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 1,585 records · Page 88Linked to original sources
PubMed2003

Dynamical properties of strongly interacting Markov chains.

Spatial interdependences of multiple stochastic units can be suitably quantified by the Kullback-Leibler divergence of the joint probability distribution from the corresponding factorized distribution. In the present paper, a generalized measure for stochastic interaction, which also captures temporal interdependences, is analysed within the setting of Markov chains. The dynamical properties of systems with strongly interacting stochastic units are analytically studied and illustrated by computer simulations. In particular, the emergence of determinism in such systems is demonstrated.

Computer Simulation↗
PubMed1999

Interhemispheric effects on map organization following simulated cortical lesions.

During recent years there has been increasing use of neural models to investigate the implications of hypotheses about brain and cognitive disorders. Here we systematically study the effects of sudden simulated lesions on cortical maps in a neural model consisting of left and right hemispheric regions connected by a corpus callosum. The model identifies conditions under which damage to one hemispheric region leads to reorganization of the contralateral, intact hemispheric region. The intact hemisphere's participation in the recovery process is found to be a function of pre-existing map lateralization/symmetry and lesion size, indicating that interpretation of future experimental work should take these factors into account.

Brain Diseases↗
PubMed1998

Incorporation of a set enumeration trees-based classifier into a hybrid computer-assisted diagnosis scheme for mass detection.

RATIONALE AND OBJECTIVES: The authors evaluated whether a hybrid classifier of two independent computer-aided diagnosis (CAD) schemes, the set enumeration (SE) trees approach and an artificial neural network (ANN), could improve the detection of masses on digitized mammograms. The potential benefits resulting from the interpretability of the SE trees model was also explored. MATERIALS AND METHODS: Two hundred thirty verified mass regions and 230 negative but suspicious regions were randomly selected from 618 digitized mammograms. Each region was represented by a 24-parameter feature vector. These features were used as input data for the SE trees and ANN-based schemes. After the positive and negative regions were randomly segmented into five exclusive partitions, a fivefold cross-validation method was applied to evaluate and compare the performance of the SE trees, ANN, and hybrid system in the identification of masses. RESULTS: The performance of the SE trees approach was comparable to that of the ANN. The average area under the receiver operating characteristic (ROC) curves for all five partitions was 0.88 (standard deviation, 0.04). Owing to the relatively low correlation between the region-based results of the SE trees and ANN methods, the hybrid classifier yielded a significantly improved performance, with an area under the ROC curve of 0.94 (standard deviation, 0.02; P < .05). CONCLUSION: The hybrid CAD scheme significantly improved performance. The amenability of the SE trees models to interpretation may aid in the assessment of the importance of specific features.

Artificial Intelligence↗
PubMed2002

Measurement of ejection fraction with standard thermodilution catheters.

Right ventricle ejection fraction (RVEF) is clinically used to evaluate right ventricular function. The thermodilution method can be modified to estimate the RVEF. However, this method requires a thermistor with a fast time response in order to yield correct estimates. Digital signal processing techniques that were developed in previous works, allow the use of industry-standard slow time response thermistors for the measurement EF. However, these algorithms were not automated, and the works did not present a complete evaluation of the method's performance. This article presents a modified automated version of these algorithms, and uses numerical and in vitro simulations to test their performance. In the simulations, the measured ejection fraction was compared to the true ejection fraction. RVEFs ranging from 0.20 to 0.80 were tested for heart rates ranging from 30 to 120 heart beats per min. Statistical analysis of data showed that the new method presents an improved performance.

Algorithms↗
PubMed2005

Computer prediction of drug resistance mutations in proteins.

Drug resistance is of increasing concern in the treatment of infectious diseases and cancer. Mutation in drug-interacting disease proteins is one of the primary causes for resistance particularly against anti-infectious drugs. Prediction of resistance mutations in these proteins is valuable both for the molecular dissection of drug resistance mechanisms and for predicting features that guide the design of new agents to counter resistant strains. Several protein structure- and sequence-based computer methods have been explored for mechanistic study and prediction of resistance mutations. These methods and their usefulness are reviewed here.

Computer Simulation↗
PubMedDate not supplied

Silicon synaptic conductances.

We have developed compact analog integrated circuits that simulate two synaptic excitatory conductances. A four-transistor circuit captures the dynamics of an excitatory postsynaptic current caused by a real AMPA conductance. A six-transistor circuit simulates the effects of a real voltage-dependent NMDA conductance. The postsynaptic current dynamics are modeled by a current mirror integrator with adjustable gain. The voltage dependence of the silicon NMDA conductance is realized by a differential pair. We show the operation of these silicon synaptic conductances and their integration with the silicon neuron (Mahowald and Douglas, 1991).

Action Potentials↗
PubMedDate not supplied

A neurodynamical model of visual attention: feedback enhancement of spatial resolution in a hierarchical system.

Human beings have the capacity to recognize objects in natural visual scenes with high efficiency despite the complexity of such scenes, which usually contain multiple objects. One possible mechanism for dealing with this problem is selective attention. Psychophysical evidence strongly suggests that selective attention can enhance the spatial resolution in the input region corresponding to the focus of attention. In this work we adopt a computational neuroscience perspective to analyze the attentional enhancement of spatial resolution in the area containing the objects of interest. We extend and apply the computational model of Deco and Schürmann (2000), which consists of several modules with feedforward and feedback interconnections describing the mutual links between different areas of the visual cortex. Each module analyses the visual input with different spatial resolution and can be thought of as a hierarchical predictor at a given level of resolution. Moreover, each hierarchical predictor has a submodule that consists of a group of neurons performing a biologically based 2D Gabor wavelet transformation at a given resolution level. The attention control decides in which local regions the spatial resolution should be enhanced in a serial fashion. In this sense, the scene is first analyzed at a coarse resolution level, and the focus of attention enhances iteratively the resolution at the location of an object until the object is identified. We propose and simulate new psychophysical experiments where the effect of the attentional enhancement of spatial resolution can be demonstrated by predicting different reaction time profiles in visual search experiments where the target and distractors are defined at different levels of resolution.

Attention↗
PubMed1997

A mobile automated mammography system.

Automated processing of mammograms has been studied for several years. Until the last few years much of the efforts have produced marginal results due to the limitations of inexpensive hardware. Even with the advanced hardware no one has yet taken an entire mammogram and determined whether it is normal or abnormal. We outline a method here to attempt to do just that.

Breast Diseases↗
PubMed2005

The dynamic nature of spatial encoding in the hippocampus.

Many hippocampal neurons (place cells) appear to represent a particular location within an environment (their place field). This property would appear to be central to hippocampal involvement in navigation based on spatial memory. Although a navigationally useful representation might also include information about distal goals, having a place field and being able to represent a distal goal would appear to be mutually exclusive place cell properties. Our simulations demonstrate, however, that information about goal direction can be simply derived from the changes in place field density that occur when place fields shift location in a goal-directed manner. Previous reports that place fields respond dynamically to shifts in goal location may, therefore, represent the operation of such a system.

Animals↗
PubMed1997

Comparisons of the abilities of a neural network and three consultant oral surgeons to make decisions about third molar removal.

AIM: To compare the performance of a computer based decision support system (a neural network) and consultant oral and maxillofacial surgeons in making decisions about the need to remove lower third molars. DESIGN AND SETTING: Receiver operating characteristic (ROC) analysis at a hospital department of oral and maxillofacial surgery. SUBJECTS AND METHODS: Three consultant oral and maxillofacial surgeons indicated on a six-point rating scale how certain they were that each of 50 documented lower third molars required removal. Similar data were obtained from the neural network following appropriate coding of the clinical information. These data were compared with gold standard treatment decisions for each tooth based on National Institutes of Health Concensus criteria using ROC analysis. MAIN OUTCOME MEASURES: The area beneath each operator ROC curve (varying between zero and one with greater areas indicating better performance). RESULTS: The network performed as well a two consultants (P = 0.12/0.18, NS) and significantly better than the third (z = 526, P < 0.01). CONCLUSIONS: This work suggests that this computer based neural network could play a useful role in supporting dental practitioners making third molar referral decisions.

Adult↗
PubMed2004

Computer-aided diagnosis in chest radiology.

Chest radiography is still a useful examination in various situations, although CT has become a modality of choice as a diagnostic examination in many cases. Current computer-aided diagnosis (CAD) schemes for chest radiographs include nodule detection, interstitial disease detection, temporal subtraction, differential diagnosis of interstitial disease, and distinction between benign and malignant pulmonary nodules. All of these schemes are demonstrated as providing potentially useful tools for radiologists when the output of these schemes is used as a "second opinion." There are some commercially available products for these schemes and more are expected to be available in the near future. The current status of CAD for CT is also discussed briefly in this article.

Diagnosis, Computer-Assisted↗
PubMed2004

Natural computation measured as a reduction of complexity.

We argue that the deeper nature of computation is to reduce the statistical obstruction against prediction. From this, we derive an explicit measure of computation for general, artificial as well as natural, systems (electronic circuits, neurons, mechanical devices, etc.). The applicability and usefulness of this concept is demonstrated using well-studied families of dynamical systems, as well as experimental time series from cortical neurons.

Action Potentials↗
PubMed1998

A neuronal model of a global workspace in effortful cognitive tasks.

A minimal hypothesis is proposed concerning the brain processes underlying effortful tasks. It distinguishes two main computational spaces: a unique global workspace composed of distributed and heavily interconnected neurons with long-range axons, and a set of specialized and modular perceptual, motor, memory, evaluative, and attentional processors. Workspace neurons are mobilized in effortful tasks for which the specialized processors do not suffice. They selectively mobilize or suppress, through descending connections, the contribution of specific processor neurons. In the course of task performance, workspace neurons become spontaneously coactivated, forming discrete though variable spatio-temporal patterns subject to modulation by vigilance signals and to selection by reward signals. A computer simulation of the Stroop task shows workspace activation to increase during acquisition of a novel task, effortful execution, and after errors. We outline predictions for spatio-temporal activation patterns during brain imaging, particularly about the contribution of dorsolateral prefrontal cortex and anterior cingulate to the workspace.

Brain↗
PubMed2005

Learning from humans: computational modeling of face recognition.

In this paper, we propose a computational architecture of face recognition based on evidence from cognitive research. Several recent psychophysical experiments have shown that humans process faces by a combination of configural and component information. Using an appearance-based implementation of this architecture based on low-level features and their spatial relations, we were able to model aspects of human performance found in psychophysical studies. Furthermore, results from additional computational recognition experiments show that our framework is able to achieve excellent recognition performance even under large view rotations. Our interdisciplinary study is an example of how results from cognitive research can be used to construct recognition systems with increased performance. Finally, our modeling results also make new experimental predictions that will be tested in further psychophysical studies, thus effectively closing the loop between psychophysical experimentation and computational modeling.

Area Under Curve↗
PubMed2003

An exact method to quantify the information transmitted by different mechanisms of correlational coding.

We derive a new method to quantify the impact of correlated firing on the information transmitted by neuronal populations. This new method considers, in an exact way, the effects of high order spike train statistics, with no approximation involved, and it generalizes our previous work that was valid for short time windows and small populations. The new technique permits one to quantify the information transmitted if each cell were to convey fully independent information separately from the information available in the presence of synergy-redundancy effects. Synergy-redundancy effects are shown to arise from three possible contributions: a redundant contribution due to similarities in the mean response profiles of different cells; a synergistic stimulus-dependent correlational contribution quantifying the information content of changes of correlations with stimulus, and a stimulus-independent correlational contribution term that reflects interactions between the distribution of rates of individual cells and the average level of cross-correlation. We apply the new method to simultaneously recorded data from somatosensory and visual cortices. We demonstrate that it constitutes a reliable tool to determine the role of cross-correlated activity in stimulus coding even when high firing rate data (such as multi-unit recordings) are considered.

Action Potentials↗
PubMed1994

PHD--an automatic mail server for protein secondary structure prediction.

By the middle of 1993, > 30,000 protein sequences has been listed. For 1000 of these, the three-dimensional (tertiary) structure has been experimentally solved. Another 7000 can be modelled by homology. For the remaining 21,000 sequences, secondary structure prediction provides a rough estimate of structural features. Predictions in three states range between 35% (random) and 88% (homology modelling) overall accuracy. Using information about evolutionary conservation as contained in multiple sequence alignments, the secondary structure of 4700 protein sequences was predicted by the automatic e-mail server PHD. For proteins with at least one known homologue, the method has an expected overall three-state accuracy of 71.4% for proteins with at least one known homologue (evaluated on 126 unique protein chains).

Algorithms↗