• Journal of Biomedical Engineering Research
• /
• v.32 no.2
• /
• pp.134-143
• /
• 2011

The selection of meaningful clinical tests and its reference values from a high-dimensional clinical data with imbalanced class distribution, one class is represented by a large number of examples while the other is represented by only a few, is an important issue for differential diagnosis between similar diseases, but difficult. For this purpose, this study introduces methods based on the concepts of both discernibility matrix and function in rough set theory (RST) with two discretization approaches, equal width and frequency discretization. Here these discretization approaches are used to define the reference values for clinical tests, and the discernibility matrix and function are used to extract a subset of significant clinical tests from the translated nominal attribute values. To show its applicability in the differential diagnosis problem, we have applied it to extract the significant clinical tests and its reference values between normal (N = 351) and abnormal group (N = 101) with either cholecystitis or cholelithiasis disease. In addition, we investigated not only the selected significant clinical tests and the variations of its reference values, but also the average predictive accuracies on four evaluation criteria, i.e., accuracy, sensitivity, specificity, and geometric mean, during l0-fold cross validation. From the experimental results, we confirmed that two discretization approaches based rough set approximation methods with relative frequency give better results than those with absolute frequency, in the evaluation criteria (i.e., average geometric mean). Thus it shows that the prediction model using relative frequency can be used effectively in classification and prediction problems of the clinical data with imbalanced class distribution.

• Geomechanics and Engineering
• /
• v.33 no.1
• /
• pp.1-9
• /
• 2023

The unconfined compression strength (UCS) of soils is commonly used either before or during the construction of geo-structures. In the pre-design stage, UCS as a mechanical property is obtained through a laboratory test that requires cumbersome procedures and high costs from in-situ sampling and sample preparation. As an alternative way, the empirical model established from limited testing cases is used to economically estimate the UCS. However, many parameters affecting the 1D soil compression response hinder employing the traditional statistical analysis. In this study, gene expression programming (GEP) is adopted to develop a prediction model of UCS with common affecting soil properties. A total of 79 undisturbed soil samples are collected, of which 54 samples are utilized for the generation of a predictive model and 25 samples are used to validate the proposed model. Experimental studies are conducted to measure the unconfined compression strength and basic soil index properties. A performance assessment of the prediction model is carried out using statistical checks including the correlation coefficient (R), the root mean square error (RMSE), the mean absolute error (MAE), the relatively squared error (RSE), and external criteria checks. The prediction model has achieved excellent accuracy with values of R, RMSE, MAE, and RSE of 0.98, 10.01, 7.94, and 0.03, respectively for the training data and 0.92, 19.82, 14.56, and 0.15, respectively for the testing data. From the sensitivity analysis and parametric study, the liquid limit and fine content are found to be the most sensitive parameters whereas the sand content is the least critical parameter.

• Computers and Concrete
• /
• v.32 no.2
• /
• pp.149-163
• /
• 2023

One of the main design parameters traditionally utilized in projects of geotechnical engineering is the uniaxial compressive strength. The present paper employed three artificial intelligence methods, i.e., the stochastic fractal search (SFS), the multi-verse optimization (MVO), and the vortex search algorithm (VSA), in order to determine the compressive strength of concrete (CSC). For the same reason, 1030 concrete specimens were subjected to compressive strength tests. According to the obtained laboratory results, the fly ash, cement, water, slag, coarse aggregates, fine aggregates, and SP were subjected to tests as the input parameters of the model in order to decide the optimum input configuration for the estimation of the compressive strength. The performance was evaluated by employing three criteria, i.e., the root mean square error (RMSE), mean absolute error (MAE), and the determination coefficient (R²). The evaluation of the error criteria and the determination coefficient obtained from the above three techniques indicates that the SFS-MLP technique outperformed the MVO-MLP and VSA-MLP methods. The developed artificial neural network models exhibit higher amounts of errors and lower correlation coefficients in comparison with other models. Nonetheless, the use of the stochastic fractal search algorithm has resulted in considerable enhancement in precision and accuracy of the evaluations conducted through the artificial neural network and has enhanced its performance. According to the results, the utilized SFS-MLP technique showed a better performance in the estimation of the compressive strength of concrete (R²=0.99932 and 0.99942, and RMSE=0.32611 and 0.24922). The novelty of our study is the use of a large dataset composed of 1030 entries and optimization of the learning scheme of the neural prediction model via a data distribution of a 20:80 testing-to-training ratio.

• Animal Bioscience
• /
• v.37 no.5
• /
• pp.817-825
• /
• 2024

Objective: The aim of this study was to identify suitable polynomial regression for modeling the average growth trajectory and to estimate the relative development of the rib eye area, scrotal circumference, and morphometric measurements of Guzerat young bulls. Methods: A total of 45 recently weaned males, aged 325.8±28.0 days and weighing 219.9±38.05 kg, were evaluated. The animals were kept on Brachiaria brizantha pastures, received multiple supplementations, and were managed under uniform conditions for 294 days, with evaluations conducted every 56 days. The average growth trajectory was adjusted using ordinary polynomials, Legendre polynomials, and quadratic B-splines. The coefficient of determination, mean absolute deviation, mean square error, the value of the restricted likelihood function, Akaike information criteria, and consistent Akaike information criteria were applied to assess the quality of the fits. For the study of allometric growth, the power model was applied. Results: Ordinary polynomial and Legendre polynomial models of the fifth order provided the best fits. B-splines yielded the best fits in comparing models with the same number of parameters. Based on the restricted likelihood function, Akaike's information criterion, and consistent Akaike's information criterion, the B-splines model with six intervals described the growth trajectory of evaluated animals more smoothly and consistently. In the study of allometric growth, the evaluated traits exhibited negative heterogeneity (b<1) relative to the animals' weight (p<0.01), indicating the precocity of Guzerat cattle for weight gain on pasture. Conclusion: Complementary studies of growth trajectory and allometry can help identify when an animal's weight changes and thus assist in decision-making regarding management practices, nutritional requirements, and genetic selection strategies to optimize growth and animal performance.

• Radiation Oncology Journal
• /
• v.23 no.3
• /
• pp.176-185
• /
• 2005

Purpose: Film dosimetry as a part of patient specific intensity modulated radiation therapy quality assurance (IMRT QA) was peformed to develop a new optimization method of film isocenter offset and to then suggest new quantitative criteria for film dosimetry. Materials and Methods: Film dosimetry was peformed on 14 IMRT patients with head and neck cancers. An optimization method for obtaining the local minimum was developed to adjust for the error in the film isocenter offset, which is the largest part of the systemic errors. Results: The adjust value of the film isocenter offset under optimization was 1 mm in 12 patients, while only two patients showed 2 mm translation. The means of absolute average dose difference before and after optimization were 2.36 and $1.56\%$, respectively, and the mean ratios over a $5\%$ tolerance were 9.67 and $2.88\%$. After optimization, the differences in the dose decreased dramatically. A low dose range cutoff (L-Cutoff) has been suggested for clinical application. New quantitative criteria of a ratio of over a $5\%$, but less than $10\%$ tolerance, and for an absolute average dose difference less than $3\%$ have been suggested for the verification of film dosimetry. Conclusion: The new optimization method was effective in adjusting for the film dosimetry error, and the newly quantitative criteria suggested in this research are believed to be sufficiently accurate and clinically useful.

• Progress in Medical Physics
• /
• v.16 no.4
• /
• pp.166-175
• /
• 2005

The purpose of this study is to develop the optimization method for adjusting the film isocenter shift and to suggest the quantitative acceptable criteria for film dosimetry after optimization In the dynamic conformal arc radiation therapy (DCAR). The DCAR planning was peformed In 7 patients with brain metastasis. Both absolute dosimetry with ion chamber and relative film dosimetry were peformed throughout the DCAR using BrainLab's micro-multileaf collimator. An optimization method for obtaining the global minimum was used to adjust for the error in the film isocenter shift, which is the largest pan of systemic errors. The mean of point dose difference between measured value using ion chamber and calculated value acquired from planning system was $0.51{\pm}0.43\%$ and maximum was $1.14\%$ with absolute dosimetry These results were within the AAPM criteria of below $5\%$. The translation values of film isocenter shift with optimization were within ${\pm}$1 mm in all patients. The mean of average dose difference before and after optimization was $1.70{\pm}0.35\%$ and $1.34{\pm}0.20\%$, respectively, and the mean ratios over $5\%$ dose difference was $4.54{\pm}3.94\%$ and $0.11{\pm}0.12\%$, respectively. After optimization, the dose differences decreased dramatically and a ratio over $5\%$ dose difference and average dose difference was less than $2\%$. This optimization method is effective in adjusting the error of the film isocenter shift, which Is the largest part of systemic errors, and the results of this research suggested the quantitative acceptable criteria could be accurate and useful in clinical application of dosimetric verification using film dosimetry as follows; film isocenter shift with optimization should be within ${\pm}$1 mm, and a ratio over $5\%$ dose difference and average dose difference were less than $2\%$.

• Economic and Environmental Geology
• /
• v.41 no.1
• /
• pp.93-113
• /
• 2008

This study presents petrochemistry and $^{40}Ar/^{39}Ar$ absolute ages of subsurface volcanic rock cores from twenty(20) boreholes in the eastern lowland (altitude loom below) of Jeju Island, Handeong-Jongdal-Udo-Susan-Samdal-Hacheon areas, and discusses topography and volcanism in the area. The subsurface volcanic rock cores are mainly basalts in composition with minor tholeiitic andesites and basaltic trachyandesites. Sequences of intercalated tholeiitic, transitional and alkalic lavas suggest that tholeiitic and transitional to alkalic lavas must have erupted contemporaneously. Especially, occurrences of trachybasalts and basaltic trachyandesites at the bases in the area imply that the volcanism in the area was initiated with slightly differentiated alkaline magma activity. The $^{40}Ar/^{39}Ar$ absolute ages of the subsurface volcanic rock cores range from $526{\pm}23ka\;to\;38{\pm}4Ka$. The lava-forming Hawaiian volcanic activities of the eastern lowland can be divided into five sequences on the basis of sediment distribution, whole rock geochemistry and $^{40}Ar/^{39}Ar$ absolute ages of the subsurface volcanic rock cores; stage I-U$(550{\sim}400Ka)$, stage II$(400{\sim}300Ka)$ and stage III$(300{\sim}200Ka)$ during syn-depositional stage of Seoguipo Formation, and stage IV$(200{\sim}100Ka)$ and stage V(younger than 100Ka) during post-depositional stage. In the eastern lowland of Jeju Island, compositional variations and local occurrences of the subsurface volcanic rocks as well as existences of various intercalated sediment layers (including hydrovolcanogenic clasts) suggest that the volcanism must have continued for long time intermittently and that the land has been progressively glowed from inland to coast by volcanic activities and sedimentation. It reveals that the subsurface volcanic rocks in the eastern lowland of Jeju Island must have erupted during relatively younger than 200Ka of stages IV and V. The results of this study are partly in contrast with those of previous studies. This study stresses the need that previous reported volcanic activities in Jeju Island based on K-Ar ages of volcanic rocks should be carefully reviewed, and that stratigraphic correlation from boreholes should be conducted by quantitative criteria combined with petrography and petrochemstry as well as radiometric studies of volcanic rock cores.

• Membrane and Water Treatment
• /
• v.9 no.6
• /
• pp.455-462
• /
• 2018

The measurement and monitoring of the biochemical oxygen demand (BOD) play an important role in the planning and operation of wastewater treatment plants. The most basic method for determining biochemical oxygen demand is direct measurement. However, this method is both expensive and takes a long time. A five-day period is required to determine the biochemical oxygen demand. This study has been carried out in a wastewater treatment plant in Turkey (Hurma WWTP) in order to estimate the biochemical oxygen demand a shorter time and with a lower cost. Estimation was performed using artificial neural network (ANN) method. There are three different methods in the training of artificial neural networks, respectively, multi-layered (ML-ANN), teaching learning based algorithm (TLBO-ANN) and artificial bee colony algorithm (ABC-ANN). The input flow (Q), wastewater temperature (t), pH, chemical oxygen demand (COD), suspended sediment (SS), total phosphorus (tP), total nitrogen (tN), and electrical conductivity of wastewater (EC) are used as the input parameters to estimate the BOD. The root mean squared error (RMSE) and the mean absolute error (MAE) values were used in evaluating performance criteria for each model. As a result of the general evaluation, the ML-ANN method provided the best estimation results both training and test series with 0.8924 and 0.8442 determination coefficient, respectively.

• Journal of the Korean Society of Clothing and Textiles
• /
• v.19 no.1
• /
• pp.115-128
• /
• 1995

The purpose of this study was to investigate the role structure of husband and wife on husband's clothing purchase. For the study, a questionnaire was developed to measure the influence structure and clothing evaluative criteria of husband's clothing purchase and psychological and demographic characteristics of husband. The Purchase influence structure can be defined by applying the concept of Wolfe's power pattern to decision making about buying behavior. Influence structure is the pattern in which influence is distributed among family buying center members for each purchasing decision making item. The analysis was conducted on the basis of 310 couples responces. The result of this study ware as follows; 1. The majority of husband's clothing items were an absolute autonomic·decision product Exceptionally underwear was wife-dominated. The purchase influence structure of husband' s clothing purchase varied on stages in the decision making process. Wife was involved considerably in gathering information search and real purchasing stage. 2. According to the degree of husband and wife influence on discussion stage and final decision stage, consumers were categorized into five types. Husband's psychological characteristics such 3s shopping interest, clothing involvement, importance of mutual satisfaction in purchasing and demographic characteristics were significantly different among types.

• Journal of The Korean Society of Agricultural Engineers
• /
• v.49 no.6
• /
• pp.105-114
• /
• 2007

This study is to predict the spatial expansion of urban areas by applying CA(Cellular Automata)-Markov technique considering MCE(multi-criteria evaluation) and MOLA(multi-objective land allocation) of factor analysis. For the 10 administration districts$(3677.3km^2)$ including the whole Anseong-cheon watershed, the past six temporal land use data(1973, 1981, 1985, 1990, 1994, 2000) from Landsat satellite images were prepared. During this period, the urban area increased $233.71km^2$. Using the 36 indices composed of topological characteristics, population and land use change, the final factor map of MOLA was produced through 5 maps of MCE. Using 1990 and 1994 land use data, the 2000 predicted urban area of CA-Markov with factor map showed 0.06% improvement of absolute error comparing with that of CA-Markov without factor map. By the CA-Markov technique considering factor map, the 2030 and 2060 urban area increased $58.94km^2(0.78%)\;and\;60.14km^2(0.81%)$ respectively comparing with 2000 urban area$(313.19km^2)$. The 2030 and 2060 paddy area decreased $93.28km^2(2.54%)\;and\;93.65km^2(2.55%)$ respectively comparing with 2000 paddy area$(1383.23km^2)$.