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 1,261 records · Page 70Linked to original sources

PubMed2000

A controlled trial of three referral methods for patients with third molars.

AIM: To evaluate the accuracy, sensitivity and specificity of three primary to secondary care referral strategies. METHOD: Thirty two primary care dental practitioners (GDPs) were randomly allocated one of three referral strategies: current practice (control strategy); a neural network embedded within a computer program and a paper-based clinical algorithm. One hundred and seven patients were assessed for lower third molar treatment: 47, 30 and 30 in each group, respectively. Clinical details were assessed by a panel of experts against a gold standard for third molar removal (the National Institutes of Health criteria). The accuracy, sensitivity, specificity, positive and negative predictive values were calculated for each strategy. RESULTS: The referral decisions made by the GDPs in the control group displayed greater accuracy and sensitivity but poorer specificity (0.83; 0.97; 0.22) compared with the neural network (0.67; 0.56; 0.79) and clinical algorithm (0.73; 0.56; 0.93). CONCLUSIONS: It was concluded that incorporation of the clinical algorithm into primary care was the most appropriate option. The computer neural network performed less well than either current practice or the clinical algorithm.

Adolescent↗

PubMed2004

Feedforward sigmoidal networks--equicontinuity and fault-tolerance properties.

Sigmoidal feedforward artificial neural networks (FFANNs) have been established to be universal approximators of continuous functions. The universal approximation results are summarized to identify the function sets represented by the sigmoidal FFANNs with the universal approximation properties. The equicontinuous properties of the identified sets is analyzed. The equicontinuous property is related to the fault tolerance of the sigmoidal FFANNs. The generally used arbitrary weight sigmoidal FFANNs are shown to be nonequicontinuous sets. A class of bounded weight sigmoidal FFANNs is established to be equicontinuous. The fault-tolerance behavior of the networks is analyzed and error bounds for the induced errors established.

Algorithms↗

PubMedDate not supplied

Neural network simulations, cortical connectivity, and schizophrenic psychosis.

Recent studies have suggested that reduced corticocortical connectivity is associated with schizophrenia. My colleagues and I have used neural network simulations to explore parallel, distributed processing systems with reduced connectivity. These systems often behaved in a "schizophrenic-like" manner. Excessively pruned attractor networks became functionally fragmented, suggesting "loose associations," and produced recurrent, intrusive representations suggestive of delusions. Pruning backpropagation simulations of speech perception networks produced spontaneous output, which provided a model of auditory hallucination or "voices." This model also suggested how dopamine-blocking drugs may reduce positive symptoms, and why negative symptoms arise in the wake of positive symptoms.

Bias↗

PubMed1997

Artificial evolution: a new path for artificial intelligence?

Recently there have been a number of proposals for the use of artificial evolution as a radically new approach to the development of control systems for autonomous robots. This paper explains the artificial evolution approach, using work at Sussex to illustrate it. The paper revolves around a case study on the concurrent evolution of control networks and visual sensor morphologies for a mobile robot. Wider intellectual issues surrounding the work are discussed, as is the use of more abstract evolutionary simulations as a new potentially useful tool in theoretical biology.

Artificial Intelligence↗

PubMed2004

Classification of mammographic masses using generalized dynamic fuzzy neural networks.

In this article, computer-aided classification of mammographic masses using generalized dynamic fuzzy neural networks (GDFNN) is presented. The texture parameters, derived from first-order gradient distribution and gray-level co-occurrence matrices, were computed from the regions of interest. A total of 343 images containing 180 benign masses and 163 malignant masses from the Digital Database for Screening Mammography were analyzed. A fast approach of automatically generating fuzzy rules from training samples was implemented to classify tumors. This work is novel in that it alleviates the problem of requiring a designer to examine all the input-output relationships of a training database in order to obtain the most appropriate structure for the classifier in a conventional computer-aided diagnosis. In this approach, not only the connection weights can be adjusted, but also the structure can be self-adaptive during the learning process. By virtue of the automatic generation of the classifier by the GDFNN learning algorithm, the area under the receiver-operating characteristic curve, Az, attains 0.868 +/- 0.020, which corresponds to a true-positive fraction of 95.0% at a false positive fraction of 52.8%. The corresponding accuracy is 70.0%, the positive predictive value is 62.0%, and the negative predictive value is 91.4%.

Adult↗

PubMed1995

On the equivalence of two-layered perceptrons with binary neurons.

We consider two-layered perceptions consisting of N binary input units, K binary hidden units and one binary output unit, in the limit N >> K > or = 1. We prove that the weights of a regular irreducible network are uniquely determined by its input-output map up to some obvious global symmetries. A network is regular if its K weight vectors from the input layer to the K hidden units are linearly independent. A (single layered) perceptron is said to be irreducible if its output depends on every one of its input units; and a two-layered perceptron is irreducible if the K + 1 perceptrons that constitute such network are irreducible. By global symmetries we mean, for instance, permuting the labels of the hidden units. Hence, two irreducible regular two-layered perceptrons that implement the same Boolean function must have the same number of hidden units, and must be composed of equivalent perceptrons.

Computer Simulation↗

PubMed2004

A hybrid neural network model for consciousness.

A new framework for consciousness is introduced based upon traditional artificial neural network models. This framework reflects explicit connections between two parts of the brain: one global working memory and distributed modular cerebral networks relating to specific brain functions. Accordingly this framework is composed of three layers, physical mnemonic layer and abstract thinking layer, which cooperate together through a recognition layer to accomplish information storage and cognition using algorithms of how these interactions contribute to consciousness: (1) the reception process whereby cerebral subsystems group distributed signals into coherent object patterns; (2) the partial recognition process whereby patterns from particular subsystems are compared or stored as knowledge; and (3) the resonant learning process whereby global workspace stably adjusts its structure to adapt to patterns' changes. Using this framework, various sorts of human actions can be explained, leading to a general approach for analyzing brain functions.

Algorithms↗

PubMed1996

Concept hierarchy memory model: a neural architecture for conceptual knowledge representation, learning, and commonsense reasoning.

This article introduces a neural network based cognitive architecture termed Concept Hierarchy Memory Model (CHMM) for conceptual knowledge representation and commonsense reasoning. CHMM is composed of two subnetworks: a Concept Formation Network (CFN), that acquires concepts based on their sensory representations; and a Concept Hierarchy Network (CHN), that encodes hierarchical relationships between concepts. Based on Adaptive Resonance Associative Map (ARAM), a supervised Adaptive Resonance Theory (ART) model, CHMM provides a systematic treatment for concept formation and organization of a concept hierarchy. Specifically, a concept can be learned by sampling activities across multiple sensory fields. By chunking relations between concepts as cognitive codes, a concept hierarchy can be learned/modified through experience. Also, fuzzy relations between concepts can now be represented in terms of the weights on the links connecting them. Using a unified inferencing mechanism based on code firing, CHMM performs an important class of commonsense reasoning, including concept recognition and property inheritance.

Algorithms↗

PubMed2002

A bi-recursive neural network architecture for the prediction of protein coarse contact maps.

Prediction of contact maps may be seen as a strategic step towards the solution of fundamental open problems in structural genomics. In this paper we focus on coarse grained maps that describe the spatial neighborhood relation between secondary structure elements (helices, strands, and coils) of a protein. We introduce a new machine learning approach for scoring candidate contact maps. The method combines a specialized noncausal recursive connectionist architecture and a heuristic graph search algorithm. The network is trained using candidate graphs generated during search. We show how the process of selecting and generating training examples is important for tuning the precision of the predictor.

Algorithms↗

PubMed1993

Data compression of the exercise ECG using a Kohonen neural network.

The goal for any data compression scheme is to maximize compression while minimizing distortion. This is particularly true for measurement sensitive electrocardiographic data. Many different approaches have been taken to achieve this goal. One common technique used in image and speech data compression, vector quantization, was selected for this study. Central to vector quantization is the creation of a codebook of vectors. Creating the best possible codebook will enable the best possible data compression. A neural network was used to create a codebook of vectors that attempt to span the low-frequency data space. Since these vectors are potentially the less critical areas of the electrocardiographic signal, less important information will be subjected to increases in distortion. The Kohonen paradigm used in this study is an unsupervised neural network that adapts the codebook vectors based on distance measurements and controls the scope of the changes based on time. This network has been shown to work well with image and speech data, but to the author's knowledge has not been used on electrocardiographic data. The compression of the signal comes from inserting the address of the codebook vector that best represents the original vector in place of the vector. A test is first performed to determine if the distortion between the original and the replacement vector is within a present limit. If it is, the address is inserted. If the distortion is too large the original vector will be retained. Typically, the QRS segment and possibly the T segment will be preserved. The new compressed file can be further reduced by lossless techniques to increase the compression ratio.(ABSTRACT TRUNCATED AT 250 WORDS)

Diagnosis, Computer-Assisted↗

PubMed2004

Potential assessment of the "support vector machine" method in forecasting ambient air pollutant trends.

Monitoring and forecasting of air quality parameters are popular and important topics of atmospheric and environmental research today due to the health impact caused by exposing to air pollutants existing in urban air. The accurate models for air pollutant prediction are needed because such models would allow forecasting and diagnosing potential compliance or non-compliance in both short- and long-term aspects. Artificial neural networks (ANN) are regarded as reliable and cost-effective method to achieve such tasks and have produced some promising results to date. Although ANN has addressed more attentions to environmental researchers, its inherent drawbacks, e.g., local minima, over-fitting training, poor generalization performance, determination of the appropriate network architecture, etc., impede the practical application of ANN. Support vector machine (SVM), a novel type of learning machine based on statistical learning theory, can be used for regression and time series prediction and have been reported to perform well by some promising results. The work presented in this paper aims to examine the feasibility of applying SVM to predict air pollutant levels in advancing time series based on the monitored air pollutant database in Hong Kong downtown area. At the same time, the functional characteristics of SVM are investigated in the study. The experimental comparisons between the SVM model and the classical radial basis function (RBF) network demonstrate that the SVM is superior to the conventional RBF network in predicting air quality parameters with different time series and of better generalization performance than the RBF model.

Air Pollutants↗

PubMed1996

Application of the fuzzy ARTMAP neural network model to medical pattern classification tasks.

This paper presents research into the application of the fuzzy ARTMAP neural network model to medical pattern classification tasks. A number of domains, both diagnostic and prognostic, are considered. Each such domain highlights a particularly useful aspect of the model. The first coronary care patient prognosis, demonstrates the ARTMAP voting strategy involving 'pooled' decision-making using a number of networks, each of which has learned a slightly different mapping of input features to pattern classes. The second domain, breast cancer diagnosis, demonstrates the model's symbolic rule extraction capabilities which support the validation and explanation of a network's predictions. The final domain, diagnosis of acute myocardial infarction, demonstrates a novel category pruning technique allowing the performance of a trained network to be altered so as to favour predictions of one class over another (e.g. trading sensitivity for specificity or vice versa). It also introduces a 'cascaded' variant of the voting strategy intended to allow identification of a subset of cases which the network has a very high certainty of classifying correctly.

Artificial Intelligence↗

PubMed2005

Stochastic resonance with differential code in feedforward network with intra-layer random connections.

We examined stochastic resonance with a differential coding scheme using a multilayer feedforward neural network which is composed of intra-layer connections. We show that the network, with random synaptic connections in each layer, encodes an input signal into a spike coherence that represents temporal differences among the inputs. We also demonstrate that both internal and external noise enhance the detection of weak signals. Finally, we discuss how the feedforward network with intra-layer random connections is similar to a membrane in its sensitivity to and amplification of a change in stimulus and suggest that the intensity of internal noise may be tuned in a real brain.

Computer Simulation↗

PubMed2000

A neural network based virtual high throughput screening test for the prediction of CNS activity.

A virtual high throughput screening test to identify potentially CNS-active drugs has been developed. Discrimination was based on the knowledge available in databases containing CNS-active (Cipsline from Prous Science) and inactive compounds (Chemical Directory from Sigma-Aldrich). Molecular structures were represented using 2D Unit y fingerprints and a feedforward neural network was trained to classify molecules regarding their CNS activity. The parameterized network was validated by reclassification of the training set elements, by the classification of a test set preselected from the Prous database, and also by the prediction of activity for known CNS drugs not used in the training set but available in the Medchem database (Daylight). These tests revealed that our neural net recognized at least 89% of CNS-active compounds and would be suitable for use in our virtual screening protocol.

Artificial Intelligence↗

PubMed1998

Adapting a feedforward heteroassociative network to Hodgkin-Huxley dynamics.

Using the original McCulloch-Pitts notion of simple on and off spike coding in lieu of rate coding, an Anderson-Kohonen artificial neural network (ANN) associative memory model was ported to a neuronal network with Hodgkin-Huxley dynamics. In the ANN, the use of 0/1 (no-spike/spike) units introduced a cross-talk term that had to be compensated by introducing balanced feedforward inhibition. The resulting ANN showed good capacity and fair selectivity (rejection of unknown input vectors). Translation to the Hodgkin-Huxley model resulted in a network that was functional but not at all robust. Evaluation of the weaknesses of this network revealed that it functioned far better using spike timing, rather than spike occurrence, as the code. The algorithm requires a novel learning algorithm for feedforward inhibition that could be sought physiologically.

Action Potentials↗

PubMed2005

New adaptive color quantization method based on self-organizing maps.

Color quantization (CQ) is an image processing task popularly used to convert true color images to palletized images for limited color display devices. To minimize the contouring artifacts introduced by the reduction of colors, a new competitive learning (CL) based scheme called the frequency sensitive self-organizing maps (FS-SOMs) is proposed to optimize the color palette design for CQ. FS-SOM harmonically blends the neighborhood adaptation of the well-known self-organizing maps (SOMs) with the neuron dependent frequency sensitive learning model, the global butterfly permutation sequence for input randomization, and the reinitialization of dead neurons to harness effective utilization of neurons. The net effect is an improvement in adaptation, a well-ordered color palette, and the alleviation of underutilization problem, which is the main cause of visually perceivable artifacts of CQ. Extensive simulations have been performed to analyze and compare the learning behavior and performance of FS-SOM against other vector quantization (VQ) algorithms. The results show that the proposed FS-SOM outperforms classical CL, Linde, Buzo, and Gray (LBG), and SOM algorithms. More importantly, FS-SOM achieves its superiority in reconstruction quality and topological ordering with a much greater robustness against variations in network parameters than the current art SOM algorithm for CQ. A most significant bit (MSB) biased encoding scheme is also introduced to reduce the number of parallel processing units. By mapping the pixel values as sign-magnitude numbers and biasing the magnitudes according to their sign bits, eight lattice points in the color space are condensed into one common point density function. Consequently, the same processing element can be used to map several color clusters and the entire FS-SOM network can be substantially scaled down without severely scarifying the quality of the displayed image. The drawback of this encoding scheme is the additional storage overhead, which can be cut down by leveraging on existing encoder in an overall lossy compression scheme.

Algorithms↗

PubMed2003

Efficient protein crystallization.

High-throughput molecular biology and crystallography advances have placed an increasing demand on crystallization, the one remaining bottleneck in macromolecular crystallography. This paper describes three experimental approaches, an incomplete factorial crystallization screen, a high-throughput nanoliter crystallization system, and the use of a neural net to predict crystallization conditions via a small sample (approximately 0.1%) of screening results. The use of these technologies has the potential to reduce time and sample requirements. Initial experimental results indicate that the incomplete factorial design detects initial crystallization conditions not previously discovered using commercial screens. This may be due to the ability of the incomplete factorial screen to sample a broader portion of "crystallization space," using a multidimensional set of components, concentrations, and physical conditions. The incomplete factorial screen is complemented by a neural network program used to model crystallization. This capability is used to help predict new crystallization conditions. An automated, nanoliter crystallization system, with a throughput of up to 400 conditions/h in 40-nl droplets (total volume), accommodates microbatch or traditional "sitting-drop" vapor diffusion experiments. The goal of this research is to develop a fully-automated high-throughput crystallization system that integrates incomplete factorial screen and neural net capabilities.

Computer Simulation↗

PubMed2005

Neurons with radial basis like rate functions.

Artificial neural networks constructed with "locally tuned processing units" and more generally referred to as "radial basis function networks" have been proposed by a number of workers. In this communication, I submit a conjecture, based on indirect experimental and direct computational evidence of the Hodgkin-Huxley model, that there may be biological neurons in nervous systems for which the rate function is locally tuned. If proved to be valid, this conjecture may simplify neurodynamic models of some functions of nervous systems.

Axons↗