• 대한의용생체공학회:의공학회지
• /
• 제42권5호
• /
• pp.232-240
• /
• 2021

Heterogeneity in cancer is the major obstacle for precision medicine and has become a critical issue in the field of a cancer diagnosis. Many attempts were made to disentangle the complexity by molecular classification. However, multi-dimensional information from dynamic responses of cancer poses fundamental limitations on biomolecular marker-based conventional approaches. Cell morphology, which reflects the physiological state of the cell, can be used to track the temporal behavior of cancer cells conveniently. Here, we first present a hybrid learning-based platform that extracts cell morphology in a time-dependent manner using a deep convolutional neural network to incorporate multivariate data. Feature selection from more than 200 morphological features is conducted, which filters out less significant variables to enhance interpretation. Our platform then performs unsupervised clustering to unveil dynamic behavior patterns hidden from a high-dimensional dataset. As a result, we visualize morphology state-space by two-dimensional embedding as well as representative morphology clusters and trajectories. This cell morphology profiling strategy by hybrid learning enables simplification of the heterogeneous population of cancer.

• Advances in biomechanics and applications
• /
• 제1권2호
• /
• pp.111-126
• /
• 2014

The microarchitecture of trabecular bone plays a significant role in mechanical strength due to its load-bearing capability. However, the complexity of trabecular microarchitecture hinders the evaluation of its morphological characteristics. We therefore propose a new classification method based on static multiscale theory and dynamic finite element method (FEM) analysis to visualize a three-dimensional (3D) trabecular network for investigating the influence of trabecular microarchitecture on load-bearing capability. This method is applied to human vertebral trabecular bone images obtained by micro-computed tomography (micro-CT) through which primary trabecular bone is successfully visualized and extracted from a highly complicated microarchitecture. The morphological features were then analyzed by viewing the percolation of load pathways in the primary trabecular bone by using the stress wave propagation method analyzed under impact loading. We demonstrate that the present method is effective for describing the morphology of trabecular bone and has the potential for morphometric measurement applications.

• Imaging Science in Dentistry
• /
• 제51권4호
• /
• pp.383-388
• /
• 2021

Purpose: The aim of this study was to evaluate the root canal morphology of mesial roots of mandibular first molars. Materials and Methods: Forty extracted mandibular first molars were used in this study. The morphological examination of root canals was conducted in accordance with the Vertucci classification using micro-computed tomography (micro-CT). Any aberrant root canal configurations not included in the Vertucci classification were recorded, and their frequency was established using descriptive statistics. Intra-observer reliability was assessed using the Wilcoxon signed-rank test, while inter-observer reliability was assessed using the Cohen kappa test. Significance was evaluated at the P<0.05 level. Results: The mesial roots of mandibular first molars had canal configurations of type I (15%), type II (7.5%), type III (25%), type IV (10%), type V (2.5%), type VI (7.5%), and type VII (7.5%). The images showed 10 (25%) additional configuration types that were not included in the Vertucci classification. These types were 1-3-2-3, 1-2-3-2-3, 2-3-1, 2-3, 1-2-3-1, 2-1-2-3, 3-2-1, 1-2-3-1, 2-3-2-3, and 1-2-1-2-1. The intra-observer differences were not statistically significant(P>0.05) and the kappa value for inter-observer agreement was found to be 0.957. Conclusion: Frequent variations were detected in mesial roots of mandibular first molars. Clinicians should take into consideration the complex structure of the root canal morphology before commencing root canal treatment procedures to prevent iatrogenic complications. Micro-CT was a highly suitable method to provide accurate 3-dimensional visualizations of root canal morphology.

• Asian Pacific Journal of Cancer Prevention
• /
• 제17권2호
• /
• pp.835-838
• /
• 2016

Background: Lung cancer remains one of the most common cancers in the world, both in terms of new cases (about 13% of total per year) and deaths (nearly one cancer death in five), because of the high case fatality. Errors in lung cancer type or malignant growth determination lead to degraded treatment efficacy, because anticancer strategy depends on tumor morphology. Materials and Methods: We have made an attempt to evaluate effectiveness of machine learning algorithms in the task of lung cancer classification based on gene expression levels. We processed four publicly available data sets. The Dana-Farber Cancer Institute data set contains 203 samples and the task was to classify four cancer types and sound tissue samples. With the University of Michigan data set of 96 samples, the task was to execute a binary classification of adenocarcinoma and non-neoplastic tissues. The University of Toronto data set contains 39 samples and the task was to detect recurrence, while with the Brigham and Women's Hospital data set of 181 samples it was to make a binary classification of malignant pleural mesothelioma and adenocarcinoma. We used the k-nearest neighbor algorithm (k=1, k=5, k=10), naive Bayes classifier with assumption of both a normal distribution of attributes and a distribution through histograms, support vector machine and C4.5 decision tree. Effectiveness of machine learning algorithms was evaluated with the Matthews correlation coefficient. Results: The support vector machine method showed best results among data sets from the Dana-Farber Cancer Institute and Brigham and Women's Hospital. All algorithms with the exception of the C4.5 decision tree showed maximum potential effectiveness in the University of Michigan data set. However, the C4.5 decision tree showed best results for the University of Toronto data set. Conclusions: Machine learning algorithms can be used for lung cancer morphology classification and similar tasks based on gene expression level evaluation.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 1999년도 추계종합학술대회 논문집
• /
• pp.637-640
• /
• 1999

In this Paper we show the LCD simulator for defect inspection using image processing algorithm and neural network. The defect inspection algorithm of the LCD consists of preprocessing, feature extraction and defect classification. Preprocess removes noise from LCD image, using morphology operator and neural network is used for the defect classification. Sample images with scratch, pinhole, and spot from real LCD color filter image are used. The proposed algorithms show that defect detected and classified in the ratio of 92.3% and 94.6 respectively.

• Journal of Plant Biology
• /
• 제17권2호
• /
• pp.99-105
• /
• 1974

Author has studied and reported on taxonomy of Korean sedges, using gross morphology, anatomy and epidermal patterns of the leaf blades(1969, 1971, 1973, 1974). This paper is the 6th report of epidermal patterns of leaf blade on sedges and includes 5 genera, Eriophorum, Fuirena, Kobresia, Rhynchospora and Scirpus. The author proposed to find epidermal patterns of leaf blades as an important taxonomic characteristic of sedges classification. The result of this study, the elements of leaf epidermis, subsidal cells, silica body, cell wall of long cell, prickles, and arrangement of the elements are considered to be significant characteristics for the identification and classification of sedge.

• 구강회복응용과학지
• /
• 제29권3호
• /
• pp.236-248
• /
• 2013

본 연구는 정상 안모이면서 정상교합자인 37명을 대상으로 이마 형태를 분류한 후 분류 기준값을 찾아보았고 이마 분류에 따른 상악 전치의 위치차이를 연구하였다. 또 이마형태에 영향을 주는 인자들과 상악 전치의 위치와의 상관관계를 조사하여 다음과 같은 결과를 얻었다. 1. 이마의 형태는 angular, round, straight, concave 형태로 구분 가능하다. 2. 이마의 형태를 분류할 수 있는 특정 기준 값은 존재하지 않았지만 S value와 이마길이(Tri-Gla)를 이용하여 이마형태의 분류 가능성이 존재하였다. 3. 이마의 형태에 따른 상악 전치의 위치는 차이는 존재하지 않았다. 4. 이마 기울기와 Andrews 분석값은 유의한 음의 상관관계를 갖는다. 즉 이마 기울기가 커질수록 상악 전치는 후방 위치하게 되며 다음과 같은 공식 Andrew analysis = -0.39*Forehead inclination으로 표현할 수 있다.

• 한국한의학연구원논문집
• /
• 제3권1호
• /
• pp.153-167
• /
• 1997

Conventionally, identification and classification methods of natural products include the morphological survey and assay of chemical disposition, sing these methods, however, is not satisfying for the precise identification of natural products because they are often valiable in the compositions and morphology To standardize the natural products identification and classification, genomic DNA analysis such as RAPD, RFLP and Amp-FLP can be adopted for this purpose. In this study, various ginsengs and bear gall bladder were tested for the development of genetic identification and classification method. Varieties of ginsengs such as, P. ginseng, P. quinquefolium, P. japonicus and P. notoginseng, were genetically analyzed by RAPD. Also, DNA isolated from Bear blood and gall bladder, Ursus thibetanus, Ursus americanus and Ursus arctos, were analyzed by the same method. The results demonstrated that the identification and classification of bear gall bladder and various ginsengs were possible by RAPD analysis. Therefore, this method was thought to be used as a additional method for the identification and classification of other natural products.

• 한국컴퓨터정보학회논문지
• /
• 제23권4호
• /
• pp.147-153
• /
• 2018

In this paper, we propose an efficient WBC 14-Diff classification which performs using the WBC-ResNet-152, a type of CNN model. The main point of view is to use Super-pixel for the segmentation of the image of WBC, and to use ResNet for the classification of WBC. A total of 136,164 blood image samples (224x224) were grouped for image segmentation, training, training verification, and final test performance analysis. Image segmentation using super-pixels have different number of images for each classes, so weighted average was applied and therefore image segmentation error was low at 7.23%. Using the training data-set for training 50 times, and using soft-max classifier, TPR average of 80.3% for the training set of 8,827 images was achieved. Based on this, using verification data-set of 21,437 images, 14-Diff classification TPR average of normal WBCs were at 93.4% and TPR average of abnormal WBCs were at 83.3%. The result and methodology of this research demonstrates the usefulness of artificial intelligence technology in the blood cell image classification field. WBC-ResNet-152 based morphology approach is shown to be meaningful and worthwhile method. And based on stored medical data, in-depth diagnosis and early detection of curable diseases is expected to improve the quality of treatment.

• 한국도서관정보학회지
• /
• 제39권3호
• /
• pp.369-386
• /
• 2008

이 연구는 식물학의 학문분류 체계와 문헌분류 체계를 비교 분석함으로써 식물학의 분류특성과 문제점을 분석하고, 이를 토대로 KDC 식물학의 분류체계를 개선할 수 있는 방안을 제시하고자 시도되었다. 이 연구결과를 요약하면 아래와 같다. 첫째, 식물학의 학문분류는 주로 식물의 연구대상에 따라 식물 형태학, 식물 생리학, 식물 생태학, 식물 계통학, 식물 유전학, 식물 진화학 등으로 구분하고 있다. 둘째, 식물의 분류는 식물 계통학이라는 하위 분과학문에서 다루고 있으며, Engler 체계 등이 일반화되어 있다. 셋째, 식물학의 문헌분류 체계는 대부분 식물학의 분과학문 위주가 아니라 식물의 분류 위주로 구성되어 있다. 이때 KDC, NDC, UDC, CC에서는 식물분류를 식물의 발달과정 즉, 진화순서에 의해 하등식물에서 고등식물로 배열되어 있는 Engler 체계를 적용하고 있으며, DDC와 LCC는 현존하는 식물에 더 중점을 두고, 고등식물에서 하등식물로 배열되어 있는 Bentham & Hooker 체계와 유사성이 있다. 넷째, KDC 식물학에서 있어서는 DDC나 CC와 같이 일반 식물학에서 다루고 있는 식물의 구조나 속성 등을 482-489에 나열되어 있는 모든 식물에 공통적으로 적용하여 세분할 수 있도록 구조화 하는 것이 바람직할 것이다.