A new method to estimate parameters of linear compartmental models using artificial neural networks

At present, the preferred tool for parameter estimation in compartmental analysis is an iterative procedure; weighted nonlinear regression. For a large number of applications, observed data can be fitted to sums of exponentials whose parameters are directly related to the rate constants/coefficients of the compartmental models. Since weighted nonlinear regression often has to be repeated for many different data sets, the process of fitting data from compartmental systems can be very time consuming. Furthermore the minimization routine often converges to a local (as opposed to global) minimum. In this paper, we examine the possibility of using artificial neural networks instead of weighted nonlinear regression in order to estimate model parameters. We train simple feed-forward neural networks to produce as outputs the parameter values of a given model when kinetic data are fed to the networks' input layer. The artificial neural networks produce unbiased estimates and are orders of magnitude faster than regression algorithms. At noise levels typical of many real applications, the neural networks are found to produce lower variance estimates than weighted nonlinear regression in the estimation of parameters from mono- and biexponential models. These results are primarily due to the inability of weighted nonlinear regression to converge. These results establish that artificial neural networks are powerful tools for estimating parameters for simple compartmental models.     

An anthropocentric approach to knowledge-based preventive maintenance

It is common knowledge today that application of sound manufacturing concepts and techniques will only provide cost-effective, long-term return when manufacturing functions are totally integrated with the operators, maintenance engineers and other associated personnel. This paper presents a human-centred methodology and its associated knowledge-based tool, to utilize equipment failure information for the purpose of machine down-time minimization. Based upon a structured and rigorous assessment of each breakdown, and combined with total employee participation, continuous improvement of maintenance function and operational effectiveness of equipment can be made a reality.     

Calculation of shift of avalanche transit time phase delay due tooptically injected carriers in indium phosphide avalanche diodes

A computer method for the calculation of the phase shift due to optically injected carriers in an InP avalanche transit time diode has been suggested using the numerically simulated negative resistance profiles in the depletion layer of the diode. The results show that the phase shift due to hole injection is larger than that due to electron injection which explains the pronounced effect of photogenerated hole leakage current in modulating the microwave properties of InP diodes     

The antimicrobial effect of wine on Listeria innocua in a model stomach system

A model stomach, containing a food matrix and a synthetic gastric fluid, was used to study the bactericidal effect of ingested wine on Listeria innocua. Volumes of wine equivalent to the ingestion of one glass and half a bottle, led, over a period of less than 2 h, to a reduction of 3 and 4 logarithmic cycles of the initial population respectively. The influence of ethanol and organic acids, wine constituents with known antimicrobial properties, was investigated. Ethanol exhibited a higher bactericidal effect than the mixture of the main wine organic acids. When testing the organic acids separately, malic and lactic acids were found to have the strongest effect. The combination of ethanol with the organic acids acted synergistically but to a lesser extent than wine itself. The results suggest that the ingestion of wine during a meal may diminish the quantity of Listeria persisting further in the alimentary tract.     

A 3-D Miniaturization Method for Low-Impedance Designs

Microstrip interconnects with a V conductor are designed, fabricated, and measured to provide a compact solution for designs requiring low characteristic impedance lines. S-parameter curves are shown up to 35 GHz for 0.5-cm-long lines. The 308-mum-deep V structure produces a 33.8-Omega line with strong standing waves and reflections under 5 dB. To further reduce the impedance, a partial shield is added that results in 6.7 times reduction of signal line width, near elimination of open-end effect, and excellent correlation with a standard 15-Omega microstrip up to 25 GHz. A filter demonstration shows near ideal behavior in the 3 dB response and low return loss when compared to a similar conventional design.     

Modeling the tensile properties in β-processed α/β Ti alloys

The development of a set of computational tools that permit microstructurally based predictions for the tensile properties of commercially important titanium alloys, such as Ti-6Al-4V, is a valuable step toward the accelerated maturation of materials. This paper will discuss the development of neural network models based on a Bayesian framework to predict the yield and ultimate tensile strengths of Ti-6Al-4V at room temperature. The development of such rules-based model requires the population of extensive databases, which in the present case are microstructurally based. The steps involved in database development include producing controlled variations of the microstructure using novel approaches to heat treatments, the use of standardized stereology protocols to characterize and quantify microstructural features rapidly, and mechanical testing of the heat-treated specimens. These databases have been used to train and test neural network models for prediction of tensile properties. In addition, these models have been used to identify the influence of individual microstructural features on the tensile properties, consequently guiding the efforts toward development of more robust mechanistically based models. Based on the neural network model, it is possible to investigate the influence of individual microstructural features on the tensile properties, and in certain cases these dependencies can point toward unrecognized phenomena. For example, the apparently unexpected trend of increase in tensile strength with increasing prior β-grain size has led to the determination of the pronounced role of the basketweave microstructure in strengthening these alloys, especially in case of larger prior β grains. This article is based on a presentation made in the symposium “Computational Aspects of Mechanical Properties of Materials,” which occurred at the 2005 TMS Annual Meeting, February 13–17, 2005, in San Francisco, CA, under the auspices of the MPMD-Computational Materials Science & Engineering (Jt. ASM-MSCTS) Committee.     

A novel fuzzy clustering algorithm for the analysis of axillary lymph node tissue sections

Recently Fourier Transform Infrared (FTIR) spectroscopic imaging has been used as a tool to detect the changes in cellular composition that may reflect the onset of a disease. This approach has been investigated as a mean of monitoring the change of the biochemical composition of cells and providing a diagnostic tool for various human cancers and other diseases. The discrimination between different types of tissue based upon spectroscopic data is often achieved using various multivariate clustering techniques. However, the number of clusters is a common unknown feature for the clustering methods, such as hierarchical cluster analysis, k-means and fuzzy c-means. In this study, we apply a FCM based clustering algorithm to obtain the best number of clusters as given by the minimum validity index value. This often results in an excessive number of clusters being created due to the complexity of this biochemical system. A novel method to automatically merge clusters was developed to try to address this problem. Three lymph node tissue sections were examined to evaluate our new method. These results showed that this approach can merge the clusters which have similar biochemistry. Consequently, the overall algorithm automatically identifies clusters that accurately match the main tissue types that are independently determined by the clinician.