A polarizable force field for water using an artificial neural network

A force field for liquid water including polarization effects has been constructed using an artificial neural network (ANN). It is essential to include a many-body polarization effect explicitly into a potential energy function in order to treat liquid water which is dense and highly polar. The new potential energy function is a combination of empirical and nonempirical potentials. The TIP4P model was used for the empirical part of the potential. For the nonempirical part, an ANN with a back-propagation of error algorithm (BPNN) was introduced to reproduce the complicated many-body interaction energy surface from ab initio quantum mechanical calculations. BPNN, described in terms of a matrix, provides enough flexibility to describe the complex potential energy surface (PES). The structural and thermodynamic properties, calculated by isobaric-isothermal (constant-NPT) Monte Carlo simulations with the new polarizable force field for water, are compatible with experimental results. Thus, the simulation establishes the validity of using our estimated PES with a polarization effect for accurate predictions of liquid state properties. Applications of this approach are simple and systematic so that it can easily be applied to the development of other force fields besides the water-water system.     

人工神经网络研究硝基苯化合物结构与毒性的关系

基于简单的化学基团描述符,应用人工神经网络研究了硝基苯类化合物对黑呆头鱼的毒性构效关系,并与多元线性回归相比较,结果显示了人工神经网络处理非线性问题的优越性.     

Simulation of aerated lagoon using artificial neural networks and multivariate regression techniques

The aim of this study was to develop an empirical model that provides accurate predictions of the biochemical oxygen demand of the output stream from the aerated lagoon at International Paper of Brazil, one of the major pulp and paper plants in Brazil. Predictive models were calculated from functional link neural networks (FLNNs), multiple linear regression, principal components regression, and partial least-squares regression (PLSR). Improvement in FLNN modeling capability was observed when the data were preprocessed using the PLSR technique. PLSR also proved to be a powerful linear regression technique for this problem, which presents operational data limitations.     

A pilot study on colonic mucosal tissues by fluorescence spectroscopy technique: Discrimination by principal component analysis (PCA) and artificial neural network (ANN) analysis

Pulsed laser‐induced autofluorescence spectra of pathologically certified normal and malignant colonic mucosal tissues were recorded at 325 nm excitation. The spectra were analysed using three different methods for discrimination purposes. First, all the spectra were subjected to the principal component analysis (PCA) and the discrimination between normal and malignant cases were achieved using parameters like, spectral residuals, Mahalanobis distance and scores of factors. Second, to understand the changes in tissue composition between the two classes (normal, and malignant), difference spectrum was constructed by subtracting mean spectrum of calibration set samples from simulated mean of all spectra of any one class (normal/malignant) and in third, artificial neural network (ANN) analysis was carried out on the same set of spectral data by training the network with spectral features like, mean, median, spectral residual, energy, standard deviation, number of peaks for different thresholds (100, 250 and 500) after carrying out 1st‐order differentiation of the training set samples and discrimination between normal and malignant conditions were achieved. The specificity and sensitivity were determined in PCA and ANN analyses and they were found to be 100 and 91.3% in PCA, and 100 and 93.47% in ANN, respectively. Copyright © 2008 John Wiley & Sons, Ltd.     

Examination of the influence of different variables on prediction of unit cell parameters in perovskites using counter‐propagation artificial neural networks

In this work, the unit cell parameter (a) of the series of cubic ABX₃ perovskites was modeled using counter‐propagation artificial neural networks, and the influence of different input variables was examined by using algorithm for automatic adjustment of the relative importance of the variables. The input variables used in this model were the ionic radii of A, B, and X as well as the oxidation state (z) and the electronegativity (χ) of the anion. The developed models have good generalization performances—good agreement between experimental and predicted values for lattice parameter. One of the important outcomes from this work is obtained from the results of the automatic adjustment of the relative importance of input variables. That is to say, this analysis gave us an insight that the most pronounced influence on the successful prediction of the unit cell parameter of the analyzed data set of cubic ABX₃ perovskites has the effective ionic radii of B‐cation. In addition to this, it may be concluded that the separation of the compounds in different regions of counter‐propagation artificial neural networks was predominantly influenced by the input variables with regard to the physical parameters of the anion. Copyright © 2012 John Wiley & Sons, Ltd.     

Fluorescence optosensor using an artificial neural network for screening of polycyclic aromatic hydrocarbons

This paper presents an optosensor for screening of four polycyclic aromatic hydrocarbons: anthracene (ANT), benzo[a]pyrene (BaP), fluoranthene (FLT), and benzo[b]fluoranthene (Bbf) using a photomultiplier device with an artificial neural network as transducer. The optosensor is based on the on-line immobilization on a non-ionic resin (Amberlite XAD-4) solid support in a continuous flow. The determination was performed in 15 mM H₂PO₄⁻/HPO₄²⁻ buffer solution at pH 7 and 25% of 1,4-dioxane. Feed forward neural networks (multiplayer perceptron) have been trained to quantify the considered Polycyclic aromatic hydrocarbons (PAHs) in mixtures under optimal conditions. The optosensor proposed was also applied satisfactorily to the determination of the considered PAHs in water samples in presence of the other 12 EPA–PAHs.     

Optimization methodology applied to feed‐forward artificial neural network parameters

This article recommends a methodology for developing a neural network with great chances to be an optimal one. The method is based on trial and error in determining the internal parameters of the network considered as having a significant influence over its performance: the number of hidden layers, activation function, number of neurons in the hidden layers, training epochs, learning rate, and momentum term. This optimization methodology is presented in two separate sections: first of them contains a series of practical considerations recommended for neural network modeling, and the second is represented by the proposed optimization algorithm, formulated in six steps and based on the practical statements. Two case studies are chosen to exemplify the use of the algorithm for finding the near optimal neural network: the dependence of the reduced and intrinsic viscosities of the siloxane‐organic copolymers of the solution concentration, temperature, and copolymer type, differing by the siloxane sequence length. The two siloxane‐organic polyazomethines resulted by the reaction of a fully aromatic bisazomethine diol with α,ω‐bis(chloromethyl)oligodimethylsiloxanes. © 2009 Wiley Periodicals, Inc. Int J Quantum Chem, 2011     

Analyses of the temperature and pH effects on the complexation of magnesium and calcium in human blood plasma: an approach using artificial neural networks

The temperature and pH effects on the equilibrium of a blood plasma model have been studied on the basis of artificial neural networks. The proposed blood plasma was modeled considering two important metals, calcium and magnesium, and six ligands, namely, alanate, carbonate, citrate, glycinate, histidinate and succinate. A large data set has been used to simulate different concentrations of magnesium and calcium as a function of temperature and pH and these data were used for training the neural network. The proposed model allowed different types of analyses, such as the effects of pH on calcium and magnesium concentrations, the competition between calcium and magnesium for ligands and the effects of temperature on calcium and magnesium concentrations. The model developed was also used to predict how the variation of calcium concentration can affect magnesium concentrations. A comparison of neural network predictions against experimental data produced errors of about 3%. Moreover, in agreement with experimental measurements (Wang et al. in Arch. Pathol. 126:947–950, 2002; Heining et al. in Scand. J. Clin. Lab. Invest. 43:709–714, 1983), the artificial neural network predicted that calcium and magnesium concentrations decrease when pH increases. Similarly, the magnesium concentrations are less sensitive than calcium concentrations to pH changes. It is also found that both calcium and magnesium concentrations decrease when the temperature increases. Finally, the theoretical model also predicted that an increase of calcium concentrations will lead to an increase of magnesium concentration almost at the same rate. These results suggest that artificial neural networks can be efficiently applied as a complementary tool for studying metal ion complexation, with especial attention to the blood plasma analysis. Figure Artificial neural networks for predicting the behavior of calcium and magnesium as a function of pH and temperature in human blood plasma     

Quantification of fructo-oligosaccharides based on the evaluation of oligomer ratios using an artificial neural network

The application of an internal standard in quantitative analysis is desirable in order to correct for variations in sample preparation and instrumental response. In mass spectrometry of organic compounds, the internal standard is preferably labelled with a stable isotope, such as ¹⁸O, ¹⁵N or ¹³C. In this study, a method for the quantification of fructo-oligosaccharides using matrix-assisted laser desorption/ionisation time-of-flight mass spectrometry (MALDI TOF MS) was proposed and tested on raftilose, a partially hydrolysed inulin with a degree of polymeration 2-7. A tetraoligosaccharide nystose, which is chemically identical to the raftilose tetramer, was used as an internal standard rather than an isotope-labelled analyte. Two mathematical approaches used for data processing, conventional calculations and artificial neural networks (ANN), were compared. The conventional data processing relies on the assumption that a constant oligomer dispersion profile will change after the addition of the internal standard and some simple numerical calculations. On the other hand, ANN was found to compensate for a non-linear MALDI response and variations in the oligomer dispersion profile with raftilose concentration. As a result, the application of ANN led to lower quantification errors and excellent day-to-day repeatability compared to the conventional data analysis. The developed method is feasible for MS quantification of raftilose in the range of 10-750 pg with errors below 7%. The content of raftilose was determined in dietary cream; application can be extended to other similar polymers. It should be stressed that no special optimisation of the MALDI process was carried out. A common MALDI matrix and sample preparation were used and only the basic parameters, such as sampling and laser energy, were optimised prior to quantification.     

Experimental study on the 3D‐printed plastic parts and predicting the mechanical properties using artificial neural networks

This study investigates the mechanical properties of 3D‐printed plastic parts fabricated using Fused Deposition Modeling (FDM). For this purpose, a 3D printer named KASAME was designed and built by the researchers. The test samples were fabricated using polylactic acid (PLA). The experiments were conducted using three melt temperatures (190°C, 205°C, and 220°C), four layer thickness values (0.06 mm, 0.10 mm, 0.19 mm, and 0.35 mm), and three raster pattern orientations (+45°/?45° [the crisscross pattern], horizontal and vertical). Tensile strength tests were performed to determine tensile strength values of the samples and fracture surfaces were also analyzed. Using artificial neural networks, a mathematical model for the tensile test results was generated corresponding to the raster pattern employed in 3D fabrication. Tensile strength tests indicated that melt temperature, layer thickness, and raster pattern orientation had a significant effect on the tensile strengths of the samples. According to the result of the experiment, the maximum average tensile strength values were observed for the samples fabricated using the crisscross raster pattern. The analysis of variance (ANOVA) table shows the raster pattern (PCR) value of 48.68% was obtained with the highest degree of influence. With respect to R ², the best performing artificial neural network model, with test and training values of 0.999199 and 0.999997, respectively, was observed to be the crisscross raster pattern. Copyright © 2016 John Wiley & Sons, Ltd.     

Study on growth rate of TiO2 nanostructured thin films: simulation by molecular dynamics approach and modeling by artificial neural network

Effects of the deposition process parameters on the thickness of TiO₂ nanostructured film were simulated using the molecular dynamics (MD) approach and modeled by the artificial neural network (ANN) and regression method. Accordingly, TiO₂ nanostructured film was prepared experimentally with the sol–gel dip‐coating method. Structural instabilities can be expected, due to short‐ and/or long‐range intermolecular forces, leading to the surface inhomogeneities. In the MD simulation, the Morse potential function was used for the inter‐atomic interactions, and equations of motion for atoms were solved by Verlet algorithm. The effect of the withdrawal velocity, drying temperature and number of deposited layers were studied in order to characterize the film thickness. The results of MD simulations are reasonably consistent with atomic force microscopy, scanning electron microscopy and Dektak surface profiler. Finally, the outputs from experimental data were analyzed by using the ANN in order to investigate the effects of deposition process parameters on the film thickness. In this case, various architectures have been checked using 75% of experimental data for training of the ANN. Among the various architectures, feed‐forward back‐propagation network with trainer training algorithm was found as the best architecture. Based on the R‐squared value, the ANN is better than the regression model in predicting the film thickness. The statistical analysis for those results was then used to verify the fitness of the complex process model. Based on the results, this modeling methodology can explain the characteristics of the TiO₂ nanostructured thin film and growth mechanism varying with process conditions. © 2013 The Authors. Surface and Interface Analysis published by John Wiley & Sons Ltd.     

Multivariate analysis of ToF‐SIMS data using mass segmented peak lists

Time‐of‐flight secondary ion mass spectrometry (ToF‐SIMS) provides detailed molecular insight into the surface chemistry of a diverse range of material types. Extracting useful and specific information from the mass spectra and reducing the dimensionality of very large datasets are a challenge that has not been fully resolved. Multivariate analysis has been widely deployed to assist in the interpretation of ToF‐SIMS data. Principal component analysis is a popular approach that requires the generation of peak lists for every spectrum. Peak list sizes and the resulting data matrices are growing, complicating manual peak selection and analysis. Here we report the generation of very large ToF‐SIMS peak lists using up‐binning, the mass segmentation of spectral data in the range 0 to 300 m/z in 0.01 m/z intervals. Time‐of‐flight secondary ion mass spectrometry data acquired from a set of 4 standard polymers (polyethylene terephthalate, polytetrafluoroethylene, poly(methyl methacrylate), and low‐density polyethylene) are used to demonstrate the efficacy of this approach. The polymer types are discriminated to a moderate extent by principal component analysis but are easily skewed with saturated species or contaminants present in ToF‐SIMS data. Artificial neural networks, in the form of self‐organising maps, are introduced and provide a non‐linear approach to classifying data and focussing on similarities between samples. The classification outcome achieved is excellent for different polymer types and for spectra from a single polymer type generated by using different primary ions. This method offers great promise for the investigation of more complex systems including polymer classes and blends and mixtures of biological materials.     

On the use of counterpropagation artificial neural networks to characterize Italian rice varieties

A counterpropagation artificial neural network (CP-ANN) approach was used to classify 1779 Italian rice samples according to their variety, using physical measurements which are routinely determined for the commercial classification of the product. If compared to the classical Principal Component Analysis, the mapping based on the Kohonen network showed a significantly better representational ability, being able to separate classes which appeared quite undistinguished in the PC space. From the classification and prediction viewpoint, the optimal CP-ANN was able to correctly predict more than 90% of the test set samples.     

An experimental design approach employing artificial neural networks for the determination of potential endocrine disruptors in food using matrix solid-phase dispersion

Matrix solid-phase dispersion (MSPD) as a sample preparation method for the determination of two potential endocrine disruptors, linuron and diuron and their common metabolites, 1-(3,4-dichlorophenyl)-3-methylurea (DCPMU), 1-(3,4-dichlorophenyl) urea (DCPU) and 3,4-dichloroaniline (3,4-DCA) in food commodities has been developed. The influence of the main factors on the extraction process yield was thoroughly evaluated. For that purpose, a 3^(4–1) fractional factorial design in further combination with artificial neural networks (ANNs) was employed. The optimal networks found were afterwards used to identify the optimum region corresponding to the highest average recovery displaying at the same time the lowest standard deviation for all analytes. Under final optimal conditions, potato samples (0.5 g) were mixed and dispersed on the same amount of Florisil. The blend was transferred on a polypropylene cartridge and analytes were eluted using 10 ml of methanol. The extract was concentrated to 50 μl of acetonitrile/water (50:50) and injected in a high performance liquid chromatography coupled to UV–diode array detector system (HPLC/UV–DAD). Recoveries ranging from 55 to 96% and quantification limits between 5.3 and 15.2 ng/g were achieved. The method was also applied to other selected food commodities such as apple, carrot, cereals/wheat flour and orange juice demonstrating very good overall performance.     

基于广义回归神经网络的微生物降解效应对助燃剂检测的影响

为探究火场土壤载体中微生物降解效应对助燃剂鉴定的影响，在普通土和培养土两种土样上注射助燃剂，以密封存放时间为变量，通过静态顶空的样品预处理方式对样品内的助燃剂残留物进行气相色谱-质谱法（GC-MS）鉴定。研究发现，微生物降解效应会改变样品内助燃剂组分，不同土样内降解结果有所不同，普通土样的降解效应较培养土样明显，C9~C12直链烷烃和单取代芳香烃更易被降解，多取代芳烃的降解难度随取代基含量的增多而增加。按土样种类采用主成分分析（PCA）的方式进行数据降维后，采用广义回归神经网络（GRNN）对不同土样结果区分，准确率达100%。     

2D bitmapping approach for identification and quantitation of common base flavor adulterants using surface acoustic wave arrays and artificial neural network data analysis

Arrays of polymer-coated surface acoustic wave microsensors are used in conjunction with a variety of signal-processing algorithms known as artificial neural networks (ANN). This format of data analysis has the capability to characterize complex mixtures of volatile and semi-volatile organic compounds commonly found in base flavors. The approach described, which minimizes the number of training sets while retaining the robustness of an ANN, utilizes a 2D bitmap matrix. The matrix is obtained by converting the time domain kinetics of sensor response into a bitmap. The high data throughput of this approach enables quantitation on the order of ppm of common base flavor adulterants.     

Direct analysis of blood serum by total reflection X-ray fluorescence spectrometry and application of an artificial neural network approach for cancer diagnosis

Iron, copper, zinc and selenium were determined directly in serum samples from healthy individuals (n=33) and cancer patients (n=27) by total reflection X-ray fluorescence spectrometry using the Compton peak as internal standard [L.M. Marcó P. et al., Spectrochim. Acta Part B 54 (1999) 1469–1480]. The standardized concentrations of these elements were used as input data for two-layer artificial neural networks trained with the generalized delta rule in order to classify such individuals according to their health status. Various artificial neural networks, comprising a linear function in the input layer, a hyperbolic tangent function in the hidden layer and a sigmoid function in the output layer, were evaluated for such a purpose. Of the networks studied, the (4:4:1) gave the highest estimation (98%) and prediction rates (94%). The latter demonstrates the potential of the total reflection X-ray fluorescence spectrometry/artificial neural network approach in clinical chemistry.     

Modeling of the relationship between electroosmotic flow and separation parameters in capillary zone electrophoresis using artificial neural networks and experimental design

The prediction of migration time of electroosmotic flow (EOF) marker was achieved by applying artificial neural networks (ANN) model based on principal component analysis (PCA) and standard normal distribution simulation to the input variables. The voltage of performance, the temperature in the capillary, the pH and the ionic strength of background electrolytes (BGE) were applied as the input variables to ANN. The range of the performance voltage studied was from 15 to 27 kV, and that of the temperature in the capillary was from 20 to 30 °C. For the pH values studied, the range was from 5.15 to 8.04. The range of the ionic strength investigated in this paper was from 0.040 to 0.097. The prediction abilities of ANN with different pre-processing procedure to the input variables were compared. Under the same performance conditions, the average prediction error of the migration time of the EOF marker was 5.46% with RSD = 1.76% according to 10 parallel runs of the optimized ANN structure by the proposed approach, and that of the 10 parallel predictions of the optimal ANN structure for the different performance conditions was 12.95% with RSD = 2.29% according to the proposed approach. The study showed that the proposed method could give better predicted results than other approaches discussed.