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 703 records · Page 39Linked to original sources
PubMed2003

Classification of Fourier transform infrared microscopic imaging data of human breast cells by cluster analysis and artificial neural networks.

Cluster analysis and artificial neural networks (ANNs) are applied to the automated assessment of disease state in Fourier transform infrared microscopic imaging measurements of normal and carcinomatous immortalized human breast cell lines. K-means clustering is used to implement an automated algorithm for the assignment of pixels in the image to cell and non-cell categories. Cell pixels are subsequently classified into carcinoma and normal categories through the use of a feed-forward ANN computed with the Broyden-Fletcher-Goldfarb-Shanno training algorithm. Inputs to the ANN consist of principal component scores computed from Fourier filtered absorbance data. A grid search optimization procedure is used to identify the optimal network architecture and filter frequency response. Data from three images corresponding to normal cells, carcinoma cells, and a mixture of normal and carcinoma cells are used to build and test the classification methodology. A successful classifier is developed through this work, although differences in the spectral backgrounds between the three images are observed to complicate the classification problem. The robustness of the final classifier is improved through the use of a rejection threshold procedure to prevent classification of outlying pixels.

Algorithms↗
PubMed1992

Limits on alpha-helix prediction with neural network models.

Using a backpropagation neural network model we have found a limit for secondary structure prediction from local sequence. By including only sequences from whole alpha-helix and non-alpha-helix structures in our training and test sets--sequences spanning boundaries between these two structures were excluded--it was possible to investigate directly the relationship between sequence and structure for alpha-helix. A group of non-alpha-helix sequences, that was disrupting overall prediction success, was indistinguishable to the network from alpha-helix sequences. These sequences were found to occur at regions adjacent to the termini of alpha-helices with statistical significance, suggesting that potentially longer alpha-helices are disrupted by global constraints. Some of these regions spanned more than 20 residues. On these whole structure sequences, 10 residues in length, a comparatively high prediction success of 78% with a correlation coefficient of 0.52 was achieved. In addition, the structure of the input space, the distribution of beta-sheet in this space, and the effect of segment length were also investigated.

Amino Acid Sequence↗
PubMed2003

Quantitation in multianalyte overlapping peaks from capillary electrophoresis runs using artificial neural networks.

The potentiality of artificial neural networks for multicomponent analysis in unresolved peaks from capillary electrophoresis (CE) is evaluated. The system chosen consists of mixtures of three ebrotidine metabolites, which cannot be successfully separated by CE. Data selected for analysis consist of UV spectra taken at the maximum of the CE peak. The most dissimilar analyte, in terms of spectral differences, is accurately quantitated in any type of mixture with an overall prediction error of 5%. Because of the strong interference of the two most overlapped compounds, a preliminary procedure for spectral data filtering based on principal component analysis is performed to improve their quantitation.

Electrophoresis, Capillary↗
PubMed2006

Prediction of target range of intact parathyroid hormone in hemodialysis patients with artificial neural network.

The application of artificial neural network (ANN) to predict outcome and explore potential relationships among clinical data is increasing being used in many clinical scenarios. The aim of this study was to validate whether an ANN is a useful tool for predicting the target range of plasma intact parathyroid hormone (iPTH) concentration in hemodialysis patients. An ANN was constructed with input variables collected retrospectively from an internal validation group (n = 129) of hemodialysis patients. Plasma iPTH was the dichotomous outcome variable, either target group (150 ng/L 300 ng/L). After internal validation, the ANN was prospectively tested in an external validation group (n = 32) of hemodialysis patients. The final ANN was a multilayer perceptron network with six predictors including age, diabetes, hypertension, and blood biochemistries (hemoglobin, albumin, calcium). The externally validated ANN provided excellent discrimination as appraised by area under the receiver operating characteristic curve (0.83 +/- 0.11, p = 0.003). The Hosmer-Lemeshow statistic was 5.02 (p= 0.08 > 0.05) which represented a good-fit calibration. These results suggest that an ANN, which is based on limited clinical data, is able to accurately forecast the target range of plasma iPTH concentration in hemodialysis patients.

Female↗
PubMed1990

Neural networks applied to quantitative structure-activity relationship analysis.

An application of the neural network to quantitative structure-activity relationship (QSAR) analysis has been studied. The new method was compared with the linear multiregression analysis in various ways. It was found that the neural network can be a potential tool in the routine work of QSAR analysis. The mathematical relationship of operation between the neural network and the multiregression analysis was described. It was shown that the neural network can exceed the level of the linear multiregression analysis.

Animals↗
PubMed1997

Neural network visual recognition for automation of the microelectromechanical systems assembly.

A neural network visual recognition system is developed. The system is intended for automation of microsystem assembly process. The recognition of microparts is insensitive on their position. This feature is enabled by calculation of the moment properties of the image during preprocessing. The system takes grey-level images and produces recognition code as the output. A feed-forward neural network is used for recognition. Learning of the neural network is performed by combination of standard backpropagation and resilient propagation rule. The system performance is satisfactory with respect to recognition accuracy and recognition time.

Algorithms↗
PubMed2006

Advanced fuzzy cellular neural network: application to CT liver images.

OBJECTIVE: To achieve better boundary integrities and recall accuracies for segmented liver images, use of the advanced fuzzy cellular neural network (AFCNN), as a variant of the fuzzy cellular neural network (FCNN), is proposed to effectively segment CT liver images. MATERIALS AND METHODS: In order to better utilize relevant contour and gray information from liver images, we have improved the FCNN [Wang S, Wang M. A new algorithm NDA based on fuzzy cellular neural networks for white blood cell detection. IEEE Trans Inform Technol Biomed, in press], which proved to be very effective for the segmentation of microscopic white blood cell images, to create the novel neural network, AFCNN. Its convergent property and global stability are proved. Based on the FCNN-based NDA algorithm [Wang S, Wang M. A new algorithm NDA based on fuzzy cellular neural networks for white blood cell detection. IEEE Trans Inform Technol Biomed, in press], we developed the AFCNN-based NDA algorithm, which we used to segment 5 CT liver images. For comparison, we also segmented the same 5 CT liver images using the FCNN-based NDA algorithm. RESULTS AND CONCLUSION: : AFCNN has distinct advantages over FCNN in both boundary integrity and recall accuracy. In particular, the performance index Binary_rate is generally much higher for AFCNN than for FCNN when applied to CT liver images.

Fuzzy Logic↗
PubMedDate not supplied

A new algorithm for learning in piecewise-linear neural networks.

Piecewise-linear (PWL) neural networks are widely known for their amenability to digital implementation. This paper presents a new algorithm for learning in PWL networks consisting of a single hidden layer. The approach adopted is based upon constructing a continuous PWL error function and developing an efficient algorithm to minimize it. The algorithm consists of two basic stages in searching the weight space. The first stage of the optimization algorithm is used to locate a point in the weight space representing the intersection of N linearly independent hyperplanes, with N being the number of weights in the network. The second stage is then called to use this point as a starting point in order to continue searching by moving along the single-dimension boundaries between the different linear regions of the error function, hopping from one point (representing the intersection of N hyperplanes) to another. The proposed algorithm exhibits significantly accelerated convergence, as compared to standard algorithms such as back-propagation and improved versions of it, such as the conjugate gradient algorithm. In addition, it has the distinct advantage that there are no parameters to adjust, and therefore there is no time-consuming parameters tuning step. The new algorithm is expected to find applications in function approximation, time series prediction and binary classification problems.

Algorithms↗
PubMed1994

Computer-assisted image classification: use of neural networks in anatomic pathology.

Artificial neural networks (ANN) implemented on digital computers have received much attention for interpretation of images in pathology and cytology. Most such images are too complex for current ANN to interpret directly; instead, ANN usually classify the images according to numeric features extracted from them. In experiments on three distinct image classification problems, ANN classifiers performed as well or better than multivariate linear discriminant analysis (a traditional parametric statistical classifier). ANN empirically define non-linear multivariate decision boundaries, and can combine non-contiguous feature areas in mapping a classification. However, many training cases are required in order to map complex area boundaries precisely and with a low risk of 'overtraining.' Careful problem selection and attention to data dimensionality and format are important for efficient ANN use.

Adenocarcinoma↗
PubMed1998

Myoelectric signal classification using neural networks.

A feed-forward neural network is used for diagnosis of spastic paralysis. It is a two-layer perceptron and it is able to classify two kinds of myoelectric signal recorded in surface electromyography: the normal EMG and the EMG in the case of spastic paralysis. The myoelectric signal was recorded with a surface electrode pair and sampled at 10 kHz. The EMG activity is stochastic and the instantaneous amplitude distribution for a fixed level of contraction is Gaussian. The signal variance is considered a measure of muscle force. We can describe any kind of this process by the AR model. For a precisely modeling of EMG there are necessary many AR model parameters. In the classification problem we have it is not necessary to use a high order AR model. We find a 4-th order AR model is good enough for this study. The Hopfield algorithm is used to calculate the parameters of the autoregressive model.

Electrodes↗
PubMed2000

Stationary and integrated autoregressive neural network processes.

We consider autoregressive neural network (AR-NN) processes driven by additive noise and demonstrate that the characteristic roots of the shortcuts-the standard conditions from linear time-series analysis-determine the stochastic behavior of the overall AR-NN process. If all the characteristic roots are outside the unit circle, then the process is ergodic and stationary. If at least one characteristic root lies inside the unit circle, then the process is transient. AR-NN processes with characteristic roots lying on the unit circle exhibit either ergodic, random walk, or transient behavior. We also analyze the class of integrated AR-NN (ARI-NN) processes and show that a standardized ARI-NN process "converges" to a Wiener process. Finally, least-squares estimation (training) of the stationary models and testing for nonstationarity is discussed. The estimators are shown to be consistent, and expressions on the limiting distributions are given.

Artifacts↗
PubMed2005

QSAR study on the Ah receptor-binding affinities of polyhalogenated dibenzo-p-dioxins using net atomic-charge descriptors and a radial basis neural network.

A radial basis function neural network (RBFN) has been used to correlate Ah receptor-binding affinities of polychlorinated, polybrominated, and polychlorinated-brominated dibenzo-p-dioxins with molecular weight and eight net atomic charge descriptors. Support vector machine (SVM) and partial least square (PLS) regression models based on the same data set have also been built. Leave-one-out cross-validation was used to train the RBFN, SVM, and PLS models. For predicting Ah receptor-binding affinities, the RBFN model with a squared cross-validation correlation coefficient (q2) of 0.8818 outperforms the SVM and PLS models and also compares favorably with any other reported quantitative structure-activity relationship model based on the same activity data set. The significance of the RBFN model with net atomic charges as descriptors suggests that electrostatic and dispersion-type interactions play important roles in governing the Ah receptor binding of polychlorinated, polybrominated, and polychlorinated-brominated dibenzo-p-dioxins.

Algorithms↗
PubMed2006

Initial biopsy outcome prediction--head-to-head comparison of a logistic regression-based nomogram versus artificial neural network.

OBJECTIVES: Nomograms and artificial neural networks (ANNs) represent alternative methodologic approaches to predict the probability of prostate cancer on initial biopsy. We hypothesized that, in a head-to-head comparison, one of the approaches might demonstrate better accuracy and performance characteristics than the other. METHODS: A previously published nomogram, which relies on age, digital rectal examination, serum prostate-specific antigen (PSA), and percent-free PSA, and an ANN, which relies on the same predictors plus prostate volume, were applied to a cohort of 3980 men, who were subjected to multicore systematic prostate biopsy. The accuracy and the performance characteristics were compared between these two approaches. RESULTS: The accuracy of the nomogram was 71% versus 67% for the ANN (p=0.0001). Graphical exploration of the performance characteristics demonstrated virtually perfect predictions for the nomogram. Conversely, the ANN underestimated the observed rate of prostate cancer. CONCLUSIONS: A 4% increase in predictive accuracy implies that the use of the nomogram instead of the ANN will result in 40 additional patients who will be correctly classified between benign and cancer.

Adult↗
PubMed2000

[Column efficiency prediction of two dimensional chromatography by artificial neural network].

This paper used artificial neural network (ANN) modeling method to study the relationship between the column efficiency and the operating conditions. This method solved the problem that it is not easy to establish a quantitative model between the column efficiency and its main effecting factors using those traditional modeling methods, as the relationship is usually quite complex and non-linear in fact. The varied-pace BP (back-propagation) learning algorithm was adopted, and a three-layer weight-connected ANN model for a typical dual column system was established. The effective plate number representing the column efficiency acted as the output vectors, while the temperature of the pre-column, the temperature of the main column, the pressure difference between the columns and the vent rate acted as the input vectors. Then the model acquired was used to predict column efficiency (characterized by "effective plate number") under different operating conditions. The results showed that the model predicting value was in consistent with the value found. This work proved that ANN modeling method was suitable for the study on the relationship between the column efficiency of two-dimensional column chromatography system and the operating conditions.

Algorithms↗
PubMed2005

Genetic neural network modeling of the selective inhibition of the intermediate-conductance Ca2+ -activated K+ channel by some triarylmethanes using topological charge indexes descriptors.

Selective inhibition of the intermediate-conductance Ca(2+)-activated K(+ )channel (IK (Ca)) by some clotrimazole analogs has been successfully modeled using topological charge indexes (TCI) and genetic neural networks (GNNs). A neural network monitoring scheme evidenced a highly non-linear dependence between the IK (Ca) blocking activity and TCI descriptors. Suitable subsets of descriptors were selected by means of genetic algorithm. Bayesian regularization was implemented in the network training function with the aim of assuring good generalization qualities to the predictors. GNNs were able to yield a reliable predictor that explained about 97% data variance with good predictive ability. On the contrary, the best multivariate linear equation with descriptors selected by linear genetic search, only explained about 60%. In spite of when using the descriptors from the linear equations to train neural networks yielded higher fitted models, such networks were very unstable and had relative low predictive ability. However, the best GNN BRANN 2 had a Q ( 2 ) of LOO of cross-validation equal to 0.901 and at the same time exhibited outstanding stability when calculating 80 randomly constructed training/test sets partitions. Our model suggested that structural fragments of size three and seven have relevant influence on the inhibitory potency of the studied IK (Ca) channel blockers. Furthermore, inhibitors were well distributed regarding its activity levels in a Kohonen self-organizing map (KSOM) built using the inputs of the best neural network predictor.

Algorithms↗
PubMed2006

Global attraction and stability for Cohen-Grossberg neural networks with delays.

We consider a class of Cohen-Grossberg neural networks with delays. We prove the existence and global asymptotic stability of an equilibrium point and estimate the region of existence. Furthermore, we show that the trajectories of the neural networks with positive initial data will stay in the positive region if the amplification function satisfies a divergent condition. We also establish the existence of a globally attracting compact set for more general networks. We estimate this compact set explicitly in terms of the network parameters from physiological and biological models. Our results can be applied to neural networks with a wide range of activation functions which are neither bounded nor globally Lipschitz continuous such as the Lotka-Volterra model. We also give some examples and simulations.

Animals↗
PubMed1994

Analysis of linear and nonlinear QSAR data using neural networks.

The use of feed forward back propagation neural networks to perform the equivalent of multiple linear regression has been examined using artificial structured data sets and real literature data. Their predictive ability has been assessed using leave-one-out cross-validation and training/test set protocols. While networks have been shown to fit data sets well, they appear to suffer from a number of disadvantages. In particular, they have performed poorly in prediction for the QSAR data examined here, they are susceptible to chance effects, and the relationships developed by the networks are difficult to interpret. This investigation reports results for one particular form of artificial neural network; other architectures and applications, however, may be more suitable.

Neural Networks, Computer↗
PubMed2005

Motif-based protein sequence classification using neural networks.

We present a system for multi-class protein classification based on neural networks. The basic issue concerning the construction of neural network systems for protein classification is the sequence encoding scheme that must be used in order to feed the neural network. To deal with this problem we propose a method that maps a protein sequence into a numerical feature space using the matching scores of the sequence to groups of conserved patterns (called motifs) into protein families. We consider two alternative ways for identifying the motifs to be used for feature generation and provide a comparative evaluation of the two schemes. We also evaluate the impact of the incorporation of background features (2-grams) on the performance of the neural system. Experimental results on real datasets indicate that the proposed method is highly efficient and is superior to other well-known methods for protein classification.

Algorithms↗