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 451 records · Page 25Linked to original sources
PubMed1996

Global effects of fluctuations in neural information processing.

We are interested in how the complex dynamics of the brain, which may include oscillations, chaos and noise, can affect the efficiency of neural information processing. Here, we consider the amplification and functional role of fluctuations, expressed as chaos or noise in the system. Using computer simulations of a neural network model of the olfactory cortex, we demonstrate how microscopic fluctuations can result in global effects at the network level. In particular, we show that the rate of information processing in associative memory tasks can be maximized for optimal noise levels. Noise can also induce transitions between different dynamical states, related to learning and memory. A chaotic-like behavior, induced by noise or by an increase in neuronal excitability, can enhance system performance if it is transient and converges to a limit cycle memory state. The level of accuracy required for correct pattern association further affects the rate of information processing. We discuss how neuromodulatory control of the cortical dynamics can shift the balance between rate and accuracy optimization, as well as between sensitivity and stability.

Acetylcholine↗
PubMed1993

Neural network approach for computer-assisted interpretation of ultrasound images of the gallbladder.

Multi-formatted films of 90 ultrasound examinations of the gallbladder (stones 56 cases, sludge 20 cases, hydrops five cases, normal gallbladder nine cases) have been digitalized and stored in a personal computer. Image data of each examination was processed to extract a 19-dimensional vector that represents the essential diagnostic information of each examination. This vector was evaluated by three different classification algorithms: (1) classical nearest neighbor principle, (2) classical linear discriminant analysis, (3) multilayered backpropagation neural network. The correct classification rate was 64% (58/90) for the nearest neighbor principle, 97% (87/90) for the linear discriminant analysis, and 99% (89/90) for the backpropagation neural network. We conclude that, (1) automated classification of ultrasound images is possible for limited diagnostic problems, (2) a neural network approach can be used successfully for that goal, and (3) the efficiency of the more flexible neural network approach is comparable to large-scale classical methods.

Cholecystography↗
PubMed2005

Synchronous neural activity in scale-free network models versus random network models.

Synchronous firing peaks at levels greatly exceeding background activity have recently been reported in neocortical tissue. A small subset of neurons is dominant in a large fraction of the peaks. To investigate whether this striking behavior can emerge from a simple model, we constructed and studied a model neural network that uses a modified Hopfield-type dynamical rule. We find that networks having a power-law ("scale-free") node degree distribution readily generate extremely large synchronous firing peaks dominated by a small subset of nodes, whereas random (Erdös-Rényi) networks do not. This finding suggests that network topology may play an important role in determining the nature and magnitude of synchronous neural activity.

Computer Simulation↗
PubMed2001

CDMA and TDMA based neural nets.

CDMA and TDMA telecommunication techniques were established long time ago, but they have acquired a renewed presence due to the rapidly increasing mobile phones demand. In this paper, we are going to see they are suitable for neural nets, if we leave the concept "connection" between processing units and we adopt the concept "messages" exchanged between them. This may open the door to neural nets with a higher number of processing units and flexible configuration.

Computer Communication Networks↗
PubMed2002

Slow feature analysis: unsupervised learning of invariances.

Invariant features of temporally varying signals are useful for analysis and classification. Slow feature analysis (SFA) is a new method for learning invariant or slowly varying features from a vectorial input signal. It is based on a nonlinear expansion of the input signal and application of principal component analysis to this expanded signal and its time derivative. It is guaranteed to find the optimal solution within a family of functions directly and can learn to extract a large number of decorrelated features, which are ordered by their degree of invariance. SFA can be applied hierarchically to process high-dimensional input signals and extract complex features. SFA is applied first to complex cell tuning properties based on simple cell output, including disparity and motion. Then more complicated input-output functions are learned by repeated application of SFA. Finally, a hierarchical network of SFA modules is presented as a simple model of the visual system. The same unstructured network can learn translation, size, rotation, contrast, or, to a lesser degree, illumination invariance for one-dimensional objects, depending on only the training stimulus. Surprisingly, only a few training objects suffice to achieve good generalization to new objects. The generated representation is suitable for object recognition. Performance degrades if the network is trained to learn multiple invariances simultaneously.

Algorithms↗
PubMedDate not supplied

Boundary detection of retinoblastoma tumors with neural networks.

Consistent and accurate measurement of retinoblastoma tumors is of important clinical value for treatment management. This paper presents an algorithm for the determination of retinoblastoma (RB) tumor to assist in the determination of tumor volume changes throughout treatment periods. The result of the development of a neural network approach for the analysis of three-dimensional ultrasound images shows that it is possible to identify retinoblastoma tumors and accurately determine the front and back boundary of the tumor. The algorithm used was a soft competitive learning network with two inputs. The outputs of the network identify the eye, the tumor, and the back of the eye.

Algorithms↗
PubMed1997

Radar image segmentation using self-adapting recurrent networks.

This paper presents a novel approach to the segmentation and integration of (radar) images using a second-order recurrent artificial neural network architecture consisting of two sub-networks: a function network that classifies radar measurements into four different categories of objects in sea environments (water, oil spills, land and boats), and a context network that dynamically computes the function network's input weights. It is shown that in experiments (using simulated radar images) this mechanism outperforms conventional artificial neural networks since it allows the network to learn to solve the task through a dynamic adaptation of its classification function based on its internal state closely reflecting the current context.

Doppler Effect↗
PubMedDate not supplied

Determining the number of centroids for CMLP network.

In our recent studies we have proposed and investigated a centroid-based multilayer perceptron (CMLP) network architecture for modelling purposes. In the CMLP network the first hidden layer is a centroid layer. We have found that the proposed hybrid can provide significant advantages over standard multilayer perceptron networks in terms of fast and efficient learning, and compact network structure in complex classification problems. Previously the number of units for the centroid layer had been determined empiricially. Here we extend our work by introducing a method for determining the minimal number of centroid units for a given problem. The proposed scheme also enables efficient initialization of the centroids units. In addition, we also propose an initialization scheme for the MLP part of the CMLP network. Our benchmark simulations show that the proposed methods significantly improve the CMLP scheme.

Computer Simulation↗
PubMed2005

Spontaneous evolution of modularity and network motifs.

Biological networks have an inherent simplicity: they are modular with a design that can be separated into units that perform almost independently. Furthermore, they show reuse of recurring patterns termed network motifs. Little is known about the evolutionary origin of these properties. Current models of biological evolution typically produce networks that are highly nonmodular and lack understandable motifs. Here, we suggest a possible explanation for the origin of modularity and network motifs in biology. We use standard evolutionary algorithms to evolve networks. A key feature in this study is evolution under an environment (evolutionary goal) that changes in a modular fashion. That is, we repeatedly switch between several goals, each made of a different combination of subgoals. We find that such "modularly varying goals" lead to the spontaneous evolution of modular network structure and network motifs. The resulting networks rapidly evolve to satisfy each of the different goals. Such switching between related goals may represent biological evolution in a changing environment that requires different combinations of a set of basic biological functions. The present study may shed light on the evolutionary forces that promote structural simplicity in biological networks and offers ways to improve the evolutionary design of engineered systems.

Algorithms↗
PubMedDate not supplied

Database mining using soft computing techniques. An integrated neural network-fuzzy logic-genetic algorithm approach.

Two different soft computing (SC) techniques (a competitive learning neural network and an integrated neural network-fuzzy logic-genetic algorithm approach) are employed in the analysis of a database subset obtained from the Cambridge Structural Database. The chemical problem chosen for study is relevant to the relationship between various metric parameters in transition metal imido (LnMdNZ, Z = carbon-based substituent) complexes and the chemical consequences of such relationships. The SC techniques confirmed and quantified the suspected relationship between the metal-nitrogen bond length and the metal-nitrogen-substituent bond angle for transition metal imidos: increased metal-nitrogen-carbon angles correlate with shortened metal-nitrogen distances. The mining effort also yielded an unexpected correlation between the NC distance and the MNC angle-shorter NC correlate with larger MNC. A fuzzy inference system is used to construct an MNred-NC-MNC hypersurface. This hypersurface suggests a complicated interdependence among NC, MNred, and the angle subtended by these two bonds. Also, major portions of the hypersurface are very flat, in regions where MNC is approaching linearity. The relationships are also seen to be influenced by whether the imido substituent is an alkyl or aryl group. Computationally, the present results are of particular interest in two respects. First, SC classification was able to isolate an "outlier" cluster. Identification of outliers is important as they may correspond to unreported experimental errors in the database or novel chemical entities, both of which warrant further investigation. Second, the SC database mining not only confirmed and quantified a suspected relationship (MNred versus MNC) within the data but also yielded a trend that was not suspected (NC versus MNC).

Journal Article↗
PubMed2005

A generalized growing and pruning RBF (GGAP-RBF) neural network for function approximation.

This paper presents a new sequential learning algorithm for radial basis function (RBF) networks referred to as generalized growing and pruning algorithm for RBF (GGAP-RBF). The paper first introduces the concept of significance for the hidden neurons and then uses it in the learning algorithm to realize parsimonious networks. The growing and pruning strategy of GGAP-RBF is based on linking the required learning accuracy with the significance of the nearest or intentionally added new neuron. Significance of a neuron is a measure of the average information content of that neuron. The GGAP-RBF algorithm can be used for any arbitrary sampling density for training samples and is derived from a rigorous statistical point of view. Simulation results for bench mark problems in the function approximation area show that the GGAP-RBF outperforms several other sequential learning algorithms in terms of learning speed, network size and generalization performance regardless of the sampling density function of the training data.

Algorithms↗
PubMed2001

Real time distributed processing of multiple associated pulse pattern sequences.

A Real Time Distributed Associative Memory Artificial Neural Network (RTANN) is described. This network associates groups of pulse pattern sequences. The subsequent reoccurrence of some sequences will cause the remainder to be regenerated. Training is carried out in real time simply by feeding pattern sequences directly into the network. The connections between units incorporate a wide range of transmission delays. During training the network enhances connection weights on units where coincidences occur between input and delayed pulses. Pattern regeneration utilises the reoccurrence of coincidences between delayed pulses. The simulation of an RTANN is presented. Continuous dual pattern sequences from notional sensors monitoring the shape and colour of an object were associated directly with a third dual pattern sequence having the form 'These objects look colour'. After training the network was able to correctly generate sentences describing combinations of object and colour not encountered during training.

Artificial Intelligence↗
PubMed1999

Artificial neural network-aided image analysis system for cell counting.

BACKGROUND: In histological preparations containing debris and synthetic materials, it is difficult to automate cell counting using standard image analysis tools, i.e., systems that rely on boundary contours, histogram thresholding, etc. In an attempt to mimic manual cell recognition, an automated cell counter was constructed using a combination of artificial intelligence and standard image analysis methods. METHODS: Artificial neural network (ANN) methods were applied on digitized microscopy fields without pre-ANN feature extraction. A three-layer feed-forward network with extensive weight sharing in the first hidden layer was employed and trained on 1,830 examples using the error back-propagation algorithm on a Power Macintosh 7300/180 desktop computer. The optimal number of hidden neurons was determined and the trained system was validated by comparison with blinded human counts. System performance at 50x and lO0x magnification was evaluated. RESULTS: The correlation index at 100x magnification neared person-to-person variability, while 50x magnification was not useful. The system was approximately six times faster than an experienced human. CONCLUSIONS: ANN-based automated cell counting in noisy histological preparations is feasible. Consistent histology and computer power are crucial for system performance. The system provides several benefits, such as speed of analysis and consistency, and frees up personnel for other tasks.

Cell Count↗
PubMed2005

Radial basis function neural networks in non-destructive determination of compound aspirin tablets on NIR spectroscopy.

The application of the second most popular artificial neural networks (ANNs), namely, the radial basis function (RBF) networks, has been developed for quantitative analysis of drugs during the last decade. In this paper, the two components (aspirin and phenacetin) were simultaneously determined in compound aspirin tablets by using near-infrared (NIR) spectroscopy and RBF networks. The total database was randomly divided into a training set (50) and a testing set (17). Different preprocessing methods (standard normal variate (SNV), multiplicative scatter correction (MSC), first-derivative and second-derivative) were applied to two sets of NIR spectra of compound aspirin tablets with different concentrations of two active components and compared each other. After that, the performance of RBF learning algorithm adopted the nearest neighbor clustering algorithm (NNCA) and the criterion for selection used a cross-validation technique. Results show that using RBF networks to quantificationally analyze tablets is reliable, and the best RBF model was obtained by first-derivative spectra.

Aspirin↗
PubMed2002

Spatial resolution enhancement of ultrasound images using neural networks.

Spatial resolution in modern ultrasound imaging systems is limited by the high cost of large aperture transducer arrays, which require a large number of transducer elements and electronic channels. A new technique to enhance the spatial resolution of pulse-echo imaging systems is presented. The method attempts to build an image that could be obtained with a transducer array aperture larger than that physically available. We consider two images of the same object obtained with two different apertures, the full aperture and a subaperture, of the same transducer. A suitable artificial neural network (ANN) is trained to reproduce the relationship between the image obtained with the transducer full aperture and the image obtained with a subaperture. The inputs of the neural network are portions of the image obtained with the subaperture (low resolution image), and the target outputs are the corresponding portions of the image produced by the full aperture (high resolution image). After the network is trained, it can produce images with almost the same resolution of the full aperture transducer, but using a reduced number of real transducer elements. All computations are carried out on envelope-detected decimated images; for this reason, the computational cost is low and the method is suitable for real-time applications. The proposed method was applied to experimental data obtained with the ultrasound synthetic aperture focusing technique (SAFT), giving quite promising results. Real-time implementation on a modern, full-digital echographic system is currently being developed.

Algorithms↗
PubMed2004

[Research on the application of neural network to diagnosis of cardiopathy].

Neural networks can fit any nonlinear function. After drawing out several characteristic parameters from the three-dimension spectrum for high frequency QRS waves, we input them into the network and trained the network. In this way, we can get a m-dimension curved surface in the m-dimension space which is constructed by those parameters, and this curved surface divides the space into two parts: the unhealthiness and the health. Now, the network can automatically distinguish between the healthiness and the unhealthiness according to their three-dimension spectrum for high frequency QRS waves.

Algorithms↗
PubMed2005

Improved prediction of critical residues for protein function based on network and phylogenetic analyses.

BACKGROUND: Phylogenetic approaches are commonly used to predict which amino acid residues are critical to the function of a given protein. However, such approaches display inherent limitations, such as the requirement for identification of multiple homologues of the protein under consideration. Therefore, complementary or alternative approaches for the prediction of critical residues would be desirable. Network analyses have been used in the modelling of many complex biological systems, but only very recently have they been used to predict critical residues from a protein's three-dimensional structure. Here we compare a couple of phylogenetic approaches to several different network-based methods for the prediction of critical residues, and show that a combination of one phylogenetic method and one network-based method is superior to other methods previously employed. RESULTS: We associate a network with each member of a set of proteins for which the three-dimensional structure is known and the critical residues have been previously determined experimentally. We show that several network-based centrality measurements (connectivity, 2-connectivity, closeness centrality, betweenness and cluster coefficient) accurately detect residues critical for the protein's function. Phylogenetic approaches render predictions as reliable as the network-based measurements, although, interestingly, the two general approaches tend to predict different sets of critical residues. Hence we propose a hybrid method that is composed of one network-based calculation--the closeness centrality--and one phylogenetic approach--the Conseq server. This hybrid approach predicts critical residues more accurately than the other methods tested here. CONCLUSION: We show that network analysis can be used to improve the prediction of amino acids critical for protein function, when utilized in combination with phylogenetic approaches. It is proposed that such improvement is due to the complementary nature of these approaches: network-based methods tend to predict as critical those residues that are highly connected and internal (i.e., non-surface), although some surface residues are indeed identified as critical by network analyses; whereas residues chosen by phylogenetic approaches display a lower overall probability of being surface inaccessible.

Computational Biology↗