Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Neural network”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 37 records · Page 2Linked to original sources
PubMed1994

Analysis of protein transmembrane helical regions by a neural network.

Neural networks were used to generalize common themes found in transmembrane-spanning protein helices. Various-sized databases were used containing nonoverlapping sequences, each 25 amino acids long. Training consisted of sorting these sequences into 1 of 2 groups: transmembrane helical peptides or nontransmembrane peptides. Learning was measured using a test set 10% the size of the training set. As training set size increased from 214 sequences to 1,751 sequences, learning increased in a nonlinear manner from 75% to a high of 98%, then declined to a low of 87%. The final training database consisted of roughly equal numbers of transmembrane (928) and nontransmembrane (1,018) sequences. All transmembrane sequences were entered into the database with respect to their lipid membrane orientation: from inside the membrane to outside. Generalized transmembrane helix and nontransmembrane peptides were constructed from the maximally weighted connecting strengths of fully trained networks. Four generalized transmembrane helices were found to contain 9 consensus residues: a K-R-F triplet was found at the inside lipid interface, 2 isoleucine and 2 other phenylalanine residues were present in the helical body, and 2 tryptophan residues were found near the outside lipid interface. As a test of the training method, bacteriorhodopsin was examined to determine the position of its 7 transmembrane helices.

Amino Acid Sequence↗
PubMed2002

Fast computation with spikes in a recurrent neural network.

Neural networks with recurrent connections are sometimes regarded as too slow at computation to serve as models of the brain. Here we analytically study a counterexample, a network consisting of N integrate-and-fire neurons with self excitation, all-to-all inhibition, instantaneous synaptic coupling, and constant external driving inputs. When the inhibition and/or excitation are large enough, the network performs a winner-take-all computation for all possible external inputs and initial states of the network. The computation is done very quickly: As soon as the winner spikes once, the computation is completed since no other neurons will spike. For some initial states, the winner is the first neuron to spike, and the computation is done at the first spike of the network. In general, there are M potential winners, corresponding to the top M external inputs. When the external inputs are close in magnitude, M tends to be larger. If M>1, the selection of the actual winner is strongly influenced by the initial states. If a special relation between the excitation and inhibition is satisfied, the network always selects the neuron with the maximum external input as the winner.

Journal Article↗
PubMed2000

Mechanisms of Modulation of a Neural Network.

Neural networks form the basis for the generation and control of various patterns of behavior. Such networks are subjected to modulatory systems that influence their operation and, thereby, the behavior. In the lamprey locomotor network, analysis on the ion channel, synaptic, and cellular levels has given new insights into the organization of such modulatory systems.

Journal Article↗
PubMed2002

An intelligent sales forecasting system through integration of artificial neural networks and fuzzy neural networks with fuzzy weight elimination.

Sales forecasting plays a very prominent role in business strategy. Numerous investigations addressing this problem have generally employed statistical methods, such as regression or autoregressive and moving average (ARMA). However, sales forecasting is very complicated owing to influence by internal and external environments. Recently, artificial neural networks (ANNs) have also been applied in sales forecasting since their promising performances in the areas of control and pattern recognition. However, further improvement is still necessary since unique circumstances, e.g. promotion, cause a sudden change in the sales pattern. Thus, this study utilizes a proposed fuzzy neural network (FNN), which is able to eliminate the unimportant weights, for the sake of learning fuzzy IF-THEN rules obtained from the marketing experts with respect to promotion. The result from FNN is further integrated with the time series data through an ANN. Both the simulated and real-world problem results show that FNN with weight elimination can have lower training error compared with the regular FNN. Besides, real-world problem results also indicate that the proposed estimation system outperforms the conventional statistical method and single ANN in accuracy.

Algorithms↗
PubMedDate not supplied

QSAR/QSPR studies using probabilistic neural networks and generalized regression neural networks.

The Probabilistic Neural Network (PNN) and its close relative, the Generalized Regression Neural Network (GRNN), are presented as simple yet powerful neural network techniques for use in Quantitative Structure-Activity Relationship (QSAR) and Quantitative Structure-Property Relationship (QSPR) studies. The PNN methodology is applicable to classification problems, and the GRNN is applicable to continuous function mapping problems. The basic underlying theory behind these probability-based methods is presented along with two applications of the PNN/GRNN methodology. The PNN model presented identifies molecules as potential soluble epoxide hydrolase inhibitors using a binary classification scheme. The GRNN model presented predicts the aqueous solubility of nitrogen- and oxygen-containing small organic molecules. For each application, the network inputs consist of a small set of descriptors that encode structural features at the molecular level. Each of these studies has also been previously addressed in this research group using more traditional techniques such as k-nearest neighbor classification, multiple linear regression, and multilayer feed-forward neural networks. In each case, the predictive power of the PNN and GRNN models was found to be comparable to that of the more traditional techniques but requiring significantly fewer input descriptors.

Bayes Theorem↗
PubMed2004

Classification capacity of a modular neural network implementing neurally inspired architecture and training rules.

A three-layer neural network (NN) with novel adaptive architecture has been developed. The hidden layer of the network consists of slabs of single neuron models, where neurons within a slab--but not between slabs--have the same type of activation function. The network activation functions in all three layers have adaptable parameters. The network was trained using a biologically inspired, guided-annealing learning rule on a variety of medical data. Good training/testing classification performance was obtained on all data sets tested. The performance achieved was comparable to that of SVM classifiers. It was shown that the adaptive network architecture, inspired from the modular organization often encountered in the mammalian cerebral cortex, can benefit classification performance.

Animals↗
PubMedDate not supplied

Neural network studies. 4. Introduction to associative neural networks.

Associative neural network (ASNN) represents a combination of an ensemble of feed-forward neural networks and the k-nearest neighbor technique. This method uses the correlation between ensemble responses as a measure of distance amid the analyzed cases for the nearest neighbor technique. This provides an improved prediction by the bias correction of the neural network ensemble. An associative neural network has a memory that can coincide with the training set. If new data becomes available, the network further improves its predictive ability and provides a reasonable approximation of the unknown function without a need to retrain the neural network ensemble. This feature of the method dramatically improves its predictive ability over traditional neural networks and k-nearest neighbor techniques, as demonstrated using several artificial data sets and a program to predict lipophilicity of chemical compounds. Another important feature of ASNN is the possibility to interpret neural network results by analysis of correlations between data cases in the space of models. It is shown that analysis of such correlations makes it possible to provide "property-targeted" clustering of data. The possible applications and importance of ASNN in drug design and medicinal and combinatorial chemistry are discussed. The method is available on-line at http://www.vcclab.org/lab/asnn.

Neural Networks, Computer↗
PubMed1993

Lattice neural network minimization. Application of neural network optimization for locating the global-minimum conformations of proteins.

A way of formulating the protein-folding problem in neural network optimization terms is presented in this paper. This is accomplished by representing the conformation of a protein as an array of the amino acid sequence versus position on a three-dimensional face-centered cubic lattice with an energy function defined in terms of the array variables. The method is called lattice neural network minimization (LNNM). Using the neural network minimization method, the energy function is minimized to locate the global minimum energy for the conformation of the protein. The energy function consisted of site exclusion and bond connectivity penalty terms and a pairwise contact energy potential. The contact energy potential used in the procedure is the united-residue potential of Miyazawa, Jernigan and Covell. The LNNM method found the global minimum for a seven-residue peptide in all of the 15 runs carried out. The time for each run was approximately 30 seconds on one processor of an IBM 3090 computer. For a nine-residue peptide, the global minimum was found in 7 out of 15 runs (47%) in approximately 50 seconds per run. For this peptide, LNNM found the global minimum or the second lowest minimum in 10 of the runs. In the same total CPU times (approximately 750 seconds), a Monte Carlo simulated annealing method found the global minimum or the second lowest minimum in only two runs, demonstrating the superiority of LNNM over the standard Monte Carlo simulated annealing method for this nine-residue peptide. Starting from a uniform array for the protein crambin (46 residues) on the lattice, the energy of the crambin array was minimized and a compact low-energy structure was found in approximately 25 minutes of CPU time. Its energy was much lower than that of the native protein, suggesting that there are inadequacies in the Miyazawa-Jernigan-Covell potential. The LNNM method was applied to the prediction of what was previously called nucleation but more properly called chain-folding initiation sites (CFIS) of a protein. LNNM correctly predicted the CFIS for the two proteins examined, RNase S and T4 lysozyme. The LNNM method was also applied to another chain optimization problem, minimization of the root-mean-square distance error (r.m.s.d.) (a measure similar to r.m.s. deviation) in fitting X-ray structures to a lattice, with good results.

Models, Chemical↗
PubMed2006

Quantitative structure-toxicity relationships (QSTRs): a comparative study of various non linear methods. General regression neural network, radial basis function neural network and support vector machine in predicting toxicity of nitro- and cyano- aromatics to Tetrahymena pyriformis.

Prediction of toxicity of 203 nitro- and cyano-aromatic chemicals to Tetrahymena pyriformis was carried out by radial basis function neural network, general regression neural network and support vector machine, in non-linear response surface methodology. Toxicity was predicted from hydrophobicity parameter (log Kow) and maximum superdelocalizability (Amax). Special attention was drawn to prediction ability and robustness of the models, investigated both in a leave-one-out and 10-fold cross validation (CV) processes. The influence that the corresponding changes in the learning sets during these CV processes could have on a common external test set including 41 compounds was also examined. This allowed us to establish the stability of the models. The non linear results slightly outperform (as expected) multilinear relationships (MLR) and also favourably compete with various other non linear approaches recently proposed by Ren (J. Chem. Inf. Comput. Sci., 43 1679 (2003)).

Animals↗
PubMed2006

Online prediction of onsets of seizure-like events in hippocampal neural networks using wavelet artificial neural networks.

It has been previously shown that wavelet artificial neural networks (WANNs) are able to classify the different states of epileptiform activity and predict the onsets of seizure-like events (SLEs) by offline processing (Ann. Biomed. Eng. 33(6):798-810, 2005) of the electrical data from the in-vitro hippocampal slice model of recurrent spontaneous SLEs. The WANN design entailed the assumption that time-varying frequency information from the biological recordings can be used to estimate the times at which onsets of SLEs would most likely occur in the future. Progressions of different frequency components were captured by the artificial neural network (ANN) using selective frequency inputs from the initial wavelet transform of the biological data. The training of the WANN had been established using 184 SLE episodes in 34 slices from 21 rats offline. Nine of these rats also exhibited periods of interictal bursts (IBs). These IBs were included as part of the training to help distinguish the difference in dynamics of bursting activities between the preictal- and interictal type. In this paper, we present the results of an online processing using WANN on 23 in-vitro rat hippocampal slices from 9 rats having 93 spontaneous SLE episodes generated under low magnesium conditions. Over the test cases, three of the nine rats exhibited over 30 min of IB activities. We demonstrated that the WANN was able to classify the different states, namely, interictal, preictal, ictal, and IB activities with an accuracy of 86.6, 72.6, 84.5, and 69.1%, respectively. Prediction of state transitions into ictal events was achieved using regression of initial "normalized time-to-onset" estimates. The SLE onsets can be estimated up to 36.4 s ahead of their actual occurrences, with a mean error of 14.3 +/- 27.0 s. The prediction errors decreased progressively as the actual time-to-onset decreased and more initial "normalized time-to-onset" estimates were used for the regression procedure.

Action Potentials↗
PubMed2002

Prediction models in the design of neural network based ECG classifiers: a neural network and genetic programming approach.

BACKGROUND: Classification of the electrocardiogram using Neural Networks has become a widely used method in recent years. The efficiency of these classifiers depends upon a number of factors including network training. Unfortunately, there is a shortage of evidence available to enable specific design choices to be made and as a consequence, many designs are made on the basis of trial and error. In this study we develop prediction models to indicate the point at which training should stop for Neural Network based Electrocardiogram classifiers in order to ensure maximum generalisation. METHODS: Two prediction models have been presented; one based on Neural Networks and the other on Genetic Programming. The inputs to the models were 5 variable training parameters and the output indicated the point at which training should stop. Training and testing of the models was based on the results from 44 previously developed bi-group Neural Network classifiers, discriminating between Anterior Myocardial Infarction and normal patients. RESULTS: Our results show that both approaches provide close fits to the training data; p = 0.627 and p = 0.304 for the Neural Network and Genetic Programming methods respectively. For unseen data, the Neural Network exhibited no significant differences between actual and predicted outputs (p = 0.306) while the Genetic Programming method showed a marginally significant difference (p = 0.047). CONCLUSIONS: The approaches provide reverse engineering solutions to the development of Neural Network based Electrocardiogram classifiers. That is given the network design and architecture, an indication can be given as to when training should stop to obtain maximum network generalisation.

Electrocardiography↗
PubMed1996

Multiple neural network response variability as a predictor of neural network accuracy for chromosome recognition.

Human chromosome classification requires all chromosome appearing in a microphotograph of a dividing human cell to be classified within the known normal or abnormal 24 chromosome types. In recent years, research has focused on the use of neural networks for classification of normal chromosomes. Experimental work in this area led us to question whether learning variability, resulting when multiple neural networks are trained to solve the same problem, could be used as a predictor of classification performance. The Copenhagen chromosome data bank, consisting of 30-component feature vectors from 8106 chromosomes isolated from 180 cells, was divided into a training and a test subsets. Back propagation neural networks with 30 input nodes, 1 to 100 nodes in the hidden layer, and 24 output nodes were trained with the same learning parameters. After training, each neural network was tested. The neural network yielding the best classification was labeled as the optimal neural network. An error variability score was calculated for each test chromosome. This score was a function of all (100) neural network outputs obtained for that chromosome. The error variability scores ranged from 0.16 to 1.31 with a mean value of 0.41 and a SD of 0.12. There was significant difference (p < 0.0001) between the variability scores from chromosomes classified correctly (mean = 0.4, SD = 0.1, n = 3804) and incorrectly (mean = 0.62, SD = 0.19, n = 241) by the optimal neural network. When the variability score was used as a threshold to decide whether or not to accept the output of the optimal neural network, a peak classification rate of 98.93% was observed for chromosomes with an error variability score < 0.35. Results indicate that the error variability of multiple neural network responses can be used as a confidence indicator for a optimal neural network.

Humans↗
PubMed2005

On the relationship between deterministic and probabilistic directed Graphical models: from Bayesian networks to recursive neural networks.

Machine learning methods that can handle variable-size structured data such as sequences and graphs include Bayesian networks (BNs) and Recursive Neural Networks (RNNs). In both classes of models, the data is modeled using a set of observed and hidden variables associated with the nodes of a directed acyclic graph. In BNs, the conditional relationships between parent and child variables are probabilistic, whereas in RNNs they are deterministic and parameterized by neural networks. Here, we study the formal relationship between both classes of models and show that when the source nodes variables are observed, RNNs can be viewed as limits, both in distribution and probability, of BNs with local conditional distributions that have vanishing covariance matrices and converge to delta functions. Conditions for uniform convergence are also given together with an analysis of the behavior and exactness of Belief Propagation (BP) in 'deterministic' BNs. Implications for the design of mixed architectures and the corresponding inference algorithms are briefly discussed.

Bayes Theorem↗
PubMed2002

Prediction of lupus nephritis in patients with systemic lupus erythematosus using artificial neural networks.

Artificial neural networks are intelligent systems that have been successfully used for prediction in different medical fields. In this study, efficiency of neural networks for prediction of lupus nephritis in patients with systemic lupus erythematosus (SLE) was compared with a logistic regression model and clinicians' diagnosis. Overall accuracy, sensitivity and specificity of the optimal neural network were 68.69, 73.77 and 62.96%, respectively. Overall accuracy of neural network was greater than the other two methods (P-value < 0.05). The neural network was more specific in predicting lupus nephritis (P-value < 0.01), but there was no significant difference between sensitivities of the three methods. Sensitivities of all three methods were greater than their specificities. We concluded that neural networks are efficient in predicting lupus nephritis in SLE patients.

Algorithms↗
PubMed1994

Identification of low frequency patterns in backpropagation neural networks.

Although neural networks have been widely applied to medical problems in recent years, their applicability has been limited for a variety of reasons. One of these barriers has been the inability to discriminate rare classes of solutions (i.e., the identification of categories that are infrequent). In this article, I demonstrate that a system of hierarchical neural networks (HNN) can overcome the problem of recognizing low frequency patterns, and therefore can improve the prediction power of neural-network systems. HNN are designed according to a divide-and-conquer approach: Triage networks are able to discriminate supersets that contain the infrequent pattern, and these supersets are then used by Specialized networks, which discriminate the infrequent pattern from the other ones in the superset. The supersets that are discriminated by the Triage networks are based on pattern similarity. The application of multilayered neural networks in more than one step allows the prior probability of a given pattern to increase at each step, provided that the predictive power of the network at the previous level is high. The method has been applied to one artificial set and one real set of data. In the artificial set, the distribution of the patterns was known and no noise was present. In this experiment, the HNN provided better discrimination than a standard neural network for all classes. In a real data set of nine thousand patients who were suspected of having thyroid disorders, the HNN also provided higher sensitivity than its corresponding standard neural network (without a corresponding decay in specificity) given the same time constraints.(ABSTRACT TRUNCATED AT 250 WORDS)

Algorithms↗
PubMed2003

Neutron spectra unfolding in Bonner spheres spectrometry using neural networks.

The neural network method has been used for the unfolding of neutron spectra in neutron spectrometry by Bonner spheres. A back propagation algorithm was used for training of neural networks. 4 mm x 4 mm bare LiI (Eu) and in a polyethylene sphere set: 2, 3, 4, 5, 6, 7, 8, 10, 12, 18 inch diameter have been used for unfolding of neutron spectra. Neural networks were trained by 199 sets of neutron spectra, which were subdivided into 6, 8, 10, 12, 15 and 20 energy bins and for each of them an appropriate neural network was designed and trained. The validation was performed by the 21 sets of neutron spectra. A neural network with 10 energy bins which had a mean value of error of 6% for dose equivalent estimation of spectra in the validation set showed the best results. The obtained results show that neural networks can be applied as an effective method for unfolding neutron spectra especially when the main target is neutron dosimetry.

Algorithms↗
PubMed1997

Diagnosis of chronic liver disease from liver scintiscans by artificial neural networks.

Artificial neural networks were used in the diagnosis of chronic liver disease based on liver scintiscanning. One hundred and thirty-seven patients with chronic liver disease (12 with chronic persistent hepatitis, 39 with chronic aggressive hepatitis, and 86 with cirrhosis) and 25 healthy controls were studied. Sixty-five subjects (10 healthy controls, 20 patients with chronic hepatitis, and 35 patients with cirrhosis of the liver) were used in the establishment of a neural network. Liver scintiscans were taken starting 20 min after the intravenous injection of 111 MBq of Tc-99m-phytate. The neural network was used to evaluate five items judged from information on liver scintiscans: the ratio of the sizes of the left and right lobes, splenomegaly, radioactivity in the bone marrow, deformity of the liver and distribution of radioactivity in the liver. The neural network was designed to distinguish between three liver conditions (healthy liver, chronic hepatitis and cirrhosis) on the basis of these five items. The diagnostic accuracy with the neural network was 86% for patients with chronic hepatitis and 93% for patients with cirrhosis. With conventional scoring, the accuracy was 77% for patients with chronic hepatitis and 87% for patients with cirrhosis. Our findings suggest that artificial neural networks may be useful for the diagnosis of chronic liver diseases from liver scintiscans.

Case-Control Studies↗
PubMed1995

Locally activated neural networks and stable neural controller design for nonlinear dynamic systems.

A stable neural control scheme using a locally activated neural network has been proposed for a class of nonlinear dynamic systems. The locally activated neural network, for a given input, essentially selects a small subset of the network hidden nodes for output computation using the CMAC-like content addressing mechanism. This network aims to maintain local representations of the system dynamics. Thus, the global control performance in the concerned state space is achieved by the cooperation of many local control efforts and furthermore, real-time control can be facilitated because only a small sized network is involved to control and learn at any given time. The proposed control scheme is composed of two stages: (1) prediction error based learning in which the network attempts to learn the nonlinear basis functions of the plant inverse dynamics by a modified backpropagation learning rule; and (2) tracking error based learning in which the network weights are further fine-tuned using the basis set obtained in (1). This basis set spans the locally partitioned vector space of the system inverse dynamics when the prediction error based learning is achieved within a prescribed error tolerance. For uniform stability, the sliding mode control is introduced as a safety mechanism when the network has not sufficiently learned the plant dynamics yet. With suitable assumptions on the controlled plant, global stability and tracking error convergence proof has been given. Finally, the proposed control scheme is verified with computer simulation.

Adaptation, Physiological↗