Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Neural Networks, Computer”

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 163 records · Page 9Linked to original sources
PubMed2001

Cortical spreading depression and the pathogenesis of brain disorders: a computational and neural network-based investigation.

This paper reviews our recent studies of the role of cortical spreading depression (CSD) in the pathogenesis of brain disorders. Our investigation is a computational one, involving the development and utilization of a complex neuro-metabolic model of the interactions assumed to occur in the cortex during the passage of multiple CSD waves. Incorporating these neuro-metabolic changes of CSD within a neural network model of normoxic cortex produces cortical activation patterns during the passage of a CSD wave that, projected onto the visual fields, resemble the visual hallucinations observed during the migraine aura. When focal ischemia is simulated with the model, the evoked CSD waves are found to affect the expansion of the infarction into the ischemic penumbra. Our findings support the hypothesis that CSD does play an important pathogenic role in these and other neurological disorders, and suggest additional experimental studies that may further substantiate it.

Animals↗
PubMed1999

Computational differences between asymmetrical and symmetrical networks.

Symmetrically connected recurrent networks have recently been used as models of a host of neural computations. However, biological neural networks have asymmetrical connections, at the very least because of the separation between excitatory and inhibitory neurons in the brain. We study characteristic differences between asymmetrical networks and their symmetrical counterparts in cases for which they act as selective amplifiers for particular classes of input patterns. We show that the dramatically different dynamical behaviours to which they have access, often make the asymmetrical networks computationally superior. We illustrate our results in networks that selectively amplify oriented bars and smooth contours in visual inputs.

Animals↗
PubMed1991

[Neuronal nets].

Neural networks are used as models of cognitive systems. On the basis of results from brain research, the processing of knowledge and information can be conceived as occurring through parallel interaction of multiple but simple and uniform processor elements in a network structure. In these 'neural networks' knowledge is stored in a distributed way throughout the network and subjected to parallel processing. According to Hebb's concept of synaptic plasticity, the process of learning is based on the strengthening of the links between the processing elements. The implementation of network models in electronic data processing systems allows for simulation of cognitive phenomena such as learning and forgetting or abilities such as recognition of acoustical or optical patterns, in a more efficient way than conventional computer programming. Neural network modelling can help to understand the abilities and disorders of the brain.

Cognition↗
PubMed2004

Integration of clinical and imaging data to predict the presence of coronary artery disease with the use of neural networks.

AIM: To establish proof of the principle that a computer-based neural network method can be employed that will enhance diagnostic accuracy vis-a-vis image analysis alone in the interpretation of treadmill exercise tests performed in conjunction with myocardial perfusion imaging. MATERIALS AND METHODS: One-hundred-and-two patients underwent myocardial perfusion imaging in association with the standard Bruce protocol. Twenty objective parameters describing each patient's exercise physiology, general clinical status and image appearance were used to train an artificial neural network. Classification accuracy of the neural network and clinical interpretation was determined by coronary angiography. We evaluated the ability of the neural network to integrate clinical, exercise and imaging data to determine the likelihood of coronary artery disease and compared these results with an optimized method of clinical image interpretation, which made use of all available clinical, angiographic and stress test data. RESULTS: The artificial neural network had a sensitivity of 88% and a specificity of 65% for detection of ischemic heart disease and was comparable to that of the optimized clinical method (sensitivity 80%, specificity 69%). Incorporation of clinical and exercise data significantly improved the predictive accuracy of the network compared to a network based on image data alone (P<0.05). CONCLUSION: The results show a computer-based neural network can perform as well as expert readers working under optimal conditions including full knowledge of the patient's clinical, prior angiographic and stress test data. Thus, the method is promising as a diagnostic aid to the recognition of ischemic heart disease in the clinical setting of treadmill exercise testing in conjunction with myocardial perfusion imaging.

Aged↗
PubMed2004

Quantitative structure-metabolism relationship modeling of metabolic N-dealkylation reaction rates.

It is widely recognized that preclinical drug discovery can be improved via the parallel assessment of bioactivity, absorption, distribution, metabolism, excretion, and toxicity properties of molecules. High-throughput computational methods may enable such assessment at the earliest, least expensive discovery stages, such as during screening compound libraries and the hit-to-lead process. As an attempt to predict drug metabolism and toxicity, we have developed an approach for evaluation of the rate of N-dealkylation mediated by two of the most important human cytochrome P450s (P450), namely CYP3A4 and CYP2D6. We have taken a novel approach by using descriptors generated for the whole molecule, the reaction centroid, and the leaving group, and then applying neural network computations and sensitivity analysis to generate quantitative structure-metabolism relationship models. The quality of these models was assessed by using the cross-validated correlation coefficients of 0.82 for CYP3A4 and 0.79 for CYP2D6 as well as external test molecules for each enzyme. The relative performance of different neural networks was also compared, and modular neural networks with two hidden layers provided the best predictive ability. Functional dependencies between the neural network input and output variables, generalization ability, and limitations of the described approach are also discussed. These models represent an initial approach to predicting the rate of P450-mediated metabolism and may be applied and integrated with other models for P450 binding to produce a systems-based approach for predicting drug metabolism.

Cytochrome P-450 CYP2D6↗
PubMed1994

Neural network for control of rearrangeable Clos networks.

Rapid evolution in the field of communication networks requires high speed switching technologies. This involves a high degree of parallelism in switching control and routing performed at the hardware level. The multistage crossbar networks have always been attractive to switch designers. In this paper a neural network approach to controlling a three-stage Clos network in real time is proposed. This controller provides optimal routing of communication traffic requests on a call-by-call basis by rearranging existing connections, with a minimum length of rearrangement sequence so that a new blocked call request can be accommodated. The proposed neural network controller uses Paull's rearrangement algorithm, along with the special (least used) switch selection rule in order to minimize the length of rearrangement sequences. The functional behavior of our model is verified by simulations and it is shown that the convergence time required for finding an optimal solution is constant, regardless of the switching network size. The performance is evaluated for random traffic with various traffic loads. Simulation results show that applying the least used switch selection rule increases the efficiency in switch rearrangements, reducing the network convergence time. The implementation aspects are also discussed to show the feasibility of the proposed approach.

Algorithms↗
PubMed2001

Learning Hough transform: a neural network model.

A single-layered Hough transform network is proposed that accepts image coordinates of each object pixel as input and produces a set of outputs that indicate the belongingness of the pixel to a particular structure (e.g., a straight line). The network is able to learn adaptively the parametric forms of the linear segments present in the image. It is designed for learning and identification not only of linear segments in two-dimensional images but also the planes and hyperplanes in the higher-dimensional spaces. It provides an efficient representation of visual information embedded in the connection weights. The network not only reduces the large space requirement, as in the case of classical Hough transform, but also represents the parameters with high precision.

Algorithms↗
PubMedDate not supplied

Searching for risk factors using multilayer neural network as a classifier.

A method to determine risk factors for particular outcomes using trained multilayer neural networks is proposed. The basic idea is to measure the partial differentials of the output with respect to input variables of the network. Differentiable activation functions and continuity of input variables is assumed.

Algorithms↗
PubMed1993

Solving Bernstein's problem: a proposal for the development of coordinated movement by selection.

In recent years, many established concepts in the theory of human motor development have undergone profound change, and our knowledge has increased greatly. Nevertheless, some outstanding problems remain unsolved. A central problem concerns the redundancy of effective movements, first pointed out by N. A. Bernstein. The human motor system is mechanically complex and can make use of a large number of degrees of freedom. The controlled operation of such a system requires a reduction of mechanical redundancy, effectively by reducing the number of degrees of freedom. More recent work has shown that this problem is hard to solve explicitly by computing solutions to the equations of motion of the system. Equally challenging to traditional computational approaches is the fact the motor systems show remarkable adaptability and flexibility in the presence of changing biomechanical properties of motor organs during development and when faced with different environmental conditions or tasks. Solutions to these problems would have a large impact on a variety of issues in child development. In this article, we stress the importance of the somatic selection of neuronal groups in maps for the progressive transformation of a primary movement repertoire into a set of motor synergies and adaptive action patterns. We present results from computer simulations of a simple motor system that works according to such selectional principles. This approach suggests a provisional solution to Bernstein's problem and provides new parameters to guide experimental approaches to the development of sensorimotor coordination.

Biomechanical Phenomena↗
PubMed2000

The forward EEG solutions can be computed using artificial neural networks.

Study of electroencenphalography (EEG) is the one of the most utilized methods in both basic brain research and clinical diagnosis of neurological disorders. Recent technological advances in computer and electronic systems have allowed the EEG to be recorded from large electrode arrays. Modeling the brain waves using a head volume conductor model provides an effective method to localize functional generators within the brain. However, the forward solutions to this model, which represent theoretical potentials in response to current sources within the volume conductor, are difficult to compute because of time-consuming numerical procedures utilized in either the boundary element method (BEM) or the finite element method (FEM). This paper presents a novel computational approach using an artificial neural network (ANN) to map two vectors of forward solutions. These two vectors correspond to different head models but with respect to the same current source. The input vector to the ANN is based on the spherical head model, which can be computed efficiently but involves large errors. The output vector from the ANN is based on the spheroidal model, which is more precise, but difficult to compute directly using the traditional means. Our experiments indicate that this ANN approach provides a remarkable improvement over the BEM and FEM methods: 1) the mean-square error of computation was only approximately 0.3% compared to the exact solution; 2) the online computation was extremely efficient, requiring only 168 floating point operations per channel to compute the forward solution, and 10.2 K-bytes of storage to represent the entire ANN. Using this approach it is possible to perform real-time EEG modeling accurately on personal computers.

Biomedical Engineering↗
PubMed2004

Dynamics and the single spike.

Responses of vestibular primary afferent neurons to head rotation exhibit fractional-order dynamics. As a consequence, the head tends to be in a localized region of its state-space at spike times of a particular neuron during arbitrary head movements, and single spikes can be interpreted as state measurements. We are developing a model of neural computations underlying trajectory prediction and control tasks, based on this experimental observation. This is a step toward a formal neural calculus in which single spikes are modeled realistically as the operands of neural computation.

Action Potentials↗
PubMed2004

The effects of multi-task learning and time-varying hemispheric asymmetry on lateralisation in a neural network model.

While various functional and cognitive capabilities appear to differ in both degree and direction of lateralisation, the factors underlying these differences are poorly understood. It is hypothesised that time-varying asymmetry in plasticity between homologous regions in the cerebral hemispheres, coupled with asynchronous development of capabilities, may account for the lateralisation differences in two ways. First, the lateralisation of an earlier acquired behaviour may influence the lateralisation of a later developing behaviour. Second, temporal changes in the underlying plasticity asymmetry may also result in differences in lateralisation for functions acquired at different times. This study examines the plausibility of these hypotheses using a computational neural network model consisting of two interacting hemispheric regions and capable of learning two tasks. Lateralisation was measured while learning rates for the two hemispheres were changed independently over time to create a time-varying asymmetry in plasticity, and while the initiation of learning for each task was also varied over time. The results suggest that the lateralisation of one behaviour/function can affect the lateralisation of another in a fundamental way, and that experimentally observed temporal differences in hemispheric development can result in functional lateralisation differences, providing support for past theories of lateralisation based on asymmetric hemispheric growth and maturation.

Cerebral Cortex↗
PubMed2004

Real-time computation at the edge of chaos in recurrent neural networks.

Depending on the connectivity, recurrent networks of simple computational units can show very different types of dynamics, ranging from totally ordered to chaotic. We analyze how the type of dynamics (ordered or chaotic) exhibited by randomly connected networks of threshold gates driven by a time-varying input signal depends on the parameters describing the distribution of the connectivity matrix. In particular, we calculate the critical boundary in parameter space where the transition from ordered to chaotic dynamics takes place. Employing a recently developed framework for analyzing real-time computations, we show that only near the critical boundary can such networks perform complex computations on time series. Hence, this result strongly supports conjectures that dynamical systems that are capable of doing complex computational tasks should operate near the edge of chaos, that is, the transition from ordered to chaotic dynamics.

Animals↗
PubMed1993

PROBE: a computer program employing an integrated neural network approach to protein structure prediction.

A computer program, PROBE, has been designed for the prediction of protein structural features from amino acid sequence. This program integrates a variety of computer-simulated neural networks, each predicting an aspect of protein structure, into a single, easy-to-use package. The surface accessibility of each residue, the presence of disulfide bonds, the overall secondary structure composition and the residue secondary structures, including beta-turn type, are predicted. In addition, the overall amino acid composition and relative hydrophobicity are used to determine whether a protein belongs to one of four common folding motifs. PROBE is able to compare and synergistically improve the predictions by allowing communication between the different networks.

Amino Acid Sequence↗
PubMed1987

Computation, rhythm, and cerebral slow waves.

The trajectories of data and modelling from formerly divergent research efforts now seem to be converging to an unexpected region of the phase space of neuroscience. Computational network theory and simulation assume that temporal rhythm may be a significant parameter for the successful organization of nonlinear analog computation effected by hierarchical sets of biological neurons and of nonbiological circuitry alike. For neurobiology, the apparently chaotic rhythms of cerebral compound field potentials--the electroencephalogram (EEG) and slow waves of event related brain potentials (ERBP)--have long been a phenomenological embarrassment, of only marginal clinical utility. But recent data from molecular biophysics, nonlinear dynamics, artificial intelligence, and scalp-conducted human electrocorticography suggest a possible functional role in the serial gating of neural network computations for the familiar theta-alpha-beta rhythms of the EEG clinic.

Auditory Perception↗
PubMed1996

Neural network simulation of visual inspection of graphs of single-subject interventions.

Judgments of the effectiveness of single-subject behavioral interventions are often based on visual examination of graphs of response data, but previous research indicates that such judgments are often unreliable and flawed. Here it is proposed that artificial neural networks could be developed to simulate the judgments of expert judges. A prototype of such a network was designed and trained in the present study, and its use in novel experiments matched the estimates of the expert whose judgments were simulated significantly better than did a prediction equation developed using a multiple regression approach.

Artificial Intelligence↗
PubMed2003

Prediction of drug metabolism: the case of cytochrome P450 2D6.

Cytochromes P450 (Cyt P450s) constitute the most important biotransformation enzymes involved in the biotransformation of drugs and other xenobiotics. Because drug metabolism by Cyt P450s plays such an important role in the disposition and in the pharmacological and toxicological effects of drugs, early consideration of ADME-properties is increasingly seen as essential for the discovery and the development of new drugs and drug candidates. The primary aim of this paper is to present various computational approaches used to rationalize and predict the activity and substrate selectivity of Cyt P450s, as well as the possibilities and limitations of these approaches, now and in the future. Attention is also paid to the experimental validation of these approaches by using high-throughput screening (HTS) of affinities to drug-drug interactions at the level of Cyt P450-isoenzymes. Since human Cyt P450 2D6 is one of the most important drug metabolizing enzymes and since in this regard much pioneering work has been done with this Cyt P450, Cyt P450 2D6 is chosen as a model for this discussion. Apart from early mechanism-based ab initio calculations on substrates of Cyt P450 2D6, pharmacophore modeling of ligands (i.e. both substrates and inhibitors) of Cyt P450 2D6 and protein homology modeling have been used successfully for the rationalisation and prediction of metabolite formation by this Cyt P450 isoenzyme. Significant protein structure-related species differences have been reported recently. It is concluded that not one computational approach is capable of rationalizing and reliably predicting metabolite formation by Cyt P450 2D6, but that it is rather the combination of the various complimentary approaches. It is moreover concluded, that experimental validation of the computational models and predictions is often still lacking. With the advent of novel, easily and well applicable in vitro based high throughput assays for ligand binding and turnover this limitation could be overcome soon, however. When effective links with other new and recent developments, such as bioinformatics, neural network computing, genomics and proteomics can be created, in silico rationalisation and prediction of drug metabolism by Cyt P450s is likely to become one of the key technologies in early drug discovery and development processes.

Animals↗
PubMed1996

Pathogenesis of schizophrenic delusions and hallucinations: a neural model.

We implement and study a computational model of Stevens' theory of the pathogenesis of schizophrenia. This theory hypothesizes that the onset of schizophrenia is associated with reactive synaptic regeneration in brain regions that receive degenerating temporal lobe projections. Concentrating on one such area, the frontal cortex, we model a frontal module as an associative memory neural network whose input synapses represent incoming temporal projections. Modeling Stevens' hypothesized pathological synaptic changes in this framework results in adverse side effects similar to hallucinations and delusions seen in schizophrenia: spontaneous, stimulus-independent retrieval of stored memories focused on just a few of the stored patterns. These could account for the delusions and hallucinations that occur in schizophrenia without any apparent external trigger and for their tendency to concentrate on a few central cognitive and perceptual themes. The model explains why the positive symptoms of schizophrenia tend to wane as the disease progresses, why delayed therapeutic intervention leads to a much slower response, and why delusions and hallucinations may persist for a long time when they do occur.

Brain↗