• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.11 no.5
• /
• pp.997-1003
• /
• 2007

In this paper, we propose a global approach for automatic inspection of defects in the polarizer film image. The proposed method does not rely on local feature of the defect. It is based on a global image reconstruction scheme using the singular value decomposition(SVD). SVD is used to decompose the image and then obtain a diagonal matrix of the singular values. Among the singular values, the first singular value is used to reconstruct a image. In reconstructed image, the normal pixels in background region have a different characteristics from the pixels in defect region. It is obtained the ratio of pixels in the reconstructed image to ones in the original image and then the defects are detected based on the the statistical process of the ratio. The experiment results show that the proposed method is efficient for defect inspection of polarizer lam image.

• The Journal of the Korea Contents Association
• /
• v.12 no.10
• /
• pp.385-394
• /
• 2012

The aim of this study was to determine the effectiveness of diagnostic range for BMD measurement tools(DEXA, QUS, and RA) to health examination in transitional ages. In standard T-score -2.5 of DEXA, cutoff value by RA is -1.675(sensitivity: 70.0%, specificity: 63.7%) and cutoff value by QUS is -1.733(sensitivity: 70.4, specificity: 59.5%), also T-score -3.0 of DEXA, cutoff value by RA is -2.325(sensitivity: 70.0%, specificity: 42.9%) and cutoff value by QUS is -2.323(sensitivity: 70.4, specificity: 56.8%). There was, however, no significant difference in standard DEXA(lumbar spine and femur) between RA and QUS by repeat measurement(precision), and correlation were without effect. ROC analysis showed that all methods are qualified for BMD measurement tools to health examination in transitional ages; however, the different sensitivities and specificities of the methods, as well as age and gender, calibration parameters for diagnostic tests have to be considered.

• Journal of Animal Science and Technology
• /
• v.44 no.1
• /
• pp.13-22
• /
• 2002

To identify the differentially expressed genes at growth stage of Hanwoo, we constructed the subtractive cDNA library from loin mRNA of 12- and 24-month old Hanwoo by PCR-based subtraction. The fourteen genes were confirmed by sequencing and reverse northern blot analysis, and they were selected as candidate of putative genes differentially expressed at the growth stage of Hanwoo. Three subtracted cDNA fragments that expressed specific signal to cDNA probe for 6-month-old loin of Hanwoo were highly homologous to those of the genes encoding EPV 20, Ca2＋ATPase, and TCTP, respectively. The nine cDNA clones showed intense signal to cDNA probe from 12-month-old loin of Hanwoo, and highly homologus to those of genes encoding VCP, HSP 70, aldolase A, MSSK1, GM-2 activator protein, ryanodine receptor, acidic ribosomal phosphoprotein p1, ADP/ATP translocase, and UCP 2, respectively. Two subtracted cDNA clones that expressed specific signal to cDNA probes for 12- and 24-month-old loin of Hanwoo were detected. One of them was highly homologus to the gene encoding ferrochelatase and the other was highly homologus to the gene encoding ADRP.

• Journal of the korean academy of Pediatric Dentistry
• /
• v.33 no.4
• /
• pp.615-623
• /
• 2006

The purpose of this study was to evaluate the specificity, sensitivity, and diagnostic power of caries activity test using LED fluorescence. The subjects of this study were 55 children of $6{\sim}7$ years old. LED light were irradiated to labial or buccal surface of all teeth. Fluorescence from initial carious lesion of teeth illuminated by an LED light was observed through barrier filter and the number of teeth showing lesion, size and position of lesion were counted. Streptococcus mutans colony counting and dDfFtT rate test were also done and their correlation was compared. And then specificity, sensitivity, diagnostic power of optical caries activity test using LED light were evaluated. 1. There was positive $correlation({\gamma}=0.43)$ between LED fluorescence test and Streptococcus mutans count(P<0.05). 2. When visual examination was defined to standard testing method, the specificity, sensitivity, diagnostic power of LED fluorescence test were 100%, 76.1%, and 100%. 3. When dDfFtT rate was defined to standard testing method, the specificity, sensitivity, diagnostic power of LED fluorescence test were 88.9%, 47.8%, and 95.7%. 4. When S. mutans colony counting was defined to standard testing method, the specificity, sensitivity, diagnostic power of LED fluorescence test were 100%, 58.7%, and 100%. Considering the above results, optical caries activity test using LED light could be regarded as a practical method because of its close relationship with microbiological caries activity test.

• Korean Journal of Microbiology
• /
• v.46 no.2
• /
• pp.219-222
• /
• 2010

In this study, we evaluated Vi-passive hemagglutination (Vi-PHA), SD Salmonella Typhi IgG/IgM ($SD^{(R)}$ kit) and Widal test for the rapid laboratory diagnosis of typhoid fever patients. A total of 36 serum samples from febrile patients in Korea from 2005 to 2006 were used. Among 36 patients, 27 were fever patients without typhoid, 9 were typhoid fever. Vi-PHA showed 8 positive results out of 9 typhoid fever patients (sensitivity 88.9%) and 1 positive and 26 negative results out of 27 febrile patients without typhoid (specificity 96.3%). The sensitivity and the specificity of $SD^{(R)}$ kit were 100% and 92.6%, respectively. However, the sensitivity and the specificity of Widal O & H tests were 88.9%, 100%, and 77.8%, 70.4%, respectively. Consequently, Widal H and $SD^{(R)}$ kit showed higher sensitivity and Vi-PHA showed higher specificity. To efficient diagnosis, Vi-PHA may be sufficient diagnosis method in acute cases and $SD^{(R)}$ kit and Widal test may be sufficient in sporadic area and high risk group.

• Journal of Animal Science and Technology
• /
• v.50 no.1
• /
• pp.9-18
• /
• 2008

Species-specific polymorphisms in chicken, pheasant, turkey, and quail were identified by cloning and sequencing of the immunoglobulin constant domain (IgLC). A set of species-specific primers were then designed on the basis of polymorphisms in the IgLC between species, as well as two additional sets of primers for the cytochrome b and tapasin genes, for the purpose of species identification. Together, the primers successfully distinguished specific species from chicken by species-specific PCR. This simple but unambiguous method may be used to screen avian inter-species germline chimeras, which are valuable models for the conservation of endangered species.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.23 no.4
• /
• pp.292-297
• /
• 2013

In this paper, unusual waveform detection algorithm based on Refractory Period in arrhythmia ECG signal is proposed. Most of arrhythmia ECG signals consist of unusual waveforms with average 10% rate. Thus tremendous benefit can be obtained in terms of time and cost by providing unusual waveform samples reduced more than 90% to medical staffs who have to monitor and analyze for a long time. The proposed algorithm detects the R-peak using the features of R wave and variable refractory period. For the detected R-peak, unusual waveforms are found using means and standard deviation of electric potential and kurtosis of the R-peaks which are not included in unusual waveform. The proposed algorithm was applied to all records of the MIT-BIH arrhythmia database and showed more than average 90% of compression ratio.

• The Korean Journal of Applied Statistics
• /
• v.29 no.5
• /
• pp.907-920
• /
• 2016

This study suggests an outlier detection algorithm that uses quantile autoregressive model in time series data, eventually applying it to actual stock manipulation cases by comparing its performance to existing methods. Studies on outlier detection have traditionally been conducted mostly in general data and those in time series data are insufficient. They have also been limited to a parametric model, which is not convenient as it is complicated with an analysis that takes a long time. Thus, we suggest a new algorithm of outlier detection in time series data and through various simulations, compare it to existing algorithms. Especially, the outlier detection algorithm in time series data can be useful in finding stock manipulation. If stock price which had a certain pattern goes out of flow and generates an outlier, it can be due to intentional intervention and manipulation. We examined how fast the model can detect stock manipulations by applying it to actual stock manipulation cases.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.40 no.4
• /
• pp.731-739
• /
• 2015

In arrhythmia ECG signal, abnormal beat that has various abnormal shape depending on the generation site and conduction disorders is included and it is very important to diagnose heart disease such as arrhythmia. In this paper, we propose a PVC abnormal beat detection algorithm associated with ventricular disease. The PVC abnormal beat is characterized by distortion of the QRS complex occurs among the components of the ECG signal. Therefore it is possible to detect PVC abnormal beat according to the degree of distortion of the QRS complex. First, quantify the distortion of the QRS complex by using the potential of the R-peak, kurtosis and period. By using the mean and standard deviation, PVC abnormal beat is detected depending on the degree of distortion from the normal beat. The proposed algorithm can detect the average over 98% of the AAMI-V class type abnormal beat associated with ventricular disease in MIT-BIH arrhythmia database.

• Journal of Intelligence and Information Systems
• /
• v.19 no.2
• /
• pp.125-140
• /
• 2013

We call a data set in which the number of records belonging to a certain class far outnumbers the number of records belonging to the other class, 'imbalanced data set'. Most of the classification techniques perform poorly on imbalanced data sets. When we evaluate the performance of a certain classification technique, we need to measure not only 'accuracy' but also 'sensitivity' and 'specificity'. In a customer churn prediction problem, 'retention' records account for the majority class, and 'churn' records account for the minority class. Sensitivity measures the proportion of actual retentions which are correctly identified as such. Specificity measures the proportion of churns which are correctly identified as such. The poor performance of the classification techniques on imbalanced data sets is due to the low value of specificity. Many previous researches on imbalanced data sets employed 'oversampling' technique where members of the minority class are sampled more than those of the majority class in order to make a relatively balanced data set. When a classification model is constructed using this oversampled balanced data set, specificity can be improved but sensitivity will be decreased. In this research, we developed a hybrid model of support vector machine (SVM), artificial neural network (ANN) and decision tree, that improves specificity while maintaining sensitivity. We named this hybrid model 'hybrid SVM model.' The process of construction and prediction of our hybrid SVM model is as follows. By oversampling from the original imbalanced data set, a balanced data set is prepared. SVM_I model and ANN_I model are constructed using the imbalanced data set, and SVM_B model is constructed using the balanced data set. SVM_I model is superior in sensitivity and SVM_B model is superior in specificity. For a record on which both SVM_I model and SVM_B model make the same prediction, that prediction becomes the final solution. If they make different prediction, the final solution is determined by the discrimination rules obtained by ANN and decision tree. For a record on which SVM_I model and SVM_B model make different predictions, a decision tree model is constructed using ANN_I output value as input and actual retention or churn as target. We obtained the following two discrimination rules: 'IF ANN_I output value <0.285, THEN Final Solution = Retention' and 'IF ANN_I output value ${\geq}0.285$, THEN Final Solution = Churn.' The threshold 0.285 is the value optimized for the data used in this research. The result we present in this research is the structure or framework of our hybrid SVM model, not a specific threshold value such as 0.285. Therefore, the threshold value in the above discrimination rules can be changed to any value depending on the data. In order to evaluate the performance of our hybrid SVM model, we used the 'churn data set' in UCI Machine Learning Repository, that consists of 85% retention customers and 15% churn customers. Accuracy of the hybrid SVM model is 91.08% that is better than that of SVM_I model or SVM_B model. The points worth noticing here are its sensitivity, 95.02%, and specificity, 69.24%. The sensitivity of SVM_I model is 94.65%, and the specificity of SVM_B model is 67.00%. Therefore the hybrid SVM model developed in this research improves the specificity of SVM_B model while maintaining the sensitivity of SVM_I model.