SEARCH · Search PubMed
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.
Using computer technique, artificial neural network and fuzzy control theory, we have explored a real-time control method for the athlete's physical workload intensity in order that the goal of physical training can be reached effectively in accordance to the exercise plan. The technique could be useful for improving the efficiency of scientific physical training.
A learning algorithm for a model binocular cell was derived according to an information maximization principle and by using a low signal-to-noise-ratio approximation. The algorithm updates cell's synaptic weights so that the information obtained from the cell's output is increased. According to the algorithm, model binocular cells were trained by using computer-generated stereo images as training data. As a result, cells tuned to various disparities were generated. Also, generated synaptic weight patterns of the cells were similar to Gabor-wavelets and receptive fields of simple cells in the visual cortex. Thus, they were orientation and spatial frequency selective as well as disparity selective. Gabor functions were used to fit the generated weight patterns. The fitting results indicated that the generated cells encode disparities in terms of phase disparity and/or position disparity. This result agrees with experimental findings by Anzai et al. [J Neurophys 82 (1999) 874] and is consistent with ICA-based theoretical results obtained [Network: Comput Neural Syst 11 (2000) 191].
This paper describes an investigation of a recurrent artificial neural network which uses association to build transform-invariant representations. The simulation implements the analytic model of Parga and Rolls [(1998). Transform-invariant recognition by association in a recurrent network. Neural Computation 10(6), 1507-1525.] which defines multiple (e.g. "view") patterns to be within the basin of attraction of a shared (e.g. "object") representation. First, it was shown that the network could store and correctly retrieve an "object" representation from any one of the views which define that object, with capacity as predicted analytically. Second, new results extended the analysis by showing that correct object retrieval could occur where retrieval cues were distorted; where there was some association between the views of different objects; and where connectivity was diluted, even when this dilution was asymmetric. The simulations also extended the analysis by showing that the system could work well with sparse patterns; and showing how pattern sparseness interacts with the number of views of each object (as a result of the statistical properties of the pattern coding) to give predictable object retrieval performance. The results thus usefully extend a recurrent model of invariant pattern recognition.
The design and blood brain barrier crossing of glycine/NMDA receptor antagonists are of significant interest in pharmaceutical research. The use of these antagonists in stroke or seizure reduction have been considered. Measuring the inhibitory concentrations, however, can be time-consuming and costly. The use of quantitative structure-activity relationships to estimate IC(50) values for these receptor antagonists is an attractive alternative compared to experimental measurement. A data set of 109 compounds with measured log(IC(50)) values ranging from -0.57 to 4.5 is used. Structural information is encoded with numerical descriptors for topological, electronic, geometric, and polar surface properties. A genetic algorithm with a computational neural network fitness evaluator is used to select the best descriptor subsets. Multiple linear regression and computational neural network models are developed. Additionally, a quantitative radial basis function neural network (QRBFNN) was developed with the intent of introducing nonlinearity at a faster speed. A genetic algorithm using the radial basis function network as a fitness evaluator was also developed to search descriptor space for optimum subsets. All models are tested using an external prediction set. The nonlinear computational neural network model has root-mean-square errors of approximately half a log unit.
A small-size model for a chaotic neural network is proposed using strange attractors for computation. This network has a chaotic ground state and is capable of responding to external stimuli by constraining the network dynamics to specific parts of the ground state attractor. For parameter optimization, bifurcation diagrams are evaluated.
Locomotor activity-based rehabilitation programs seem to bring encouraging recovery to incomplete spinal cord injury paralytics. But in "complete" injuries recovery is long delayed and in small increments. Developmental and computer neural networks studies show that motor relearning requires the participation of redundant numbers of and activity-dependent competition of synapses among spinal interneurons. The present programs do not meet this prerequisite. They focus mainly on the mechanical (retraining) part while the neural (relearning) part plays a rather passive role. Here we suggest how retraining can be integrated with relearning by inductive lability procedures (Krishnan, 2003a, 2003b) with the objective of enhancing and hastening recovery.
A computational neural network model is presented that explains how the hippocampus can contribute to transitive inference performance observed in rats (Dusek and Eichenbaum, 1997. Proc Natl Acad Sci U S A 94:7109-7114; Van Elzakker et al., 2003. Hippocampus 12:this issue). In contrast to existing theories that emphasize the idea that the hippocampus contributes by flexibly relating previously encoded memories, we find that the hippocampus contributes by altering the elemental associative weights of individual stimulus elements during learning. We use this model to account for a range of existing data and to make a number of distinctive predictions that clearly contrast these two views.
Mathematical models are developed to find quantitative structure-activity relationships that correlate chemical structure and inhibition toward three carbonic anhydrase (CA) isozymes: CA I, II, and IV. Numerical descriptors are generated to encode important topological, geometric, and electronic features of molecular structure. After descriptor generation, multiple linear regression, and computational neural network (CNN) analyses are performed on various descriptor subsets to find superior models for prediction. Committees of five CNNs were utilized to average final predicted values for the 142-compound data set. For inhibitors of CA I, an 8-5-1 CNN committee produced a training set rms error of 0.105 log K(i) (r(2) = 0.994) and prediction set rms error of 0.208 log K(i) (r(2) = 0.980). Training and prediction set rms errors of 0.140 log K(i) (r(2) = 0.992) and 0.231 log K(i) (r(2) = 0.971), respectively, were produced by a 9-5-1 CNN committee for inhibitors of CA II. For prediction of CA IV inhibitors, an 8-5-1 CNN committee produced training and prediction set rms errors of 0.147 log K(i) (r(2) = 0.992) and 0.211 log K(i) (r(2) = 0.991), respectively. In addition, classification models were built using k-nearest neighbor (kNN) analysis to solve two- and three-class problems for inhibitors of CA IV. A three-descriptor classification model proved superior in labeling compounds as active or inactive inhibitors for the two-class problem. Training and prediction set percent classification rates of 100% and 87.1%, respectively, were obtained. For the three-class (active/moderate/inactive) problem, a five-descriptor model was deemed optimal producing a training set percent classification rate of 98.8% and prediction set rate of 79.0%.
Experimental IC(50) data for 314 selective cyclooxygenase-2 (COX-2) inhibitors are used to develop quantitation and classification models as a potential screening mechanism for larger libraries of target compounds. Experimental log(IC(50)) values ranged from 0.23 to > or = 5.00. Numerical descriptors encoding solely topological information are calculated for all structures and are used as inputs for linear regression, computational neural network, and classification analysis routines. Evolutionary optimization algorithms are then used to search the descriptor space for information-rich subsets which minimize the rms error of a diverse training set of compounds. An eight-descriptor model was identified as a robust predictor of experimental log(IC(50)) values, producing a root-mean-square error of 0.625 log units for an external prediction set of inhibitors which took no part in model development. A k-nearest neighbor classification study of the data set discriminating between active and inactive members produced a nine-descriptor model able to accurately classify 83.3% of the prediction set compounds correctly.
The absorption of a drug compound through the human intestinal cell lining is an important property for potential drug candidates. Measuring this property, however, can be costly and time-consuming. The use of quantitative structure-property relationships (QSPRs) to estimate percent human intestinal absorption (%HIA) is an attractive alternative to experimental measurements. A data set of 86 drug and drug-like compounds with measured values of %HIA taken from the literature was used to develop and test a QSPR mode. The compounds were encoded with calculated molecular structure descriptors. A nonlinear computational neural network model was developed by using the genetic algorithm with a neural network fitness evaluator. The calculated %HIA (cHIA) model performs wells, with root-mean-square (rms) errors of 9.4%HIA units for the training set, 19.7%HIA units for the cross-validation (CV) set, and 16.0%HIA units for the external prediction set.
A data set of 348 urea-like compounds that inhibit the soluble epoxide hydrolase enzyme in mice and humans is examined. Compounds having IC(50) values ranging from 0.06 to >500 microM (murine) and 0.10 to >500 microM (human) are categorized as active or inactive for classification, while quantitation is performed on smaller compound subsets ranging from 0.07 to 431 microM (murine) and 0.11 to 490 microM (human). Each compound is represented by calculated structural descriptors that encode topological, geometrical, electronic, and polar surface features. Multiple linear regression (MLR) and computational neural networks (CNNs) are employed for quantitative models. Three classification algorithms, k-nearest neighbor (kNN), linear discriminant analysis (LDA), and radial basis function neural networks (RBFNN), are used to categorize compounds as active or inactive based on selected data split points. Quantitative modeling of human enzyme inhibition results in a nonlinear, five-descriptor model with root-mean-square errors (log units of IC(50) [microM]) of 0.616 (r(2) = 0.66), 0.674 (r(2) = 0.61), and 0.914 (r(2) = 0.33) for training, cross-validation, and prediction sets, respectively. The best classification results for human and murine enzyme inhibition are found using kNN. Human classification rates using a seven-descriptor model for training and prediction sets are 89.1% and 91.4%, respectively. Murine classification rates using a five-descriptor model for training and prediction sets are 91.5% and 88.6%, respectively.
A group contribution method has been developed to correlate the acute toxicity (96-h LC50) to the fathead minnow (Pimephales promelas) for 397 organic chemicals. Multilinear regression and computational neural networks (CNNs) were used for model building. The models were able to achieve a fairly good correlation of the data (r2 > 0.9). The linear model, which included four specific interaction terms, provided a rapid means of predicting the toxicity of a compound. The CNN model was able to yield virtually the same predictions with or without the four interaction terms that were included in the multilinear model.
Interest in the prediction of toxicity without the use of experimental data is growing, and quantitative structure-activity relationship (QSAR) methods are valuable for such predictions. A QSAR study of acute aqueous toxicity of 375 diverse organic compounds has been developed using only calculated structural features as independent variables. Toxicity is expressed as -log(LD(50)) with the units -log(millimoles per liter) and ranges from -3 to 6. Multiple linear regression and computational neural networks (CNNs) are utilized for model building. The best model is a nonlinear CNN model based on eight calculated molecular structure descriptors. The root-mean-square log(LD(50)) errors for the training, cross-validation, and prediction sets of this CNN model are 0.71, 0.77, and 0.74 -log(mmol/L), respectively. These results are compared to a previous study with the same data set which included many more descriptors and used experimental data in the descriptor pool.
A quantitative structure-activity relationship (QSAR) investigation was done for the acute oral mammalian toxicity (LD50) of a set of 54 organophosphorus pesticide compounds. The compounds were represented with calculated molecular structure descriptors, which encoded their topological, electronic, and geometrical features. Feature selection was done with a genetic algorithm to find subsets of descriptors that would support a high quality computational neural network (CNN) model to link the structural descriptors to the -log(mmol/kg) values for the compounds. The best seven-descriptor non-linear CNN model found had an rms error of 0.22 log units for the training set compounds and 0.25 log units for the prediction set compounds.
The frontal cortex and the basal ganglia interact via a relatively well understood and elaborate system of interconnections. In the context of motor function, these interconnections can be understood as disinhibiting, or "releasing the brakes," on frontal motor action plans: The basal ganglia detect appropriate contexts for performing motor actions and enable the frontal cortex to execute such actions at the appropriate time. We build on this idea in the domain of working memory through the use of computational neural network models of this circuit. In our model, the frontal cortex exhibits robust active maintenance, whereas the basal ganglia contribute a selective, dynamic gating function that enables frontal memory representations to be rapidly updated in a task-relevant manner. We apply the model to a novel version of the continuous performance task that requires subroutine-like selective working memory updating and compare and contrast our model with other existing models and theories of frontal-cortex-basal-ganglia interactions.
A classical quantitative structure-activity relationship (Hansch) study and artificial neural networks (ANNs) have been applied to a training set of 32 substituted phenylpiperazines with affinity for 5-HT(1A) and alpha(1)-adrenergic receptors, to evaluate the structural requirements that are responsible for 5-HT(1A)/alpha(1) selectivity. The resulting models provide a significant correlation of electronic, steric, and hydrophobic parameters with the biological affinities. Although the derived linear Hansch correlations give good statistics and acceptable predictions, the introduction of nonlinear relationships in the analysis gives more solid models and more accurate predictions. In the ANN models on the basis of the obtained 3D plots, the 5-HT(1A) affinity has a nonlinear dependence on F, V(o), V(m), and pi(o), although the nonlinear relationship is not far from a planar one. The alpha(1)-adrenergic receptor affinity has a clear nonlinear dependence on F, V(o), V(m), pi(o), and pi(m). A comparison of both analyses gives an additional understanding for 5-HT(1A)/alpha(1) selectivity: (a) high F values increase the binding affinity for 5-HT(1A) receptors and decrease the affinity for alpha(1) sites; (b) the hydrophobicity at the meta-position has only influence for the alpha(1)-adrenergic receptor; (c) the meta-position seems to be implicated in the 5-HT(1A)/alpha(1) selectivity. While the 5-HT(1A) receptor is able to accommodate bulky substituents in the region of its active site, the steric requirements of the alpha(1)-adrenergic receptor at this position are more restricted. This information was used for the design of the new ligand EF-7412 (33) (5-HT(1A): K(i exptl) = 27 nM, alpha(1): K(i exptl) > 1000 nM; 5-HT(1A): K(i pred) (ANN) = 36 nM, alpha(1): K(i pred ANN) = 2745 nM) which was characterized as an antagonist in vivo in pre- and postsynaptic 5-HT(1A)R sites. Computational simulations of the complex between EF-7412 (33) and a 3D model of the transmembrane domain of the 5-HT(1A) receptor allowed us to define the molecular details of the ligand-receptor interaction that includes: (i) the ionic interaction between the protonated amine of the ligand and Asp 3.32; (ii) the hydrogen bonds between the m-NHSO(2)Et group of the ligand and Asn 7.39; and the hydrogen bonds between the hydantoin moiety of the ligand and (iii) Thr 3.37, (iv) Ser 5.42, and (v) Thr 5.43. These QSAR and ANN results in combination with computational simulations of ligand recognition will be useful for the design of potent selective 5-HT(1A) ligands.
Artificial neural network is a complex system capable of gathering and synthesizing data and capable of learning using particular given parameters. This technology was applied to studying cytological diagnostics of thyroid diseases to differentiate between follicular cancer and follicular adenoma. A complex of quantitative cytological characteristics was used to describe cytological preparations by examples of known cytological diagnosis to train the neural network. Among offered cytological examples, 96% were recognized correctly during testing.
The accurate detection of malignant neoplasms whilst they are still small is recognised as one of the main factors increasing chances of survival. Neural networks have many biomedical applications and they have been applied to neoplasia but their use in oral pathology has only recently been documented. The objectives of this study were to train networks to discriminate between normal and dysplastic mucosa. Each network was trained by back propagation, internal cross validation and tested on additional data. The data were derived by analysing 348 intra-oral smears and included mean nuclear and mean cytoplasmic areas of the smears measured by image analysis. A neural network differentiated between normal/non-dysplastic mucosa and dysplastic/malignant mucosa (specificity 0.82, sensitivity 0.76). These early results suggest that integrating neural networks and image analysis, as well as investigating additional criteria, could enhance automation and accuracy of smear techniques in diagnosing oral malignancy.