SEARCH · PubMed Health

Results for “Multilayer Perceptrons”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 235 records · Page 13Linked to original sources

PubMed2004

The use of neural networks for predicting the result of endoscopic treatment for vesico-ureteric reflux.

OBJECTIVE: To create an artificial neural network (ANN) to aid in predicting the results of endoscopic treatment for vesico-ureteric reflux (VUR). MATERIALS AND METHODS: During 1999-2001 we used endoscopic treatment in 261 ureteric units with VUR of all grades and causes. An ANN based on multilayer perceptron architecture was created using an 11 x 6 x 1 structure, taking the following as variables: the cause and grade of VUR, the patient's age and sex, the type of implanted substance and its volume, the number of treatments, the affected ureter, the endoscopic findings, and the type of cystography used. In all, 174 cases were used as training samples for the ANN and 87 to validate it. We calculated the sensitivity, specificity, positive (PPV) and negative predictive values (NPV), and the success rate (%) of the system. RESULTS: In the training group the ANN gave a sensitivity of 86.4%, a specificity of 89.5%, a PPV of 76% and NPV of 94%, with a success rate of 88.6%. In the same training group logistic regression (LR) gave respective values of 68.2%, 58.8%, 39%, 82.7% and 61.4%. In the validation group the respective values for the ANN were 71.4%, 81.6%, 58.8%, 88.6% and 78.9%, and in the same validation group the LR gave 64.4%, 50%, 32.1%, 79.2% and 53.9%. The Wilcoxon test confirmed the independence of both methods (P < 0.001). CONCLUSION: The ANN is an effective tool for assisting the urologist in indicating and applying endoscopic treatments for VUR.

Adolescent↗

PubMedDate not supplied

Filter banks and neural network-based feature extraction and automatic classification of electrogastrogram.

Dysrhythmia in gastric myoelectrical activity has been frequently observed in patients with gastric motor disorders and gastrointestinal symptoms. The assessment of the regularity of gastric myoelectrical activity is of great clinical significance. The aim of this study was to develop an automated assessment method for the regularity of gastric myoelectrical activity from the surface electrogastrogram (EGG). The method proposed in this paper was based on the filter bank and neural network. First, the EGG signal was divided into frequency subbands using filter bank analysis. Second, a parameter called the subband energy ratio (SER) was computed for each subband signal. A multilayer perceptron neural network was then used to automatically classify the EGG signal into four categories: bradygastria, normal, tachygastria, and arrhythmia, using the SER as the input. The EGG recording was made using the standard method of electrogastrography by placing electrodes on the abdominal surface. The study was performed in 40 patients with various gastric motor disorders, ten healthy adults, and ten healthy children. The neural network was trained and tested using the EGG data obtained from the patients. The regularity of gastric myoelectrical activity was assessed based on the classification of the minute-by-minute EGG segments. Using the running spectral analysis method as a gold standard, the proposed automated method had an accuracy of 100% for the training set and 97% for the test set. It was concluded that the proposed method provides an accurate and automatic assessment of the regularity of gastric myoelectrical activity from the EGG.

Adolescent↗

PubMed1993

Speaker-independent consonant classification in continuous speech with distinctive features and neural networks.

This paper provides experimental evidence to the assertion that the design of appropriate neural networks (NN) for speech recognition should be inspired by acoustic and phonetic knowledge, and not only by knowledge in pattern recognition. Rather than investigating the NN learning paradigm, the paper is focused on the influence of the input parameters, of the internal structure, and of the desired output representation on the classification performance of recurrent multilayer perceptrons. As an instructive example, the paper analyzes the problem of classifying ten stop and nasal consonants in continuous speech independently of the speaker. Experiments are reported for the TIMIT database, using 343 speakers in the training set and 77 different speakers in the test set. Comparative experiments show that good performance is obtained when many input acoustic parameters are used, including a time/frequency gradient operator related to transitions of the second formant, and when the desired outputs represent context-dependent articulatory features. Classification is performed by principal component analysis of the NN outputs. Refinement of the design parameters yield increasingly better performance on the test set, ranging from 45% errors for a perceptron without hidden nodes to 23.3% errors for the best NN.

Female↗

PubMed1995

On the feasibility of using neural nets to derive hearing-aid prescriptive procedures.

A neural net is a "black box" information processing system that can be used for pattern matching, optimal prediction, or functional approximation. A neural net requires a minimal amount of a priori knowledge about the problem to be solved, but can require large amounts of data to converge to a solution. For a hearing-aid fitting procedure, a multilayer perceptron net was trained to generate an optimum match between a set of input pure-tone audiograms and the corresponding best frequency response and gain for each subject. The feasibility of using neural nets to select hearing-aid response characteristics was tested using both simulated and real audiometric data. The simulation results indicate that a neural net can be successfully trained to reproduce a fitting rule such as the NAL-R procedure, and that a minimum of about 50 sets of audiometric response data are needed for the net to converge to a generalized solution. When used to predict, from the pure-tone audiograms, the best frequency response characteristics determined for subjects having severe-to-profound hearing losses, the neural net was more accurate than the NAL-R fitting procedure derived from the same data.

Audiometry, Pure-Tone↗

PubMed1999

Automated localisation of the optic disc, fovea, and retinal blood vessels from digital colour fundus images.

AIM: To recognise automatically the main components of the fundus on digital colour images. METHODS: The main features of a fundus retinal image were defined as the optic disc, fovea, and blood vessels. Methods are described for their automatic recognition and location. 112 retinal images were preprocessed via adaptive, local, contrast enhancement. The optic discs were located by identifying the area with the highest variation in intensity of adjacent pixels. Blood vessels were identified by means of a multilayer perceptron neural net, for which the inputs were derived from a principal component analysis (PCA) of the image and edge detection of the first component of PCA. The foveas were identified using matching correlation together with characteristics typical of a fovea-for example, darkest area in the neighbourhood of the optic disc. The main components of the image were identified by an experienced ophthalmologist for comparison with computerised methods. RESULTS: The sensitivity and specificity of the recognition of each retinal main component was as follows: 99.1% and 99.1% for the optic disc; 83.3% and 91.0% for blood vessels; 80.4% and 99.1% for the fovea. CONCLUSIONS: In this study the optic disc, blood vessels, and fovea were accurately detected. The identification of the normal components of the retinal image will aid the future detection of diseases in these regions. In diabetic retinopathy, for example, an image could be analysed for retinopathy with reference to sight threatening complications such as disc neovascularisation, vascular changes, or foveal exudation.

Algorithms↗

PubMed2000

Inductive learning approaches to rainfall-runoff modelling.

Trying to model the rainfall-runoff process is a complex activity as it is influenced by a number of implicit and explicit factors--for example, precipitation distribution, evaporation, transpiration, abstraction, watershed topography, and soil types. However, this kind of forecasting is particularly important as it is used to predict serious flooding, estimate erosion and identify problems associated with low flow. Inductive learning approaches (e.g. decision trees and artificial neural networks) are particularly well suited to problems of this nature as they can often interpret underlying factors (such as seasonal variations) which cannot be modelled by other techniques. In addition, these approaches can easily be trained on the explicit factors (e.g. rainfall) and the inexplicit factors (e.g. abstraction) that affect river flow. Inductive learning approaches can also be extended to account for new factors that emerge over a period of time. This paper evaluates the application of decision trees and two artificial neural network models (the multilayer perceptron and the radial basis function network) to river flow forecasting in two flood prone UK catchments using real hydrometric data. Comparisons are made between the performance of these approaches and conventional flood forecasting systems.

Algorithms↗

PubMed2000

Cooperative coevolution of neural representations.

A genetic algorithm (GA) is used to search for a set of local feature detectors or hidden units. These are in turn employed as a representation of the input data for neural learning in the upper layer of a multilayer perceptron (MLP) which performs an image classification task. Three different methods of encoding hidden unit weights in the chromosome of the GA are presented, including one which coevolves all the feature detectors in a single chromosome, and two which promote the cooperation of feature detectors by encoding them in their own individual chromosomes. The fitness function measures the MLP classification accuracy together with the confidence of the networks.

Algorithms↗

PubMed2001

Improved long-term temperature prediction by chaining of neural networks.

When an artificial neural network (ANN) is trained to predict signals p steps ahead, the quality of the prediction typically decreases for large values of p. In this paper, we compare two methods for prediction with ANNs: the classical recursion of one-step ahead predictors and a new kind of chain structure. When applying both techniques to the prediction of the temperature at the end of a blast furnace, we conclude that the chaining approach leads to an improved prediction of the temperature and avoidance of instabilities, since the chained networks gradually take the prediction of their predecessors in the chain as an extra input. It is observed that instabilities might occur in the iterative case, which does not happen with the chaining approach. To select relevant inputs and decrease the number of weights in this approach, Automatic Relevance Determination (ARD) for multilayer perceptrons is applied.

Bayes Theorem↗

PubMed2001

Neural network based automatic diagnosis of children with brain dysfunction.

This paper proposes the use of multilayer perceptron for brain dysfunction diagnosis. The performance of MLP was better than that of Discriminant Analysis and Decision Tree classifiers, with an 85% accuracy rate in an experimental test involving 332 subjects. In addition, the neural network employing Bayesian learning was able to identify the most important input variable. These two results demonstrate that the neural network can be effectively used in the diagnosis of children with brain dysfunction.

Attention Deficit Disorder with Hyperactivity↗

PubMed2001

A fast and convergent stochastic MLP learning algorithm.

We propose a stochastic learning algorithm for multilayer perceptrons of linear-threshold function units, which theoretically converges with probability one and experimentally exhibits 100% convergence rate and remarkable speed on parity and classification problems with typical generalization accuracy. For learning the n bit parity function with n hidden units, the algorithm converged on all the trials we tested (n=2 to 12) after 5.8 x 4.1(n) presentations for 0.23 x 4.0(n-6) seconds on a 533MHz Alpha 21164A chip on average, which is five to ten times faster than Levenberg-Marquardt algorithm with restarts. For a medium size classification problem known as Thyroid in UCI repository, the algorithm is faster in speed and comparative in generalization accuracy than the standard backpropagation and Levenberg-Marquardt algorithms.

Algorithms↗

PubMed2006

A neural network model for the simulation of word production errors of Finnish nouns.

We constructed a connectionist model with multilayer perceptron networks for the simulation of word production errors in the Finnish language. Such errors occur as slips of the tongue in healthy subjects and even more often in aphasic patients suffering from brain damage. Word production errors are of theoretical interest as they open an avenue to inner workings of the language production system. Here we present our coding schemes for semantic and phoneme information of words, investigate the general properties of the model, and compare the results of the model against empirical naming error data from ten aphasic patients. The model was reasonable in simulating word production errors in this heterogeneous group of aphasic patients.

Aphasia, Broca↗

PubMed1993

Analogue synaptic noise--implications and learning improvements.

We analyse the effects of analogue noise on the synaptic arithmetic during multilayer perceptron training by expanding the cost function to include noise-mediated penalty terms. Predictions are made in the light of these calculations which suggest that fault tolerance, generalisation ability and learning trajectory should be improved by such noise-injection. Extensive simulation experiments on two distinct classification problems substantiate the claims. The results appear to be perfectly general for all training schemes where weights are adjusted incrementally, and have wide-ranging implications for all applications, particularly those involving "inaccurate" analogue neural VLSI.

Algorithms↗

PubMed1994

Predictive control of nonlinear systems based on identification by backpropagation networks.

Using the property of universal approximation of multilayer perceptron neural network, a class of discrete nonlinear dynamical systems are modeled by a perceptron with two hidden layers. A backpropagation algorithm is then used to train the model to identify the nonlinear systems to a desired level of accuracy. Based on the identified model, a one-step-ahead predictive control scheme is proposed in which the future control inputs are obtained through some nonlinear optimization process. Making use of the online learning properties of neural networks, the predictive control scheme is further developed into an adaptive one which is robust to the incompleteness of identification. Simulation results show that this neural control scheme works well even for some very complicated nonlinear systems.

Mathematics↗

PubMed1995

Training neural networks by means of genetic algorithms working on very long chromosomes.

In the neural network/genetic algorithm community, rather limited success in the training of neural networks by genetic algorithms has been reported. In a paper by Whitley et al. (1991), he claims that, due to "the multiple representations problem", genetic algorithms will not effectively be able to train multilayer perceptrons, whose chromosomal representation of its weights exceeds 300 bits. In the following paper, by use of a "real-life problem", known to be non-trivial, and by a comparison with "classic" neural net training methods, I will try to show, that the modest success of applying genetic algorithms to the training of perceptrons, is caused not so much by the "multiple representations problems" as by the fact that problem-specific knowledge available is often ignored, thus making the problem unnecessarily tough for the genetic algorithm to solve. Special success is obtained by the use of a new fitness function, which takes into account the fact that the search performed by a genetic algorithm is holistic, and not local as is usually the case when perceptrons are trained by traditional methods.

Algorithms↗

PubMed1995

Nonlinear gated experts for time series: discovering regimes and avoiding overfitting.

In the analysis and prediction of real-world systems, two of the key problems are nonstationarity (often in the form of switching between regimes) and overfitting (particularly serious for noisy processes). This article addresses these problems using gated experts, consisting of a (nonlinear) gating network, and several (also nonlinear) competing experts. Each expert learns to predict the conditional mean, and each expert adapts its width to match the noise level in its regime. The gating network learns to predict the probability of each expert, given the input. This article focuses on the case where the gating network bases its decision on information from the inputs. This can be contrasted to hidden Markov models where the decision is based on the previous state(s) (i.e. on the output of the gating network at the previous time step), as well as to averaging over several predictors. In contrast, gated experts soft-partition the input space, only learning to model their region. This article discusses the underlying statistical assumptions, derives the weight update rules, and compares the performance of gated experts to standard methods on three time series: (1) a computer-generated series, obtained by randomly switching between two nonlinear processes; (2) a time series from the Santa Fe Time Series Competition (the light intensity of a laser in chaotic state); and (3) the daily electricity demand of France, a real-world multivariate problem with structure on several time scales. The main results are: (1) the gating network correctly discovers the different regimes of the process; (2) the widths associated with each expert are important for the segmentation task (and they can be used to characterize the sub-processes); and (3) there is less overfitting compared to single networks (homogeneous multilayer perceptrons), since the experts learn to match their variances to the (local) noise levels. This can be viewed as matching the local complexity of the model to the local complexity of the data.

Computers↗

PubMed1996

Hardware prototypes of a Boolean neural network and the simulated annealing optimization method.

Boolean Neural Network is a neural network that operates with binary weight values of "1" and "0". Otherwise it is formally analogous to the Multilayer Perceptron (MLP). Simulated Annealing is a stochastic optimization methods that is suitable for performing nonlinear multivariable optimization tasks. Training a Boolean Neural Network is a well-suited problem to this algorithm. However, the Simulated Annealing method is computationally heavy, which makes the training procedure slow. The training speed can be improved by using custom designed hardware for the whole system including the optimization method and the neural network. Hardware prototypes of a Boolean Neural Network and the Simulated Annealing optimization method have been designed using discrete components. The Boolean Neural Network implementation is basically a dynamically configurable feedforward network of Boolean logic gates of two inputs. The Simulated Annealing implementation is a general purpose hardware tool for multivariable optimization tasks. Here it is applied to do supervised training of the Boolean Neural Network hardware.

Algorithms↗

PubMedDate not supplied

Neural solution to the correction of miss distance in gun fire control system.

Multilayer perceptrons trained with the backpropagation algorithm are tested in gun fire control system for error correction and are compared to optimal algorithms based on minimum mean square error. The structure of the proposed neural controller is described and performance results are shown.

Algorithms↗

PubMed1999

Feedforward neural network with adaptive reference pattern layer.

A hybrid neural network architecture is investigated for modeling purposes. The proposed hybrid is based on the multilayer perceptron (MLP) network. In addition to the usual hidden layers, the first hidden layer is selected to be an adaptive reference pattern layer. Each unit in this new layer incorporates a reference pattern that is located somewhere in the space spanned by the input variables. The outputs of these units are the component wise-squared differences between the elements of a reference pattern and the inputs. The reference pattern layer has some resemblance to the hidden layer of the radial basis function (RBF) networks. Therefore the proposed design can be regarded as a sort of hybrid of MLP and RBF networks. The presented benchmark experiments show that the proposed hybrid can provide significant advantages over standard MLPs and RBFs in terms of fast and efficient learning, and compact network structure.

Feedback↗