• Nuclear Medicine and Molecular Imaging
• /
• v.41 no.4
• /
• pp.299-308
• /
• 2007

Purpose: The aim of this study was to develop a bioinformatics software and to test it in serum samples of papillary thyroid cancer using mass spectrometry (SELDI-TOF-MS). Materials and Methods: Development of 'Protein analysis' software performing decision tree analysis was done by customizing C4.5. Sixty-one serum samples from 27 papillary thyroid cancer, 17 autoimmune thyroiditis, 17 controls were applied to 2 types of protein chips, CM10 (weak cation exchange) and IMAC3 (metal binding - Cu). Mass spectrometry was performed to reveal the protein expression profiles. Decision trees were generated using 'Protein analysis' software, and automatically detected biomarker candidates. Validation analysis was performed for CM10 chip by random sampling. Results: Decision tree software, which can perform training and validation from profiling data, was developed. For CM10 and IMAC3 chips, 23 of 113 and 8 of 41 protein peaks were significantly different among 3 groups (p<0.05), respectively. Decision tree correctly classified 3 groups with an error rate of 3.3% for CM10 and 2.0% for IMAC3, and 4 and 7 biomarker candidates were detected respectively. In 2 group comparisons, all cancer samples were correctly discriminated from non-cancer samples (error rate = 0%) for CM10 by single node and for IMAC3 by multiple nodes. Validation results from 5 test sets revealed SELDI-TOF-MS and decision tree correctly differentiated cancers from non-cancers (54/55, 98%), while predictability was moderate in 3 group classification (36/55, 65%). Conclusion: Our in-house software was able to successfully build decision trees and detect biomarker candidates, therefore it could be useful for biomarker discovery and clinical follow up of papillary thyroid cancer.

• Applied Biological Chemistry
• /
• v.47 no.4
• /
• pp.414-421
• /
• 2004

3D QSAR studies for protox inhibition activities against root and shoot of the rice plant (Orysa sativa L.) and barnyardgrass (Echinochloa crus-galli) by a series of new 1-(5-methyl-3-phenylisoxazolin-5-yl)methoxy-2-chloro-4-fluorobenzene derivatives were conducted based on the results (Sung, N. D. et al.'s, (2004) J. Korean Soc. Appl. Biol. Chem. 47(3), 351-356) using comparative molecular similarity indices analysis (CoMSIA) methodology. Four CoMSIA models, without hydrogen bond donor field for the protox inhibition activities against root and shoot of the two plants, were derived from the combination of several fields using steric field, hydrophobic field, hydrogen bond acceptor field, LUMO molecular orbital field, dipole moment (DM) and molar refractivity (MR) as additional descriptors. The predictabilities and fitness of CoMSIA models for protox inhibition activities against barnyard-grass were higher than that of rice plant. The statistical results of these models showed the best predictability of the protox inhibition activities against barnyard-grass based on the cross-validated value $r^2\;_{cv}\;(q^2=0.635{\sim}0.924)$, non cross-validated, conventional coefficient $r^2\;_{ncv.}$ value $(r^2=0.928{\sim}0.977)$ and PRESS value $(0.255{\sim}0.273)$. The protox inhibition activities exhibited a strong correlation with the steric $(5.4{\sim}15.7%)$ and hydrophobic $(68.0{\sim}84.3%)$ factors of the molecules. Particularly, the CoMSIA models indicated that the groups of increasing steric bulk at ortho-position on the C-phenyl ring will enhance the protox inhibition activities against barnyard-grass and subsequently increase the selectivity.

• Journal of Hospice and Palliative Care
• /
• v.21 no.1
• /
• pp.9-13
• /
• 2018

Purpose: The purpose of this study was to assess the factors influencing the rescue medication decisions for breakthrough cancer patients and evaluate treatments using the factors. Methods: Based on the results of an online survey conducted by the Korean Society of Hospice and Palliative Care from September 2014 through December 2014, we assessed the level of agreement on nine factors influencing rescue medication preference. The same factors were used to evaluate oral transmucosal fentanyl lozenge, oral oxycodone and intravenous morphine. Results: Agreed by 77 physicians, a rapid onset of action was the most important factor for their decision of rescue medication. Other important factors were easy administration, strong efficacy, predictable efficacy and less adverse effects. Participants agreed that intravenous morphine produced a rapid onset of action and strong and predictable efficacy and cited difficulty of administration and adverse effects as negative factors. Oral oxycodone was desirable in terms of easy administration and less adverse effects. However, its onset of action was slower than intravenous morphine. While many agreed to easy administration of oral transmucosal fentanyl lozenge, the level of agreement was low for strength and predictability of its efficacy, long-term durability and sleep improvement. Conclusion: Rapid onset of action is one of the important factors that influence physicians' selection of rescue medication. Physicians' assessment of rescue medication differed by medication.

• The Korean Journal of Pesticide Science
• /
• v.8 no.3
• /
• pp.151-161
• /
• 2004

Three dimensional quantitative structure-activity relationships (3D-QSAR) studies for the protox inhibition activities against root and shoot of rice plant (Orysa sativa L.) and barnyardgrass (Echinochloa crus-galli) by a series of new A=3,4,5,6-tetrahydrophthalimino, B=3-chloro-4,5,6,7-tetrahydro-2H-indazolyl and C=3,4-dimethylmaleimino group, and R-group substituted on the phenyl ring in 1-(5-methyl-3-phenylisoxazolin-5-yl)methoxy-2chloro-4-fluorobenzene derivatives were performed using comparative molecular field analyses (CoMFA) methodology with Gasteiger-Huckel charge. Four CoMFA models for the protox inhibition activities against root and shoot of the two plants were generated using 46 molecules as training set and the predictive ability of the each models was evaluated against a test set of 8 molecules. And the statistical results of these models with combination (SIH) of standard field, indicator field and H-bond field showed the best predictability of the protox inhibition activities based on the cross-validated value $r^2_{cv.}$ $(q^2=0.635\sim0.924)$, conventional coefficient $(r^2_{ncv.}=0.928\sim0.977)$ and PRESS value $(0.091\sim0.156)$, respectively. The activities exhibited a strong correlation with steric $(74.3\sim87.4%)$, electrostatic $(10.10\sim18.5%)$ and hydrophobic $(1.10\sim8.30%)$ factors of the molecules. The steric feature of molecule may be an important factor for the activities. We founded that an novel selective and higher protox inhibitors between the two plants may be designed by modification of X-subsitutents for barnyardgrass based upon the results obtained from CoMFA analyses.

• The Korean Journal of Pesticide Science
• /
• v.9 no.1
• /
• pp.26-34
• /
• 2005

3D-QSAR studies for the fungicidal activities against resistance phytophthora blight (RPC; 95CC7303) and sensitive phytophthora blight (Phytopthora capsici) (SPC; 95CC7105) by a series of new 2-alkoxyphenyl-3-phenylthioisoindoline-1-one derivatives (A & B) were studieded using comparative molecular similarity indices analyses (CoMSIA) methodology. From the based on the results, the two CoMSIA models, R5 and S1: as the best models were derivated. The statistical results of the models showed the best predictability and fitness for the fungicidal activities based on the cross- validated value ($q^2=0.714{\sim}0.823$) and non cross-validated, value ($r^2_{ncv.}=0.918{\sim}0.954$), respectively. The model R5 for fungicidal activity of RPC generated from the field fit alignment and combination of electrostatic field, H-bond acceptor field and LUMO molecular orbital field. The model S1 (or S5) for fungicidal activity of SPC generated from the atom based fit alignment and combination of steric field and HOMO molecular orbital field. The models also shows that inclusion of H-bond acceptor field (A) improved the statistical significance of the models. From the based graphical analyses of CoMSIA contribution maps, it was revealed that the novel selective character for fungicidal activities between the two fungi by modify of X-sub-stituent on the N-phenyl group and R-substituent on the S-phenyl group will be able to achivement.

• Korean journal of applied entomology
• /
• v.50 no.3
• /
• pp.235-245
• /
• 2011

Temperature-related parameters of Panonychus citri (McGregor) (Acarina: Tetranychidae) development were estimated and a stage-structured matrix model was developed. The lower threshold temperatures were estimated as $8.4^{\circ}C$ for eggs, $9.9^{\circ}C$ for larvae, $9.2^{\circ}C$ for protonymphs, and $10.9^{\circ}C$ for deutonymphs. Thermal constants were 113.6, 29.1, 29.8, and 33.4 degree days for eggs, larvae, protonymphs, and deutonymphs, respectively. Non-linear development models were established for each stage of P. citri. In addition, temperature-dependent total fecundity, age-specific oviposition rate, and age-specific survival rate models were developed for the construction of an oviposition model. P. citri age was categorized into five stages to construct a matrix model: eggs, larvae, protonymphs, deutonymphs and adults. For the elements in the projection matrix, transition probabilities from an age class to the next age class or the probabilities of remaining in an age class were obtained from development rate function of each stage (age classes). Also, the fecundity coefficients of adult population were expressed as the products of adult longevity completion rate (1/longevity) by temperature-dependent total fecundity. To evaluate the predictability of the matrix model, model outputs were compared with actual field data in a cool early season and hot mid to late season in 2004. The model outputs closely matched the actual field patterns within 30 d after the model was run in both the early and mid to late seasons. Therefore, the developed matrix model can be used to estimate the population density of P. citri for a period of 30 d in citrus orchards.

• Journal of Korean Society of Forest Science
• /
• v.109 no.3
• /
• pp.259-270
• /
• 2020

Extreme weather events, such as heat and drought, have occurred frequently over the past two decades. This has led to continuous reports of cases of forest damage due to physiological stress, not pest damage. In 2014, pine trees were collectively damaged in the forest genetic resources reserve of Sogwang-ri, Uljin, South Korea. An investigation was launched to determine the causes of the dieback, so that a forest management plan could be prepared to deal with the current dieback, and to prevent future damage. This study aimedto 1) understand the topographic and structural characteristics of the area which experienced pine tree dieback, 2) identify the main causes of the dieback, and 3) predict future risk areas through the use of machine-learning techniques. A model for identifying risk areas was developed using 14 explanatory variables, including location, elevation, slope, and age class. When three machine-learning techniques-Decision Tree, Random Forest (RF), and Support Vector Machine (SVM) were applied to the model, RF and SVM showed higher predictability scores, with accuracies over 93%. Our analysis of the variable set showed that the topographical areas most vulnerable to pine dieback were those with high altitudes, high daily solar radiation, and limited water availability. We also found that, when it came to forest stand characteristics, pine trees with high vertical stand densities (5-15 m high) and higher age classes experienced a higher risk of dieback. The RF and SVM models predicted that 9.5% or 115 ha of the Geumgang Pine Forest are at high risk for pine dieback. Our study suggests the need for further investigation into the vulnerable areas of the Geumgang Pine Forest, and also for climate change adaptive forest management steps to protect those areas which remain undamaged.

• Journal of Intelligence and Information Systems
• /
• v.16 no.4
• /
• pp.99-112
• /
• 2010

Ensemble learning is a method for improving the performance of classification and prediction algorithms. It is a method for finding a highly accurateclassifier on the training set by constructing and combining an ensemble of weak classifiers, each of which needs only to be moderately accurate on the training set. Ensemble learning has received considerable attention from machine learning and artificial intelligence fields because of its remarkable performance improvement and flexible integration with the traditional learning algorithms such as decision tree (DT), neural networks (NN), and SVM, etc. In those researches, all of DT ensemble studies have demonstrated impressive improvements in the generalization behavior of DT, while NN and SVM ensemble studies have not shown remarkable performance as shown in DT ensembles. Recently, several works have reported that the performance of ensemble can be degraded where multiple classifiers of an ensemble are highly correlated with, and thereby result in multicollinearity problem, which leads to performance degradation of the ensemble. They have also proposed the differentiated learning strategies to cope with performance degradation problem. Hansen and Salamon (1990) insisted that it is necessary and sufficient for the performance enhancement of an ensemble that the ensemble should contain diverse classifiers. Breiman (1996) explored that ensemble learning can increase the performance of unstable learning algorithms, but does not show remarkable performance improvement on stable learning algorithms. Unstable learning algorithms such as decision tree learners are sensitive to the change of the training data, and thus small changes in the training data can yield large changes in the generated classifiers. Therefore, ensemble with unstable learning algorithms can guarantee some diversity among the classifiers. To the contrary, stable learning algorithms such as NN and SVM generate similar classifiers in spite of small changes of the training data, and thus the correlation among the resulting classifiers is very high. This high correlation results in multicollinearity problem, which leads to performance degradation of the ensemble. Kim,s work (2009) showedthe performance comparison in bankruptcy prediction on Korea firms using tradition prediction algorithms such as NN, DT, and SVM. It reports that stable learning algorithms such as NN and SVM have higher predictability than the unstable DT. Meanwhile, with respect to their ensemble learning, DT ensemble shows the more improved performance than NN and SVM ensemble. Further analysis with variance inflation factor (VIF) analysis empirically proves that performance degradation of ensemble is due to multicollinearity problem. It also proposes that optimization of ensemble is needed to cope with such a problem. This paper proposes a hybrid system for coverage optimization of NN ensemble (CO-NN) in order to improve the performance of NN ensemble. Coverage optimization is a technique of choosing a sub-ensemble from an original ensemble to guarantee the diversity of classifiers in coverage optimization process. CO-NN uses GA which has been widely used for various optimization problems to deal with the coverage optimization problem. The GA chromosomes for the coverage optimization are encoded into binary strings, each bit of which indicates individual classifier. The fitness function is defined as maximization of error reduction and a constraint of variance inflation factor (VIF), which is one of the generally used methods to measure multicollinearity, is added to insure the diversity of classifiers by removing high correlation among the classifiers. We use Microsoft Excel and the GAs software package called Evolver. Experiments on company failure prediction have shown that CO-NN is effectively applied in the stable performance enhancement of NNensembles through the choice of classifiers by considering the correlations of the ensemble. The classifiers which have the potential multicollinearity problem are removed by the coverage optimization process of CO-NN and thereby CO-NN has shown higher performance than a single NN classifier and NN ensemble at 1% significance level, and DT ensemble at 5% significance level. However, there remain further research issues. First, decision optimization process to find optimal combination function should be considered in further research. Secondly, various learning strategies to deal with data noise should be introduced in more advanced further researches in the future.

• Journal of Bio-Environment Control
• /
• v.29 no.2
• /
• pp.161-170
• /
• 2020

There is systematic spatial variations in environmental properties due to sensitive reaction to external conditions at plastic greenhouse occupied 99.2% of domestic agricultural facilities. In order to construct 3 dimensional distribution of temperature, relative humidity, CO₂ and illuminance, measurement matrix as 3 by 3 by 5 in direction of width, height and length, respectively, dividing indoor space of greenhouse was designed and tested at experimental site. Linear regression analysis was conducted to evaluate optimal estimation method in terms with horizontal and vertical variations. Even though sole measurement point for temperature and relative humidity could be feasible to assess indoor condition, multiple measurement matrix is inevitably required to improve spatial precision at certain time domain such as period of sunrise and sunset. In case with CO₂, multiple measurement matrix could not successfully improve the spatial predictability during a whole experimental period. In case with illuminance, prediction performance was getting smaller after a time period of sunrise due to systematic interference such as indoor structure. Thus, multiple sensing methodology was proposed in direction of length at higher height than growing bed, which could compensate estimation error in spatial domain. Appropriate measurement matrix could be constructed considering the transition of stability in indoor environmental properties due to external variations. As a result, optimal measurement matrix should be carefully designed considering flexibility of construction relevant with the type of property, indoor structure, the purpose of crop and the period of growth. For an instance, partial cooling and heating system to save a consumption of energy supplement could be successfully accomplished by the deployment of multiple measurement matrix.

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.19 no.6
• /
• pp.915-936
• /
• 2017

In this study, current status of Korean hazard mitigation guideline for tunnel operation is summarized. It shows that requirement for CCTV installation has been gradually stricted and needs for tunnel incident detection system in conjunction with the CCTV in tunnels have been highly increased. Despite of this, it is noticed that mathematical algorithm based incident detection system, which are commonly applied in current tunnel operation, show very low detectable rates by less than 50%. The putative major reasons seem to be (1) very weak intensity of illumination (2) dust in tunnel (3) low installation height of CCTV to about 3.5 m, etc. Therefore, an attempt in this study is made to develop an deep-learning based tunnel incident detection system, which is relatively insensitive to very poor visibility conditions. Its theoretical background is given and validating investigation are undertaken focused on the moving vehicles and person out of vehicle in tunnel, which are the official major objects to be detected. Two scenarios are set up: (1) training and prediction in the same tunnel (2) training in a tunnel and prediction in the other tunnel. From the both cases, targeted object detection in prediction mode are achieved to detectable rate to higher than 80% in case of similar time period between training and prediction but it shows a bit low detectable rate to 40% when the prediction times are far from the training time without further training taking place. However, it is believed that the AI based system would be enhanced in its predictability automatically as further training are followed with accumulated CCTV BigData without any revision or calibration of the incident detection system.