• Proceedings of the Korean Society of Radiological Science
• /
• 2002.10a
• /
• pp.25-25
• /
• 2002

• Proceedings of the KSCN Conference
• /
• 2006.05a
• /
• pp.81-81
• /
• 2006

• Journal of the Korean Association of Geographic Information Studies
• /
• v.24 no.3
• /
• pp.1-9
• /
• 2021

In this study, we evaluated the accuracy of deep learning-based tree species classification model trained by using high-resolution images. We selected five species classed, i.e., pine, birch, larch, korean pine, mongolian oak for classification. We created 5,000 datasets using high-resolution orthophoto and forest type map. CNN deep learning model is used to tree species classification. We divided training data, verification data, and test data by a 5:3:2 ratio of the datasets and used it for the learning and evaluation of the model. The overall accuracy of the model was 89%. The accuracy of each species were pine 95%, birch 89%, larch 80%, korean pine 86% and mongolian oak 98%.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2006.05a
• /
• pp.39-42
• /
• 2006

심전도는 각종심장질환 들을 예측하는데 널리 사용되고 있다. 이러한 심전도에서 ST-분절은 허혈성 심장 질환, 확장성 심근성, 비후성 심근증 등을 예측하는데 이용되고 있다. 이 논문에서는 환자들의 임상 정보와 심전도로부터 심장 질환 예측을 위한 중요 파라미터인 ST-부절을 추출하였다. 그리고 이러한 추출된 데이터 분석을 위해서 데이터마이닝 기법을 적용한다. 데이터마이닝의 분류 알고리즘인 베이지안 네트워크를 적용 심장 질환을 효율적으로 분류하기 위한 방법을 제시 하였다.

• Proceedings of the Korean Association of Geographic Inforamtion Studies Conference
• /
• 2010.06a
• /
• pp.291-293
• /
• 2010

최근 발생하고 있는 국지성 집중호우 및 돌발홍수로 인해 강원도와 경상북도 등을 중심으로 산지하천유역의 산사태 피해가 급증하고 있으며 발생면적은 연평균 402ha에 이르며 연평균 피해면적은 80년대에 비해 2000년대 들어 3배 이상 증가한 것으로 보고 되고 있다. 본 연구에서는 산지하천 유역의 토사유출재해 취약성 분석을 위해 GEOMania GMMap 기반으로 구동되는 산사태 분석모듈(HyGIS-Landslide)을 개발하였다 HyGIS-Landslide는 산림청의 산사태 위험지도 제작에 사용된 위험지역 평가기준을 참조하였으며 DEM을 이용하여 경사인자 및 사면인자를 생성하고 수치지질도, 수치임상도 산림입지도 등과의 연산을 통해 위험등급에 대한 분류결과를 제시한다. 또한, 과거 산사태 발생지역에 대한 맵핑 경과가 존재하는 경우 산사태 위험지역 분류결과를 과거 사상과 중첩하여 분류정확도를 확인할 수 있도록 제작되었다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2004.04b
• /
• pp.595-597
• /
• 2004

본 연구에서는 베이지안망을 기초로 불임환자의 임상 데이터에 대한 다양한 실험을 전개한다. 실험을 통해 임신여부에 영향을 주는 요인들간의 상호 의존성을 분석하고. 또 제약조건이 다른 다양한 베이지안망의 대표적 유형으로 나이브 베이지안망(NBN), 베이지안망으로 확장한 나이브 베이지안망(BAN), 일반 베이지안앙(GBN) 분류기들의 분류성능을 서로 비교 분석한다. 베이지안망을 적응할 때 변수의 수가 많아짐에 따라 베이지안망의 구조를 학습하는데 탐색공간이 넓어져 시간의 요구량이 급격히 많아진다. 따라서 이런 탐색공간을 효율적으로 줄이기 위하여 클래스 노드의 Markov blanket에 속한 특징들로 집합을 축소하는 것을 제안하고, 실험을 통해 이 특징 축소 방법이 베이지안망 분류기들의 성능을 높여 줄 수 있는지 알아본다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2010.06c
• /
• pp.205-210
• /
• 2010

심장질환은 암 다음으로 높은 사망 원인으로 초기 진단은 치료에 매우 중요한 문제로 대두 되고 있다. 심장병을 분석하기 위해서는 임상 데이터에 대해 자세히 알고 분석 하는 것이 중요하다. 본 논문에서는 심장 질환 데이터를 효율적으로 분석하기 위해 배깅 알고리즘을 사용하여 중요 검사 항목을 추출해내고 분석하는 방법을 제안한다. 데이터를 분석하는 과정에 있어서 분류자들을 생성하고 앙상블 하는 과정에 효과적인 결과를 얻기 위해서 다양한 알고리즘들을 결합해야 구성해야한다. 앙상블을 이용하여 가장 좋은 의 분류 효과를 얻기 위해서는 수천가지의 분류자들을 훈련시켜 성능이 좋은 앙상블을 구성한다.

• Journal of the Korean Society of Radiology
• /
• v.82 no.4
• /
• pp.808-816
• /
• 2021

Pulmonary emphysema is a cause of chronic obstructive pulmonary disease. Emphysema can be accurately diagnosed via CT. The severity of emphysema can be assessed using visual interpretation or quantitative analysis. Various studies on emphysema using deep learning have also been conducted. Although the classification of emphysema has proven clinically useful, there is a need to improve the reliability of the measurement.

• Clinical and Experimental Pediatrics
• /
• v.49 no.11
• /
• pp.1158-1166
• /
• 2006

Purpose : The purpose of this study is to make a diagnostic classification and discuss a diagnostic strategy of floppy infants by investigating clinical, neurological, electrophysiological, and genetic analysis of infants admitted to intensive care units with the complaint of hypotonia. Methods : A retrospective study was performed from Jan. 1993 to Dec. 2005 in neonatal and pediatric intensive care units of Seoul National University Children's Hospital. Clinical features and all tests related to hypotonia were investigated. Results : There were 21 cases of floppy infants admitted to intensive care units. Final diagnosis was classified as centra (7 cases[33.3 percent]), peripheral (11 cases [52.4 percent]), and unspecified (3 cases [14.3 percent]). Among the central group, three patients were diagnosed as hypoxic ischemic encephalopathy, two patients as Prader-Willi syndrome, one patient as chromosomal disorder, and one patient as transient hypotonia. Among the peripheral group, four patients were diagnosed as myotubular myopathy, three patients as SMA type 1, two patients as congenital myotonic dystrophy, one patient as congenital muscular dystrophy, and one as unspecified motor-neuron disease. Motor power was above grade 3 on average, and deep tendon reflex was brisk in the central group. Among investigations, electromyography showed 66 percent sensitivity in the peripheral group, and muscle biopsy was all diagnostic in the peripheral group. Brain image was diagnostic in the central group, and Prader-Willi FISH or karyotyping was helpful in diagnosis in central group. Morbidity and mortality was more severe in the peripheral group Conclusion : Classification of diagnosis by clinical characteristics in this study, and application of investigations step by step, may provide an effective diagnostic strategy.

• Journal of dental hygiene science
• /
• v.16 no.5
• /
• pp.393-399
• /
• 2016

The objective of this study was to realize research demands by gathering the research agenda from the Korean Dental Hygienists Association and affiliated groups through a Delphi survey. The study subjects were the heads of eight groups. The Delphi survey was conducted by e-mail after a researcher explained the objective of the study to the heads. The research agenda was classified according to the Korean dental hygiene academic classification and the National Dental Hygiene Research Agenda classification. To compare the number of research agendas according to classification among the groups, we conducted a chi-squared test by using IBM SPSS 20.0. In total, 316 agendas were submitted. According to the Korean dental hygiene academic classification, there was a significant difference among the groups, and the agendas associated with clinical dental hygiene (36.7%) were the best. According to the National Dental Hygiene Research Agenda classification, there was no significant difference and the agendas associated with health promotion and disease prevention (27.8%) were the best. In conclusion, the Korean Dental Hygienists Association and affiliated groups demanded similar research agendas, corresponding to the purpose of each establishment. However, the groups did not seem to show unique characteristics according to the purpose of each establishment. In other words, the groups should corroborate such characteristics and suggest a corresponding agenda to establish specific areas of research in dental hygiene. Also, dental hygiene researchers should mark the academic classification of the studies they want to submit.