• Communications for Statistical Applications and Methods
• /
• v.30 no.4
• /
• pp.389-402
• /
• 2023

Multivariate or clustered failure time data often occur in many medical, epidemiological, and socio-economic studies when survival data are collected from several research centers. If the data are periodically observed as in a longitudinal study, survival times are often subject to various types of interval-censoring, creating multivariate interval-censored data. Then, the event times of interest may be correlated among individuals who come from the same cluster. In this article, we propose a unified linear regression method for analyzing multivariate interval-censored data. We consider a semiparametric multivariate accelerated failure time model as a statistical analysis tool and develop a generalized Buckley-James method to make inferences by imputing interval-censored observations with their conditional mean values. Since the study population consists of several heterogeneous clusters, where the subjects in the same cluster may be related, we propose a generalized estimating equations approach to accommodate potential dependence in clusters. Our simulation results confirm that the proposed estimator is robust to misspecification of working covariance matrix and statistical efficiency can increase when the working covariance structure is close to the truth. The proposed method is applied to the dataset from a diabetic retinopathy study.

• Journal of Korea Spatial Information System Society
• /
• v.8 no.3
• /
• pp.105-115
• /
• 2006

A sensor network which uses DCS(Data-Centric Storage) stores the same data into the same sensor node. Thus it has a hot spot problem when the sensor network grows and the same data arise frequently. In the past researches of the sensor network using DCS, the hot spot problem caused by growing the sensor network was ignored because they only concentrated on managing stored sensor data efficiently. In this paper, we proposed a non-equal region split method that supports efficient scalability on storing multi-dimensional sensor data. This method can reduce the storing cost, as the sensor network is growing, by dividing whole space into regions which have the same number of sensor nodes according to the distribution of sensor nodes, and storing and managing sensor data within each region. Moreover, this method can distribute the energy consumption of sensor nodes by increasing the number of regions according to the size of the sensor network, the number of sensor nodes within the sensor network, and the quantity of sensor data. Therefore it can help to increase the life time and the scalability of the sensor network.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.22 no.9
• /
• pp.2080-2089
• /
• 1997

Recent evolution in compute technology makesthe multimedia processing widely availiable. Conventional storage systems do not meet the requirements of multimedia data. Several approaches were suggested to improve disk storing methods for them. Bocheck proposed a disk partitioning technique for multiple steams assuming that all steams have same retrieval intervals with the same amount data for each access. While Bocheck's one provides a good method for same period, it does not consider the case of different periods of continous media streams. This paper proposes a new partitioning technique in which a fixed number of blocks are assigned for stresms with different retrieval periodicity. The analysis shows this problem is the same as the one scheduling the steams into a given sequence. The simulation was done to compare the proposed m-sequence merge method with the conventional Scan-EDF and Partitioning methods.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.33B no.9
• /
• pp.49-61
• /
• 1996

Back-propagation(BP) algorithm needs a lot of time to train the artificial neural network (ANN) to get high accuracy level in classification tasks. So there have been extensive researches to process back-propagation algorithm on parallel processors. This paper prsents a linear systolic array which calculates forward-backward propagation of BP algorithm at the same time using effective space-time transformation and PE structure. First, we analyze data flow of forwared and backward propagations and then, represent the BP algorithm into data dapendency graph (DG) which shows parallelism inherent in the BP algorithm. Next, apply space-time transformation on the DG of ANN is turn with orthogonal direction projection. By doing so, we can get a snakelike systolic array. Also we calculate the interval of input for parallel processing, calculate the indices to make the right datas be used at the right PE when forward and bvackward propagations are processed in the same PE. And then verify the correctness of output when forward and backward propagations are executed at the same time. By doing so, the proposed system maximizes parallelism of BP algorithm, minimizes th enumber of PEs. And it reduces the execution time by 2 times through making idle PEs participate in forward-backward propagation at the same time.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.9 no.3
• /
• pp.79-89
• /
• 2013

This paper proposes a watermarking scheme to protect ownership of a digital hologram, an ultra-high value-added content. It performs pre-defined levels of quad-tree Fresnelet transforms. The relationship among the same-positional-blocks is extracted as the digital pre-watermark. For the relationship, we use properties of a digital hologram that a hologram pixel retains all the information of the object and that the same size of partial holograms reconstructs the same size of object but different in their view points. Also we mix a set of private data with the pre-watermark and the result is encrypted by a block cipher algorithm with a private key. Experimental results showed that the proposed scheme is very robust for the various malicious and non-malicious attacks. Also because it extracts the watermarking data instead of inserting, the watermarking process does not harm the original hologram data. So, it is expected to be used effectively for invisible and robust watermark for digital holograms.

• Journal of The Korean Association For Science Education
• /
• v.30 no.4
• /
• pp.402-411
• /
• 2010

The purposes of this study are to analyze an argumentation between an astronomers group and a counter astronomers group that have concluded different results by using the same methods and to find an implication for applying to school science. For this study, two science research papers, which have been interpreted differently, were selected and analyzed in spite of observing same area in Titan and using same data process method. Their key issues are involved in interpretation and explanation, and the credibility of observed data. From this result, scientific argumentation accompanied with the credibility evaluation about the justification process of scientific explanation and experiment results needs to be developed.

• Korean Journal of Head & Neck Oncology
• /
• v.26 no.2
• /
• pp.204-211
• /
• 2010

Background and Objectives : Boundary detection and drawing are essential in 3D reconstruction of neck mass. Manual tracing methods are popular for drawing head and neck tumor. To improve manual tracing, circular boundaries overlapping was tried. Materials and Methods : Twenty patients with neck tumors were recruited for study. Representative frames were examined for shapes of outline. They were all single closed curves. Circular boundaries were added to fill the outlines of the tumors. Inserted circles were merged to form single closed curves(Circular boundary overlapping, CBO). After surface rendering, 3 dimensional images with volumes and area data were made. Same procedures were performed with manual tracing from same cases. 3D images were compared with surgical photographs of tumors for shape similarity by 2 doctors. All data were evaluated with Mann-Whitney test(p<0.05). Results : Shapes of boundaries from CBO were similar with boundaries from manual tracing. Tumor outlines could be filled with multiple circular boundaries., While both boundary tracing gave same results in small tumors, the bigger tumors showed different data. Two raters gave the similar high scores for both manual and CBO methods. Conclusion : Circular boundary overlapping is time saver in 3 dimensional reconstruction of CT images.

• Journal of the Korean Society of Industry Convergence
• /
• v.23 no.1
• /
• pp.41-48
• /
• 2020

Currently, the industry uses MODBUS communication method using RS485 as a communication method used when constructing distributed equipment and networks. However, this method has a rather good transmission and reception distance, but has a disadvantage of being a half-duplex communication method. Therefore, there is a great need for a full-duplex communication method that can simultaneously transmit and receive two-wire communications. Therefore, in this paper, we propose a new communication hardware equipment that can implement full-duplex communication method by communication signal level to overcome the disadvantage of communication speed. The proposed communication hardware is a structure that can transmit and receive at the same time on the same communication line in two equipments communicating in a two-wire system. The characteristic of this communication hardware is that the transmitter generates two-level signal for data transmission, but the receiver generates three-level electric signal according to the status of the transmission data generated by the receiver. The data transmission signal information of the receiver is present at the same time. Therefore, the receiving side can analyze the received signal based on the information on the current transmission signal and can analyze the signal only in the two communicating devices, so it can be seen that the communication security is very excellent.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2009.01a
• /
• pp.99-102
• /
• 2009

In this paper a hierarchical stereo matching algorithm based on feature extraction is proposed. The boundary (edge) as feature point in an image is first obtained by segmenting an image into red, green, blue and white regions. With the obtained boundary information, disparities are extracted by matching window on the image boundary, and the initial disparity map is generated when assigned the same disparity to neighbor pixels. The final disparity map is created with the initial disparity. The regions with the same initial disparity are classified into the regions with the same color and we search the disparity again in each region with the same color by changing block size and search range. The experiment results are evaluated on the Middlebury data set and it show that the proposed algorithm performed better than a phase based algorithm in the sense that only about 14% of the disparities for the entire image are inaccurate in the final disparity map. Furthermore, it was verified that the boundary of each region with the same disparity was clearly distinguished.

• Proceedings of the Korean Society for Bioinformatics Conference
• /
• 2000.11a
• /
• pp.59-60
• /
• 2000

The past few years have seen a dramatic increase in gene expression data on the basis of DNA microarrays or DNA chips. Going beyond a generic view on the genome, microarray data are able to distinguish between gene populations in different tissues of the same organism and in different states of cells belonging to the same tissue. This affords a cell-wide view of the metabolic and regulatory processes under different conditions, building an effective basis for new diagnoses and therapies of diseases. In this talk we present machine learning techniques for effective mining of DNA microarray data. A brief introduction to the research field of machine learning from the computer science and artificial intelligence point of view is followed by a review of recently-developed learning algorithms applied to the analysis of DNA chip gene expression data. Emphasis is put on graphical models, such as Bayesian networks, latent variable models, and generative topographic mapping. Finally, we report on our own results of applying these learning methods to two important problems: the identification of cell cycle-regulated genes and the discovery of cancer classes by gene expression monitoring. The data sets are provided by the competition CAMDA-2000, the Critical Assessment of Techniques for Microarray Data Mining.