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 1,657 records · Page 92Linked to original sources
PubMed2007

Support vector machine based training of multilayer feedforward neural networks as optimized by particle swarm algorithm: application in QSAR studies of bioactivity of organic compounds.

Multilayer feedforward neural networks (MLFNNs) are important modeling techniques widely used in QSAR studies for their ability to represent nonlinear relationships between descriptors and activity. However, the problems of overfitting and premature convergence to local optima still pose great challenges in the practice of MLFNNs. To circumvent these problems, a support vector machine (SVM) based training algorithm for MLFNNs has been developed with the incorporation of particle swarm optimization (PSO). The introduction of the SVM based training mechanism imparts the developed algorithm with inherent capacity for combating the overfitting problem. Moreover, with the implementation of PSO for searching the optimal network weights, the SVM based learning algorithm shows relatively high efficiency in converging to the optima. The proposed algorithm has been evaluated using the Hansch data set. Application to QSAR studies of the activity of COX-2 inhibitors is also demonstrated. The results reveal that this technique provides superior performance to backpropagation (BP) and PSO training neural networks.

Algorithms↗
PubMed2004

Investigation of color constancy with a neural network.

The perceptual stability of an object's color under different illuminants is called color constancy. We created a neural network to investigate this phenomenon. The net consisted of one input channel for the background and one for the test object. Each channel had a set of three (L, M, and S) receptors that were transmitting to three opponent neurons. The signals from the opponent neurons were passed to hidden neurons, which were connected to the output neurons. The output signal was generated from the three components of a color vector. The neural net was trained to identify the color of Munsell samples under various illuminants using the back-propagation algorithm. Our study investigated the properties of a successfully trained neural network. Based on the cross-neuron weight analysis, we report that the successfully trained neural net calculates color differences between the test object and the background. By comparing the human visual system to the neural net, we conclude that to satisfy the color constancy phenomenon, the human visual system has to contain two separate components: one to approximate the background color and the other to estimate the color difference between the object and the background.

Color↗
PubMedDate not supplied

The impact of available experimental data on the prediction of 1H NMR chemical shifts by neural networks.

Two different ways were explored to incorporate new available experimental data into previously trained ensembles of feed-forward neural networks, for the structure-based prediction of (1)H NMR chemical shifts of organic compounds. One approach used the new data as the memory of an associative neural network (ASNN) system. For an independent prediction set of 952 cases, a mean average error of 0.19 ppm was achieved (0.13 ppm for 90% of the cases). This approach advantageously avoids retraining the networks, and the predictions compared favorably with those obtained by available commercial software packages. Excellent predictions could also be achieved by retraining the networks with the new data, but only if the training sets were selected so as to be balanced or if the retraining started with the weights of the previously trained networks.

Magnetic Resonance Spectroscopy↗
PubMed1979

Effect of boundaries on the response of a neural network.

The effect an abrupt boundary has upon the dynamical response of a neural network is investigated. The retina of the Limulus eye is used as a model system for studying this effect. A theoretical technique is presented for the quantitative prediction of the manner in which this neural network responds in the vicinity of its boundary. Corresponding experimental measurements of the response to moving stimuli by single optic neurons located near retinal boundaries are presented. Theory and experiment show detailed quantitative agreement.

Animals↗
PubMed1998

Peptide design by artificial neural networks and computer-based evolutionary search.

A technique for systematic peptide variation by a combination of rational and evolutionary approaches is presented. The design scheme consists of five consecutive steps: (i) identification of a "seed peptide" with a desired activity, (ii) generation of variants selected from a physicochemical space around the seed peptide, (iii) synthesis and testing of this biased library, (iv) modeling of a quantitative sequence-activity relationship by an artificial neural network, and (v) de novo design by a computer-based evolutionary search in sequence space using the trained neural network as the fitness function. This strategy was successfully applied to the identification of novel peptides that fully prevent the positive chronotropic effect of anti-beta1-adrenoreceptor autoantibodies from the serum of patients with dilated cardiomyopathy. The seed peptide, comprising 10 residues, was derived by epitope mapping from an extracellular loop of human beta1-adrenoreceptor. A set of 90 peptides was synthesized and tested to provide training data for neural network development. De novo design revealed peptides with desired activities that do not match the seed peptide sequence. These results demonstrate that computer-based evolutionary searches can generate novel peptides with substantial biological activity.

Amino Acid Sequence↗
PubMed2002

Predicting CNS permeability of drug molecules: comparison of neural network and support vector machine algorithms.

Two different machine-learning algorithms have been used to predict the blood-brain barrier permeability of different classes of molecules, to develop a method to predict the ability of drug compounds to penetrate the CNS. The first algorithm is based on a multilayer perceptron neural network and the second algorithm uses a support vector machine. Both algorithms are trained on an identical data set consisting of 179 CNS active molecules and 145 CNS inactive molecules. The training parameters include molecular weight, lipophilicity, hydrogen bonding, and other variables that govern the ability of a molecule to diffuse through a membrane. The results show that the support vector machine outperforms the neural network. Based on over 30 different validation sets, the SVM can predict up to 96% of the molecules correctly, averaging 81.5% over 30 test sets, which comprised of equal numbers of CNS positive and negative molecules. This is quite favorable when compared with the neural network's average performance of 75.7% with the same 30 test sets. The results of the SVM algorithm are very encouraging and suggest that a classification tool like this one will prove to be a valuable prediction approach.

Algorithms↗
PubMed1997

Automated interpretation of myocardial SPECT perfusion images using artificial neural networks.

UNLABELLED: The purpose of this study was to develop a computer-based method for automatic detection and localization of coronary artery disease (CAD) in myocardial bull's-eye scintigrams. METHODS: A population of 135 patients who had undergone both myocardial 99mTc-sestamibi rest-stress scintigraphy and coronary angiography within 3 mo was studied. Different image data reduction methods, including pixel averaging and two-dimensional Fourier transform, were applied to the bull's-eye scintigrams. After a quantitative and qualitative evaluation of these methods, 30 Fourier components were chosen as inputs to multilayer perceptron artificial neural networks. The networks were trained to detect CAD in two vascular territories, using coronary angiography as gold standard. A "leave one out" procedure was used for training and evaluation. The performance of the networks was compared to those of two human experts. RESULTS: One of the human experts detected CAD in one of two vascular territories, with a sensitivity of 54.4% at a specificity of 70.5%. The sensitivity of the networks was significantly higher at that level of specificity (77.2%, p = 0.0022). The other expert had a sensitivity of 63.2% at a specificity of 61.5%. The networks had a sensitivity of 77.2% (p = 0.038) at this specificity level as well. The differences in sensitivity between human experts and networks for the other vascular territory were all less than 6% and were not statistically significant. CONCLUSION: Artificial neural networks can detect CAD in myocardial bull's-eye scintigrams with such a high accuracy that the application of neural networks as clinical decision support tools appears to have significant potential.

Coronary Angiography↗
PubMed1999

Comparison of viral load and human leukocyte antigen statistical and neural network predictive models for the rate of HIV-1 disease progression across two cohorts of homosexual men.

We compared the performance of HIV-1 RNA and models based on human leukocyte antigen (HLA) in predicting the rate of HIV-1 disease progression using both linear regression and neural network models across two different cohorts of homosexual men. In all, 139 seroconverters from the Multicenter AIDS Cohort Study were used as the training set and 97 seroconverters from the District of Columbia Gay (DCG) cohort were used for validation to assess the generalizability of trained predictive models. Both viral load and HLA markers were strongly predictive of disease progression (p < .0001 and p = .001, respectively), with viral load superior to HLA (change in -2 log likelihood [-2LL] 26.7 and 10.2, respectively, in proportional hazards models). Consideration of both HLA markers and viral load offered no significant predictive advantage over viral load alone in most cases; however, HLA-based predictions obtained from neural networks modeling improved the discrimination among patients with high viral load (p = .02). Viral load, HLA scores, and rapid disease progression were moderately correlated (p < .01 for all three pairs of these variables). The median viral load was 10(3.70) copies/ml among DCG patients who had more favorable than unfavorable HLA markers and 10(4.66) copies/ml among patients with more unfavorable than favorable HLA markers. Viral load is a simpler, stronger predictor of disease progression than early developed HLA models, but neural network methods and further refined HLA models may offer additional prognostic information, especially for rapid progressors. The correlation between viral load and HLA markers suggests a possible HLA effect on setting viral load levels.

Cohort Studies↗
PubMed2002

Use of self-training artificial neural networks in modeling of gas chromatographic relative retention times of a variety of organic compounds.

A quantitative structure-activity relationship study based on multiple linear regression (MLR), artificial neural network (ANN), and self-training artificial neural network (STANN) techniques was carried out for the prediction of gas chromatographic relative retention times of 13 different classes of organic compounds. The five descriptors appearing in the selected MLR model are molecular density, Winer number, boiling point, polarizability and square of polarizability. A 5-6-1 ANN and a 5-4-1 STANN were generated using the five descriptors appearing in the MLR model as inputs. Comparison of the standard errors and correlation coefficients shows the superiority of ANN and STANN over the MLR model. This is due to the fact that the retention behaviors of molecules show non-linear characteristics. Inspection of the results of STANN and ANN shows there are few differences between these methods. However, optimization of STANN is much faster and the number of adjustable parameters for this technique is much less compared with those of the conventional ANN.

Chromatography, Gas↗
PubMed2004

Unsupervised fuzzy clustering using Weighted Incremental Neural Networks.

A new more efficient variant of a recently developed algorithm for unsupervised fuzzy clustering is introduced. A Weighted Incremental Neural Network (WINN) is introduced and used for this purpose. The new approach is called FC-WINN (Fuzzy Clustering using WINN). The WINN algorithm produces a net of nodes connected by edges, which reflects and preserves the topology of the input data set. Additional weights, which are proportional to the local densities in input space, are associated with the resulting nodes and edges to store useful information about the topological relations in the given input data set. A fuzziness factor, proportional to the connectedness of the net, is introduced in the system. A watershed-like procedure is used to cluster the resulting net. The number of the resulting clusters is determined by this procedure. Only two parameters must be chosen by the user for the FC-WINN algorithm to determine the resolution and the connectedness of the net. Other parameters that must be specified are those which are necessary for the used incremental neural network, which is a modified version of the Growing Neural Gas algorithm (GNG). The FC-WINN algorithm is computationally efficient when compared to other approaches for clustering large high-dimensional data sets.

Algorithms↗
PubMed1991

Applications of neural networks in quantitative structure-activity relationships of dihydrofolate reductase inhibitors.

Back propagation neural networks is a new technology useful for modeling nonlinear functions of several variables. This paper explores their applications in the field of quantitative structure-activity relationships. In particular, their ability to fit biological activity surfaces, predict activity, and determine the "functional forms" of its dependence on physical properties is compared to well-established methods in the field. A dataset of 256 5-phenyl-3,4-diamino-6,6-dimethyldihydrotriazines that inhibit dihydrofolate reductase enzyme is used as a basis for comparison. It is found that neural networks lead to enhanced surface fits and predictions relative to standard regression methods. Moreover, they circumvent the need for ad hoc indicator variables, which account for a significant part of the variance in linear regression models. Additionally, they lead to the elucidation of nonlinear and "cross-products" effects that correspond to trade-offs between physical properties in their effect on biological activity. This is the first demonstration of the latter two findings. On the other hand, due to the complexity of the resulting models, an understanding of the local, but not the global, structure-activity relationships is possible. The latter must await further developments. Furthermore, the longer computational time required to train the networks is somewhat inconveniencing, although not restrictive.

Animals↗
PubMedDate not supplied

Modeling memory: what do we learn from attractor neural networks?

In this paper we summarize some of the main contributions of models of recurrent neural networks with associative memory properties. We compare the behavior of these attractor neural networks with empirical data from both physiology and psychology. This type of network could be used in models with more complex functions.

Humans↗
PubMed2006

Assisting the diagnosis of thyroid diseases with Bayesian-type and SOM-type neural networks making use of routine test data.

Patients with hyperthyroidism sometimes take much time to receive the final diagnosis. To improve patient QOL, simple screening for hyperthyroidism by thyroid non-specialists at the physical check-up is highly expected. Therefore, we applied both Bayesian-type and SOM-type neural networks since we assured the approach useful in analysing thyroid function diagnosis in the previous work. Routine test (14 parameters) data from 66 subjects with a known diagnosis (18 patients with hyperthyroidism and 48 healthy volunteers) were adopted as learning data, and then 142 individuals who also received the same routine tests at the Tohoku University Hospital were screened to predict patients with hyperthyroidism. Both neural networks using 14 parameters predicted several patients as having hyperthyroidism with high probability, including all three hyperthyroid patients diagnosed later by the physician. Further detailed analysis of the routine test parameters that were important for classification found that screening with a set of three parameters (alkaline phosphatase, serum creatinine and total cholesterol) or plus aspartate aminotransferase allowed for quite accurate screening. These results showed that the same neural networks as previous work allows simple screening of patients for hyperthyroidism on the basis of routine test data, and that physicians not specializing in the thyroid can rapidly identify individuals suspected of having hyperthyroidism, to permit a rapid referral for examination and treatment by thyroid specialists.

Bayes Theorem↗
PubMed2001

On preprocessing of protein sequences for neural network prediction of polyproline type II secondary structures.

Polyproline type II stretches are somewhat rare on proteins. The backbone of this secondary structural element folds to a triangular form instead of the normal alpha-helix with 3.6 residues per turn. It is a very challenging task to try to detect them computationally from protein sequence. Here, we have studied the preprocessing phase in particular, which is important for any machine learning method. Preprocessing included selection of relevant data from the Protein Data Bank and investigation of learnability properties. These properties show whether the material is suitable for neural network computing. The complexity of algorithms in connection with preprocessing was briefly considered. We found that feedforward perceptron neural networks were appropriate for the prediction of polyproline type II and also relatively efficient in this task. The problem is very difficult because of the great similarity of the two classes present in the classification. Nevertheless, neural networks were able to recognize and predict about 75% of secondary structures.

Algorithms↗
PubMed2004

Neural networks as robust tools in drug lead discovery and development.

Empirical methods for building predictive models of the relationships between molecular structure and useful properties are becoming increasingly important. This has arisen because drug discovery and development have become more complex. A large amount of biological target information is becoming available though molecular biology. Automation of chemical synthesis and pharmacological screening has also provided a vast amount of experimental data. Tools for designing libraries and extracting information from molecular databases and high-throughput screening (HTS) experiments robustly and quickly enable leads to be discovered more effectively. As drug leads progress down the development pipeline, the ability to predict physicochemical, pharmacokinetic, and toxicological properties of these leads is becoming increasingly important in reducing the number of expensive, late-development failures. Neural network methods have much to offer in these areas. This review introduces the concepts behind neural networks applied to quantitative structure-activity relationships (QSARs), points out problems that may be encountered, suggests ways of avoiding the pitfalls, and introduces several exciting new neural network methods discovered during the last decade.

Animals↗
PubMed2002

Prediction of risk of coronary events in middle-aged men in the Prospective Cardiovascular Münster Study (PROCAM) using neural networks.

BACKGROUND: Logistic regression (LR) is commonly used to estimate risk of coronary heart disease. We investigated if neural networks improved on the risk estimate of LR by analysing data from the Prospective Cardiovascular Münster Study (PROCAM), a large prospective epidemiological study of risk factors for coronary heart disease among men and women at work in northern Germany. METHODS: We used a multi-layer perceptron (MLP) and probabilistic neural networks (PNN) to estimate the risk of myocardial infarction or acute coronary death (coronary events) during 10 years' follow-up among 5159 men aged 35-65 years at recruitment into PROCAM. In all, 325 coronary events occurred in this group. We assessed the performance of each procedure by measuring the area under the receiver-operating characteristics curve (AUROC). RESULTS: The AUROC of the MLP was greater than that of the PNN (0.897 versus 0.872), and both exceeded the AUROC for LR of 0.840. If 'high risk' is defined as an event risk >20% in 10 years, LR classified 8.4% of men as high risk, 36.7% of whom suffered an event in 10 years (45.8% of all events). The MLP classified 7.9% as high risk, 64.0% of whom suffered an event (74.5% of all events), while with the PNN, only 3.9% were at high risk, 58.6% of whom suffered an event (33.5% of all events). CONCLUSION: Intervention trials indicate that about one in three coronary events can be prevented by 5 years of lipid-lowering treatment. Our analysis suggests that use of the MLP to identify high-risk individuals as candidates for drug treatment would allow prevention of 25% of coronary events in middle-aged men, compared to 15% and 11% with LR and the PNN, respectively.

Adult↗
PubMed1997

Prediction of protein supersecondary structures based on the artificial neural network method.

The sequence patterns of 11 types of frequently occurring connecting peptides, which lead to a classification of supersecondary motifs, were studied. A database of protein supersecondary motifs was set up. An artificial neural network method, i.e. the back propagation neural network, was applied to the predictions of the supersecondary motifs from protein sequences. The prediction correctness ratios are higher than 70%, and many of them vary from 75 to 82%. These results are useful for the further study of the relationship between the structure and function of proteins. It may also provide some important information about protein design and the prediction of protein tertiary structure.

Amino Acid Sequence↗
PubMed1997

Identification and classification of autoantibody repertoires (Western blots) with a pattern recognition algorithm by an artificial neural network.

The screening of sera for autoantibodies with Western blots reveals complex repertoires. the compostion of such repertoires depends on genetic control of autoantibody-producing cells, the individual's history of exposure to its own and to foreign antigens, and also on the presence of autoimmune diseases. Our method shows how staining patterns of Western blots can be recoded as binary or grey-value vectors. Vectors are transferred to artificial neural networks for learning. Artificial neural networks are able to recognize group-specific antibody binding patterns. Staining patterns can be attributed to diagnostic groups. This may support diagnostic procedures.

Algorithms↗