Automatic detection of electroencephalographic changes using adaptive neuro-fuzzy inference system employing Lyapunov exponents

This paper describes the application of adaptive neuro-fuzzy inference system (ANFIS) model for automatic detection of electroencephalographic changes. Decision making was performed in two stages: feature extraction by computation of Lyapunov exponents and classification by the ANFIS trained with the backpropagation gradient descent method in combination with the least squares method. Five types of electroencephalogram (EEG) signals were classified by five ANFIS classifiers. To improve diagnostic accuracy, the sixth ANFIS classifier (combining ANFIS) was trained using the outputs of the five ANFIS classifiers as input data. The proposed ANFIS model combined the neural network adaptive capabilities and the fuzzy logic qualitative approach. Some conclusions concerning the saliency of features on classification of the EEG signals were obtained through analysis of the ANFIS. The performance of the ANFIS model was evaluated in terms of training performance and classification accuracies and the results confirmed that the proposed ANFIS model has potential in classifying the EEG signals.     

A new mixed integer linear programming model for product development using quality function deployment

Quality function deployment (QFD) is a product development process performed to maximize customer satisfaction. In the QFD, the design requirements (DRs) affecting the product performance are primarily identified, and product performance is improved to optimize customer needs (CNs). For product development, determining the fulfillment levels of design requirements (DRs) is crucial during QFD optimization. However, in real world applications, the values of DRs are often discrete instead of continuous. To the best of our knowledge, there is no mixed integer linear programming (MILP) model in which the discrete DRs values are considered. Therefore, in this paper, a new QFD optimization approach combining MILP model and Kano model is suggested to acquire the optimized solution from a limited number of alternative DRs, the values of which can be discrete. The proposed model can be used not only to optimize the product development but also in other applications of QFD such as quality management, planning, design, engineering and decision-making, on the condition that DR values are discrete. Additionally, the problem of lack of solutions in integer and linear programming in the QFD optimization is overcome. Finally, the model is illustrated through an example.     

A novel microbial BOD biosensor developed by the immobilization of P. Syringae in micro‐cellular polymers

BACKGROUND: A biochemical oxygen demand (BOD) sensor, based on an immobilized Pseudomonas syringae in highly porous micro‐cellular polymer (MCP) in combination with a dissolved oxygen electrode, has been developed for the analysis of biodegradable organic compounds in aqueous samples. Microorganisms were immobilized in a molded MCP disk and a wastewater sample was injected into the biocomposite disk by a flow injection system. Dissolved oxygen (DO) changes as a measure of soluble BOD was read with a DO probe placed into a flow cell carrying biocatalytically activated disk. RESULTS: Optimal response of the MCP BOD sensor was obtained at pH 6.8 and 25 °C with a typical response time of 3–5 min for a 2 mm thick molded polymeric disk. The sensor showed detection linearity over the range 5–100 mg L^?1 BOD₅ (r² > 0.99) at a flow rate of 0.6 mL min^?1. The repeatability and reproducibility of the sensor response were found to be 3.08% and 7.77%, respectively. BOD values produced with this biosensor for various municipal and industrial wastewaters correlated well with those determined by the conventional 5‐day BOD test. CONCLUSION: This new biosensor was different from present amperometric BOD biosensor configurations in which the biocatalyst (microbial/enzymatic) is placed between cellulose and Teflon membranes installed on a DO probe. The use of a molded MCP disk coniainng microbial activity offers better stability and lifetime for commercial use in environmental monitoring. Copyright © 2008 Society of Chemical Industry     

Termite Resistance of MDF Panels Treated with Various Boron Compounds

In this study, the effects of various boron compounds on the termite resistance of MDF panels were evaluated. Either borax (BX), boric acid (BA), zinc borate (ZB), or sodium perborate tetrahydrate (SPT) were added to urea-formaldehyde (UF) resin at target contents of 1%, 1.5%, 2% and 2.5% based on dry fiber weight. The panels were then manufactured using 12% urea-formaldehyde resin and 1% NH₄Cl. MDF samples from the panels were tested against the subterranean termites, Coptotermes formosanus Shiraki. Laboratory termite resistance tests showed that all samples containing boron compounds had greater resistance against termite attack compared to untreated MDF samples. At the second and third weeks of exposure, nearly 100% termite mortalities were recorded in all boron compound treated samples. The highest termite mortalities were determined in the samples with either BA or BX. Also, it was found that SPT showed notable performance on the termite mortality. As chemical loadings increased, termite mortalities increased, and at the same time the weight losses of the samples decreased.     

Multi-machine earliness and tardiness scheduling problem: an interconnected neural network approach

This paper addresses the problem of scheduling a set of independent jobs with sequence-dependent setups and distinct due dates on non-uniform multi-machines to minimize the total weighted earliness and tardiness, and explores the use of artificial neural networks as a valid alternative to the traditional scheduling approaches. The objective is to propose a dynamical gradient neural network, which employs a penalty function approach with time varying coefficients for the solution of the problem which is known to be NP-hard. After the appropriate energy function was constructed, the dynamics are defined by steepest gradient descent on the energy function. The proposed neural network system is composed of two maximum neural networks, three piecewise linear and one log-sigmoid network all of which interact with each other. The motivation for using maximum networks is to reduce the network complexity and to obtain a simplified energy function. To overcome the tradeoff problem encountered in using the penalty function approach, a time varying penalty coefficient methodology is proposed to be used during simulation experiments. Simulation results of the proposed approach on a scheduling problem indicate that the proposed coupled network yields an optimal solution which makes it attractive for applications of larger sized problems.     

Examination of the ECG signal classification technique DEA-ELM using deep convolutional neural network features

Multimedia Tools and Applications - The accurate separation of ECG signals has become crucial to identify heart diseases. Machine learning methods are widely used to separate ECG signals. The aim...     

Reduction of Salmonella enterica in Turkey breast slices kept under aerobic and vacuum conditions by application of lactic acid,a bacteriophage,and ultrasound

The effects of different antimicrobial applications against Salmonella enterica subs. enterica DSM 18522 strain in turkey breast slices kept under either aerobic or anaerobic-vacuum conditions for different keeping times of 30 min and 6 hr at 4°C were evaluated. In this study, it was aimed to compare chemical (lactic acid, 3%), biopreservation (bacteriophage Felix O1, 1 × 10⁸ pfu/ml), and physical (ultrasound, 20 kHz) antimicrobial applications in reducing a high initial S. enterica DSM 18522 count (~6 log cfu/g) in turkey breast slices at different conditions. The highest reduction in S. enterica DSM 18522 counts (1.62 log units) was observed in turkey breast slices kept under aerobic condition for 6 hr after being treated with lactic acid. The bacteriophage application significantly (p < .05) reduced S. enterica DSM 18522 counts 1.22 and 1.41 log cfu/g in turkey breast slices that kept under aerobic and anaerobic-vacuum conditions for 30 min, respectively. There was no significant effect of keeping time under aerobic or anaerobic-vacuum conditions on S. enterica DSM 18522 counts in turkey breast slices treated with bacteriophage or ultrasound. However, the reduction of S. enterica DSM 18522 counts in turkey breast slices treated with lactic acid increased with keeping time under aerobic or anaerobic-vacuum conditions. Regardless of keeping conditions, the lowest reductions in S. enterica DSM 18522 counts were detected in turkey breast slices treated with ultrasound. The results of this study show that lactic acid and bacteriophage applications are effective in reducing S. enterica DSM 18522 counts in turkey breast slices.     

Investigating Neutronic Parameters of a Thorium Fusion Breeder with Recurrent Neural Networks

Artificial neural networks (ANNs) have recently been introduced to the nuclear engineering applications as a fast and flexible vehicle to modeling, simulation and optimization. In this paper, a new approach based on recurrent neural networks (RNNs) was presented for the neutronic parameters of a thorium fusion breeder. The results of the RNNs implemented for the tritium breeding ratio computation, energy multiplication factor and net ²³³U production in a thorium fusion breeder and the results available in the literature obtained by using Scale 4.3 were compared. The drawn conclusions confirmed that the proposed RNNs could provide an accurate computation of the tritium breeding ratio computation, the energy multiplication factor and the net ²³³U production of the thorium fusion breeder.     

Analysis of Serum Advanced Glycation Endproducts Reveals Methylglyoxal-Derived Advanced Glycation MG-H1 Free Adduct Is a Risk Marker in Non-Diabetic and Diabetic Chronic Kidney Disease

Accumulation of advanced glycation endproducts (AGEs) is linked to decline in renal function, particularly in patients with diabetes. Major forms of AGEs in serum are protein-bound AGEs and AGE free adducts. In this study, we assessed levels of AGEs in subjects with and without diabetes, with normal renal function and stages 2 to 4 chronic kidney disease (CKD), to identify which AGE has the greatest progressive change with decline in renal function and change in diabetes. We performed a cross-sectional study of patients with stages 2–4 CKD, with and without diabetes, and healthy controls (n = 135). Nine protein-bound and free adduct AGEs were quantified in serum. Most protein-bound AGEs increased moderately through stages 2–4 CKD whereas AGE free adducts increased markedly. Methylglyoxal-derived hydroimidazolone MG-H1 free adduct was the AGE most responsive to CKD status, increasing 8-fold and 30-fold in stage 4 CKD in patients without and with diabetes, respectively. MG-H1 Glomerular filtration flux was increased 5-fold in diabetes, likely reflecting increased methylglyoxal glycation status. We conclude that serum MG-H1 free adduct concentration was strongly related to stage of CKD and increased in diabetes status. Serum MG-H1 free adduct is a candidate AGE risk marker of non-diabetic and diabetic CKD.     

IPM: A computer Interactive Project Management teaching tool

Project management deals with all phases of a project, from the initiation of the planning procedure to the physical completion of the entire effort. The process of teaching the quantitative approaches to project management (i.e. Critical-Path Method and Program Evaluation and Review Technique) is a fairly straightforward task that has been performed for many years. However, teaching the principles of project management is a different task altogether. The project management process involves a sequence of planning, decision making, follow-up activities and reevaluation followed by a repeat of the sequence. That is, in order to truly examine and teach the process of project management (i.e. not just the tools), it is necessary to view the activities in a dynamic projectlike environment. With this idea in mind, the Interactive Project Management game (IPM) was developed. IPM was created with the intent of teaching project management principles in a simulated project environment using the computer. This paper describes IPM, its characteristics and its capabilities, and includes an IPM gaming session.