Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Neural network”

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 1,243 records · Page 69Linked to original sources
PubMed2003

Multistability analysis for recurrent neural networks with unsaturating piecewise linear transfer functions.

Multistability is a property necessary in neural networks in order to enable certain applications (e.g., decision making), where monostable networks can be computationally restrictive. This article focuses on the analysis of multistability for a class of recurrent neural networks with unsaturating piecewise linear transfer functions. It deals fully with the three basic properties of a multistable network: boundedness, global attractivity, and complete convergence. This article makes the following contributions: conditions based on local inhibition are derived that guarantee boundedness of some multistable networks, conditions are established for global attractivity, bounds on global attractive sets are obtained, complete convergence conditions for the network are developed using novel energy-like functions, and simulation examples are employed to illustrate the theory thus developed.

Linear Models↗
PubMed1997

Classification of mild Alzheimer's disease by artificial neural network analysis of SPET data.

An evaluation of the performance of artificial neural networks (ANNs) for the classification of probable Alzheimer's disease (pAD) patients was undertaken using data extracted from four regions of interest constructed on single photon emission tomographic (SPET) cerebral perfusion images. Two studies using feed-forward neural networks (FFNNs) were undertaken. The first was to determine if it would be possible to classify pAD patients and normal subjects in a mixed group, comprising 29 patients diagnosed as having pAD varying in severity from mild, established dementia to moderate dementia and 10 healthy control subjects. The second was to determine if the networks generated in the first study could prospectively classify 15 additional patients with very mild or mild cognitive impairment. The results were compared to those obtained using the same data and discriminant analysis. The relative performances of the two analysis techniques were assessed on the basis of the area under receiver operating characteristics (ROC) curves. The FFNN successfully classified all datasets in the first study, achieving an area under the ROC curve of 1.00, whereas discriminant analysis achieved 0.94. When tested on data from the second group, the areas under the ROC curves varied between 0.86 and 1.00 for the FFNN, whereas that for discriminant analysis was 0.99. We conclude that FFNNs can accurately classify pAD patients with mild to moderate dementia using data obtained from SPET cerebral perfusion images.

Aged↗
PubMed2000

Mean field theory for asymmetric neural networks.

The computation of mean firing rates and correlations is intractable for large neural networks. For symmetric networks one can derive mean field approximations using the Taylor series expansion of the free energy as proposed by Plefka. In asymmetric networks, the concept of free energy is absent. Therefore, it is not immediately obvious how to extend this method to asymmetric networks. In this paper we extend Plefka's approach to asymmetric networks and in fact to arbitrary probability distributions. The method is based on an information geometric argument. The method is illustrated for asymmetric neural networks with sequential dynamics. We compare our approximate analytical results with Monte Carlo simulations for a network of 100 neurons. It is shown that the quality of the approximation for asymmetric networks is as good as for symmetric networks.

Models, Neurological↗
PubMed2004

Analysis of global exponential stability and periodic solutions of neural networks with time-varying delays.

In this paper, a general class of recurrent neural networks with time-varying delays is studied. Some novel and sufficient conditions are given to guarantee the global exponential stability of the equilibrium point and the existence of periodic solutions for such delayed neural networks. Comparing with some previous literature, in which the time-varying delays were assumed to be differentiable and their derivatives were simultaneously required to be not greater than 1, the restrictions on the time-varying delays are removed. Therefore, our results obtained here improve and extend some previously related results. Finally, two numerical examples are provided to illustrate our theorems.

Algorithms↗
PubMed2004

Applying an artificial neural network to warfarin maintenance dose prediction.

BACKGROUND: Oral anticoagulation with warfarin can lead to life-threatening events as a result of either over-anticoagulation or undertreatment. One of the main contributors to an undesirable warfarin effect is the need to adjust its daily dose for a specific patient. The dose is adjusted empirically based on the experience of the clinician, a method that is often imprecise. There is currently no other well-accepted method for predicting the maintenance dose of warfarin. OBJECTIVE: To describe the application of an artificial neural network to the problem of warfarin maintenance dose prediction. METHODS: We designed a neural network that predicts the maintenance dose of warfarin. Data on 148 patients attending a large anticoagulant clinic were collected by file review. Using correlational analysis of the patients' data we selected the best input variables. The network was trained by using the back-propagation algorithm on a subset of our data and the results were validated against the rest of the data. We used a multivariate linear regression to create a comparable model. RESULTS: The neural network generated reasonable predictions of the maintenance dose (r = 0.823). The results of the linear regression model were similar (r = 0.800). CONCLUSION: Neural networks can be applied successfully for warfarin maintenance dose prediction. The results are promising, but further investigation is needed.

Adolescent↗
PubMed2005

A neural network model of memory and higher cognitive functions.

I first describe a neural network model of associative memory in a small region of the brain. The model depends, unconventionally, on disinhibition of inhibitory links between excitatory neurons rather than long-term potentiation (LTP) of excitatory projections. The model may be shown to have advantages over traditional neural network models both in terms of information storage capacity and biological plausibility. The learning and recall algorithms are independent of network architecture, and require no thresholds or finely graded synaptic strengths. Several copies of this local network are then connected by means of many, weak, reciprocal, excitatory projections that allow one region to control the recall of information in another to produce behaviors analogous to serial memory, classical and operant conditioning, secondary reinforcement, refabrication of memory, and fabrication of possible future events. The network distinguishes between perceived and recalled events, and can predicate its response on the absence as well as the presence of particular stimuli. Some of these behaviors are achieved in ways that seem to provide instances of self-awareness and imagination, suggesting that consciousness may emerge as an epiphenomenon in simple brains.

Algorithms↗
PubMed1999

Effect of dominant features on neural network performance in the classification of mammographic lesions.

Two different classifiers, an artificial neural network (ANN) and a hybrid system (one step rule-based method followed by an artificial neural network) have been investigated to merge computer-extracted features in the task of differentiating between malignant and benign masses. A database consisting of 65 cases (38 malignant and 26 benign) was used in the study. A total of four computer-extracted features--spiculation, margin sharpness and two density-related measures--was used to characterize these masses. Results from our previous study showed that the hybrid system performed better than the ANN classifier. In our current study, to understand the difference between the two classifiers, we investigated their learning and decision-making processes by studying the relationships between the input features and the outputs. A correlation study showed that the outputs from the ANN-alone method correlated strongly with one of the input features (spiculation), yielding a correlation coefficient of 0.91, whereas the correlation coefficients (absolute value) for the other features ranged from 0.19 to 0.40. This strong correlation between the ANN output and spiculation measure indicates that the learning and decision-making processes of the ANN-alone method were dominated by the spiculation measure. Three-dimensional plots of the computer output as functions of the input features demonstrate that the ANN-alone method did not learn as effectively as the hybrid system in differentiating non-spiculated malignant masses from benign masses, thus resulting in an inferior performance at the high sensitivity levels. We found that with a limited database it is detrimental for an ANN to learn the significance of other features in the presence of a dominant feature. The hybrid system, which initially applied a rule concerning the value of the spiculation measure prior to employing an ANN, prevents over-learning from the dominant feature and performed better than the ANN-alone method in merging the computer-extracted features into a correct diagnosis regarding the malignancy of the masses.

Breast Diseases↗
PubMed1996

Potential of the back propagation neural network in the discrimination of benign from malignant gastric cells.

OBJECTIVE: To investigate the potential value of morphometry and the back propagation neural network for the discrimination of benign and malignant gastric lesions. STUDY DESIGN: The study group consisted of 23 cases of cancer, 19 of gastritis and 58 of ulcer. Images of routinely processed gastric smears stained by the Papanicolaou technique were processed by a custom image analysis system. Analysis of the images gave a data set of 11,024 cells. Two different neural net architectures were used to classify benign from malignant cells based on the extracted morphometric and textural features. The data from 2,500 randomly selected cells were used as a training set, and the data from the remaining 8,524 cells were applied as a test set. RESULTS: Application of the back propagation neural network permitted the correct classification of 97.6% of benign cells and 95% of malignant cells with overall accuracy 97.3%. CONCLUSION: These results indicate that neural networks and image morphometry may offer useful information about the potential for malignancy in gastric cells.

Adult↗
PubMed2006

Neural networks for detection and classification of walking pattern changes due to ageing.

With age, gait functions reflected in the walking patterns degenerate and threaten the balance control mechanisms of the locomotor system. The aim of this paper is to explore applications of artificial neural networks for automated recognition of gait changes due to ageing from their respective gait-pattern characteristics. The ability of such discrimination has many advantages including the identification of at-risk or faulty gait. Various gait features (e.g., temporal-spatial, foot-ground reaction forces and lower limb joint angular data) were extracted from 12 young and 12 elderly participants during normal walking and these were utilized for training and testing on three neural network algorithms (Standard Backpropagation: Scaled Conjugate Gradient; and Backpropagation with Bayesian Regularization, BR). Receiver operating characteristics plots, sensitivity and specificity results as well as accuracy rates were used to evaluate performance of the three classifiers. Cross-validation test results indicate a maximum generalization performance of 83.3% in the recognition of the young and elderly gait patterns. Out of the three neural network algorithms, BR performed superiorly in the test results with best sensitivity, selectivity and detection rates. With the help of a feature selection technique, the maximum classification accuracy of the BR attained 100%, when trained with a small subset of selected gait features. The results of this study demonstrate the capability of neural networks in the detection of gait changes with ageing and their potentials for future applications as gait diagnostics.

Adult↗
PubMed1997

Neural network analysis of biochemical markers for early assessment of acute myocardial infarction.

Neural network analysis was applied for early diagnosis/exclusion of acute myocardial infarction and prediction of infarct size. Eighty-eight patients admitted with onset of chest pain within 8 hours were included. Frequent blood samples for measurement of myoglobin, CK-MB and troponin-T were obtained and used in the development of a set of neural network components of a decision support system. The results indicate that this approach could provide useful support for assessment of patients with suspected AMI.

Biomarkers↗
PubMed2005

A new method for sleep apnea classification using wavelets and feedforward neural networks.

OBJECTIVES: This paper presents a novel approach for sleep apnea classification. The goal is to classify each apnea in one of three basic types: obstructive, central and mixed. MATERIALS AND METHODS: Three different supervised learning methods using a neural network were tested. The inputs of the neural network are the first level-5-detail coefficients obtained from a discrete wavelet transformation of the samples (previously detected as apnea) in the thoracic effort signal. In order to train and test the systems, 120 events from six different patients were used. The true error rate was estimated using a 10-fold cross validation. The results presented in this work were averaged over 100 different simulations and a multiple comparison procedure was used for model selection. RESULTS: The method finally selected is based on a feedforward neural network trained using the Bayesian framework and a cross-entropy error function. The mean classification accuracy, obtained over the test set was 83.78+/-1.90%. CONCLUSION: The proposed classifier surpasses, up to the author's knowledge, other previous results. Finally, a scheme to maintain and improve this system during its clinical use is also proposed.

Algorithms↗
PubMed2003

[Study on meteorological factors-based neural network model of malaria].

OBJECTIVE: In order to provide reliable data for strategies development on prevention, a meteorological factors-based predicating model for malaria forecast was studied. METHODS: Data on malaria occurrence and climate changes from 1994 to 1999 in counties in Yunnan province was collected and analyzed with software packages of FoxPro 6.0 and Excel 5.0. The forecasting model for malaria occurrence was established, using the Neural Network Toolbox of Matlab 6.1 software package. In the studies of forecasting model, data of malaria and meteorological factors from 1994 to 1999 in Honghe state in Yunnan province was chosen. The meteorological factors included average monthly pressure, air temperature, relative humidity, monthly maximum air temperature, minimum air temperature, rainfall, rainday, evaporation and sunshine hours in the study. The established forecasting model was also tested and verified. RESULTS: The BP network model was established according to data of diseases and meteorological factors from Honghe state in Yunnan province. After training the neural network for 100 times, the error of performance decreased from 3.23608 to 0.035862. Verified by fact data of malaria, the efficiency of malaria forecasting was 84.85%. CONCLUSION: Neural network model was effective for forecasting malaria. It showed advantages as: strong ability for analysis, lower claim for data, convenient and easy to apply etc. Neural network model might be used as a new method for malaria forecasting.

Humans↗
PubMed1996

Application of artificial neural networks for the classification of liver lesions by image texture parameters.

Ultrasound imaging is a powerful tool for characterizing the state of soft tissues; however, in some cases, where only subtle differences in images are seen as in certain liver lesions such as hemangioma and malignancy, existing B-scan methods are inadequate. More detailed analyses of image texture parameters along with artificial neural networks can be utilized to enhance differentiation. From B-scan ultrasound images, 11 texture parameters comprising of first, second and run length statistics have been obtained for normal, hemangioma and malignant livers. Tissue characterization was then performed using a multilayered backpropagation neural network. The results for 113 cases have been compared with a classification based on discriminant analysis. For linear discriminant analysis, classification accuracy is 79.6% and with neural networks the accuracy is 100%. The present results show that neural networks classify better than discriminant analysis, demonstrating a much potential for clinical application.

Diagnosis, Differential↗
PubMed1999

Feedforward neural network with adaptive reference pattern layer.

A hybrid neural network architecture is investigated for modeling purposes. The proposed hybrid is based on the multilayer perceptron (MLP) network. In addition to the usual hidden layers, the first hidden layer is selected to be an adaptive reference pattern layer. Each unit in this new layer incorporates a reference pattern that is located somewhere in the space spanned by the input variables. The outputs of these units are the component wise-squared differences between the elements of a reference pattern and the inputs. The reference pattern layer has some resemblance to the hidden layer of the radial basis function (RBF) networks. Therefore the proposed design can be regarded as a sort of hybrid of MLP and RBF networks. The presented benchmark experiments show that the proposed hybrid can provide significant advantages over standard MLPs and RBFs in terms of fast and efficient learning, and compact network structure.

Feedback↗
PubMed2000

Predictions of coronary artery stenosis by artificial neural network.

Data from angiography patient records comprised 14 input variables of a neural network. Outcomes (coronary artery stenosis or none) formed both supervisory and output variables. The network was trained by backpropagation on 332 records, optimized on 331 subsequent records, and tested on final 100 records. If 0.40 was chosen as the output distinguishing stenosis from no stenosis, 81 patients who had stenosis would have been identified, while 9 of 19 patients who did not have stenosis might have been spared angiography. The results demonstrated that artificial neural networks could identify some patients who do not need coronary angiography.

Adult↗
PubMed1995

Neural network reconstruction of single-photon emission computed tomography images.

An artificial neural network (ANN) trained on high-quality medical tomograms or phantom images may be able to learn the planar data-to-tomographic image relationship with very high precision. As a result, a properly trained ANN can produce comparably accurate image reconstruction without the high computational cost inherent in some traditional reconstruction techniques. We have previously shown that a standard backpropagation neural network can be trained to reconstruct sections of single photon emission computed tomography (SPECT) images based on the planar image projections as inputs. In this study, we present a method of deriving activation functions for a backpropagation ANN that make it readily trainable for full SPECT image reconstruction. The activation functions used for this work are based on the estimated probability density functions (PDFs) of the ANN training set data. The statistically tailored ANN and the standard sigmoidal backpropagation ANN methods are compared both in terms of their trainability and generalization ability. The results presented show that a statistically tailored ANN can reconstruct novel tomographic images of a quality comparable with that of the images used to train the network. Ultimately, an adequately trained ANN should be able to properly compensate for physical photon transport effects, background noise, and artifacts while reconstructing the tomographic image.

Algorithms↗