PubMed Health⌕ Search

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 1,639 records · Page 91Linked to original sources
PubMed2006

Intrinsic bursters increase the robustness of rhythm generation in an excitatory network.

The pre-Botzinger complex (pBC) is a vital subcircuit of the respiratory central pattern generator. Although the existence of neurons with pacemaker-like bursting properties in this network is not questioned, their role in network rhythmogenesis is unresolved. Modeling is ideally suited to address this debate because of the ease with which biophysical parameters of individual cells and network architecture can be manipulated. We modeled the parameter variability of experimental data from pBC bursting pacemaker and nonpacemaker neurons using a modified version of our previously developed pBC neuron and network models. To investigate the role of pacemakers in networkwide rhythmogenesis, we simulated networks of these neurons and varied the fraction of the population made up of pacemakers. For each number of pacemaker neurons, we varied the amount of tonic drive to the network and measured the frequency of synchronous networkwide bursting produced. Both excitatory networks with all-to-all coupling and sparsely connected networks were explored for several levels of synaptic coupling strength. Networks containing only nonpacemakers were able to produce networkwide bursting, but with a low probability of bursting and low input and output ranges. Our results indicate that inclusion of pacemakers in an excitatory network increases robustness of the network by more than tripling the input and output ranges compared with networks containing no pacemakers. The largest increase in dynamic range occurs when the number of pacemakers in the network is greater than 20% of the population. Experimental tests of our model predictions are proposed.

Afferent Pathways↗
PubMed2006

Propagation and control of stochastic signals through universal learning networks.

The way of propagating and control of stochastic signals through Universal Learning Networks (ULNs) and its applications are proposed. ULNs have been already developed to form a superset of neural networks and have been applied as a universal framework for modeling and control of non-linear large-scale complex systems. However, the ULNs cannot deal with stochastic variables. Deterministic signals can be propagated through a ULN, but the ULN does not provide any stochastic characteristics of the signals propagating through it. The proposed method named Probabilistic Universal Learning Networks (PrULNs) can process stochastic variables and can train network parameters so that the signals behave with the pre-specified stochastic properties. As examples of applications of the proposed method, control and identification of non-linear dynamic systems with noises are studied, and it is shown that the method are useful for dealing with the control and identification of the non-linear stochastic systems contaminated with noises.

Artificial Intelligence↗
PubMed2003

A guide to building image-centric databases.

There is a paucity of image-centric neuroinformatics infrastructure within the individual investigator's laboratory despite the obvious need for automation and integration of experimental results. Yet, solutions can often be readily built using off-the-shelf databases and associated tools. Doing so simplifies day-to-day research operation and increases throughput. Proper construction of in-house solutions may also expedite community-wide integration of private and public data repositories. Here we describe neuroinformatics approaches at different levels of functionality, required expertise, and size of image datasets. The simplest approach offers ease of image browsing and rudimentary searching. More sophisticated systems provide powerful search capabilities, a means of tracking analysis, and even automated serial processing pipelines. In this practicum, we provide guidance in selecting among the different options.

Animals↗
PubMed2004

GenSo-EWS: a novel neural-fuzzy based early warning system for predicting bank failures.

Bank failure prediction is an important issue for the regulators of the banking industries. The collapse and failure of a bank could trigger an adverse financial repercussion and generate negative impacts such as a massive bail out cost for the failing bank and loss of confidence from the investors and depositors. Very often, bank failures are due to financial distress. Hence, it is desirable to have an early warning system (EWS) that identifies potential bank failure or high-risk banks through the traits of financial distress. Various traditional statistical models have been employed to study bank failures [J Finance 1 (1975) 21; J Banking Finance 1 (1977) 249; J Banking Finance 10 (1986) 511; J Banking Finance 19 (1995) 1073]. However, these models do not have the capability to identify the characteristics of financial distress and thus function as black boxes. This paper proposes the use of a new neural fuzzy system [Foundations of neuro-fuzzy systems, 1997], namely the Generic Self-organising Fuzzy Neural Network (GenSoFNN) [IEEE Trans Neural Networks 13 (2002c) 1075] based on the compositional rule of inference (CRI) [Commun ACM 37 (1975) 77], as an alternative to predict banking failure. The CRI based GenSoFNN neural fuzzy network, henceforth denoted as GenSoFNN-CRI(S), functions as an EWS and is able to identify the inherent traits of financial distress based on financial covariates (features) derived from publicly available financial statements. The interaction between the selected features is captured in the form of highly intuitive IF-THEN fuzzy rules. Such easily comprehensible rules provide insights into the possible characteristics of financial distress and form the knowledge base for a highly desired EWS that aids bank regulation. The performance of the GenSoFNN-CRI(S) network is subsequently benchmarked against that of the Cox's proportional hazards model [J Banking Finance 10 (1986) 511; J Banking Finance 19 (1995) 1073], the multi-layered perceptron (MLP) and the modified cerebellar model articulation controller (MCMAC) [IEEE Trans Syst Man Cybern: Part B 30 (2000) 491] in predicting bank failures based on a population of 3635 US banks observed over a 21 years period. Three sets of experiments are performed-bank failure classification based on the last available financial record and prediction using financial records one and two years prior to the last available financial statements. The performance of the GenSoFNN-CRI(S) network as a bank failure classification and EWS is encouraging.

Accidents↗
PubMed1993

AIDS2: a decision-support tool for decreasing physicians' uncertainty regarding patient eligibility for HIV treatment protocols.

We have developed a decision-support tool, the AIDS Intervention Decision-Support System (AIDS2), to assist in the task of matching patients to therapy-related research protocols. The purposes of AIDS2 are to determine the initial eligibility status of HIV-infected patients for therapy-related research protocols, and to suggest additional data-gathering activities that will decrease uncertainty related to the eligibility status. AIDS2 operates in either a patient-driven or protocol-driven mode. We represent the system knowledge in three combined levels: a classification level, where deterministic knowledge is represented; a belief-network level, where probabilistic knowledge is represented; and a control level, where knowledge about the system's operation is stored. To determine whether the design specifications were met, we presented a series of 10 clinical cases based on actual patients to the system. AIDS2 provided meaningful advice in all cases.

Acquired Immunodeficiency Syndrome↗
PubMed2004

SINBAD: a neocortical mechanism for discovering environmental variables and regularities hidden in sensory input.

We propose that a top priority of the cerebral cortex must be the discovery and explicit representation of the environmental variables that contribute as major factors to environmental regularities. Any neural representation in which such variables are represented only implicitly (thus requiring extra computing to use them) will make the regularities more complex and therefore more difficult, if not impossible, to learn. The task of discovering such important environmental variables is not an easy one since their existence is only indirectly suggested by the sensory input patterns the cortex receives--these variables are "hidden". We present a candidate computational strategy for (i) discovering regularity-simplifying environmental variables, (ii) learning the regularities, and (iii) using regularities in perceptual and decision-making tasks. The SINBAD computational model discovers useful environmental variables through a search for different, but nevertheless highly correlated, functions of any kind over nonoverlapping subsets of the known variables, this being indicative of some important environmental variable that is responsible for the correlation. We suggest that such a search is performed in the neocortex by the dendritic trees of individual pyramidal cells. According to the SINBAD model, the basic function of each pyramidal cell is to (i) discover and represent one of the regularity-simplifying environmental variables and (ii) learn to infer the state of its variable from the states of other variables, represented by other pyramidal cells. A network of such cells--each cell just attending to representation of its variable--can function as a sophisticated and useful inferential model of the outside world.

Association Learning↗
PubMed2005

Data classification with radial basis function networks based on a novel kernel density estimation algorithm.

This paper presents a novel learning algorithm for efficient construction of the radial basis function (RBF) networks that can deliver the same level of accuracy as the support vector machines (SVMs) in data classification applications. The proposed learning algorithm works by constructing one RBF subnetwork to approximate the probability density function of each class of objects in the training data set. With respect to algorithm design, the main distinction of the proposed learning algorithm is the novel kernel density estimation algorithm that features an average time complexity of O(n log n), where n is the number of samples in the training data set. One important advantage of the proposed learning algorithm, in comparison with the SVM, is that the proposed learning algorithm generally takes far less time to construct a data classifier with an optimized parameter setting. This feature is of significance for many contemporary applications, in particular, for those applications in which new objects are continuously added into an already large database. Another desirable feature of the proposed learning algorithm is that the RBF networks constructed are capable of carrying out data classification with more than two classes of objects in one single run. In other words, unlike with the SVM, there is no need to resort to mechanisms such as one-against-one or one-against-all for handling datasets with more than two classes of objects. The comparison with SVM is of particular interest, because it has been shown in a number of recent studies that SVM generally are able to deliver higher classification accuracy than the other existing data classification algorithms. As the proposed learning algorithm is instance-based, the data reduction issue is also addressed in this paper. One interesting observation in this regard is that, for all three data sets used in data reduction experiments, the number of training samples remaining after a naive data reduction mechanism is applied is quite close to the number of support vectors identified by the SVM software. This paper also compares the performance of the RBF networks constructed with the proposed learning algorithm and those constructed with a conventional cluster-based learning algorithm. The most interesting observation learned is that, with respect to data classification, the distributions of training samples near the boundaries between different classes of objects carry more crucial information than the distributions of samples in the inner parts of the clusters.

Algorithms↗
PubMed2005

Decision feedback recurrent neural equalization with fast convergence rate.

Real-time recurrent learning (RTRL), commonly employed for training a fully connected recurrent neural network (RNN), has a drawback of slow convergence rate. In the light of this deficiency, a decision feedback recurrent neural equalizer (DFRNE) using the RTRL requires long training sequences to achieve good performance. In this paper, extended Kalman filter (EKF) algorithms based on the RTRL for the DFRNE are presented in state-space formulation of the system, in particular for complex-valued signal processing. The main features of global EKF and decoupled EKF algorithms are fast convergence and good tracking performance. Through nonlinear channel equalization, performance of the DFRNE with the EKF algorithms is evaluated and compared with that of the DFRNE with the RTRL.

Algorithms↗
PubMed2000

Cerebellar learning of accurate predictive control for fast-reaching movements.

Long conduction delays in the nervous system prevent the accurate control of movements by feedback control alone. We present a new, biologically plausible cerebellar model to study how fast arm movements can be executed in spite of these delays. To provide a realistic test-bed of the cerebellar neural model, we embed the cerebellar network in a simulated biological motor system comprising a spinal cord model and a six-muscle two-dimensional arm model. We argue that if the trajectory errors are detected at the spinal cord level, memory traces in the cerebellum can solve the temporal mismatch problem between efferent motor commands and delayed error signals. Moreover, learning is made stable by the inclusion of the cerebello-nucleo-olivary loop in the model. It is shown that the cerebellar network implements a nonlinear predictive regulator by learning part of the inverse dynamics of the plant and spinal circuit. After learning, fast accurate reaching movements can be generated.

Arm↗
PubMed1999

Improved performance in protein secondary structure prediction by inhomogeneous score combination.

MOTIVATION: In many fields of pattern recognition, combination has proved efficient to increase the generalization performance of individual prediction methods. Numerous systems have been developed for protein secondary structure prediction, based on different principles. Finding better ensemble methods for this task may thus become crucial. Furthermore, efforts need to be made to help the biologist in the post-processing of the outputs. RESULTS: An ensemble method has been designed to post-process the outputs of discriminant models, in order to obtain an improvement in prediction accuracy while generating class posterior probability estimates. Experimental results establish that it can increase the recognition rate of protein secondary structure prediction methods that provide inhomogeneous scores, even though their individual prediction successes are largely different. This combination thus constitutes a help for the biologist, who can use it confidently on top of any set of prediction methods. Moreover, the resulting estimates can be used in various ways, for instance to determine which areas in the sequence are predicted with a given level of reliability. AVAILABILITY: The prediction is freely available over the Internet on the Network Protein Sequence Analysis (NPS@) WWW server at http://pbil.ibcp.fr/NPSA/npsa_server.ht ml. The source code of the combiner can be obtained on request for academic use.

Amino Acid Sequence↗
PubMed2006

A hybrid neuro-fuzzy filter for edge preserving restoration of images corrupted by impulse noise.

A new operator for restoring digital images corrupted by impulse noise is presented. The proposed operator is a hybrid filter obtained by appropriately combining a median filter, an edge detector, and a neuro-fuzzy network. The internal parameters of the neuro-fuzzy network are adaptively optimized by training. The training is easily accomplished by using simple artificial images that can be generated in a computer. The most distinctive feature of the proposed operator over most other operators is that it offers excellent line, edge, detail, and texture preservation performance while, at the same time, effectively removing noise from the input image. Extensive simulation experiments show that the proposed operator may be used for efficient restoration of digital images corrupted by impulse noise without distorting the useful information in the image.

Algorithms↗
PubMed2000

Individual and developmental differences in semantic priming: empirical and computational support for a single-mechanism account of lexical processing.

Existing accounts of single-word semantic priming phenomena incorporate multiple mechanisms, such as spreading activation, expectancy-based processes, and postlexical semantic matching. The authors provide empirical and computational support for a single-mechanism distributed network account. Previous studies have found greater semantic priming for low- than for high-frequency target words as well as inhibition following unrelated primes only at long stimulus-onset asynchronies (SOAs). A series of experiments examined the modulation of these effects by individual differences in age or perceptual ability. Third-grade, 6th-grade, and college students performed a lexical-decision task on high- and low-frequency target words preceded by related, unrelated, and nonword primes. Greater priming for low-frequency targets was exhibited only by participants with high perceptual ability. Moreover, unlike the college students, the children showed no inhibition even at the long SOA. The authors provide an account of these results in terms of the properties of distributed network models and support this account with an explicit computational simulation.

Adult↗
PubMed2004

A discrete approach for a model of temporal learning by the cerebellum: in silico classical conditioning of the eyeblink reflex.

Cerebellar cortex is known to be involved in acquisition and expression of eyeblink conditioned reflex. These phenomena imply temporal intervals of learning. Several cellular and network mechanisms have been proposed to produce the eyeblink. In this paper we briefly review the main theories concerning temporal coding, and we propose an alternative way of producing and storing delays and signal sequences after supervised learning. A network of Leaky Integrate-and-Fire (LIF) neurons is built, taking into account several cerebellar features. This network is then trained to produce simple or multiple eyeblink delays using (i) the classical conditioning paradigm and (ii) known data on cerebellar spike timing dependent plasticity (STDP). The resulting model behaves like an adaptive temporal filter. It improves cell subpopulations effects according to their mean firing rate. This rate based selection allows robust supervised learning of temporal events (i.e., delayed signals) and gives the network ability to react with anticipation on the arousal of a noxious event.

Action Potentials↗
PubMed1997

Symmetric GroEL-GroES complexes can contain substrate simultaneously in both GroEL rings.

Incubation of rhodanese with hche aperonins GroEL and GroES (1:2 GroEL14:GroES7 molar ratio) under functional and steady state conditions for ATP leads to the formation of a high proportion of rhodanese-bound symmetric complexes (GroEL14(GroES7)2), as revealed by native electrophoresis. Aliquots of such samples were observed under the electron microscope, and the symmetric particles were classified using neuronal networks and multivariate statistical analysis. Three different populations of symmetric particles were obtained which contained substrate in none, one or both GroEL cavities, respectively. The presence of substrate in the symmetric complexes under functional conditions supports their role as active intermediates in the protein folding cycle. These results also suggest that symmetric GroEL-GroES complexes can use both rings simultaneously for folding, probably increasing the efficiency of the reaction.

Adenosine Triphosphate↗
PubMed2004

Retinally-generated saccadic suppression of a locust looming-detector neuron: investigations using a robot locust.

A fundamental task performed by many visual systems is to distinguish apparent motion caused by eye movements from real motion occurring within the environment. During saccadic eye movements, this task is achieved by inhibitory signals of central and retinal origin that suppress the output of motion-detecting neurons. To investigate the retinally-generated component of this suppression, we used a computational model of a locust looming-detecting pathway that experiences saccadic suppression. This model received input from the camera of a mobile robot that performed simple saccade-like movements, allowing the model's response to simplified real stimuli to be tested. Retinally-generated saccadic suppression resulted from two inhibitory mechanisms within the looming-detector's input architecture. One mechanism fed inhibition forward through the network, inhibiting the looming-detector's initial response to movement. The second spread inhibition laterally within the network, suppressing the looming-detector's maintained response to movement. These mechanisms prevent a looming-detector model response to whole-field visual stimuli. In the locust, this mechanism of saccadic suppression may operate in addition to centrally-generated suppression. Because lateral inhibition is a common feature of early visual processing in many organisms, we discuss whether the mechanism of retinally-generated saccadic suppression found in the locust looming-detector model may also operate in these species.

Animals↗
PubMed2003

Parametric stability evaluation in computer experiments on the mathematical model of Drosophila control gene subnetwork.

Using the method of generalized threshold models, the problem is formulated and solved to evaluate the parametric stability of the model of a gene subnetwork controlling the early ontogenesis of the fruit fly Drosophila melanogaster. Computer experiments have been performed to test the parametric stability of the model. Quantitative evaluations have been obtained for parametric stability of the Drosophila gene subnetwork in nuclei along the embryo's anterior-posterior axis. The results of computer experiments have been compared with the previous research data on "sensitivity" of functioning regimes to random changes of the parameters in the models of prokaryotic and eukaryotic systems, namely the system controlling the lambda-phage development and the subsystem controlling the flower morphogenesis of Arabidopsis thaliana. The obtained results confirm high parametric stability of gene networks that control the development of organisms.

Animals↗
PubMed1995

The assessment of LH surge for predicting ovulation time using clinical, hormonal, and ultrasonic indices in infertile women with an ensemble of neural networks.

An ensemble of independently trained neural networks (NN) is proposed for the assessment of luteinizing hormone (LH) surge for predicting ovulation time in infertile but ovulating women. The proposed ensemble involves a number of parallel NN modules. Each pair of the NNs learn specific data that are previously collected for monitoring timing function of LH estradiol (pg ml-1), and follicle diameter (mm) are used to train NN pairs to approximate the function of the LH values. A reasonable and accurate estimation places ovulation approximately 10-12 h after the LH peak. The double-valued (bi-phasic) regions of training data are separated into two single-valued (bi-phasic) regions of training data are separated into two single-valued parts (not exactly preovulatory, postovulatory division) that can be learned by each module of the NN pair. During testing, after the initial decision to have single-valued sides, the assessment is obtained by a linear opinion pool (consensus rule) using the decisions of NNs on the corresponding side without waiting. The network ensemble has various desirable properties: high assessment accuracy of a double-valued multisource data, minimized learning and recall times, and a parallel structure. The ovulation time can be predicted through the assessment of LH peak with a better precision and fewer number of tests.

Diagnosis, Computer-Assisted↗