jeong Song ) 한국치위생학회(구 한국치위생교육학회) 2011 한국치위생학회지 Vol.11 No.6
Objectives: This study is designed to identify the need for an efficient infection control plan in the curriculum of Dept. of Dental Hygiene and encourage dental hygienists to be more attentive to infection control by analyzing their personal protective measures, hand-washing frequency and other factors relating to infection control. Methods: A survey was conducted of 471 dental hygienists in general hospitals, dental clinics and dental hospitals in City of Busan and Gyeongsangnam-do from December 20, 2009 to February 21, 2010. Results: 1. Among those who worked for more than 11 years, personal protection (3.79) and hand-washing frequency (3.90) factors were rated high (p<0.05). 2. When surveyedby workplace, personal protection was high in general hospitals (3.75) while hand-washing frequency was high in dental clinics (3.74). 3. Among those who received infection control education more than three times, personal protection (4.07) and hand-washing frequency (3.80) were high (p<0.05). 4. Personal protection-related factors were general hospital workers, necessity of infection control education and awareness on infectious disease (p<0.05), while factors associated with handwashing frequency were number of patients per day, whether or not infection control-related class is given at school, necessity of infection control education and awareness on infectious disease (p<0.05). Conclusions: To enhance awareness of infection control, it may be necessary to addan infection control-related class to the school curriculum and provide the necessity of conducting continuous and systematic infection control education through newly entering staff training and education of dental practitioners. (J Korean Soc Dent Hygiene 2011; 11(6):993-1003)
국부적 영역에서의 특징 공간 속성을 이용한 다중 인식기 선택
신동국(Dong-Kuk Shin),송혜정(Hye-Jeong Song),김백섭(BaekSop Kim) 한국정보과학회 2004 정보과학회논문지 : 소프트웨어 및 응용 Vol.31 No.12
본 논문은 시험 표본 주위의 영역에 대한 속성을 이용한 다중 인식기 선택 방법을 제안한다. 기존의 DCS-LA 동적 인식기 선택 방법은 시험 표본 주위의 학습표본들을 사용해서 각 인식기의 국부적정확성을 계산하여 인식기를 동적으로 선택하기 때문에 인식 시간이 오래 걸린다. 본 논문에서는 특징공간에서 국부적인 속성을 계산해서 그 속성값에 적합한 인식기를 미리 선정해서 저장해 놓은 후 시험 표본이 들어오면 그 주변의 속성값에 따라 저장된 인식기에서 선택을 하기 때문에 인식시간을 줄일 수 있다. 국부적인 속성으로는 표본 주위의 작은 영역에 대한 엔트로피와 밀도를 계산하여 사용하였으며 이들을 특징공간속성(Feature Space Attribute)라고 하였다. 이들 두 속성으로 이루어지는 속성 공간을 규칙적인 사각형 셀로 나누어, 학습과정에서 각각의 학습표본에 대해 계산된 속성값이 어떤 셀에 속하는지를 구한다. 또한 각 셀에 속하는 학습표본들에 대해 각 인식기의 국부적 정확도를 구하여 셀에 저장한다. 시험 과정에서 시험표본에 대해 속성값 계산을 통해 그 표본이 속하는 셀을 구한 후 그 셀에서 국부적 정확도가 가장 높은 인식기로 인식한다. Elena 데이타베이스를 사용해서 기존의 방법과 제안된 방법을 비교하였다. 제안된 방법은 기존의 DCS-LA와 거의 같은 인식률을 나타내지만 인식속도는 약 4배 가까이 빨라짐을 실험을 통해 확인할 수 있었다. This paper presents a method for classifier selection that uses distribution information of the training samples in a small region surrounding a sample. The conventional DCS-LA(Dynamic Classifier Selection - Local Accuracy) selects a classifier dynamically by comparing the local accuracy of each classifier at the test time, which inevitably requires long classification time. On the other hand, in the proposed approach, the best classifier in a local region is stored in the FSA(Feature Space Attribute) table during the training time, and the test is done by just referring to the table. Therefore, this approach enables fast classification because classification is not needed during test. Two feature space attributes are used : entropy and density of k training samples around each sample. Each sample in the feature space is mapped into a point in the attribute space made by two attributes. The attribute space is divided into regular rectangular cells in which the local accuracy of each classifier is appended. The cells with associated local accuracy comprise the FSA table. During test, when a test sample is applied, the cell to which the test sample belongs is determined first by calculating the two attributes, and then, the most accurate classifier is chosen from the FSA table. To show the effectiveness of the proposed algorithm, it is compared with the conventional DCS-LA using the Elena database. The experiments show that the accuracy of the proposed algorithm is almost same as DCS-LA, but the classification time is about four times faster than that.
서현기(Hyun Gi Seo),송혜정(Hye Jeong Song),김백섭(Baek Sop Kim) 한국정보과학회 2010 정보과학회논문지 : 소프트웨어 및 응용 Vol.37 No.10
본 논문에서는 초음파 영상에서 끊어진 구조의 연결성을 향상시키는 방법을 제안한다. 본 제안방법은 대상 픽셀의 밝기값과 주변픽셀의 그라디언트 정보를 사용하여 연결 여부를 결정하게 된다. 끊어진 구조로 판단되어진 픽셀은 그 밝기값을 증가시켜 연관성 있는 구조로 연결시키게 된다. 이후 시각적 자연스러운 표현을 위하여 비등방성 확산 필터링을 적용하여 수행하게 된다. 연산속도의 이득을 위하여 제안하는 연결성 향상 방법의 구조행렬은 그라디언트 정보와 확산 방향을 결정하는 확산 필터의 구조행렬과 동일한 구조행렬을 사용하였다. 실험은 끊어진 구조가 나타나는 실제 초음파 영상에 대해 수행하였으며 제안된 방법에 의해 끊어진 구조가 연결되어 졌음을 확인할 수 있다. This paper is to propose a method to enhance the connectivity of the broken structures. The pixels to be connected are found by a decision rule which is based on the intensity and gradient information of the neighboring pixels. The intensity of the pixel, if it is decided to be filled, is increased so that the pixel together with its neighbors looks connected. The anisotropic diffusion follows to make the connected structural region look more natural. The same structure matrices have been used both to get gradient information and to decide the direction of diffusion to improve the computational speed. It has been shown by the experiments on the real ultrasound images that the broken structural regions can be connected by the proposed method.