Input Data For Artificial Neural Network Research Articles

Fast Nondestructive Measurement and Discrimination of Tablet Based on Vis/NIR Spectroscopy and Chemometrics

This paper presents methods based on chemometrics analysis to select the optimal model for variety discrimination of ginkgo (Ginkgo biloba L.) tablets by using a visible/short-wave near-infrared spectroscopy (Vis/NIRS) system. The tablet varieties used in the research include Da na kang, Xin bang, Tian bao ning, Yi kang, Hua na xing, Dou le, Lv yuan, Hai wang, and Ji yao. All samples (n=270) were scanned in the Vis/NIR region between 325-1075nm using a spectrograph. Principal component artificial neural network (PC-ANN) was used to identify the tablet varieties. In PC-ANN models, the scores of the principal components were chosen as the input nodes for the input layer of ANN. Independent component analysis (ICA) was executed to select several optimal wavelengths based on loading weights. The absorbance values log (1/R), corresponding to the wavelengths of 481nm, 1000nm, 460nm, 572nm, 658nm, 401nm, 998nm, 996nm, 468nm and 661nm were then chosen as the input data of artificial neural network (IC-ANN), and the discrimination rate was reached at 95.6%, which was better than PC-ANN. The results indicated that ginkgo tablets discrimination was good based on the both methods.

The Possibility of Using Artificial Neural Networks for the Estimation of Mass Composition of High-Energy Primary Cosmic Ray

This paper shows that the artificial neural networks (ANN) can be used for determining the type of particles of high-energy primary cosmic ray (i.e. its mass composition) initiating the EAS. The approach implemented here can be used, e.g., in the Auger experiment. We describe the details of the ANN construction and demonstrate that the program is correct and can be further used to solve physical problems. The network was taught and tested based on the data for the maximum of the EAS development (X_max) and primary energy of a particle initiating this EAS (lg(E0)). The identification of particles based on X_max and lg(E0) resulted in around 80% of correct answers for the light mass composition and 99% for the heavy one. We have a correct answer for the mass composition with domination of one type of particles, i.e. light or heavy. Otherwise, additional parameters should be included as ANN input data.

Predicting of blasting vibrations in Sarcheshmeh copper mine by neural network

Artificial Neural Networks (ANN) have proven to be an effective tool for solving complex engineering problems requiring estimation, pattern recognition, and classification of variables. Ground vibration caused by blasting imposes damages and financial penalties and as such must be predicted accurately. In this study, the potentials of ANN are investigated in prediction of ground vibrations due to blasting in open pit mines. Real vibration data is recorded using PDAS 100 seismometers, and used as input data for ANN. Using back propagation algorithm and performance function, root mean square error (RMSE) the network, containing four hidden layers and 23 data sets, was trained. Six sets of data were used to make sure that correct training had been carried out. This produced the coefficient correlation of 0.99355.

Automated quality assessment of autonomously acquired microscopic images of fluorescently stained bacteria

Quality assessment of autonomously acquired microscopic images is an important issue in high-throughput imaging systems. For example, the presence of low quality images (>or=10%) in a dataset significantly influences the counting precision of fluorescently stained bacterial cells. We present an approach based on an artificial neural network (ANN) to assess the quality of such images. Spatially invariant estimators were extracted as ANN input data from subdivided images by low level image processing. Different ANN designs were compared and >400 ANNs were trained and tested on a set of 25,000 manually classified images. The optimal ANN featured a correct identification rate of 94% (3% false positives, 3% false negatives) and could process about 10 images per second. We compared its performance with the image quality assessment by different humans and discuss the difficulties in assigning images to the correct quality class. The computer program and the documented source code (VB.NET) are provided under General Public Licence.

Artificial neural networks for fish-species identification

Abstract Cabreira, A. G., Tripode, M., and Madirolas, A. 2009. Artificial neural networks for fish-species identification. – ICES Journal of Marine Science, 66: 1119–1129. Acoustic fish detection is a valuable tool for the continuous monitoring of fish schools. However, changes in species composition or mixed multispecies situations still complicate the analysis of acoustic data. Validation of echo recordings is usually accomplished by trawling, but only at point locations. However, species proportions and size distributions in the catch can be biased because of gear selectivity and fish avoidance. In this paper, techniques involving training and testing of artificial neural networks (ANNs) are applied for the automatic recognition and classification of digital echo recordings of schools in the Southwest Atlantic. Energetic, morphometric, and bathymetric school descriptors were extracted from the echo-recordings as the input for the ANNs. Several pelagic and demersal fish species known to aggregate into schools were considered, including anchovy, rough scad, longtail hoki, sprat, and blue whiting. Different types of ANNs were tested. Best performances were obtained by levelling the input data (number of schools) per species. Correct classification rates up to 96% were obtained, depending on the species, type of network, and the number of school descriptors utilized. Some of these species inhabit areas geographically distant from each other. Hence, the contribution of the school position as a descriptor was investigated. By deleting the geographical location of the schools from the ANN input data, the average performance decreased to some extent but was still satisfactory, proving the networks were able usually to recognize fish species based only on the intrinsic characteristics of the school. The results have encouraged further testing of this method as a useful tool for scrutinizing echograms.

Evaluation of an optical phenolic biosensor signal employing artificial neural networks

Abstract This paper presents artificial neural network (ANN)-based evaluation in signal processing of an optical phenolic biosensor. The biosensor was developed based on stacked immobilization of 3-methyl-2-benzothiazolinone hydrazone (MBTH) in hybrid Nafion/sol–gel silicate and tyrosinase in chitosan. The biosensor signal was simulated employing a feed-forward neural network with three layers and trained using back-propagation (BP) algorithm. Spectra generated from an optical phenolic biosensor at selected wavelengths were used as input data for ANN. The network architecture of 5 inputs neurons, 21 hidden neurons and 1 output neuron was found suitable for this application. The results show very good agreement between phenol concentration values obtained by using the developed biosensor and those predicted by ANN.

Heuristic extraction of rules in pruned artificial neural networks models used for quantifying highly overlapping chromatographic peaks.

The suitability of an approach for extracting heuristic rules from trained artificial neural networks (ANNs) pruned by a regularization method and with architectures designed by evolutionary computation for quantifying highly overlapping chromatographic peaks is demonstrated. The ANN input data are estimated by the Levenberg-Marquardt method in the form of a four-parameter Weibull curve associated with the profile of the chromatographic band. To test this approach, two N-methylcarbamate pesticides, carbofuran and propoxur, were quantified using a classic peroxyoxalate chemiluminescence reaction as a detection system for chromatographic analysis. Straightforward network topologies (one and two outputs models) allow the analytes to be quantified in concentration ratios ranging from 1:7 to 5:1 with an average standard error of prediction for the generalization test of 2.7 and 2.3% for carbofuran and propoxur, respectively. The reduced dimensions of the selected ANN architectures, especially those obtained after using heuristic rules, allowed simple quantification equations to be developed that transform the input variables into output variables. These equations can be easily interpreted from a chemical point of view to attain quantitative analytical information regarding the effect of both analytes on the characteristics of chromatographic bands, namely profile, dispersion, peak height, and residence time.

Identification of input variables for feature based artificial neural networks-saccade detection in EOG recordings

Though artificial neural networks (ANN) are excellent tools for pattern recognition problems when signal to noise ratio is low, the identification of decision relevant features for ANN input data is still a crucial issue. The experience of the ANN designer and the existing knowledge and understanding of the problem seem to be the only links for a specific construction. In the present study a backpropagation ANN based on modified raw data inputs showed encouraging results. Investigating the specific influences of prototypical input patterns on a specially designed ANN led to a new sparse and efficient input data presentation. This data coding obtained by a semiautomatic procedure combining existing expert knowledge and the internal representation structures of the raw data based ANN yielded a list of feature vectors, each representing the relevant information for saccade identification. The feature based ANN produced a reduction of the error rate of nearly 40% compared with the raw data ANN. An overall correct classification of 92% of so far unknown data was realized. The proposed method of extracting internal ANN knowledge for the production of a better input data representation is not restricted to EOG recordings, and could be used in various fields of signal analysis.

Estimation of Parameters of Kinetic Compartmental Models by Use of Computational Neural Networks

A novel methodological approach to the estimation of parameters involved in multicomponent kinetic determinations from real kinetic data by use of computational or artificial neural networks (ANNs) is proposed. The ANN input data used are also estimates obtained by using the Levenberg−Marquardt method in the form of an approximate nonlinear function that is the sum of the two expressions associated with the pseudo-first-order kinetics of the two mixture components. The performance of the optimized network architecture, 2:4s:2l, was tested at variable rate constant ratios. The reduced dimensions of the network input space obtained using the Kolmogorov−Sprecher theorem result in improved limits of precision in estimating parameters at near-unity rate constant ratios. Experiments with real kinetic data provided a relative standard error of prediction of 2.47% and 4.23% for the two mixture components. These errors are much smaller than those obtained with existing alternative methods, particularly at the low ...

What can artificial neural networks teach us about neurodegenerative disorders with extrapyramidal features?

Artificial neural networks (ANNs), computer paradigms that can learn, excel in pattern recognition tasks such as disease diagnosis. Artificial neural networks operate in two different learning modes: supervised, in which a known diagnostic outcome is presented to the ANN, and unsupervised, in which the diagnostic outcome is not presented. A supervised learning ANN could emulate human expert diagnostic performance and identify relevant predictive markers in the diagnostic task, while an unsupervised learning ANN could suggest reasonable alternative diagnostic classification criteria. In the present study, we used ANN methodology to try to overcome the neuropathological difficulties in differentiating the subtypes of progressive supranuclear palsy (PSP), and in differentiating PSP from postencephalitic parkinsonism (PEP) and corticobasal degeneration, or Pick's disease from corticobasal degeneration. First, we applied supervised learning ANN to classify 62 cases of these disorders and to identify diagnostic markers that distinguish them. In a second experiment, we used unsupervised learning ANN to investigate possible alternative nosological classifications. Artificial neural networks input data for each case consisted of values representing histological features, including neurofibrillary tangles, neuronal loss and gliosis found in multiple brain sampling areas. The supervised learning ANN achieved excellent accuracy in classifying PSP but had difficulty classifying the other disorders. This method identified a few features that might help to differentiate PEP, supported currently proposed criteria for Pick's disease, corticobasal degeneration and typical PSP, but detected no features to characterize the atypical subtype of PSP. In general, unsupervised learning ANN supported the present nosological classification for PSP, PEP, Pick's disease and corticobasal degeneration, although it overlapped some groups. Artificial neural networks methodology appears promising for studying neurodegenerative disorders.

Reactor Handbook, volume I, materials

In the present paper, neutron spectrum is reconstructed using the Artificial Neural Network (ANN) and Modified Least Square (MLSQR) methods. The detector's response (pulse height distribution) as a required data for unfolding of energy spectrum is calculated using the developed MCNPX-ESUT computational code (MCNPX-Energy engineering of Sharif University of Technology). Unlike the usual methods that apply inversion procedures to unfold the energy spectrum from the Fredholm integral equation, the MLSQR method uses the direct procedure. Since liquid organic scintillators like NE-213 are well suited and routinely used for spectrometry of neutron sources, the neutron pulse height distribution is simulated/measured in the NE-213 detector. The response matrix is calculated using the MCNPX-ESUT computational code through the simulation of NE-213 detector's response to monoenergetic neutron sources. For known neutron pulse height distribution, the energy spectrum of the neutron source is unfolded using the MLSQR method. In the developed multilayer perception neural network for reconstruction of the energy spectrum of the neutron source, there is no need for formation of the response matrix. The multilayer perception neural network is developed based on logsig, tansig and purelin transfer functions. The developed artificial neural network consists of two hidden layers of type hyperbolic tangent sigmoid transfer function and a linear transfer function in the output layer. The motivation of applying the ANN method may be explained by the fact that no matrix inversion is needed for energy spectrum unfolding. The simulated neutron pulse height distributions in each light bin due to randomly generated neutron spectrum are considered as the input data of ANN. Also, the randomly generated energy spectra are considered as the output data of the ANN. Energy spectrum of the neutron source is identified with high accuracy using both MLSQR and ANN methods. The results obtained from MLSQR and ANN methods for 252Cf and 241Am-9Be source are validated against the ISO spectrum. The unfolded neutron energy spectra from both MLSQR and ANN methods show a good agreement with the actual spectrum of 252Cf and 241Am-9Be source.