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 1,369 records · Page 76Linked to original sources
PubMed1995

Extraction of fuzzy rules using neural networks with structure level adaptation--verification to the diagnosis of hepatobiliary disorders.

This paper presents the reasoning and learning method for fuzzy rules using structure level adaptation of neural networks. In a usual neural network mechanism, we can observe some behaviors during the learning process. Based on such behaviors of neuron activity, we can generate or annihilate the specified neurons respectively in hidden layer to achieve an overall good system. In the method that we have proposed, we have used a procedure to derive the neuron generation/annihilation automatically, and applied such a procedure to the learning system where the experimental data related to hepatobiliary disorders were used. After learning by using randomly chosen data, the proposed system correctly diagnosed over 70% of cases. According to these results, we can find that fuzzy rules have some relationship with the degree of the input weight vector. As a result, we can assume that fuzzy rules for hepatobiliary disorders are extracted from this learned network.

Algorithms↗
PubMed1991

Evaluation of a parallel implementation of the learning portion of the backward error propagation neural network: experiments in artifact identification.

Various methods have been proposed in an attempt to solve problems in artifact and/or alarm identification including expert systems, statistical signal processing techniques, and artificial neural networks (ANN). ANNs consist of a large number of simple processing units connected by weighted links. To develop truly robust ANNs, investigators are required to train their networks on huge training data sets, requiring enormous computing power. We implemented a parallel version of the backward error propagation neural network training algorithm in the widely portable parallel programming language C-Linda. A maximum speedup of 4.06 was obtained with six processors. This speedup represents a reduction in total run-time from approximately 6.4 hours to 1.5 hours. We conclude that use of the master-worker model of parallel computation is an excellent method for obtaining speedups in the backward error propagation neural network training algorithm.

Algorithms↗
PubMed2004

Global asymptotic stability of Hopfield neural network involving distributed delays.

In the paper, we study dynamical behaviors of Hopfield neural networks system with distributed delays. Some new criteria ensuring the existence and uniqueness, and the global asymptotic stability (GAS) of equilibrium point are derived. In the results, we do not assume that the signal propagation functions satisfy the Lipschitz condition and do not require them to be bounded, differentiable or strictly increasing. Moreover, the symmetry of the connection matrix is not also necessary. Thus, we improve some previous works of other researchers. These conditions are presented in terms of system parameters and have importance leading significance in designs and applications of the GAS for Hopfield neural networks system with distributed delays. Two examples are also worked out to demonstrate the advantages of our results.

Computer Simulation↗
PubMed2000

A hierarchical neural network algorithm for robust and automatic windowing of MR images.

A novel hierarchical neural network based algorithm for automatic adjustment of display window width and center for a wide range of magnetic resonance (MR) images is presented in this paper. The algorithm consists of a feature generator utilizing both wavelet histogram and compact spatial statistical information computed from a MR image, a competitive layer based neural network for clustering MR images into different subclasses, two pairs of a radial basis function (RBF) network and a bi-modal linear estimator for each subclass, as well as a data fusion process using estimates from both estimators to compute the final display parameters. Both estimators can adapt to new kinds of MR images simply by training them with those images, which make the algorithm adaptive and extendable. The RBF based estimator performs very well for images that are similar to those in the training data set. The bi-modal linear estimator provides reasonable estimations for a wide range of images that may not be included in the training data set. The data fusion step makes the final estimation of the display parameters accurate for trained images and robust for the unknown images. The algorithm has been tested on a wide range of MR images and has shown satisfactory results.

Algorithms↗
PubMed2001

The roles of co-transmission in neural network modulation.

Neuromodulation provides considerable flexibility to the output of neural networks. In spite of the extensive literature documenting the presence of modulatory peptide co-transmitters in many neurons, considerably less is known about the specific roles of co-transmission in circuit function. This review describes some of the potential consequences of peptide co-transmission in functional circuits, using specific examples from recent work on the actions of identified peptidergic projection neurons acting on the multifunctional neural network within the crustacean stomatogastric ganglion. This system reveals that co-transmission provides projection neurons with a rich assortment of strategies for eliciting multiple outputs from a multifunctional network.

Animals↗
PubMedDate not supplied

Logistic regression and artificial neural network classification models: a methodology review.

Logistic regression and artificial neural networks are the models of choice in many medical data classification tasks. In this review, we summarize the differences and similarities of these models from a technical point of view, and compare them with other machine learning algorithms. We provide considerations useful for critically assessing the quality of the models and the results based on these models. Finally, we summarize our findings on how quality criteria for logistic regression and artificial neural network models are met in a sample of papers from the medical literature.

Classification↗
PubMed2001

Volume learning algorithm artificial neural networks for 3D QSAR studies.

The current study introduces a new method, the volume learning algorithm (VLA), for the investigation of three-dimensional quantitative structure-activity relationships (QSAR) of chemical compounds. This method incorporates the advantages of comparative molecular field analysis (CoMFA) and artificial neural network approaches. VLA is a combination of supervised and unsupervised neural networks applied to solve the same problem. The supervised algorithm is a feed-forward neural network trained with a back-propagation algorithm while the unsupervised network is a self-organizing map of Kohonen. The use of both of these algorithms makes it possible to cluster the input CoMFA field variables and to use only a small number of the most relevant parameters to correlate spatial properties of the molecules with their activity. The statistical coefficients calculated by the proposed algorithm for cannabimimetic aminoalkyl indoles were comparable to, or improved, in comparison to the original study using the partial least squares algorithm. The results of the algorithm can be visualized and easily interpreted. Thus, VLA is a new convenient tool for three-dimensional QSAR studies.

Algorithms↗
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

Potential assessment of a neural network model with PCA/RBF approach for forecasting pollutant trends in Mong Kok urban air, Hong Kong.

The forecasting of air pollutant trends has received much attention in recent years. It is an important and popular topic in environmental science, as concerns have been raised about the health impacts caused by unacceptable ambient air pollutant levels. Of greatest concern are metropolitan cities like Hong Kong. In Hong Kong, respirable suspended particulates (RSP), nitrogen oxides (NOx), and nitrogen dioxide (NO2) are major air pollutants due to the dominant usage of diesel fuel by commercial vehicles and buses. Hence, the study of the influence and the trends relating to these pollutants is extremely significant to the public health and the image of the city. The use of neural network techniques to predict trends relating to air pollutants is regarded as a reliable and cost-effective method for the task of prediction. The works reported here involve developing an improved neural network model that combines both the principal component analysis technique and the radial basis function network and forecasts pollutant tendencies based on a recorded database. Compared with general neural network models, the proposed model features a more simple network architecture, a faster training speed, and a more satisfactory prediction performance. The improved model was evaluated with hourly time series of RSP, NOx and NO2 concentrations monitored at the Mong Kok Roadside Gaseous Monitory Station in Hong Kong during the year 2000 and proved to be effective. The model developed is a potential tool for forecasting air quality parameters and is superior to traditional neural network methods.

Air Pollutants↗
PubMed2004

Artificial neural networks (ANNs) for differential diagnosis of interstitial lung disease: results of a simulation test with actual clinical cases.

RATIONALE AND OBJECTIVES: To evaluate the performance of an artificial neural network (ANN) scheme with use of consecutive clinical cases and its effect on radiologists with an observer test. MATERIALS AND METHODS: Artificial neural networks were designed to distinguish among 11 interstitial lung diseases on the basis of 26 inputs (16 radiologic findings, 10 clinical parameters). Chest radiographs of 96 consecutive cases with interstitial lung disease were used. Five radiologists independently rated their radiologic findings on the 96 chest radiographs. Based on their ratings of radiologic findings and clinical parameters obtained from the hospital information system, the output values indicating the likelihood of each of the 11 interstitial lung diseases were determined. Subsequently, 30 cases were selected from these 96 cases for an observer test. Five radiologists marked their confidence levels for diagnosis of 11 possible diseases in each case without and with ANN output. The performance of ANNs and radiologists was evaluated by receiver operating characteristic analysis based on their outputs and on confidence levels, respectively. RESULTS; The average Az value (area under the receiver operating characteristic curve) indicating ANN performance for the 96 consecutive cases was 0.85 +/- 0.03. The average Az values indicating radiologists' performance without and with ANN outputs were 0.81 +/- 0.11 and 0.87 +/- 0.06, respectively. The diagnostic accuracy was improved significantly when radiologists read chest radiographs with ANN outputs (P < .05). CONCLUSION: Artificial neural networks for differential diagnosis of interstitial lung disease may be useful in clinical situations, and radiologists may be able to utilize the ANN output to their advantage in the differential diagnosis of interstitial lung disease on chest radiographs.

Adult↗
PubMed2001

A use of a neural network to evaluate contrast enhancement curves in breast magnetic resonance images.

For the diagnosis of breast cancer using magnetic resonance imaging (MRI), one of the most important parameters is the analysis of contrast enhancement. A three-dimensional MR sequence is applied before and five times after bolus injection of paramagnetic contrast medium (Gd-DTPA). The dynamics of absorption are described by a time/intensity enhancement curve, which reports the mean intensity of the MR signal in a small region of interest (ROI) for about 8 minutes after contrast injection. The aim of our study was to use an artificial neural network to automatically classify the enhancement curves as "benign" or "malignant." We used a classic feed-forward back-propagation neural network, with three layers: five input nodes, two hidden nodes, and one output node. The network has been trained with 26 pathologic curves (10 invasive carcinoma [K], two carcinoma-in-situ [DCIS], and 14 benign lesion [B]). The trained network has been tested with 58 curves (36 K, one DCIS, 21 B). The network was able to correctly identify the test curves with a sensitivity of 76% and a specificity of 90%. For comparison, the same set of curves was analyzed separately by two radiologists (a breast MR expert and a resident radiologist). The first correctly interpreted the curves with a sensitivity of 76% and a specificity of 90%, while the second scored 59% for sensitivity and 90% for specificity. These results demonstrate that a trained neural network recognizes the pathologic curves at least as well as an expert radiologist. This algorithm can help the radiologist attain rapid and affordable screening of a large number of ROIs. A complete automatic computer-aided diagnosis support system should find a number of potentially interesting ROIs and automatically analyze the enhancement curves for each ROI by neural networks, reporting to the radiologist only the potentially pathologic ROIs for a more accurate, manual, repeated evaluation.

Breast↗
PubMed2025

Detecting Interspecific Positive Selection Using Convolutional Neural Networks.

Traditional statistical methods using maximum likelihood and Bayesian inference can detect positive selection from an interspecific phylogeny and a codon sequence alignment based on model assumptions, but they are prone to false positives due to alignment errors and can lack power. These problems are particularly pronounced when faced with high levels of indels and divergence. To address these issues, we trained and tested convolutional neural network models on simulated data and achieved higher accuracy in detecting selection across a specific range of phylogenetic scenarios and evolutionary modes. This advantage is particularly evident when performing inference on noisy data prone to misalignments. Our method shows some ability to account for these errors, where most statistical frameworks fail to do so in a tractable manner. We explore the generalizability of our convolutional neural network models to unseen evolutionary scenarios and identify future avenues to achieve broader utility. Once trained, our convolutional neural network model is faster at test time, making it a scalable alternative to traditional statistical methods for large-scale, multigene analyses. In addition to binary classification (inference of the presence or absence of positive selection during the evolution of the sequences), we use saliency maps to understand what the model learns and observe how this could be leveraged for sitewise inference of positive selection.

Neural Networks, Computer↗
PubMed2002

The use of a neural network to forecast daily grass pollen concentration in a Mediterranean region: the southern part of the Iberian Peninsula.

BACKGROUND: Pollen allergy is a common disease causing hayfever in 15% of the population in Europe. Medical studies report that a prior knowledge of pollen content in the air can be useful in the management of pollen-related diseases. OBJECTIVES: The aim of this work was to forecast daily Poaceae pollen concentrations in the air by using meteorological data and pollen counts from previous days as independent variables. METHODS: Linear regression models and co-evolutive neural network models were used for this study. Pollen was monitored by a Hirst-type spore trap using standard techniques. The data were obtained from the Spanish Aerobiology Network database, University of Cordoba Monitoring Unit. The set of data includes a series of 20 years, from 1982 to 2001. A classification of the years according to their allergenic potential was made using a K-mean cluster analysis with pollen and meteorological parameters. Statistical analysis was applied to all the years of each class with the exception of the most recent year, which was used for model validation. RESULTS: It was observed that cumulative variables and pollen values from previous days are the most important factors in the models. In general, neural network equations produce better results than linear regression equations. CONCLUSION: Co-evolutive neural network models, which obtain the best forecasts (an almost 90% "good" classification), make it possible to predict daily airborne Poaceae pollen concentrations. This new system based on neural network models is a step toward the automation of the pollen forecast process.

Environmental Pollution↗
PubMed2006

[Building artificial neural networks model on portable NIR integrity wheat component measuring apparatus].

The quantitative analysis model of protein in integrity wheat was built by three layers back propagation artificial neural networks for portable near infrared (NIR) integrity wheat component measuring apparatus. The structure diagram of integrity wheat component measuring apparatus, light route structure of apparatus and the spectrum of integrity wheat were given in the present paper. The theory of artificial neural network was briefly introduced and the results of quantitative analysis model of protein were given. For calibration set and prediction set, the correlation coefficient was 0.90 and 0.96 respectively; the relative standard deviation is 3.77% and 4.46% respectively. Because of the influence of light route structure, electrical circuit, and integrity sample forms on the measuring apparatus, some nonlinearity exists between the spectral parameters and chemical values. The results of artificial neural networks nonlinear model were superior to linear model.

Neural Networks, Computer↗
PubMed1994

Removing the assumption of conditional independence from Bayesian decision models by using artificial neural networks: some practical techniques and a case study.

The article describes how artificial neural networks with special designs can be applied to approximate a subjective Bayesian decision model without the assumption of conditional independence. New techniques are proposed to resolve some of the practical difficulties during the processes of problem structuring, knowledge elicitation, quantitative modeling, and model interpretation. A Bayesian model considering the conditional dependencies to predict a teenager's marijuana use was constructed by experts using these techniques, and compared to another conventional Bayesian model which assumed conditional independence. The new approach without the assumption of conditional independence had predictive power (r = 0.7) in the test of linearity compared to the conventional approach (r = 0.58) on a data set (n = 129). Its receiver operating characteristic curve dominated the alternative approach within the range (true positive fraction > 0.7) that we were interested in. The interpretations of the possible conditional dependencies provided by the artificial neural network after the training process were consistent with the expert's descriptions.

Adolescent↗
PubMed2000

Use of Neural Networks to Correlate Spine and Rib Deformity in Scoliosis.

Artificial neural networks (ANN's) recognize patterns relating input and output data in a manner analogous to the function of biological neurons. Here, we show that ANN's can predict rib deformity in scoliosis more accurately than regression analysis. ANN's and linear regression models were developed to predict rib rotation from several combinations of input spinal indices including Cobb angle, vertebral rotation, apex location and orientation of the plane of maximal curvature. ANN's averaged 60% correct predictions compared to 34% for regression analysis. This study provides evidence for the utility of artificial neural networks in scoliosis research. These data lend credence to the use of ANN's in future work on the prediction of scoliotic spinal deformity from torso surface data, which would permit assessment of scoliosis severity with minimal use of harmful X-rays.

Journal Article↗
PubMed1999

A comparison of ICU mortality prediction using the APACHE II scoring system and artificial neural networks.

The aim of this study was to compare the ability of artificial neural networks and the Acute Physiology and Chronic Health Evaluation II score to predict mortality in adult intensive care units. The same physiological variables were used in both predictive models to predict hospital mortality from a data set of 8796 patients collected from 26 adult intensive care units in the United Kingdom and Ireland as part of the Intensive Care Society study. The results from the two models were compared with the actual outcome. The overall prediction accuracy and the overall goodness-of-fit of all the models were assessed. Both predictive models showed similar goodness-of-fit and prediction discrimination. The overall predictive and classification performance of the artificial neural network developed matched and in some aspects was better than that of Acute Physiology and Chronic Health Evaluation II.

APACHE↗
PubMed2006

An artificial neural network approach to the drug efficacy of interferon treatments.

INTRODUCTION: Interferon taken alone or in combination with ribavirin can be used for the treatment of persons with chronic hepatitis C. It is highly desirable, both clinically and economically, to establish tools to distinguish responders from nonresponders and to predict possible outcomes of the treatments. In this work, our goal is to develop a prediction model resulting from the analysis of chronic hepatitis C patients' single nucleotide polymorphisms, viral genotype, viral load, age and gender, to predict the responsiveness of interferon combination treatment. MATERIALS AND METHODS: We collected blood samples from 523 chronic hepatitis C patients that had received interferon and ribavirin combination therapy. Based on the current treatment strategy for chronic hepatitis C patients, we focused our search for candidate genes involved in pathways related to interferon signaling and immunomodulation. With artificial neural network algorithms, we then developed pattern recognition methodologies to achieve predictions among the patients. The artificial neural network model was trained by an algorithm with an adaptive momentum and learning rate. RESULTS: There were seven single nucleotide polymorphisms selected from six candidate genes including adenosine deaminase-RNA-specific, caspase 5, interferon consensus sequence binding protein 1, interferon-induced protein 44, phosphoinositide-3-kinase catalytic gamma polypeptide and transporter 2 ATP-binding cassette subfamily B genes. We further applied the artificial neural network algorithms with these seven single nucleotide polymorphisms, viral genotype, viral load, age and gender information to build tools for predicting the responsiveness of interferon. Based on the fivefold cross-validation method to evaluate the performance, the model achieved a high success rate of prediction. CONCLUSION: We demonstrated that a trained artificial neural network model is a promising method for providing the inference from factors such as single nucleotide polymorphisms, viral genotype, viral load, age and gender to the responsiveness of interferon.

Aging↗