• Journal of the Korean Society of Radiology
• /
• v.15 no.4
• /
• pp.547-554
• /
• 2021

In order to quickly and accurately diagnose pneumonia on a chest X-ray image, different batch sizes of 4, 8, 16, and 32 were applied to the same Xception deep learning model, and modeling was performed 3 times, respectively. As a result of the performance evaluation of deep learning modeling, in the case of modeling to which batch size 32 was applied, the results of accuracy, loss function value, mean square error, and learning time per epoch showed the best results. And in the accuracy evaluation of the Test Metric, the modeling applied with batch size 8 showed the best results, and the precision evaluation showed excellent results in all batch sizes. In the recall evaluation, modeling applied with batch size 16 showed the best results, and for F1-score, modeling applied with batch size 16 showed the best results. And the AUC score evaluation was the same for all batch sizes. Based on these results, deep learning modeling with batch size 32 showed high accuracy, stable artificial neural network learning, and excellent speed. It is thought that accurate and rapid lesion detection will be possible if a batch size of 32 is applied in an automatic diagnosis study for feature extraction and classification of pneumonia in chest X-ray images using deep learning in the future.

• Journal of the Korean Society of Radiology
• /
• v.16 no.5
• /
• pp.595-602
• /
• 2022

This study tried to suggest the most efficient learning rate for accurate and efficient automatic diagnosis of medical images for chest X-ray pneumonia images using deep learning. After setting the learning rates to 0.1, 0.01, 0.001, and 0.0001 in the Inception V3 deep learning model, respectively, deep learning modeling was performed three times. And the average accuracy and loss function value of verification modeling, and the metric of test modeling were set as performance evaluation indicators, and the performance was compared and evaluated with the average value of three times of the results obtained as a result of performing deep learning modeling. As a result of performance evaluation for deep learning verification modeling performance evaluation and test modeling metric, modeling with a learning rate of 0.001 showed the highest accuracy and excellent performance. For this reason, in this paper, it is recommended to apply a learning rate of 0.001 when classifying the presence or absence of pneumonia on chest X-ray images using a deep learning model. In addition, it was judged that when deep learning modeling through the application of the learning rate presented in this paper could play an auxiliary role in the classification of the presence or absence of pneumonia on chest X-ray images. In the future, if the study of classification for diagnosis and classification of pneumonia using deep learning continues, the contents of this thesis research can be used as basic data, and furthermore, it is expected that it will be helpful in selecting an efficient learning rate in classifying medical images using artificial intelligence.

• Molecular & Cellular Toxicology
• /
• v.1 no.3
• /
• pp.157-163
• /
• 2005

Because of their widespread occurrence and substantial biological activity, halogenated aromatic hydrocarbons are one of the important classes of contaminants in the environment. We have performed comparative molecular field analysis (CoMFA) on structurally diverse ligands of Ah (dioxin) receptor to explore the physico-chemical requirements for binding. All CoMFA models have given $q^{2}$ value of more than 0.5 and $r^{2}$ value of more than 0.83. The predictive ability of the models was validated by an external test set, which gave satisfactory predictive $r^{2}$ values. Best predictions were obtained with CoMFA model of combined modified training set ($q^{2}=0.631,\;r^{2}=0.900$), giving predictive residual value = 0.002 log unit for the test compound. We have suggested a model comprises of four structurally different compounds, which offers a good predictability for various ligands. Our QSAR model is consistent with all previously established QSAR models with less structurally diverse ligands. The implications of the CoMFA/QSAR model presented herein are explored with respect to quantitative hazard identification of potential toxicants.

• The Transactions of the Korean Institute of Electrical Engineers P
• /
• v.61 no.2
• /
• pp.87-92
• /
• 2012

Voltage and current phasor estimation has been executed by GPS-based synchronized PMU, which has become an important way of wide-area blackout protection for the prevention of expending faults in a power system. The PMU technique can not easily get the field data and it is impossible to share information, so that there has been used a FNET(Frequency Monitoring Network) method for the wide-area intelligent protection in USA. It consists of FDR(Fault Disturbance Recorder) and IMS(Information Management System). Therefore, FDR must provide an optimal frequency estimation method that is robust to noise and failure. In this paper, we present comparative studies for the frequency estimation method using IRDWT(Improved Recursive Discrete Wavelet Transform), FRDWT(Fast Recursive Discrete Wavelet Transform), and DFT(Discrete Fourier Transform). The Republic of Korea345[kV] power system modeling data by EMTP-RV are used to evaluate the performance of the proposed two kinds of RDWT(Recursive Discrete Wavelet Transform) and DFT. The simulation results show that the proposed frequency estimation technique using FRDWT could be the optimal frequency measurement method, and thus be applied to FDR.

• Computers and Concrete
• /
• v.26 no.3
• /
• pp.285-292
• /
• 2020

In engineering structures, to having the projected structure to serve all the engineering purposes, the theory to be used during the modeling stage is also of great importance. In the present work, an analytical solution of the free vibration of the beam composed of functionally graded materials (FGMs) is presented utilizing different beam theories. The comparison of supposed beam theory for free vibration of functionally graded (FG) beam is examined. For this aim, Euler-Bernoulli, Rayleigh, Shear, and Timoshenko beam theories are employed. The functionally graded material properties are assumed to vary continuously through the thickness direction of the beam with respect to the volume fraction of constituents. The governing equations of free vibration of FG beams are derived in the frameworks of four beam theories. Resulting equations are solved versus simply supported boundary conditions, analytically. To verify the results, comparisons are carried out with the available results. Parametrical studies are performed for discussing the effects of supposed beam theory, the variation of beam characteristics, and FGM properties on the free vibration of beams. In conclusion, it is found that the interaction between FGM properties and the supposed beam theory is of significance in terms of free vibration of the beams and that different beam theories need to be used depending on the characteristics of the beam in question.

• The Korean Journal of Physiology and Pharmacology
• /
• v.14 no.6
• /
• pp.441-447
• /
• 2010

B13, a ceramide analogue, is a ceramidase inhibitor and induces apoptosis to give potent anticancer activity. A series of thiourea B13 analogues was evaluated for their in vitro cytotoxic activities against human renal cancer Caki-2 and leukemic cancer HL-60 in the MTT assay. Some compounds (12, 15, and 16) showed stronger cytotoxicity than B13 and C6-ceramide against both tumor cell lines, and compound (12) gave the most potent activity with $IC_{50}$ values of 36 and $9\;{\mu}M$, respectively. Molecular modeling of thiourea B13 analogues was carried out by comparative molecular field analysis (CoMFA) and comparative molecular similarity indices analysis (CoMSIA). We obtained highly reliable and predictive CoMSIA models with cross-validated $q^2$ values of 0.707 and 0.753 and CoMSIA contour maps to show the structural requirements for potent activity. These data suggest that the amide group of B13 could be replaced by thiourea, that the stereochemistry of 1,3-propandiol may not be essential for activity and that long alkyl chains increase cytotoxicity.

• The Korean Journal of Physiology and Pharmacology
• /
• v.15 no.6
• /
• pp.423-429
• /
• 2011

B13 analogues are being considered as therapeutic agents for cancer cells, since B13 is a ceramide analogue and inhibits ceramidase to promote apoptosis in cancer cells. B13 sulfonamides are assumed to have biological activity similar to B13, since they are made by bioisosterically substituting the carboxyl moiety of B13 with sulfone group. Twenty B13 sulfonamides were evaluated for their in vitro cytotoxicities against human colon cancer HT-29 and lung cancer A549 cell lines using MTT assays. Replacement of the amide group with a sulfonamide group increased cytotoxicity in both cancer cell lines. The sulfonamides with long alkyl chains exhibited activities two to three times more potent than that of B13 and compound (15) had the most potent activity with $IC_{50}$ values of 27 and $28.7{\mu}M$ for HT-29 and A549, respectively. The comparative molecular field analysis (CoMFA) and comparative molecular similarity indices analysis (CoMSIA) were used to carry out QSAR molecular modeling of these compounds. The predictive CoMSIA models for HT-29 and A549 gave cross-validated q2 values of 0.703 and 0.830, respectively. From graphical analysis of these models, we suppose that the stereochemistry of 1,3-propandiol is not important for activity and that introduction of a sulfonamide group and long alkyl chains into B13 can increase cytotoxicity.

• Journal of Integrative Natural Science
• /
• v.8 no.2
• /
• pp.89-98
• /
• 2015

Chemoattractant receptor homologous molecule expressed on Th2 cells (CRTh2) is a G-protein coupled receptor targeted for inflammatory diseases such as asthma, allergic rhinitis and atopic dermatitis. In this study, pharmacophore modeling and comparative molecular field analysis (CoMFA) were performed on the series of 2-(2-(benzylthio)-1H-benzo[d]imidazol-1-yl) acetic acids derivatives. Five highly active compounds were used for generation of pharmacophore models using GASP module. The best pharmacophore model was selected and used as template for the alignment of compounds which was used for CoMFA analysis. The best predictions obtained for CoMFA was $q^2=0.545$, $r^2=0.756$. The predictive ability of the model was investigated using 15 test set compounds. Contour maps suggested that presence of bulky substituents at $5^{th}$ position of benzene ring connected to suphur atoms attached to imidazol ring will increase the activity of the compounds. The results obtained from this study will be useful to design more potent CRTh2 antagonists.

• Transactions of Materials Processing
• /
• v.26 no.1
• /
• pp.18-27
• /
• 2017

A comparative study was performed on strain-compensated Arrhenius-type constitutive models established with two regression methods: polynomial regression and regression Kriging. For measurements at high temperatures, experimental data of 70Cr3Mo steel were adopted from previous research. An Arrhenius-type constitutive model necessitates strain compensation for material constants to account for strain effect. To associate the material constants with strain, we first evaluated them at a set of discrete strains, then capitalized on surrogate modeling to represent the material constants as a function of strain. As a result, disparate flow stress models were formed via the two different regression methods. The constructed constitutive models were examined systematically against measured flow stresses by validation methods. The predicted material constants were found to be quite accurate compared to the actual material constants. However, notable mismatches between measured and predicted flow stresses were revealed by the proposed validation techniques, which carry out validation with not the entire, but a single tensile test case.

• Smart Structures and Systems
• /
• v.22 no.4
• /
• pp.399-411
• /
• 2018

The difficulty in modeling complex nonlinear structures lies in the presence of significant sources of uncertainties mainly attributed to sudden changes in the structure's behavior caused by regular aging factors or extreme events. Quantifying these uncertainties and accurately representing them within the complex mathematical framework of Structural Health Monitoring (SHM) are significantly essential for system identification and damage detection purposes. This study highlights the importance of uncertainty quantification in SHM frameworks, and presents a comparative analysis between intrusive and non-intrusive techniques in quantifying uncertainties for SHM purposes through two different variations of the Kalman Filter (KF) method, the Ensemble Kalman filter (EnKF) and the Polynomial Chaos Kalman Filter (PCKF). The comparative analysis is based on a numerical example that consists of a four degrees-of-freedom (DOF) system, comprising Bouc-Wen hysteretic behavior and subjected to El-Centro earthquake excitation. The comparison is based on the ability of each technique to quantify the different sources of uncertainty for SHM purposes and to accurately approximate the system state and parameters when compared to the true state with the least computational burden. While the results show that both filters are able to locate the damage in space and time and to accurately estimate the system responses and unknown parameters, the computational cost of PCKF is shown to be less than that of EnKF for a similar level of numerical accuracy.