• Journal of Korea Multimedia Society
• /
• v.24 no.5
• /
• pp.667-675
• /
• 2021

The treatments for symptoms of Alzheimer's disease are being provided and for the early diagnosis several researches are undergoing. In this regard, by using T1-weighted images several classification techniques had been proposed to distinguish among AD, MCI, and Healthy Control (HC) patients. In this paper, we also used some traditional Machine Learning (ML) approaches in order to diagnose the AD. This paper consists of an improvised feature selection method which is used to reduce the model complexity which accounted an issue while utilizing the ML approaches. In our presented work, combination of subcortical and cortical features of 308 subjects of ADNI dataset has been used to diagnose AD using structural magnetic resonance (sMRI) images. Three classification experiments were performed: binary classification. i.e., AD vs eMCI, AD vs lMCI, and AD vs HC. Proposed Feature Selection method consist of a combination of Principal Component Analysis and Recursive Feature Elimination method that has been used to reduce the dimension size and selection of best features simultaneously. Experiment on the dataset demonstrated that SVM is best suited for the AD vs lMCI, AD vs HC, and AD vs eMCI classification with the accuracy of 95.83%, 97.83%, and 97.87% respectively.

• Journal of Mechanical Science and Technology
• /
• v.14 no.1
• /
• pp.84-92
• /
• 2000

A new bounce back boundary method of the second order in error is proposed for the lattice Boltzmann fluid simulation. This new method can be used for the arbitrarily irregular lattice geometry of a non-slip boundary. The traditional bounce back boundary condition for the lattice Boltzmann simulation is of the first order in error. Since the lattice Boltzmann method is the second order scheme by itself, a boundary technique of the second order has been desired to replace the first order bounce back method. This study shows that, contrary to the common belief that the bounce back boundary condition is unilaterally of the first order, the second order bounce back boundary condition can be realized. This study also shows that there exists a generalized bounce back technique that can be characterized by a single interpolation parameter. The second order bounce back method can be obtained by proper selection of this parameter in accordance with the detailed lattice geometry of the boundary. For an illustrative purpose, the transient Couette and the plane Poiseuille flows are solved by the lattice Boltzmann simulation with various boundary conditions. The results show that the generalized bounce back method yields the second order behavior in the error of the solution, provided that the interpolation parameter is properly selected. Coupled with its intuitive nature and the ease of implementation, the bounce back method can be as good as any second order boundary method.

• Nuclear Engineering and Technology
• /
• v.49 no.2
• /
• pp.404-410
• /
• 2017

In order to help decision-makers in the early design phase to improve and make more cost-efficient system safety and reliability baselines of aircraft design concepts, a method (Multi-objective Optimization for Safety and Reliability Trade-off) that is able to handle trade-offs such as system safety, system reliability, and other characteristics, for instance weight and cost, is used. Multi-objective Optimization for Safety and Reliability Trade-off has been developed and implemented at SAAB Aeronautics. The aim of this paper is to demonstrate how the implemented method might work to aid the selection of optimal design alternatives. The method is a three-step method: step 1 involves the modelling of each considered target, step 2 is optimization, and step 3 is the visualization and selection of results (results processing). The analysis is performed within Architecture Design and Preliminary Design steps, according to the company's Product Development Process. The lessons learned regarding the use of the implemented trade-off method in the three cases are presented. The results are a handful of solutions, a basis to aid in the selection of a design alternative. While the implementation of the trade-off method is performed for companies, there is nothing to prevent adapting this method, with minimal modifications, for use in other industrial applications.

• Industrial Engineering and Management Systems
• /
• v.11 no.2
• /
• pp.155-164
• /
• 2012

As the global market becomes more competitive, manufacturing industries face relentless pressure caused by a growing tendency of greater varieties of products, shorter manufacturing cycles and more sophisticated customer requirements. Efficient and effective supplier selection and order allocation decisions are, therefore, important decisions for a manufacturer to ensure stable material flows in a highly competitive supply chain, in particular, when customers are willing to accept products with less desirable product attributes (e.g., color, delivery date) for economic reasons. This paper attempts to solve optimally the challenging problem of supplier selection and order allocation, taking into consideration the customer flexibility for a manufacturer producing multi-products to satisfy the customers' demands in a multi period planning horizon. A new mixed integer programming model is developed to describe the behavior of the supply chain. The objective is to maximize the manufacturer's total profit subject to various operating constraints of the supply chain. Due to the complexity and non-deterministic polynomial-time (NP)-hard nature of the problem, an improved genetic approach is proposed to solve the problem optimally. This approach differs from a canonical genetic algorithm in three aspects: a new selection method to reduce the chance of premature convergence and two problem-specific repair heuristics to guarantee feasibility of the solutions. The results of applying the proposed approach to solve a set of randomly generated test problems clearly demonstrate its excellent performance. When compared with applying the canonical genetic algorithm to locate optimal solutions, the average improvement in the solution quality amounts to as high as ten percent.

• Health Policy and Management
• /
• v.21 no.2
• /
• pp.263-278
• /
• 2011

As the competition among hospitals become intensified, hospital management is required to carry out more effective marketing and positioning of the hospital. Successful positioning of a hospital requires the knowledge about how the concrete attributes inherent in the medical service associated with the customer values that customers seek eventually in the medical service. Thus, it is required for hospital management to understand which hospital attributes should be emphasized in order to improve the customer values. The hierarchical structure of service attributes can provide valuable information about effective positioning and advertising. To obtain such knowledge, we employed the order analysis technique as an objective means-end chain method. Order analysis is useful for identifying the causal structure among attributes. Thus, we can examine the underlying causal relationship and eventually the vertical structure of hospital selection attributes. For this study, we conducted a survey to obtain 370 responses for the analysis. The result suggests that hospital management had better increase the scale of hospital, improve the services of health care providers, and modernize the hospital facilities and equipments so as to enhance customer values and eventually to lead them to recommend the hospital to others. Our findings would provide the valuable information for hospital management to develop the more effective hospital positioning and marketing strategies.

• Wind and Structures
• /
• v.31 no.5
• /
• pp.473-482
• /
• 2020

The first-order method for estimating the extreme wind pressure on building envelopes with consideration of the directionality of wind speed and wind pressure is improved to enhance its computational efficiency. In this improved method, the result is obtained directly from the empirical distribution of a random selection of annual maximum wind pressure samples generated by a Monte Carlo method, rather than from the previously utilized extreme wind pressure probability distribution. A discussion of the relationship between the first- and full-order methods indicates that when extreme wind pressures in a non-typhoon climate with a high return period are estimated with consideration of directionality, using the relatively simple first-order method instead of the computationally intensive full-order method is reasonable. The validation of this reasonableness is equivalent to validating two assumptions to improve its computational efficiency: 1) The result obtained by the full-order method is conservative when the extreme wind pressure events among different sectors are independent. 2) The result obtained by the first-order method for a high return period is not significantly affected when the extreme wind speeds among the different sectors are assumed to be independent. These two assumptions are validated by examples in different regions and theoretical derivation.

• Journal of the Korean Data and Information Science Society
• /
• v.21 no.3
• /
• pp.427-435
• /
• 2010

In this study, we develop clinical pathways for test and medical treatment items of acute appendicitis patients in emergency room. In order to develop the clinical pathways, we first employ the order selection method by Park et al. (2010), and then display the selected orders in such a way that associated orders are tied together. More specifically the order selection method that we employ is based on lift and, starting from the orders with higher frequencies, sequentially removes the negatively associated orders with lift values less than 0.9. The way associated orders are selected and tied together is similar to that of the order selection method. More specifically, starting from the selected orders with higher frequencies, the orders with lift values greater than 1.1 are considered associated and displayed to the right of the corresponding order. Therefore, in the diagram of clinical pathways, the orders at the left hand side, in principle, are not associated with each other and upper orders have higher frequencies, and associated orders are located to the right of corresponding order at the left hand side and more left orders, in principle, have higher frequencies.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.6 no.10
• /
• pp.2587-2600
• /
• 2012

It has been demonstrated that choosing an appropriate relay node can improve the transmission rate for the system. However, such system improvement brought by the relay selection may be degraded with the presence of the malicious relay nodes, which are selected but refuse to cooperate for transmissions deliberately. In this paper, we formulate the relay selection issue as a restless bandit problem with the objective to maximize the average rate, while considering the credibility of each relay node, which may be different at each time instant. Then the optimization problem is solved by using the priority-index heuristic method effectively. Furthermore, a low complexity algorithm is offered in order to facilitate the practical implementations. Simulation results are conducted to demonstrate the effectiveness of the proposed trust-based relay selection scheme.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.38 no.2
• /
• pp.82-90
• /
• 2015

Location selection of distribution centers is a crucial task for logistics operators and key decision makers of an organization. This is a multi-criteria decision making (MCDM) process which includes both quantitative and qualitative criteria. In order to propose an optimized location selection model, this research suggests a hierarchical group of evaluation criteria : 5 major criteria with 15 sub-criteria. The MCDM approach presented in this research, by integrating Grey Relational Analysis (GRA) with Analytic Hierarchy Process (AHP), tends to rectify the overall quality and uncertainty of the values of evaluation criteria. An example of a location selection case in Korea is illustrated in this study to show the effectiveness of this method.

• Smart Structures and Systems
• /
• v.5 no.5
• /
• pp.531-546
• /
• 2009

In the present analysis, the spline finite strip with higher-order shear deformation is formulated for the static analysis of piezoelectric composite plates. The proposed method incorporates Reddy's third-order shear deformation theory, Touratier's "Sine" model, Afaq's exponential model, Cho's higher-order zigzag laminate theory, as well as the classic plate theory and the first-order plate theory. Thus, the analysis can be conducted based on any of the above-mentioned theories. The selection of a specific method is done by simply changing a few terms in a 2 by 2 square matrix and the results, obtained according to different plate theories, can be compared to each other. Numerical examples are presented for piezoelectric composite plates subjected to mechanical loading. The results based on different shear deformation theories are compared with the three-dimensional solutions. The behaviours of piezoelectric composite plates with different length-to-thickness ratios, fibre orientations, and boundary conditions are also investigated in these examples.