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 1,153 records · Page 64Linked to original sources
PubMed2006

Application of artificial neural networks to assess pesticide contamination in shallow groundwater.

In this study, a feed-forward back-propagation neural network (BPNN) was developed and applied to predict pesticide concentrations in groundwater monitoring wells. Pesticide concentration data are challenging to analyze because they tend to be highly censored. Input data to the neural network included the categorical indices of depth to aquifer material, pesticide leaching class, aquifer sensitivity to pesticide contamination, time (month) of sample collection, well depth, depth to water from land surface, and additional travel distance in the saturated zone (i.e., distance from land surface to midpoint of well screen). The output of the neural network was the total pesticide concentration detected in the well. The model prediction results produced good agreements with observed data in terms of correlation coefficient (R=0.87) and pesticide detection efficiency (E=89%), as well as good match between the observed and predicted "class" groups. The relative importance of input parameters to pesticide occurrence in groundwater was examined in terms of R, E, mean error (ME), root mean square error (RMSE), and pesticide occurrence "class" groups by eliminating some key input parameters to the model. Well depth and time of sample collection were the most sensitive input parameters for predicting the pesticide contamination potential of a well. This infers that wells tapping shallow aquifers are more vulnerable to pesticide contamination than those wells tapping deeper aquifers. Pesticide occurrences during post-application months (June through October) were found to be 2.5 to 3 times higher than pesticide occurrences during other months (November through April). The BPNN was used to rank the input parameters with highest potential to contaminate groundwater, including two original and five ancillary parameters. The two original parameters are depth to aquifer material and pesticide leaching class. When these two parameters were the only input parameters for the BPNN, they were not able to predict contamination potential. However, when they were used with other parameters, the predictive performance efficiency of the BPNN in terms of R, E, ME, RMSE, and pesticide occurrence "class" groups increased. Ancillary data include data collected during the study such as well depth and time of sample collection. The BPNN indicated that the ancillary data had more predictive power than the original data. The BPNN results will help researchers identify parameters to improve maps of aquifer sensitivity to pesticide contamination.

Environmental Monitoring↗
PubMed1990

Neural networks in radiologic diagnosis. I. Introduction and illustration.

Artificial neural networks (NNs) process information in a manner similar to the way the human brain is thought to process information. Neural networks have potential application in radiology as an artificial intelligence technique that can provide computer-aided diagnostic assistance for the practicing radiologist. The basic characteristics of NNs and the manner in which information propagates through an NN are discussed in nontechnical language, to assist the diagnostic radiologist in understanding the basic principles of neurocomputing. Computer-aided diagnosis selection in pediatric chest radiography using NNs is discussed in a companion article.

Artificial Intelligence↗
PubMed2006

A bio-basis function neural network for protein peptide cleavage activity characterisation.

This paper presents a novel neural learning algorithm for analysing protein peptides which comprise amino acids as non-numerical attributes. The algorithm is derived from the radial basis function neural networks (RBFNNs) and is referred to as a bio-basis function neural network (BBFNN). The basic principle is to replace the radial basis function used by RBFNNs with a bio-basis function. Each basis in BBFNN is supported by a peptide. The bases collectively form a feature space, in which each basis represents a feature dimension. A linear classifier is constructed in the feature space for characterising a protein peptide in terms of functional status. The theoretical basis of BBFNN is that peptides, which perform the same function will have similar compositions of amino acids. Because of this, the similarity between peptides can have statistical significance for modelling while the proposed bio-basis function can well code this information from data. The application to two real cases shows that BBFNN outperformed multi-layer perceptrons and support vector machines.

Algorithms↗
PubMed2003

Using improved neural network model to analyze RSP, NOx and NO2 levels in urban air in Mong Kok, Hong Kong.

As the health impact of air pollutants existing in ambient addresses much attention in recent years, forecasting of air pollutant parameters becomes an important and popular topic in environmental science. Airborne pollution is a serious, and will be a major problem in Hong Kong within the next few years. In Hong Kong, Respirable Suspended Particulate (RSP) and Nitrogen Oxides NOx and NO2 are major air pollutants due to the dominant diesel fuel usage by public transportation and heavy vehicles. Hence, the investigation and prediction of the influence and the tendency of these pollutants are of significance to public and the city image. The multi-layer perceptron (MLP) neural network is regarded as a reliable and cost-effective method to achieve such tasks. The works presented here involve developing an improved neural network model, which combines the principal component analysis (PCA) technique and the radial basis function (RBF) network, and forecasting the pollutant levels and tendencies based in the recorded data. In the study, the PCA is firstly used to reduce and orthogonalize the original input variables (data), these treated variables are then used as new input vectors in RBF neural network model established for forecasting the pollutant tendencies. Comparing with the general neural network models, the proposed model possesses simpler network architecture, faster training speed, and more satisfactory predicting performance. This improved model is evaluated by using hourly time series of RSP, NOx and NO2 concentrations collected at Mong Kok Roadside Gaseous Monitory Station in Hong Kong during the year 2000. By comparing the predicted RSP. NOx and NO2 concentrations with the actual data of these pollutants recorded at the monitory station, the effectiveness of the proposed model has been proven. Therefore, in authors' opinion, the model presented in the paper is a potential tool in forecasting air quality parameters and has advantages over the traditional neural network methods.

Air Pollutants↗
PubMed1995

Neural network modeling of a dolphin's sonar discrimination capabilities.

The capability of an echolocating dolphin to discriminate differences in the wall thickness of cylinders was previously modeled by a counterpropagation neural network using only spectral information from the echoes. In this study, both time and frequency information were used to model the dolphin discrimination capabilities. Echoes from the same cylinders were digitized using a broadband simulated dolphin sonar signal with the transducer mounted on the dolphin's pen. The echoes were filtered by a bank of continuous constant-Q digital filters and the energy from each filter was computed in time increments of 1/bandwidth. Echo features of the standard and each comparison target were analyzed in pairs by a counterpropagation neural network, a backpropagation neural network, and a model using Euclidean distance measures. The backpropagation network performed better than both the counterpropagation network, and the Euclidean model, using either spectral-only features or combined temporal and spectral features. All models performed better using features containing both temporal and spectral information. The backpropagation network was able to perform better than the dolphins for noise-free echoes with Q values as low as 2 and 3. For a Q of 2, only temporal information was available. However, with noisy data, the network required a Q of 8 in order to perform as well as the dolphin.

Animals↗
PubMed1995

The cerebral hemispheres and neural network simulations: design considerations.

The conclusions concerning hemispheric specializations based on neural network simulations, which were previously reported by Kosslyn, Chabris, Marsolek, and Koenig (1992), are shown not to be valid. Differences in network performance on tasks said to be "categorical" and "coordinate spatial" in nature were due to imbalances in the input stimuli and cannot, in principle, be related to differences in performance on such tasks in human subjects. The use of truth tables and correlation coefficients in the design of neural networks is discussed.

Brain↗
PubMed2003

Combinative neural network and its applications.

A new approach named combinative neural network (CN) using partial least squares (PLS) analysis to modify the hidden layer in the multi-layered feed forward (MLFF) neural networks (NN) was proposed in this paper. The significant contributions of PLS in the proposed CN are to reorganize the outputs of hidden nodes such that the correlation of variables could be circumvented, to make the network meet the non-linear relationship best between the input and output data of the NN, and to eliminate the risk of over-fitting problem at the same time. The performance of the proposed approach was demonstrated through two examples, a well defined nonlinear approximation problem, and a practical nonlinear pattern classification problem with unknown relationship between the input and output data. The results were compared with those by conventional MLFF NNs. Good performance and time-saving implementation make the proposed method an attractive approach for non-linear mapping and classification.

Journal Article↗
PubMed2006

Application of artificial neural network in prediction of bladder outlet obstruction: a model based on objective, noninvasive parameters.

OBJECTIVES: An artificial neural network model previously described by us that was based on lower urinary tract symptoms yielded a modest prediction of bladder outlet obstruction. The aim of this study was to establish another model, using more objective parameters, that could better predict for bladder outlet obstruction. METHODS: The records of 457 patients were used in the construction of the model. Of the 457 records, 300 were allocated to the training phase and 157 to the testing phase. All patients had the average flow rate, maximal flow rate, postvoid residual urine volume (PVR), and total prostate volume recorded. The results of the pressure flow study of those patients were considered the reference standard against which the artificial neural network was tested. RESULTS: Three models were tested. Models 1 and 2 were based on a three-output design (ie, nonobstructed, equivocal, and obstructed). The only difference was the number of iterations. The accuracy of the first model was 60.5% compared with 46.5% for the second. For a third model, in which the equivocal pressure flow study results were added to the nonobstructed group, the accuracy rose to 72%. Deletion of equivocal cases (around 19% of the total) was associated with an accuracy of 76% in the prediction of obstruction. CONCLUSIONS: An artificial neural network based on objective and noninvasive parameters could replace the pressure flow study in only 72% of cases. An accuracy of 76% in the detection of bladder outlet obstruction is rather impractical, because an equivocal zone has always been available on pressure flow study nomograms.

Humans↗
PubMed2002

A probabilistic neural network approach for modeling and classification of bacterial growth/no-growth data.

In this paper, we propose to use probabilistic neural networks (PNNs) for classification of bacterial growth/no-growth data and modeling the probability of growth. The PNN approach combines both Bayes theorem of conditional probability and Parzen's method for estimating the probability density functions of the random variables. Unlike other neural network training paradigms, PNNs are characterized by high training speed and their ability to produce confidence levels for their classification decision. As a practical application of the proposed approach, PNNs were investigated for their ability in classification of growth/no-growth state of a pathogenic Escherichia coli R31 in response to temperature and water activity. A comparison with the most frequently used traditional statistical method based on logistic regression and multilayer feedforward artificial neural network (MFANN) trained by error backpropagation was also carried out. The PNN-based models were found to outperform linear and nonlinear logistic regression and MFANN in both the classification accuracy and ease by which PNN-based models are developed.

Bacteria↗
PubMed1992

Recognition of distantly related protein sequences using conserved motifs and neural networks.

A sensitive technique for protein sequence motif recognition based on neural networks has been developed. It involves three major steps. (1) At each appropriate alignment position of a set of N matched sequences, a set of N aligned oligopeptides is specified with preselected window length. N neural nets are subsequently and successively trained on N-1 amino acid spans after eliminating each ith oligopeptide. A test for recognition of each of the ith spans is performed. The average neural net recognition over N such trials is used as a measure of conservation for the particular windowed region of the multiple alignment. This process is repeated for all possible spans of given length in the multiple alignment. (2) The M most conserved regions are regarded as motifs and the oligopeptides within each are used to train intensively M individual neural networks. (3) The M networks are then applied in a search for related primary structures in a databank of known protein sequences. The oligopeptide spans in the database sequence with strongest neural net output for each of the M networks are saved and then scored according to the output signals and the proper combination that follows the expected N- to C-terminal sequence order. The motifs from the database with highest similarity scores can then be used to retrain the M neural nets, which can be subsequently utilized for further searches in the databank, thus providing even greater sensitivity to recognize distant familial proteins. This technique was successfully applied to the integrase, DNA-polymerase and immunoglobulin families.

Aldehyde Dehydrogenase↗
PubMed1999

Application of artificial neural networks for detection of abnormal fetal heart rate pattern: a comparison with conventional algorithms.

Cardiotocography signals were sampled during labour in 53 patients. A recurrent artificial neural network with hidden layer feedback was trained and performance was compared with that of several conventional systems. Correct and false positive rates of all systems tested were calculated. To ensure that the performance of neural networks was not just caused by using different cut-off levels, the threshold used for conventional methods were also adapted and optimised. The correct positives rate of neural networks was between 0.72 and 0.9, and the false positive rate between 0.2 and 0.4. Before optimising, conventional algorithms produced a very low correct positive (0.02-0.5) and a low false positive rate (0.0-0.08). After adjusting the parameters, the tested neural networks still performed better than optimised conventional systems.

Journal Article↗
PubMed1997

Recognition of patient anaesthetic levels: neural network systems, principal components analysis, and canonical discriminant variates.

The goal of this study was to examine the ability of Neural Networks to recognise the levels of anaesthetic state of a patient. Data obtained under different levels of anaesthesia have been modelled for the purpose. It is shown that inferential parameters can be used to recognise the levels of anaesthesia. In addition to demonstrating the ability of neural networks for classification we were interested in understanding the classification strategy discovered by the neural networks. Multivariate data analysis techniques, namely Principal Components Analysis and Canonical Discriminant Variates, were applied to analyse the resultant networks.

Anesthesia↗
PubMed1998

Serial VCG/ECG analysis using neural networks.

Serial ECG analysis is an important diagnostic tool in which two or more successive ECG recordings from the same patient are compared in order to find changes due to, e.g. myocardial infarction. The present study investigates a new approach to serial analysis which is based on artificial neural networks. Interrecording changes are sometimes falsely detected due to electrode misplacement or positional changes of the heart. In order to compensate for such problems, a new technique for VCG loop alignment was employed. A study population of 1000 patients with two recordings was used and manually scrutinized by three experienced ECG interpreters. Pathological changes indicating newly developed infarcts were found in 256 patients. Different combinations of VCG/ECG measurements served as input data to the neural network. The best performance of the neural network was obtained when ECG and VCG measurements were combined and the resulting sensitivity was 69% at a specificity of 90%. The use of only ECG or VCG measurements reduced the sensitivity to 63% and 60%, respectively. The results indicated that serial analysis based on neural networks did not improve significantly when VCG loop alignment was included.

Case-Control Studies↗
PubMed1999

Predicting active pulmonary tuberculosis using an artificial neural network.

BACKGROUND: Nosocomial outbreaks of tuberculosis (TB) have been attributed to unrecognized pulmonary TB. Accurate assessment in identifying index cases of active TB is essential in preventing transmission of the disease. OBJECTIVES: To develop an artificial neural network using clinical and radiographic information to predict active pulmonary TB at the time of presentation at a health-care facility that is superior to physicians' opinion. DESIGN: Nonconcurrent prospective study. SETTING: University-affiliated hospital. PARTICIPANTS: A derivation group of 563 isolation episodes and a validation group of 119 isolation episodes. INTERVENTIONS: A general regression neural network (GRNN) was used to develop the predictive model. MEASUREMENTS: Predictive accuracy of the neural network compared with clinicians' assessment. RESULTS: Predictive accuracy was assessed by the c-index, which is equivalent to the area under the receiver operating characteristic curve. The GRNN significantly outperformed the physicians' prediction, with calculated c-indices (+/- SEM) of 0.947 +/- 0.028 and 0.61 +/- 0.045, respectively (p < 0.001). When the GRNN was applied to the validation group, the corresponding c-indices were 0. 923 +/- 0.056 and 0.716 +/- 0.095, respectively. CONCLUSION: An artificial neural network can identify patients with active pulmonary TB more accurately than physicians' clinical assessment.

AIDS-Related Opportunistic Infections↗
PubMed1997

Detection of malignancy associated changes in cervical cell nuclei using feed-forward neural networks.

Normal cells in the presence of a precancerous lesion undergo subtle changes of their DNA distribution when observed by visible microscopy. These changes have been termed Malignancy Associated Changes (MACs). Using statistical models such as neural networks and discriminant functions it is possible to design classifiers that can separate these objects from truly normal cells. The correct classification rate using feed-forward neural networks is compared to linear discriminant analysis when applied to detecting MACs. Classifiers were designed using 53 nuclear features calculated from images for each of 25,360 normal appearing cells taken from 344 slides diagnosed as normal or containing severe dysplasia. A linear discriminant function achieved a correct classification rate of 61.6% on the test data while neural networks scored as high as 72.5% on a cell-by-cell basis. The cell classifiers were applied to a library of 93,494 cells from 395 slides, and the results were jackknifed using a single slide feature. The discriminant function achieved a correct classification rate of 67.6% while the neural networks managed as high as 76.2%.

Female↗
PubMed2005

Sesquiterpene lactone-based classification of three Asteraceae tribes: a study based on self-organizing neural networks applied to chemosystematics.

This work describes an application of artificial neural networks on a small data set of sesquiterpene lactones (STLs) of three tribes of the family Asteraceae. Structurally different types of representative STLs from seven subtribes of the tribes Eupatorieae, Heliantheae and Vernonieae were selected as input data for self-organizing neural networks. Encoding the 3D molecular structures of STLs and their projection onto Kohonen maps allowed the classification of Asteraceae into tribes and subtribes. This approach allowed the evaluation of structural similarities among different sets of 3D structures of sesquiterpene lactones and their correlation with the current taxonomic classification of the family. Predictions of the occurrence of STLs from a plant species according to the taxa they belong to were also performed by the networks. The methodology used in this work can be applied to chemosystematic or chemotaxonomic studies of Asteraceae.

Asteraceae↗
PubMed2003

[Simultaneous determination of tryptophan and tyrosine in injection by artificial neural network and fluorescence spectrum method].

A new fluorescence method by artificial neural network for the simultaneous determination of tryptophan and tyrosine was developed. The determination was carried out in the KH2PO4-K2HPO4 buffer solution (pH = 7.15) and at the EX (excitation) wavelength of 224 nm. In the range of 290-400 nm, the fluorescence intensities at fourteen wavelengths were taken as characteristic parameters of the artificial neural network, and the samples were arranged by the method of equality design. The mean recoveries of tryptophan and tyrosine were 100.9% and 101.6% respectively. The RSDs of the results were 4.18% and 4.17%. The method has been applied to the determination of tryptophan and tyrosine in compound amino acid injection, and the relative errors were 4.0% and 2.6%, respectively. The results were better than those of other networks in training speed and accuracy. In conclusion, the new network spectrofluorimetry is a good choice for multicomponent resolving analysis.

Amino Acids↗
PubMed2002

Prediction of Caco-2 cell permeability using a combination of MO-calculation and neural network.

In the present study, we developed an approach involving a combination of molecular orbital (MO) calculation and neural network to predict Caco-2 cell permeability (logP(app)) from the molecular structure of compounds. For a total of 87 compounds with logP(app) values obtained from the literature, three-dimensional molecular structures were determined by MO-calculation, and then five molecular descriptors were obtained, namely, the dipole moment, polarizability, sum of charges of nitrogen atoms (sum(N)), oxygen atoms (sum(O)), and hydrogen atoms bonding to nitrogen or oxygen atoms (sum(H)). The correlation between these five molecular descriptors and logP(app) was analyzed using a feed-forward back-propagation neural network with a configuration of 5-4-1 for input, hidden, and output layers found suitable for predicting Caco-2 cell permeability. A leave-one-out cross-validation procedure revealed that the neural network model possesses a fairly good predictability as far as Caco-2 cell permeability is concerned (predictive root mean square error (RMSE)=0.507), and better than the simple and quadratic regression model (predictive RMSE=0.584 and 0.568, respectively).

Caco-2 Cells↗