• Nuclear Medicine and Molecular Imaging
• /
• v.41 no.6
• /
• pp.553-560
• /
• 2007

Purpose: We evaluated the standard uptake value (SUV) of F-18 FDG at PET/CT for differentiation of benign from malignant tumor in primary musculoskeletal tumors. Materials and Methods: Forty-six tumors (11 benign and 12 malignant soft tissue tumors, 9 benign and 14 malignant bone tumors) were examined with F-18 FDG PET/CT (Discovery ST, GE) prior to tissue diagnosis. The maxSUV(maximum value of SUV) were calculated and compared between benign and malignant lesions. The lesion analysis was based on the transverse whole body image. The maxSUV with cutoff of 4.1 was used in distinguishing benign from malignant soft tissue tumor and 3.05 was used in bone tumor by ROC curve. Results: There was a statistically significant difference in maxSUV between benign (n=11; maxSUV $3.4{\pm}3.2$) and malignant (n=12; maxSUV $14.8{\pm}12.2$) lesions in soft tissue tumor (p=0.001). Between benign bone tumor (n=9; maxSUV $5.4{\pm}4.0$) and malignant bone tumor (n=14; maxSUV $7.3{\pm}3.2$), there was not a significant difference in maxSUV. The sensitivity and specificity for differentiating malignant from benign soft tissue tumor was 83% and 91%, respectively. There were four false positive malignant bone tumor cases to include fibrous dysplasia, Langerhans-cell histiocytosis (n=2) and osteoid osteoma. Also, one false positive case of malignant soft tissue tumor was nodular fasciitis. Conclusion: The maxSUV was useful for differentiation of benign from malignant lesion in primary soft tissue tumors. In bone tumor, the low maxSUV correlated well with benign lesions but high maxSUV did not always mean malignancy.

• Journal of Intelligence and Information Systems
• /
• v.17 no.1
• /
• pp.53-69
• /
• 2011

Vision and voice-based technologies are commonly utilized for human-robot interaction. But it is widely recognized that the performance of vision and voice-based interaction systems is deteriorated by a large margin in the real-world situations due to environmental and user variances. Human users need to be very cooperative to get reasonable performance, which significantly limits the usability of the vision and voice-based human-robot interaction technologies. As a result, touch screens are still the major medium of human-robot interaction for the real-world applications. To empower the usability of robots for various services, alternative interaction technologies should be developed to complement the problems of vision and voice-based technologies. In this paper, we propose the use of accelerometer-based gesture interface as one of the alternative technologies, because accelerometers are effective in detecting the movements of human body, while their performance is not limited by environmental contexts such as lighting conditions or camera's field-of-view. Moreover, accelerometers are widely available nowadays in many mobile devices. We tackle the problem of classifying acceleration signal patterns of 26 English alphabets, which is one of the essential repertoires for the realization of education services based on robots. Recognizing 26 English handwriting patterns based on accelerometers is a very difficult task to take over because of its large scale of pattern classes and the complexity of each pattern. The most difficult problem that has been undertaken which is similar to our problem was recognizing acceleration signal patterns of 10 handwritten digits. Most previous studies dealt with pattern sets of 8~10 simple and easily distinguishable gestures that are useful for controlling home appliances, computer applications, robots etc. Good features are essential for the success of pattern recognition. To promote the discriminative power upon complex English alphabet patterns, we extracted 'motion trajectories' out of input acceleration signal and used them as the main feature. Investigative experiments showed that classifiers based on trajectory performed 3%~5% better than those with raw features e.g. acceleration signal itself or statistical figures. To minimize the distortion of trajectories, we applied a simple but effective set of smoothing filters and band-pass filters. It is well known that acceleration patterns for the same gesture is very different among different performers. To tackle the problem, online incremental learning is applied for our system to make it adaptive to the users' distinctive motion properties. Our system is based on instance-based learning (IBL) where each training sample is memorized as a reference pattern. Brute-force incremental learning in IBL continuously accumulates reference patterns, which is a problem because it not only slows down the classification but also downgrades the recall performance. Regarding the latter phenomenon, we observed a tendency that as the number of reference patterns grows, some reference patterns contribute more to the false positive classification. Thus, we devised an algorithm for optimizing the reference pattern set based on the positive and negative contribution of each reference pattern. The algorithm is performed periodically to remove reference patterns that have a very low positive contribution or a high negative contribution. Experiments were performed on 6500 gesture patterns collected from 50 adults of 30~50 years old. Each alphabet was performed 5 times per participant using $Nintendo{(R)}$ $Wii^{TM}$ remote. Acceleration signal was sampled in 100hz on 3 axes. Mean recall rate for all the alphabets was 95.48%. Some alphabets recorded very low recall rate and exhibited very high pairwise confusion rate. Major confusion pairs are D(88%) and P(74%), I(81%) and U(75%), N(88%) and W(100%). Though W was recalled perfectly, it contributed much to the false positive classification of N. By comparison with major previous results from VTT (96% for 8 control gestures), CMU (97% for 10 control gestures) and Samsung Electronics(97% for 10 digits and a control gesture), we could find that the performance of our system is superior regarding the number of pattern classes and the complexity of patterns. Using our gesture interaction system, we conducted 2 case studies of robot-based edutainment services. The services were implemented on various robot platforms and mobile devices including $iPhone^{TM}$. The participating children exhibited improved concentration and active reaction on the service with our gesture interface. To prove the effectiveness of our gesture interface, a test was taken by the children after experiencing an English teaching service. The test result showed that those who played with the gesture interface-based robot content marked 10% better score than those with conventional teaching. We conclude that the accelerometer-based gesture interface is a promising technology for flourishing real-world robot-based services and content by complementing the limits of today's conventional interfaces e.g. touch screen, vision and voice.

• Journal of Intelligence and Information Systems
• /
• v.17 no.4
• /
• pp.157-173
• /
• 2011

As the Internet use explodes recently, the malicious attacks and hacking for a system connected to network occur frequently. This means the fatal damage can be caused by these intrusions in the government agency, public office, and company operating various systems. For such reasons, there are growing interests and demand about the intrusion detection systems (IDS)-the security systems for detecting, identifying and responding to unauthorized or abnormal activities appropriately. The intrusion detection models that have been applied in conventional IDS are generally designed by modeling the experts' implicit knowledge on the network intrusions or the hackers' abnormal behaviors. These kinds of intrusion detection models perform well under the normal situations. However, they show poor performance when they meet a new or unknown pattern of the network attacks. For this reason, several recent studies try to adopt various artificial intelligence techniques, which can proactively respond to the unknown threats. Especially, artificial neural networks (ANNs) have popularly been applied in the prior studies because of its superior prediction accuracy. However, ANNs have some intrinsic limitations such as the risk of overfitting, the requirement of the large sample size, and the lack of understanding the prediction process (i.e. black box theory). As a result, the most recent studies on IDS have started to adopt support vector machine (SVM), the classification technique that is more stable and powerful compared to ANNs. SVM is known as a relatively high predictive power and generalization capability. Under this background, this study proposes a novel intelligent intrusion detection model that uses SVM as the classification model in order to improve the predictive ability of IDS. Also, our model is designed to consider the asymmetric error cost by optimizing the classification threshold. Generally, there are two common forms of errors in intrusion detection. The first error type is the False-Positive Error (FPE). In the case of FPE, the wrong judgment on it may result in the unnecessary fixation. The second error type is the False-Negative Error (FNE) that mainly misjudges the malware of the program as normal. Compared to FPE, FNE is more fatal. Thus, when considering total cost of misclassification in IDS, it is more reasonable to assign heavier weights on FNE rather than FPE. Therefore, we designed our proposed intrusion detection model to optimize the classification threshold in order to minimize the total misclassification cost. In this case, conventional SVM cannot be applied because it is designed to generate discrete output (i.e. a class). To resolve this problem, we used the revised SVM technique proposed by Platt(2000), which is able to generate the probability estimate. To validate the practical applicability of our model, we applied it to the real-world dataset for network intrusion detection. The experimental dataset was collected from the IDS sensor of an official institution in Korea from January to June 2010. We collected 15,000 log data in total, and selected 1,000 samples from them by using random sampling method. In addition, the SVM model was compared with the logistic regression (LOGIT), decision trees (DT), and ANN to confirm the superiority of the proposed model. LOGIT and DT was experimented using PASW Statistics v18.0, and ANN was experimented using Neuroshell 4.0. For SVM, LIBSVM v2.90-a freeware for training SVM classifier-was used. Empirical results showed that our proposed model based on SVM outperformed all the other comparative models in detecting network intrusions from the accuracy perspective. They also showed that our model reduced the total misclassification cost compared to the ANN-based intrusion detection model. As a result, it is expected that the intrusion detection model proposed in this paper would not only enhance the performance of IDS, but also lead to better management of FNE.

• Analytical Science and Technology
• /
• v.33 no.3
• /
• pp.115-124
• /
• 2020

Marker pens belong to school things that are controlled by the regulation system called safety confirmation under special act on the safety of products for children with the formaldehyde criteria of 20 mg/kg. With nine marker pens available commercially, formaldehyde in marker pen ink was analyzed by present test standard where marking on a fabric swatch with a pen and extracting the swatch in water and derivatization with Nash reagent followed by UV/Vis spectrophotometeric measurement (Nash-UV/Vis method), giving not detected results or a false positive result in case of a colored water extract. However, the contents of formaldehyde in ink of nine marker pens were determinded to range between 3.2 ~ 93.2 mg/kg with three results above the safety criteria of 20 mg/kg by HPLC/DAD measurements on DNPH derivatives of formaldehyde (DNPH-HPLC/DAD method) in ink dissolved directly in water using an ultrasonic bath. Therefore, the DNPH-HPLC/DAD method with the extraction of ultrasonic dissolving ink in water is proposed as a proper method for analyzing formaldehyde in ink. The proposed method has advantages of lower detection limit and accuracy with colored extracts as well as a simple and fast extraction. The accuracy and precision of this method was estimated to be 90.1 ~ 105.4 % and 0.6 ~ 3.3 %, respectively by spiking tests in the ranges of 20 mg/kg and 40 mg/kg using matrixes such as highlighter pen ink, board marker ink, chalk marker pen ink and painter marker ink.

• The Korean Journal of Nuclear Medicine
• /
• v.31 no.1
• /
• pp.50-56
• /
• 1997

Gated SPECT can evaluate the regional wall motion of the heart. We evaluated the regional wall motion of the perfusion abnormality in conventional perfusion SPECT with gated SPECT. In case of suspicious perfusion abnormalities, we tried to differentiate the artifact from true abnormality in coronary vascular disease using gated SPECT. We thought that artifacts would have normal wall motion, whereas fixed defects with decreased wall motion would probably represent coronary artery disease. A total of 275 patients who were performed coronary angiography and T1-201 rest/Tc-99m MIBI dipyridamole stress gated SPECT within 2 months were enrolled. In coronary angiography, stenosis more than 50% was considered as coronary artery disease. After injection of 111MBq T1-201 rest image was obtained on triple head SPECT system. 370MBg Tc-99m MIBI was used for the stress image. Eight-frame per-cardiac-cycle gated Tc-99m SPECT studies were done. All the images were analyzed visually. Using perfusion SPECT, the overall sensitivity and specificity were 87% and 55% respectively. Regarding artery territory, sensitivity and specificity were 68% and 73% for left anterior descending artery(LAD), 62% and 78% for right coronary artery(RCA), 42% and 90% for left ciramflex artery(LCX). Using gated SPECT, the overall sensitivity and specificity were 87% and 66% respectively. Sensitivity and specificity were 68% and 78% for LAD, 62% and 79% for RCA, 42% and 90% for LCX. Among 21 false positive cases in perfusion SPECT, 5 cases were interpreted as true negative with gated SPECT. We conclude that gated SPECT provides a valuable adjunct to perfusion SPECT in characterizing perfusion abnormalities and to improve specificity.

• Korean Security Journal
• /
• no.28
• /
• pp.207-228
• /
• 2011

In the situation that crime is diversified qualitatively and quantitatively, the mutual cooperation system between police and private security is very important to cope with the change of policing environment properly. The primary purpose of this study is to suggest the method to improve the policing service quality of police and advance sound security industry by analyzing the case of close mutual cooperation system between police and private security of USA, which can be called the best country in the field of private security. When considering the cases of USA synthetically, we can know that as the most basic characteristic, the discussion over actual cooperation method is performed on the basis of positive recognition over mutual existence of police or private security. It means that the mutual relationship is based on the basic recognition over partner relationship to meet citizen's desire for safety. While reviewing the cases of USA, Korea shall refer to the fact that social safety activities of advanced countries display effects because of many factors such as various mutual cooperation programs between police and private security, efforts of local government, high quality of private security, reliance of citizen and general understanding of police, private security and citizen to solve crime problem. As shown in the review of USA cases, when considering the fact that the mutual cooperation between police and private security is performed in the level of autonomous police, Korea shall perform autonomous police system to provide better policing service, which is close to citizen.

• Applied Chemistry for Engineering
• /
• v.33 no.3
• /
• pp.235-241
• /
• 2022

According to the World Water Development Report 2015 released by the United Nations, drinking water is expected to decrease by 40% by 2030. This does not mean that the amount of water decreases, but rather that the water source is contaminated due to environmental pollution. Because microbes are deeply related to water quality, the analysis of microbe is very important for water quality management. While the most common method currently used for microbial analysis is microscopic examination of the shape and feature after cell culture, as the gene analysis technology advances, quantitative polymerase chain reaction (qPCR) can be applied to the microscopic microbiological analysis, and the application method has been studied. Among them, a reverse transcription (RT) step enables the analysis of RNA by RT-PCR. Integrated cell culture (ICC)-qPCR shortens the test time by using it with microbial culture analysis, and viability qPCR can reduce the false positive errors of samples collected from natural water source. Multiplex qPCR for improved throughput, and microfluidic qPCR for analysis with limited amount of sample has been developed In this paper, we introduce the case, principle and development direction of the qPCR method applied to the analysis of microorganisms.

• Journal of Genetic Medicine
• /
• v.5 no.1
• /
• pp.47-54
• /
• 2008

Purpose : Fluorescence in situ hybridization (FISH) on uncultured amniotic fluid cells offers the opportunity for rapid screening of aneuploidies and has become an integral part of the current practice in many clinical cytogenetics laboratories. Here, we retrospectively analyzed the results of interphase FISH in 943 amniotic fluid samples and assessed the efficiency of FISH for rapid detection of aneuploidies. Methods : Interphase FISH for chromosome 13, 18, and 21 was performed in 943 consecutive amniotic fluid samples for rapid diagnosis of aneuploidies referred from 2004 to 2006. Karyotypes from standard cytogenetic analysis were compared to the FISH results. Results : A total of 45 chromosomal rearrangements (4.8%) were found after conventional cytogenetic analysis of the 943 amniotic fluid. After exclusion of known familiar chromosomal rearrangements and inversions (2.1%, 20/943), 2.7% (25/943) were found to have chromosomal abnormalities. Of this group, 0.7% (6/943) were chromosomal abnormalities not detectable by FISH and 2.0% (19/943) were numerical abnormalities detectable by FISH. All 14 cases of Down syndrome (Classic type, 13 cases; Robertsonian type, 1 case) and 5 cases of trisomy 18 were diagnosed and detected by FISH and there were no false-positive or -negative results (specificity and sensitivity=100%). Conclusion : The present study demonstrates that FISH can provide a rapid and sensitive clinical method for prenatal identification of chromosome aneuploidies. However, careful genetic counseling is essential to explain the limitations of FISH, including the inability to detect all chromosomal abnormalities and the possibilities of uninformative or false-negative results in some cases.

• Journal of Intelligence and Information Systems
• /
• v.18 no.1
• /
• pp.125-141
• /
• 2012

These days, the malicious attacks and hacks on the networked systems are dramatically increasing, and the patterns of them are changing rapidly. Consequently, it becomes more important to appropriately handle these malicious attacks and hacks, and there exist sufficient interests and demand in effective network security systems just like intrusion detection systems. Intrusion detection systems are the network security systems for detecting, identifying and responding to unauthorized or abnormal activities appropriately. Conventional intrusion detection systems have generally been designed using the experts' implicit knowledge on the network intrusions or the hackers' abnormal behaviors. However, they cannot handle new or unknown patterns of the network attacks, although they perform very well under the normal situation. As a result, recent studies on intrusion detection systems use artificial intelligence techniques, which can proactively respond to the unknown threats. For a long time, researchers have adopted and tested various kinds of artificial intelligence techniques such as artificial neural networks, decision trees, and support vector machines to detect intrusions on the network. However, most of them have just applied these techniques singularly, even though combining the techniques may lead to better detection. With this reason, we propose a new integrated model for intrusion detection. Our model is designed to combine prediction results of four different binary classification models-logistic regression (LOGIT), decision trees (DT), artificial neural networks (ANN), and support vector machines (SVM), which may be complementary to each other. As a tool for finding optimal combining weights, genetic algorithms (GA) are used. Our proposed model is designed to be built in two steps. At the first step, the optimal integration model whose prediction error (i.e. erroneous classification rate) is the least is generated. After that, in the second step, it explores the optimal classification threshold for determining intrusions, which minimizes the total misclassification cost. To calculate the total misclassification cost of intrusion detection system, we need to understand its asymmetric error cost scheme. Generally, there are two common forms of errors in intrusion detection. The first error type is the False-Positive Error (FPE). In the case of FPE, the wrong judgment on it may result in the unnecessary fixation. The second error type is the False-Negative Error (FNE) that mainly misjudges the malware of the program as normal. Compared to FPE, FNE is more fatal. Thus, total misclassification cost is more affected by FNE rather than FPE. To validate the practical applicability of our model, we applied it to the real-world dataset for network intrusion detection. The experimental dataset was collected from the IDS sensor of an official institution in Korea from January to June 2010. We collected 15,000 log data in total, and selected 10,000 samples from them by using random sampling method. Also, we compared the results from our model with the results from single techniques to confirm the superiority of the proposed model. LOGIT and DT was experimented using PASW Statistics v18.0, and ANN was experimented using Neuroshell R4.0. For SVM, LIBSVM v2.90-a freeware for training SVM classifier-was used. Empirical results showed that our proposed model based on GA outperformed all the other comparative models in detecting network intrusions from the accuracy perspective. They also showed that the proposed model outperformed all the other comparative models in the total misclassification cost perspective. Consequently, it is expected that our study may contribute to build cost-effective intelligent intrusion detection systems.

• The Korean Journal of Cytopathology
• /
• v.3 no.1
• /
• pp.1-11
• /
• 1992

From the one hundred forty eight patients with evidence of biliary tract obstruction, 275 bile samples were obtained from percutaneously placed biliary drainage catheters. Of the 148 patients, ova of Clonorchis sinensis were demonstrated in 17 patients (11.5%), with the epithelial cells. Among them, one case also demonstrated coexisting adenocarcinoma. In 105 patients, the medical records were available for review and the clinical diagnoses were malignancy in 99 patients and benign lesion in 6 patients. Of the 99 patients in which clinico-radiologic diagnosis were malignant, cytologic results were positive in 23.2%. Dividing the patients Into two groups, the ones with tumor of bile duct origin (group I) and the others with tumors producing extrinsic compression of bile duct, such as periampullary carcinoma, pancreas head carcinoma or metastatic carcinoma in lymph nodes from tumors of adjacent organs (group II), the cytologic results were positive in 37% and 11.6%, respectively. In patients with histologic confirmation, the positive correlation was found in 50% and 20% in group I and group II, respectively, with remarkable difference between two groups. There were no false positives in cytologic diangosis. The overall concordance rate of cytologic diagnosis with diagnosis of clinical investigation in both benign and malignant lesions was 27.6% and the diagnostic specificity was 100%.