• Progress in Medical Physics
• /
• v.30 no.2
• /
• pp.49-58
• /
• 2019

Purpose: Automated analytical systems have begun to emerge as a database system that enables the scanning of medical images to be performed on computers and the construction of big data. Deep-learning artificial intelligence (AI) architectures have been developed and applied to medical images, making high-precision diagnosis possible. Materials and Methods: For diagnosis, the medical images need to be labeled and standardized. After pre-processing the data and entering them into the deep-learning architecture, the final diagnosis results can be obtained quickly and accurately. To solve the problem of overfitting because of an insufficient amount of labeled data, data augmentation is performed through rotation, using left and right flips to artificially increase the amount of data. Because various deep-learning architectures have been developed and publicized over the past few years, the results of the diagnosis can be obtained by entering a medical image. Results: Classification and regression are performed by a supervised machine-learning method and clustering and generation are performed by an unsupervised machine-learning method. When the convolutional neural network (CNN) method is applied to the deep-learning layer, feature extraction can be used to classify diseases very efficiently and thus to diagnose various diseases. Conclusions: AI, using a deep-learning architecture, has expertise in medical image analysis of the nerves, retina, lungs, digital pathology, breast, heart, abdomen, and musculo-skeletal system.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.19 no.4
• /
• pp.568-573
• /
• 2009

Marine accidents have always caused huge economic losses, as well as environmental pollution. Prevention of marine accidents has become a focus of argumentation. The analysis of past accident cases, reviewing the experience and lessons, is important and necessary for preventing marine accidents. With the same subject above, the Korean Maritime Safety Tribunal provides for past marine accidents' written judgments and analysis of judgment and associated retrieval system on its homepage. In these systems, the name of the ship, accident occurrence time, accident pattern or related keywords are used as search conditions. However, most of the marine events' happening were not due to a single reason, but multiple ones. In addition, one marine event could often come under several categories. In this case, now the retrieval systems' DB is used on the Korean Maritime Safety Tribunal homepage was built based on single category and failed to be able to retrieve according to multiple reasons or multiple categories. In order to solve this problem, a more practical retrieval approach might be needed. Therefore, in this paper, a new retrieval system will be proposed, which using the linguistic label to describe the cluster after analyzing the relational properties between marine accidents and clustering by FCM algorithm, and then adding an interface to allow users to get the results they want through choosing multiple reasons or multiple categories.

• The Mathematical Education
• /
• v.29 no.1
• /
• pp.21-34
• /
• 1990

This is a study on an instruction of estimation for error correction in the calculation with a computer or a calculator. The aim of this study is to survey a new aspect of calaulation teaching and the teaching strategy of estimation and finally to frame a new curriculum model of estimation instruction. This research required a year and the outcomes of the research can be listed as follows: 1. Social utilities of estimation were made clear, and a new trend of calculation teaching related to estimation instruction was shown. 2. The definition of estimation was given and actual examples of conducting an estimation among pupils in lower grades were given for them to have abundant experiences. 3. The ways of finding estimating values in fraction and decimal fraction were presented for the pupils to be able to conduct an estimation. 4. The following contents were given as a basic strategy for estimation. 1) Front-end strategy 2) Clustering strategy 3) Rounding strategy 4) Compatible numbers strategy 5) Special numbers strategy 5. In an instuction of estimation the meaning, method. and process of calculation and calculating algorithm were reviewed for the cultivation of children's creativity through promoting their basic skill, mathematical thinking and problem-solving ability. 6. The following contents were also covered as an estimation strategy for measurement 1) Calculating the sense of quantity on the size of unit. 2) Estimating the total quantity by frequent repetition of unit quantity. 3) Estimating the length and the volume by weighing. 4) Estimating unknown quantity based on the quatity already known. 5) Estimating the area by means of equivalent area transformation. 7. The ways of instructing mental computation were presented. 8. Reviews were made on the curricular and the textbook contents concerning estimation instructions both in Korea and Japan. and a new model of curriculum was devised with reference to estimation instruction data of the United States.

• Journal of the Korean Society of School Health
• /
• v.22 no.2
• /
• pp.169-181
• /
• 2009

Purpose: Mood disorders such as depression and bipolar disorder are a major mental health problem in college students. We investigate the prevalence of depression and bipolar disorder and the relevance of risk factors for these mood disorders among one college freshmen. Methods: The subjects were 2,865 college students who entered one university located in Seoul and Ansung in 2009. We used BDI (Beck Depression Inventory) for depression assessment and K-MDQ (Mood Disorder Questionnaire) for bipolar disorder assessment. Demographic and socioeconomic factors were measured by questionnaire. Height, weight, blood pressure, total cholesterol, complete blood cell count, and liver function test data were obtained by physical examination for freshmen. Chi-square test and multiple logistic regression were performed to analyze the possible risk factors for depression and bipolar disorder. Results: With different BDI cutoff value, 16 and 21, the prevalence of depression was 8.7% (male: 7.6%, female: 10.1%) and 2.4% (male: 2.5%, female: 2.3%), separately. 'Low economic status', 'urban birth place', and 'low grade at entrance' were significantly associated with depression. Using the original cutoff criterion, defined as clustering of 7 or more symptoms that caused moderate or severe problems, the prevalence of bipolar disorder was 1.3% (male: 1.4%, female: 1.1%). The risk factor of bipolar disorder was academic fields (art fields). Conclusion: Depression and bipolar disorder are common disease in college freshmen. Therefore, Campus-based mental health service program is needed to help with prevention of and early intervention of these mood disorders.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.13 no.6
• /
• pp.635-642
• /
• 2003

Reinforcement learning got successful results in a lot of applications such as control and scheduling. Various function approximation methods have been studied in order to improve the learning speed and to solve the shortage of storage in the standard reinforcement learning algorithm of Q-Learning. Most function approximation methods remove some special quality of reinforcement learning and need prior knowledge and preprocessing. Fuzzy Q-Learning needs preprocessing to define fuzzy variables and Local Weighted Regression uses training examples. In this paper, we propose a function approximation method, Fuzzy Q-Map that is based on on-line fuzzy clustering. Fuzzy Q-Map classifies a query state and predicts a suitable action according to the membership degree. We applied the Fuzzy Q-Map, CMAC and LWR to the mountain car problem. Fuzzy Q-Map reached the optimal prediction rate faster than CMAC and the lower prediction rate was seen than LWR that uses training example.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.49 no.2
• /
• pp.96-102
• /
• 2012

An advanced form of the Partitioned Feature-based Classifier with Expertise Table(PFC-ET) is proposed in this paper. As is the case with the PFC-ET, the proposed classifier model, called Classifier Integration Model(CIM), does not use the entire feature vectors extracted from the original data in a concatenated form to classify each datum, but rather uses groups of features related to each feature vector separately. The proposed CIM utilizes a proportion of selected cluster members instead of the expertise table in PFC-ET to minimize the error in confusion table. The proposed CIM is applied to the classification problem on two data sets, Caltech data set and collected terrain data sets. When compared with PFC model and PFC-ET model. the proposed CIM shows improvements in terms of classification accuracy and post processing efforts.

• Journal of Information Technology and Architecture
• /
• v.10 no.1
• /
• pp.123-135
• /
• 2013

This paper presents a two-dimensional grouping index(2DG-index) for efficient processing of XML filtering queries. Recently, many index techniques have been suggested for the efficient processing of structural relationships among the elements in the XML database such as an ancestor- descendant and a parent-child relationship. However, these index techniques focus on simple path queries, and don't consider the path queries that include a condition value for filtering. The 2DG-index is an index structure that deals with the problem of clustering index entries in the twodimensional domain space that consists of a XML path identifier domain and a filtering data value domain. For performance evaluation, we have compared our proposed 2DG-index with the conventional one dimensional index structure such as the data grouping index (DG-index) and the path grouping index (PG-index). As the result of the performance evaluations, we have verified that our proposed 2DG-index can efficiently support the query processing in XML databases according to the query types.

• The KIPS Transactions:PartC
• /
• v.13C no.1 s.104
• /
• pp.103-112
• /
• 2006

Since sensor node has a limited energy supply in a wireless sensor network, it is very important to maximize the network lifetime through energy-efficient routing. Thus, many routing protocols have been developed for wireless sensor networks and can be classified into flat and hierarchical routing protocols. Recent researches focus on hierarchical routing scheme and LEACH is a representative hierarchical routing protocol. In this paper, we investigated the problems of the LEACH and proposed a novel energy efficient routing scheme, called ENTER(ENergy efficient Two-tiEr Routing protocol), to resolve the problem. ENTER reduces an energy consumption and increases a network lifetime by organizing clusters by the same distributed algerian as in the LEACH and establishing paths among cluster-heads to transmit the aggregated data to the sink node. We compared the performance of the ENTER with the LEACH through simulation and showed that the ENTER could enhance the network lifetime by utilizing the resources more efficiently.

• Journal of Digital Contents Society
• /
• v.14 no.1
• /
• pp.65-72
• /
• 2013

Current and future parallel computing model has been suggested for running and solving large-scale application problems such as climate, bio, cryptology, and astronomy, etc. Parallel computing is a form of computation in which many calculations are carried out simultaneously. And we are able to shorten the execution time of the program, as well as can extend the scale of the problem that can be solved. In this paper, we perform the actual cryptographic algorithms through parallel processing and evaluate its efficiency. Length of the key, which is stable criterion of cryptographic algorithm, judged according to the amount of complete enumeration computation. So we present a detailed procedure of DES key search cryptographic algorithms for executing of enumeration computation in parallel processing environment. And then, we did the simulation through applying to clustering system. As a result, we can measure the safety and solidity of cryptographic algorithm.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.20 no.1
• /
• pp.1-11
• /
• 2002

This study set up a purpose in the efficient utilization of security target objects. This purpose is the following: Firstly, this study analyzed problem about deleted areas for security described on aerial photography image. Secondly, this study made clustering and labeling to recognize censored areas of image. Finally, this study tried to maximize various utilizability of digital image data through postprocessing algorithm. Based on these courses, the results of this study appeared that brightness value of image increased depending on topography and quantities of topographic features. It was estimated that these was able to utilized by useful estimative data in judging information of topography and topographic features included in the total image. Besides, in the image recognition and postprocessing, the better result value was not elicited than in a mountainous region. Because it was included that a lots of topography and topographic features was similarly recognized with the process for deletion of the existing security target objects in urban and suburb region. This result appeared that the topography and quantities of topographic features absolutely affected the recognition and processing of image.