SEARCH · PubMed Health

Results for “Neural Networks, Computer”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

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

At least 883 records · Page 49Linked to original sources

PubMed2007

Functional organization within a neural network trained to update target representations across 3-D saccades.

The goal of this study was to understand how neural networks solve the 3-D aspects of updating in the double-saccade task, where subjects make sequential saccades to the remembered locations of two targets. We trained a 3-layer, feed-forward neural network, using back-propagation, to calculate the 3-D motor error the second saccade. Network inputs were a 2-D topographic map of the direction of the second target in retinal coordinates, and 3-D vector representations of initial eye orientation and motor error of the first saccade in head-fixed coordinates. The network learned to account for all 3-D aspects of updating. Hidden-layer units (HLUs) showed retinal-coordinate visual receptive fields that were remapped across the first saccade. Two classes of HLUs emerged from the training, one class primarily implementing the linear aspects of updating using vector subtraction, the second class implementing the eye-orientation-dependent, non-linear aspects of updating. These mechanisms interacted at the unit level through gain-field-like input summations, and through the parallel "tweaking" of optimally-tuned HLU contributions to the output that shifted the overall population output vector to the correct second-saccade motor error. These observations may provide clues for the biological implementation of updating.

Animals↗

PubMed2005

Mapping the dimensionality, density and topology of data: the growing adaptive neural gas.

Self-organized maps are commonly applied for tasks of cluster analysis, vector quantization or interpolation. The artificial neural network model introduced in this paper is a hybrid model of the growing neural gas model introduced by Fritzke (Fritzke, in Advances in Neural Information Processing Systems 7, MIT Press, Cambridge MA, 1995) and the adaptive resolution clustering modification for self-organized maps proposed by Firenze (Firenze et al., in International Conference on Artificial Neural Networks, Springer-Verlag, London, 1994). The hybrid model is capable of mapping the distribution, dimensionality and topology of the input data. It has a local performance measure that enables the network to terminate growing in areas of the input space that is mapped by units reaching a performance goal. Therefore the network can accurately map clusters of data appearing on different scales of density. The capabilities of the algorithm are tested using simulated datasets with similar spatial spread but different local density distributions, and a simulated multivariate MR dataset of an anatomical human brain phantom with mild multiple sclerosis lesions. These tests demonstrate the advantages of the model compared to the growing neural gas algorithm when adaptive mapping of areas with low sample density is desirable.

Algorithms↗

PubMed1995

Fuzzy neural networks: between functional equivalence and applicability.

Research in fuzzy neural networks, which started from application oriented fuzzy system tuning, then moving to the automatic generation of fuzzy systems from data, is reaching a more mature stage, especially after the proof of functional equivalence of certain fuzzy models and neural networks. It is essential that the applicability of such developments is explored emphasizing the directions that research should follow. It can be shown that the nearest prototype classifier is functionally equivalent to an alternative fuzzy classifier model. Efficient, hardware friendly training algorithms are developed for dynamic generation of an optimum number of nearest prototypes for neural classifiers which enable the generation of fuzzy systems in real time. These systems are tested with complex applications showing the simulation results.

Bayes Theorem↗

PubMed2006

Solving quadratic programming problems by delayed projection neural network.

In this letter, the delayed projection neural network for solving convex quadratic programming problems is proposed. The neural network is proved to be globally exponentially stable and can converge to an optimal solution of the optimization problem. Three examples show the effectiveness of the proposed network.

Algorithms↗

PubMed1996

Optimizing synaptic conductance calculation for network simulations.

High computational requirements in realistic neuronal network simulations have led to attempts to realize implementation efficiencies while maintaining as much realism as possible. Since the number of synapses in a network will generally far exceed the number of neurons, simulation of synaptic activation may be a large proportion of total processing time. We present a consolidating algorithm based on a recent biophysically-inspired simplified Markov model of the synapse. Use of a single lumped state variable to represent a large number of converging synaptic inputs results in substantial speed-ups.

Algorithms↗

PubMed2005

Design and analysis of a general recurrent neural network model for time-varying matrix inversion.

Following the idea of using first-order time derivatives, this paper presents a general recurrent neural network (RNN) model for online inversion of time-varying matrices. Different kinds of activation functions are investigated to guarantee the global exponential convergence of the neural model to the exact inverse of a given time-varying matrix. The robustness of the proposed neural model is also studied with respect to different activation functions and various implementation errors. Simulation results, including the application to kinematic control of redundant manipulators, substantiate the theoretical analysis and demonstrate the efficacy of the neural model on time-varying matrix inversion, especially when using a power-sigmoid activation function.

Algorithms↗

PubMedDate not supplied

A new algorithm for learning in piecewise-linear neural networks.

Piecewise-linear (PWL) neural networks are widely known for their amenability to digital implementation. This paper presents a new algorithm for learning in PWL networks consisting of a single hidden layer. The approach adopted is based upon constructing a continuous PWL error function and developing an efficient algorithm to minimize it. The algorithm consists of two basic stages in searching the weight space. The first stage of the optimization algorithm is used to locate a point in the weight space representing the intersection of N linearly independent hyperplanes, with N being the number of weights in the network. The second stage is then called to use this point as a starting point in order to continue searching by moving along the single-dimension boundaries between the different linear regions of the error function, hopping from one point (representing the intersection of N hyperplanes) to another. The proposed algorithm exhibits significantly accelerated convergence, as compared to standard algorithms such as back-propagation and improved versions of it, such as the conjugate gradient algorithm. In addition, it has the distinct advantage that there are no parameters to adjust, and therefore there is no time-consuming parameters tuning step. The new algorithm is expected to find applications in function approximation, time series prediction and binary classification problems.

Algorithms↗

PubMed1995

Artificial neural network for diagnosis of acute pulmonary embolism: effect of case and observer selection.

PURPOSE: To compare the diagnostic performance of an artificial neural network (ANN) with that of physicians in patients with suspected pulmonary embolism (PE). MATERIALS AND METHODS: An ANN was developed to predict PE by using findings from ventilation-perfusion lung scans and chest radiographs. First, the network was evaluated on 1,064 cases from the Prospective Investigation of Pulmonary Embolism Diagnosis (PIOPED) study that had a definitive angiographic outcome. An upper and lower bound of its diagnostic performance was provided depending on case difficulty. Then, the network was tested on 104 patients with suspected PE in whom pulmonary angiography was essential for diagnosis. The diagnostic performance of the ANN was compared with that of (a) two nuclear medicine physicians who read the scans for the needs of this study and (b) the nuclear medicine physicians who originally read the scans. The effects of case and observer selection on performance were addressed. RESULTS: The ANN outperformed the physicians when they used the PIOPED criteria for categoric assessment, and it performed as well as the two study physicians on the basis of their probability assessments. CONCLUSION: The ANN can detect or exclude PE in a highly selected group of difficult cases with a consistency equivalent to that of very experienced physicians.

Algorithms↗

PubMed1998

Spatial and temporal pattern analysis via spiking neurons.

Spiking neurons, receiving temporally encoded inputs, can compute radial basis functions (RBFs) by storing the relevant information in their delays. In this paper we show how these delays can be learned using exclusively locally available information (basically the time difference between the pre- and postsynaptic spikes). Our approach gives rise to a biologically plausible algorithm for finding clusters in a high-dimensional input space with networks of spiking neurons, even if the environment is changing dynamically. Furthermore, we show that our learning mechanism makes it possible that such RBF neurons can perform some kind of feature extraction where they recognize that only certain input coordinates carry relevant information. Finally we demonstrate that this model allows the recognition of temporal sequences even if they are distorted in various ways.

Action Potentials↗

PubMed1996

Statistical approach to neural network model building for gentamicin peak predictions.

Feed forward neural networks are flexible, nonlinear modeling tools that are an extension of traditional statistical techniques. The hypothesis that feed forward neural network models can be built in a similar fashion as a statistical model was tested. Feed forward neural network models were built using forward and backward variable selection, and zero to five hidden nodes, and tanh and linear transfer functions were used. Gentamicin serum concentrations were predicted as a model drug for testing these methods. Peak observations from 392 patients were used to train, test, and validate the feed forward neural network. Inputs were demographic and drug dosing information. Model selection was performed using the Akaike information criteria (AIC), Bayesian information criteria (BIC), and a method of stopped training. The models with lowest root mean square (rms) error were those with all 10 inputs and five hidden nodes. Average rms error in the validation set was lowest for stopped training (1.46), then AIC (1.51), and finally BIC (1.56). Larger models tended to result in the best predictions. Overfitting can occur in models that are too large, either by using too many nodes in the hidden layer (rms = 1.49) or by using too many inputs with little information associated with them (rms = 1.70). We conclude that neural networks can be built using a large number of parameters that have good predictive performance. Care must be used during training to avoid overfitting the data. A stopped training method resulted in the network with the lowest rms error.

Adult↗

PubMed2006

Feature selection using a piecewise linear network.

We present an efficient feature selection algorithm for the general regression problem, which utilizes a piecewise linear orthonormal least squares (OLS) procedure. The algorithm 1) determines an appropriate piecewise linear network (PLN) model for the given data set, 2) applies the OLS procedure to the PLN model, and 3) searches for useful feature subsets using a floating search algorithm. The floating search prevents the "nesting effect." The proposed algorithm is computationally very efficient because only one data pass is required. Several examples are given to demonstrate the effectiveness of the proposed algorithm.

Algorithms↗

PubMed1998

Learning viewpoint-invariant face representations from visual experience in an attractor network.

In natural visual experience, different views of an object or face tend to appear in close temporal proximity as an animal manipulates the object or navigates around it, or as a face changes expression or pose. A set of simulations is presented which demonstrate how viewpoint-invariant representations of faces can be developed from visual experience by capturing the temporal relationships among the input patterns. The simulations explored the interaction of temporal smoothing of activity signals with Hebbian learning in both a feedforward layer and a second, recurrent layer of a network. The feedforward connections were trained by competitive Hebbian learning with temporal smoothing of the post-synaptic unit activities. The recurrent layer was a generalization of a Hopfield network with a low-pass temporal filter on all unit activities. The combination of basic Hebbian learning with temporal smoothing of unit activities produced an attractor network learning rule that associated temporally proximal input patterns into basins of attraction. These two mechanisms were demonstrated in a model that took grey-level images of faces as input. Following training on image sequences of faces as they changed pose, multiple views of a given face fell into the same basin of attraction, and the system acquired representations of faces that were approximately viewpoint-invariant.

Animals↗

PubMed2002

Noise and coupling affect signal detection and bursting in a simulated physiological neural network.

Signal detection in the CNS relies on a complex interaction between the numerous synaptic inputs to the detecting cells. Two effects, stochastic resonance (SR) and coherence resonance (CR) have been shown to affect signal detection in arrays of basic neuronal models. Here, an array of simulated hippocampal CA1 neurons was used to test the hypothesis that physiological noise and electrical coupling can interact to modulate signal detection in the CA1 region of the hippocampus. The array was tested using varying levels of coupling and noise with different input signals. Detection of a subthreshold signal in the network improved as the number of detecting cells increased and as coupling was increased as predicted by previous studies in SR; however, the response depended greatly on the noise characteristics present and varied from SR predictions at times. Careful evaluation of noise characteristics may be necessary to form conclusions about the role of SR in complex systems such as physiological neurons. The coupled array fired synchronous, periodic bursts when presented with noise alone. The synchrony of this firing changed as a function of noise and coupling as predicted by CR. The firing was very similar to certain models of epileptiform activity, leading to a discussion of CR as a possible simple model of epilepsy. A single neuron was unable to recruit its neighbors to a periodic signal unless the signal was very close to the synchronous bursting frequency. These findings, when viewed in comparison with physiological parameters in the hippocampus, suggest that both SR and CR can have significant effects on signal processing in vivo.

Algorithms↗

PubMed1997

Optimality: from neural networks to universal grammar.

Can concepts from the theory of neural computation contribute to formal theories of the mind? Recent research has explored the implications of one principle of neural computation, optimization, for the theory of grammar. Optimization over symbolic linguistic structures provides the core of a new grammatical architecture, optimality theory. The proposition that grammaticality equals optimality sheds light on a wide range of phenomena, from the gulf between production and comprehension in child language, to language learnability, to the fundamental questions of linguistic theory: What is it that the grammars of all languages share, and how may they differ?

Adult↗

PubMed2002

Prediction of tumoricidal activity and accumulation of photosensitizers in photodynamic therapy using multiple linear regression and artificial neural networks.

The biological activities of a congeneric series of pyropheophorbides used as sensitizers in photodynamic therapy have been predicted on the basis of their molecular structures, using multiple linear regression and artificial neural network (ANN) computations. Theoretical descriptors (a total of 81) were calculated by the 3DNET program based on the three-dimensional structure (3D) of the geometry-optimized molecules. These input descriptors were tested as independent variables and used for model building. Systematic descriptor selections yielded models with one, two or three descriptors with good cross-validation results. The predictive abilities of the best fitting models were checked by shuffling and cross-validation procedures. ANN was suitable for building models for both linear and nonlinear relationships. Lipophilicity was sufficient to predict the accumulation of the sensitizers in the target tissue. Weighted holistic invariant molecular descriptors weighted by atomic mass, Van der Waals volume or electronegativity were also needed to predict photodynamic activity properly. Our models were able to predict the biological activities of 13 pyropheophorbide derivatives solely on the basis of their 3D molecular structures. Moreover, linear and nonlinear variable selection methods were compared in models built linearly and nonlinearly. It is expedient to use the same method (linear or nonlinear) for variable selection as for parameter estimation.

Antineoplastic Agents↗

PubMed1997

Recognition of temporally changing action potentials in multiunit neural recordings.

We present a method to iteratively train an artificial neural network (ANN) or other supervised pattern classifier in order to adaptively recognize and track temporally changing patterns. This method uses recently acquired data and the existing classifier to create new training sets, from which a new classifier is then trained. The procedure is repeated periodically using the most recently trained classifier. This scheme was evaluated by applying it to simulated situations that arise in chronic recordings of multiunit neural activity from peripheral nerves. The method was able to track the changes in these simulated chronic recordings and to provide better unit recognition rates than an unsupervised clustering method suited to this problem.

Action Potentials↗

PubMed2005

Coarse-graining and self-dissimilarity of complex networks.

Can complex engineered and biological networks be coarse-grained into smaller and more understandable versions in which each node represents an entire pattern in the original network? To address this, we define coarse-graining units as connectivity patterns which can serve as the nodes of a coarse-grained network and present algorithms to detect them. We use this approach to systematically reverse-engineer electronic circuits, forming understandable high-level maps from incomprehensible transistor wiring: first, a coarse-grained version in which each node is a gate made of several transistors is established. Then the coarse-grained network is itself coarse-grained, resulting in a high-level blueprint in which each node is a circuit module made of many gates. We apply our approach also to a mammalian protein signal-transduction network, to find a simplified coarse-grained network with three main signaling channels that resemble multi-layered perceptrons made of cross-interacting MAP-kinase cascades. We find that both biological and electronic networks are "self-dissimilar," with different network motifs at each level. The present approach may be used to simplify a variety of directed and nondirected, natural and designed networks.

Algorithms↗

PubMed2005

Motif-based protein sequence classification using neural networks.

We present a system for multi-class protein classification based on neural networks. The basic issue concerning the construction of neural network systems for protein classification is the sequence encoding scheme that must be used in order to feed the neural network. To deal with this problem we propose a method that maps a protein sequence into a numerical feature space using the matching scores of the sequence to groups of conserved patterns (called motifs) into protein families. We consider two alternative ways for identifying the motifs to be used for feature generation and provide a comparative evaluation of the two schemes. We also evaluate the impact of the incorporation of background features (2-grams) on the performance of the neural system. Experimental results on real datasets indicate that the proposed method is highly efficient and is superior to other well-known methods for protein classification.

Algorithms↗