• Imaging Science in Dentistry
• /
• v.49 no.4
• /
• pp.281-286
• /
• 2019

Purpose: The purpose of this study was to evaluate the morphology of the mandibular cortex in cases of medication-related osteonecrosis of the jaw (MRONJ) in patients with osteoporosis or bone metastases using a computer programme. Materials and Methods: Fifty-four patients with MRONJ (35 with osteoporosis and 19 with bone metastases) were examined using panoramic radiography. The morphology of the mandibular cortex was evaluated using a computer programme that scanned the mandibular inferior cortex and automatically assessed the mandibular cortical index (MCI) according to the thickness and roughness of the mandibular cortex, as follows: normal (class 1), mildly to moderately eroded (class 2), or severely eroded (class 3). The MCI classifications of MRONJ patients with osteoporosis or bone metastases were evaluated with the Pearson chi-square test. In these analyses, a 5% significance level was used. Results: The MCI of MRONJ patients with osteoporosis(class 1: 6, class 2: 15, class 3: 14) tended to be higher than that of patients with bone metastases(class 1: 14, class 2: 5, class 3: 0)(P=0.000). Conclusion: The use of a computer programme to assess mandibular cortex morphology may be an effective technique for the objective and quantitative evaluation of the MCI in MRONJ patients with osteoporosis or bone metastases.

• Journal of The Korean Astronomical Society
• /
• v.39 no.4
• /
• pp.139-145
• /
• 2006

We have made a comprehensive statistical study on the coronal mass ejections(CMEs) associated with helmet streamers. A total number of 3810 CMEs observed by SOHO/LASCO coronagraph from 1996 to 2000 have been visually inspected. By comparing their LASCO images and running difference images, we picked out streamer-associated CMEs, which are classified into two sub-groups: Class-A events whose morphological shape seen in the LASCO running difference image is quite similar to that of the pre-existing streamer, and Class-B events whose ejections occurred in a part of the streamer. The former type of CME may be caused by the destabilization of the helmet streamer and the latter type of CME may be related to the eruption of a filament underlying the helmet streamer or narrow CMEs such as streamer puffs. We have examined the distributions of CME speed and acceleration for both classes as well as the correlation between their speed and acceleration. The major results from these investigations are as follows. First, about a quarter of all CMEs are streamer-associated CMEs. Second, their mean speed is 413 km $s^{-1}$ for Class-A events and 371 km $s^{-1}$ for Class-B events. And the fraction of the streamer-associated CMEs decreases with speed. Third, the speed-acceleration diagrams show that there are no correlations between two quantities for both classes and the accelerations are nearly symmetric with respect to zero acceleration line. Fourth, their mean angular width are about $60^{\circ}$, which is similar to that of normal CMEs. Fifth, the fraction of streamer-associated CMEs during the solar minimum is a little larger than that during the solar maximum. Our results show that the kinematic characteristics of streamer-associated CMEs, especially Class-A events, are quite similar to those of quiescent filament-associated CMEs.

• Journal of the Korean Society of Environmental Restoration Technology
• /
• v.10 no.5
• /
• pp.20-30
• /
• 2007

Environmental Conservation Value Assessment Map (ECVAM) is the class map to divide the national land into conservation areas and development areas based on legal and ecological assessment criteria. It contributes to enhancements of the efficiency and the scientificity when framing a policy in various fields including the environment. However, it is impossible to understand the multiphase vegetation structure as data on judging the national forest class in ECVAM are restricted to areal information of Ecological Nature Status, Degree of Green Naturality and Forest Map. This point drops the reliability of ECVAM. Therefore we constructed vegetation information using LiDAR (Light Detection And Raging) technology. We generated Biomass Class Maps as final results of this study, to introduce the new forest assessment criterion in ECVAM that alternates or makes up for existing forest assessment criteria. And then, we compared these with Forest Map and Landsat TM NDVI image. As a result, biomass classes are generally higher than stand age classes and DBH classes of Vegetation Map, and lower than NDVI of Landsat TM image because of the difference of time on data construction. However distributions between these classes are mostly similar. Therefore we estimates that it is possible to apply the biomass item to the new forest assessment criterion of ECVAM. The introduction of the biomass in ECVAM makes it useful to detect the vegetation succession, to adjust the class of the changed zone since the production of Vegetation Map and to rectify the class error of Vegetation Map because variations on tree heights, forest area, gaps between trees, vegetation vitality and so on are acquired as interim findings in process of computing biomass.

• Journal of The Korean Association For Science Education
• /
• v.28 no.4
• /
• pp.350-358
• /
• 2008

The purpose of this study was to investigate the effects of interview with a scientist and engineer on service performance assessment on science career orientation and image of scientists. Science track students in the 11th grade carried out the interviews and made powerpoint presentations. After the students' presentation in the chemistry class, the teacher made comments on the contents of the interviews. Students gave presentation in each class for a year. Before starting this assessment, students took science career orientation questionnaire and DAST (draw-a-scientist-test). These two tests were conducted again at the end of the year. The results of this study showed that there was no significant difference between pre- and post-test score for the science career orientation. However, a significant difference was observed in the 'preference for science learning' category. These results showed that the career decision of a high school student has already been fixed rigidly. On the other hand, there was a significant difference (p < 0.01) between pre- and post-test on the image of scientists. This demonstrated that the stereotypic image for a scientist was reduced by the interview performance assessment and that, students came to have an affirmative perception of scientists on service.

• Journal of the Korea Convergence Society
• /
• v.10 no.9
• /
• pp.23-29
• /
• 2019

In this paper, the mood of images was classified into eight categories through a deep convolutional neural network and video was automatically generated using proper background music. Based on the collected image data, the classification model is learned using a multilayer perceptron (MLP). Using the MLP, a video is generated by using multi-class classification to predict image mood to be used for video generation, and by matching pre-classified music. As a result of 10-fold cross-validation and result of experiments on actual images, each 72.4% of accuracy and 64% of confusion matrix accuracy was achieved. In the case of misclassification, by classifying video into a similar mood, it was confirmed that the music from the video had no great mismatch with images.

• Journal of Korean Society for Geospatial Information Science
• /
• v.1 no.1 s.1
• /
• pp.181-192
• /
• 1993

Image is composed of the digital numbers including information on natural phenomena, their condition and the kind of objects. Digital numbers change in geometric correction(that is preprocessing). This change of digital numbers gave an effect on results of land-cover classification. We intend to know the influence of resampling as classifying land-cover using the image reconstructed by geometric correction in this paper. Chun-cheon basin was selected the study area having most variable land-cover pattern in North-Han river valley and made on use of RESTEC data resampled in preprocessing. Land-cover is classified as six classes of LEVEL I using maximum likelyhood classification method. We classified land-cover using the image resampled by two methods in this study. Bilinear interpolation method was most accurate in five classes except bear-land in the result of comparing each class with topographic map. We should choose the method of resampling according to the class in which we put the importance in the image resampling of geometric correction. And if we use four-season's image, we may classify more accurately in case of the confusion in case of the confusion in borders of rice field and farm.

• Advances in nano research
• /
• v.12 no.2
• /
• pp.185-195
• /
• 2022

Deep learning is another field of artificial intelligence (AI) utilized for computer aided diagnosis (CAD) and image processing in scientific research. Considering numerous mechanical repetitive tasks, reading image slices need time and improper with geographical limits, so the counting of image information is hard due to its strong subjectivity that raise the error ratio in misdiagnosis. Regarding the highest mortality rate of Lung cancer, there is a need for biopsy for determining its class for additional treatment. Deep learning has recently given strong tools in diagnose of lung cancer and making therapeutic regimen. However, identifying the pathological lung cancer's class by CT images in beginning phase because of the absence of powerful AI models and public training data set is difficult. Convolutional Neural Network (CNN) was proposed with its essential function in recognizing the pathological CT images. 472 patients subjected to staging FDG-PET/CT were selected in 2 months prior to surgery or biopsy. CNN was developed and showed the accuracy of 87%, 69%, and 69% in training, validation, and test sets, respectively, for T1-T2 and T3-T4 lung cancer classification. Subsequently, CNN (or deep learning) could improve the CT images' data set, indicating that the application of classifiers is adequate to accomplish better exactness in distinguishing pathological CT images that performs better than few deep learning models, such as ResNet-34, Alex Net, and Dense Net with or without Soft max weights.

• Journal of radiological science and technology
• /
• v.23 no.1
• /
• pp.103-111
• /
• 2000

In recent years, medical procedures have become more complex, while financial pressures for shortened hospital stays and increased efficiency in patient care have increased. As a result, several shortcomings of present film-based systems for managing medical images have become apparent. Maintaining film space is labor intensive and consumes valuable space. Because only single copies of radiological examinations exist, they are prone to being lost or misplaced, thereby consuming additional valuable time and expense. In this paper, MINI-PACS for image archiving, transmission, and viewing offers a solution to these problems. Proposed MINI-PACS consists of mainly four parts such as Web Module, Client-Server Module, Internal Module, Acquisition Module. In addition, MINI-PACS system includes DICOM Converter that Non-DICOM file format converts standard file format. In Client-Server Module case, Proposed system is combined both SCU(Service Class User: Client) part and SCP(Service Class Provider: Server)part therefore this system provides the high resolution image processing techniques based on windows platform. Because general PACS system is too expensive for Medium and Small hospitals to install and operate the full-PACS. Also, we constructed Web Module for database connection through the WWW.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2000.10b
• /
• pp.887-890
• /
• 2000

본 연구에서는 내용기반 영상 데이터 검색을 위하여 변환 영역에서 위치 정보와 주파수 정보를 가지는 웨이블릿 성질을 이용하여 영상을 압축한 후에 저주파 성분에 의한 객체들의 특징을 추출하는 방안으로 Vector Quantization 을 이용한 class 별 영상 검색을 제시한다 내용기반 영상 검색의 주요특징들은 색상, 질감, 그리고 영상의 공간적인 특징을 고려한 특징 값 둥이 사용된다. 먼저 검색의 효율성을 높이기 위해 영상을 구성하는 특징 치 중에서 가장 빈도가 많은 class 부터 영상의 유사도를 검색한 후에 다음으로 영상을 구성하는 빈도가 큰 순서대로 DB 내에 저장되어 있는 영상과 비교를 하게 된다. DB내 영상 검색은 빈도수가 우선인 5개의 class를 기준으로 유사도를 측정해서 검색을 이룬다. 이러한 영상의 특징들을 어떻게 결합하고 특징 추출을 하느냐에 따라 검색의 효율성에 영향을 준다. 따라서 본 연구에서는 영상의 위치 정보와 주파수 정보를 가지는 웨이블릿 변환 후 얻어지는 저대역 부밴드에서의 공간적인 특성을 고려한 특징 값을 이용하여 Vector Quantization 알고리즘에 의해 정지영상의 객체 대표 특징들을 마르게 검색하고자 한다. 본 연구에서는 Haar Wavelet과 Vector Quantization 에서 색상과 질감의 가중치를 적용한 후 DB 에 저장된 영상과 유사도를 검색하는 방법을 취하고자 한다.

• Journal of Integrative Natural Science
• /
• v.5 no.2
• /
• pp.120-126
• /
• 2012

This paper describes the algorithm that lowers the dimension, maintains the gesture recognition and significantly reduces the eigenspace configuration time by combining the higher correlation feature information and Principle Component Analysis. Since the suggested method doesn't require a lot of computation than the method using existing geometric information or stereo image, the fact that it is very suitable for building the real-time system has been proved through the experiment. In addition, since the existing point to point method which is a simple distance calculation has many errors, in this paper to improve recognition rate the recognition error could be reduced by using several successive input images as a unit of recognition with K-Nearest Neighbor which is the improved Class to Class method.