• 한국근적외분광분석학회:학술대회논문집
• /
• 한국근적외분광분석학회 2001년도 NIR-2001
• /
• pp.1042-1042
• /
• 2001

In linear discriminant analysis there are two important properties concerning the effectiveness of discriminant function modeling. The first is the separability of the discriminant function for different classes. The separability reaches its optimum by maximizing the ratio of between-class to within-class variance. The second is the stability of the discriminant function against noises present in the measurement variables. One can optimize the stability by exploring the discriminant variates in a principal variation subspace, i. e., the directions that account for a majority of the total variation of the data. An unstable discriminant function will exhibit inflated variance in the prediction of future unclassified objects, exposed to a significantly increased risk of erroneous prediction. Therefore, an ideal discriminant function should not only separate different classes with a minimum misclassification rate for the training set, but also possess a good stability such that the prediction variance for unclassified objects can be as small as possible. In other words, an optimal classifier should find a balance between the separability and the stability. This is of special significance for multivariate spectroscopy-based classification where multicollinearity always leads to discriminant directions located in low-spread subspaces. A new regularized discriminant analysis technique, the principal discriminant variate (PDV) method, has been developed for handling effectively multicollinear data commonly encountered in multivariate spectroscopy-based classification. The motivation behind this method is to seek a sequence of discriminant directions that not only optimize the separability between different classes, but also account for a maximized variation present in the data. Three different formulations for the PDV methods are suggested, and an effective computing procedure is proposed for a PDV method. Near-infrared (NIR) spectra of blood plasma samples from daily monitoring of two Japanese cows have been used to evaluate the behavior of the PDV method in comparison with principal component analysis (PCA), discriminant partial least squares (DPLS), soft independent modeling of class analogies (SIMCA) and Fisher linear discriminant analysis (FLDA). Results obtained demonstrate that the PDV method exhibits improved stability in prediction without significant loss of separability. The NIR spectra of blood plasma samples from two cows are clearly discriminated between by the PDV method. Moreover, the proposed method provides superior performance to PCA, DPLS, SIMCA md FLDA, indicating that PDV is a promising tool in discriminant analysis of spectra-characterized samples with only small compositional difference.

• 대한치과보철학회지
• /
• 제55권2호
• /
• pp.198-204
• /
• 2017

악안면부의 결손은 선천적으로 결손을 가지고 있거나 외상이나 수술적인 절제 등에 의한 후천적 원인에 의해 발생할 수 있다. 특히 구강 내 결손을 가진 환자 중 상악의 결손과 연관되어 있는 경우가 높은 비중을 차지하고 있으며 보철적 치료의 필요성이 높다. 하악의 부분적 결손을 가질 경우 기능적 회복에 상당한 한계점을 보이나 양측 턱관절이 정상적으로 남아 있을 경우 국소의치를 이용한 악안면 보철물을 제작함으로써 양호한 결과를 기대할 수 있다. 본 증례의 환자는 58세 남자 환자로서 안면부 총상으로 인한 구개부 및 좌측 하악 구치부 결손을 가진 분으로 obturator와 하악 RPD 재제작을 위해 내원하였다. 환자의 상악 결손 범위는 Aramany 분류법 Class IV에 해당되는 상태이며, 하악은 Cantor와 Curtis 분류법 Type V resection 상태이다. 상, 하악의 우측 구치부가 잔존하였으나, 수술 후 악골 변형으로 인해 서로 교합되지 않았으며 치주상태 불량으로 인한 동요도도 존재하는 상태였다. Obturator의 유지를 위해 잔존 지대치를 최대한 활용하였고 안정적인 교합이 형성되도록 치아를 배열하였다. 하악 RPD는 하악 절제술로 변형된 연조직에 적합하도록 제한된 범위를 피개하는 RPD를 제작하였다. 이에 환자분의 저작, 연하, 발음 기능이 많이 개선되었으며 만족할 만한 결과를 보여주어 본 증례를 발표하는 바이다.

• Advances in pediatric surgery
• /
• 제12권2호
• /
• pp.137-146
• /
• 2006

Pediatric surgery could establish a definitive position in the medical field on the basis of a stable patient population. Neonatal surgery, the core of pediatric surgery, requires highly skilled surgeons. However, recent advancement of prenatal diagnosis followed by intervention and decreased birth rate has resulted in a significant decrease in the neonatal surgical population and the number of surgical operations. The purpose of this study is to examine the outcome of neonatal surgeries and to propose a guide for the future surgeries. A total of 359 neonatal surgical patients operated upon at the Department of Surgery, Ewha Medical Center, during past 21 years were studied. The study period hasbeen divided into two time periods: from 1983 to 1993 and from 1994 to 2004. Analysis was based on the Clinical Classification System and mortality pattern, frequency of disorders, occurrence and cause of death, and other changes. Neonatal surgery was 6.4 % of all pediatric surgery during the total 21 year period, 9.9 % in the first period and 4.8 % in the second. Male to female ratio increased from 2.7:1 to 2.1:1. The overall mortality was 6.7 %, and there was significant decrease from 7.4 % in the first period to 6.0 % in the second. The clinical classification system (CCS) for death cases included class II 2, III 4, and IV 7 during the first period and class III 3, and IV 8 during the second, respectively. According to the mortality pattern by Hazebroek, there were 6 preventable death cases during the first period, and only one in the second, and 2 non-preventable death cases during the first period and 8 in second, respectively. Although the patients in the second period had more serious diseases, surgical mortality has been decreased in the second period, which may be the result of improved surgery methods for newborns and advanced patient care.

• 대한치과교정학회지
• /
• 제52권2호
• /
• pp.112-122
• /
• 2022

Objective: This study aimed to present and evaluate a new deep learning model for determining cervical vertebral maturation (CVM) degree and growth spurts by analyzing lateral cephalometric radiographs. Methods: The study sample included 890 cephalograms. The images were classified into six cervical stages independently by two orthodontists. The images were also categorized into three degrees on the basis of the growth spurt: pre-pubertal, growth spurt, and post-pubertal. Subsequently, the samples were fed to a transfer learning model implemented using the Python programming language and PyTorch library. In the last step, the test set of cephalograms was randomly coded and provided to two new orthodontists in order to compare their diagnosis to the artificial intelligence (AI) model's performance using weighted kappa and Cohen's kappa statistical analyses. Results: The model's validation and test accuracy for the six-class CVM diagnosis were 62.63% and 61.62%, respectively. Moreover, the model's validation and test accuracy for the three-class classification were 75.76% and 82.83%, respectively. Furthermore, substantial agreements were observed between the two orthodontists as well as one of them and the AI model. Conclusions: The newly developed AI model had reasonable accuracy in detecting the CVM stage and high reliability in detecting the pubertal stage. However, its accuracy was still less than that of human observers. With further improvements in data quality, this model should be able to provide practical assistance to practicing dentists in the future.

• Childhood Kidney Diseases
• /
• 제7권2호
• /
• pp.150-156
• /
• 2003

목적 : IgA 신병증의 임상 및 병리학적 소견의 비교를 통해서 IgA 신병증의 신조직 소견을 예측할 수 있는 인자를 찾아봄으로써 신생검 시기를 결정하는데 도움이 될 수 있는 지표를 찾아본다. 방법 : 원주기독병원에서 신생검상 IgA 신병증으로 확인된 40례를 대상으로 후향적 연구를 시행하였다. 결과 : 40명의 환자 중 크레아티닌이 1.5 mg/dL 이상을 보인 1명은 추적 관찰이 안되었고 나머지 39명 중 2명에서는 혈중크레아티닌이 1.5mg/dL 이상으로 진행하였으며 그 중의 한명은 진단 후 2년 7개월만에 말기신부전으로 진행하였다. WHO 분류에 따르면 class I이 15명, class II가 14명, class III가 7명, classs IV가 3명, class V 0명이었다. 육안적 혈뇨군에서 현미경적 혈뇨군보다 class I과 class II의 비율이 의미있게 높았다(P<0.02). 세뇨관간질 소견만을 분류해 보았을 때 grade 1은 24명, grade 2는 4명, grade 3는 8명, grade 4는 3명이었다. 세뇨관간질의 변화가 심해질수록 24시간 요단백/알부민 비율이 증가하였다(r=-0.32, P<0.05) 혈청 크레아티닌치가 0.79 mg/dL 이하인 경우 세뇨관간질소견이 경하였으며, 크레아티닌치가 1.13 mg/dL 보다 큰 경우 세뇨관간질병변이 심하였다. 육안적 혈뇨군에서 혈청 크레아티닌치는 현미경적 혈뇨군에 비해 낮았고(0.78 vs 1.09 mg/dL)(P=0.027), 혈청 IgA 치가 높았으며(316.3 vs 198.8mg/dL)(P=0.027), class I과 II의 낮은 WHO 분류 class를 보였다(23 vs 4)(P=0.029). 결론 : IgA 신병증 환자에서 혈청크레아티닌이 0.79 mg/dL 이하, 육안적 혈뇨, 높은 24시간 urine protein/albumin 비율을 보이는 경우 경한 신조직 소견을 보임으로 이런 소견을 보이는 경우 신조직 생검 시기를 미룰 수 있을 것으로 사료된다.

• 대한치과교정학회지
• /
• 제23권4호
• /
• pp.633-644
• /
• 1993

There has been so much controversies about the position of upper and lower jaws, and their first permanent molars in normal occlusion and Angle's class $I{\cdot}II{\cdot}III$ malocclusions. So, the purpose of this study is to compare the position of upper and lower jaws, and their first molars in normal occlusion and Angle's class $I{\cdot}II{\cdot}III$ malocclusions by lateral cephalometric analysis. The sample consisted of one hundred and twenty girls(thirty in each group) who had completed growth. The findings of this study were as follows : 1. In class I malocclusion, both maxilla and mandible were slightly posterior position than normal occlusion, but they showed harmonious relationship. 2. In class II malocclusion, the mandible was greatly retruded, and the maxilla was also slightly retruded to the cranial base as compared with normal occlusion. 3. In class III malocclusion, the maxilla was significantly retruded to the cranial base, but no significant difference was found in mandibular position as compared with normal occlusion. 4. The maxillary first molar was located at posterior position in class II malocclusion, and anterior position in class III malocclusion to the cranium, so that the rotation of mandible was influenced by that. 5. The mandibular first molar showed constant relationship to the mandible in all four groups, but different position to the cranial base in direct proportion to the mandibular position. 6. On the treatment planning of class III malocclusion, it seems to be better to promote the mandibular horizontal growth by inhibiting the vertical growth of maxillary molar area, and on the treatment planning of class III malocclusion, it seems to be better to promote the antero-inferior growth of maxilla mi to promote the mandibular vertical growth by inducing the vertical growth of maxillary molar area.

• 대한본초학회지
• /
• 제32권4호
• /
• pp.33-38
• /
• 2017

Objectives : Termitomyces albuminosus (Berk.) Heim is one of the famous wild edible mushrooms in the southern part of China. It is known that Termitomyces albuminosus, like Aconitum koreanum used in Korean traditional medicine, contains a kind of cerebroside, termitomycesphin, causing a pharmacologic effect on the neuron system. The pharmacologic effect of Termitomyces albuminosus can be used to possibly replace Aconitum koreanum. However, It needs to be certified as safe before it can be used. Here, a single-oral toxicity test and safety classification was conducted to obtain acute information of the toxicity of dried-Termitomyces albuminosus powder and to secure its safety in clinical applications. Methods : In order to calculate approximate lethal dose(ALD), test substance was orally administered to male and female SD-rat at dose levels of 5,000 and 0 (vehicle control) mg/kg (body weight). Based on the result of this toxicity, also the estimation of safety classification was calculated using the HED-based (human equivalent dose) MOS (margin of safety). Results : There were no mortalities, test substances treatment-related clinical signs, no changes in the body or organ weights, and no gross or histopathological findings at 14 days after treatment with test substance. Thus, the approximate lethal dose of dried-Termitomyces albuminosus powder was considered over 5,000 mg/kg in both female and male mice. Conclusions : Based on the limit dose, 5000 mg/kg, it was estimated that dried-Termitomyces albuminosus powder is classified as "Specified class B" indicating that clinical dose is not limited to patients as safe as food.

• Journal of Species Research
• /
• 제8권2호
• /
• pp.197-214
• /
• 2019

The taxonomic hierarchy of the phylum Proteobacteria was assessed, after which the isolation and classification state of Proteobacteria species with valid names for Korean indigenous isolates were studied. The hierarchical taxonomic system of the phylum Proteobacteria began in 1809 when the genus Polyangium was first reported and has been generally adopted from 2001 based on the road map of Bergey's Manual of Systematic Bacteriology. Until February 2018, the phylum Proteobacteria consisted of eight classes, 44 orders, 120 families, and more than 1,000 genera. Proteobacteria species isolated from various environments in Korea have been reported since 1999, and 644 species have been approved as of February 2018. In this study, all novel Proteobacteria species from Korean environments were affiliated with four classes, 25 orders, 65 families, and 261 genera. A total of 304 species belonged to the class Alphaproteobacteria, 257 species to the class Gammaproteobacteria, 82 species to the class Betaproteobacteria, and one species to the class Epsilonproteobacteria. The predominant orders were Rhodobacterales, Sphingomonadales, Burkholderiales, Lysobacterales and Alteromonadales. The most diverse and greatest number of novel Proteobacteria species were isolated from marine environments. Proteobacteria species were isolated from the whole territory of Korea, with especially large numbers from the regions of Chungnam/Daejeon, Gyeonggi/Seoul/Incheon, and Jeonnam/Gwangju. Most Halomonadaceae species isolated from Korean fermented foods and solar salterns were halophilic or halotolerant. Air-borne members of the genera Microvirga, Methylobacterium, and Massilia had common characteristics in terms of G+C content, major respiratory quinones, and major polar lipids.

• 대한치과교정학회지
• /
• 제52권2호
• /
• pp.85-101
• /
• 2022

Objective: To classify facial asymmetry (FA) phenotypes in adult patients with skeletal Class III (C-III) malocclusion. Methods: A total of 120 C-III patients who underwent orthognathic surgery (OGS) and whose three-dimensional computed tomography images were taken one month prior to OGS were evaluated. Thirty hard tissue landmarks were identified. After measurement of 22 variables, including cant (°, mm), shift (mm), and yaw (°) of the maxilla, maxillary dentition (Max-dent), mandibular dentition, mandible, and mandibular border (Man-border) and differences in the frontal ramus angle (FRA, °) and ramus height (RH, mm), K-means cluster analysis was conducted using three variables (cant in the Max-dent [mm] and shift [mm] and yaw [°] in the Manborder). Statistical analyses were conducted to characterize the differences in the FA variables among the clusters. Results: The FA phenotypes were classified into five types: 1) non-asymmetry type (35.8%); 2) maxillary-cant type (14.2%; severe cant of the Max-dent, mild shift of the Man-border); 3) mandibular-shift and yaw type (16.7%; moderate shift and yaw of the Man-border, mild RH-difference); 4) complex type (9.2%; severe cant of the Max-dent, moderate cant, severe shift, and severe yaw of the Man-border, moderate differences in FRA and RH); and 5) maxillary reverse-cant type (24.2%; reverse-cant of the Max-dent). Strategic decompensation by pre-surgical orthodontic treatment and considerations for OGS planning were proposed according to the FA phenotypes. Conclusions: This FA phenotype classification may be an effective tool for differential diagnosis and surgical planning for Class III patients with FA.

• 방송공학회논문지
• /
• 제25권5호
• /
• pp.750-757
• /
• 2020

공연에 사용하는 로봇이 인간과의 상호작용하기 위한 기본 성능 중 하나는 인간의 행동을 빠르고 정확하게 파악하는 것이다. 따라서 로봇이 인간의 자세를 추정할 때 자세 인식의 정확도를 높임과 동시에 가능한 빠른 속도로 인식할 수 있어야 한다. 그러나 현재 인공지능 기술의 대표적인 방식인 딥 러닝을 사용하여 인간의 자세를 추정할 경우, 인식의 정확도와 속도라는 두 가지 성능을 동시에 만족하지 못하고 있다. 따라서 사용 목적에 따라 추론정확도가 높은 하향식 자세추정과 처리속도가 빠른 상향식 자세추정 중 하나를 선택해서 사용하는 것이 일반적이다. 본 논문에서는 앞서 언급한 두 가지 방식이 가진 장점을 모두 포함하면서 단점을 보완한 두 가지 방식을 제안한다. 첫 번째는 다중 그래픽 처리 장치를 활용해 상향식 자세추정과 물체검출을 병렬로 사용하는 방식이고, 두 번째는 상향식 자세추정과 단항분류를 융합하는 방식이다. 실험을 통해 두 가지 방식 모두 속도가 개선됨을 증명했다. 공연로봇에 이 두 가지 방식 중 하나를 사용한다면, 관객과 신뢰도 높으며 보다 빠른 상호작용을 수행할 수 있을 것으로 기대된다.