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 1,711 records · Page 95Linked to original sources
PubMed2000

Prediction of spontaneous ureteral calculous passage by an artificial neural network.

PURPOSE: Although a consensus exists that small stones presenting in the distal ureter have a good probability of spontaneous passage, it is difficult to predict in individuals whether a particular ureteral stone would pass or require intervention. If an accurate judgment were made at presentation on the likelihood of stone passage, patients would receive immediate intervention for the stone or be notified of a more appropriate time at which to expect passage. We used an artificial neural network to evaluate data in patients with ureteral calculi to predict whether a stone would pass spontaneously or require intervention. MATERIALS AND METHODS: Data were collected from the records of 181 patients presenting with colic due to a ureteral calculus. Patient input factors included age, sex, race, marital status, insurance, stone side, level and size, hydronephrosis and obstruction grades, duration of symptoms before presentation, serum creatinine, history of stone passage or intervention and nausea, vomiting or fever. Outcomes evaluated were stone passage or intervention. Data were entered into a neural network created using commercially available computer software. RESULTS: A set of 125 patients from the database was used for training the network. The network correctly predicted outcome in 38 of the remaining 55 patients (76%) used for testing. In the 25 cases in which stones passed spontaneously sensitivity was 100%. Duration of symptoms before presentation was the most influential factor in network ability to predict accurately stone passage, followed by hydronephrosis grade. CONCLUSIONS: An artificial neural network may be used to predict accurately the probability of spontaneous ureteral stone passage. Using such a model at presentation may help to determine whether a patient should receive early intervention for a stone or expect a lengthy interval before stone passage.

Adolescent↗
PubMed2000

Using neural networks for prediction of subcellular location of prokaryotic and eukaryotic proteins.

T. Kohonen's self-organization model, a typical neural network model, was applied to predict the subcellular location of proteins from their amino acid composition. The Reinhardt and Hubbard database was used to examine the performance of the neural network method. The rates of correct prediction for the three possible subcellular location of prokaryotic proteins were 96.1% by the self-consistency test and 84.4% by the jackknife test. The rates of correct prediction for the four possible subcellular location of eukaryotic proteins were 95.6% by the self-consistency test and 70.6% by the jackknife test.

Algorithms↗
PubMed1996

Neural-network-based classification of Space Acceleration Measurement Systems (SAMS) data via supervised learning.

This paper illustrates the applicability of neural networks in classifying events using Space Acceleration Measurement System (SAMS) data. Computer programs have been written in the MATLAB environment for the following purposes: automatic retrieval of SAMS data from NASA CDROM disks, computation of power spectral densities for SAMS data and construction of input patterns for the training of a multi-layer neural network (MNN). The MNN has been trained using the backpropagation learning algorithm and the SAMS data collected on the STS-50 Space Shuttle mission for three crew exercise events. It is found that the trained MNN is highly successful in classifying events. In addition, the performance of MNN is found to be better than that of the nearest neighbor classifier.

Acceleration↗
PubMed1995

Neural network analysis of quantitative histological factors to predict pathological stage in clinical stage I nonseminomatous testicular cancer.

A great deal of controversy exists in staging clinical stage I (CSI) nonseminomatous testicular germ cell tumors (NSGCT) because of the difficulty of distinguishing true stage I patients from those with occult retroperitoneal or distant metastases. The goal of this study was to quantitate primary tumor histologic factors and to apply these in a neural network computer analysis to determine if more accurate staging could be achieved. All available primary tumor histological slides from 93 CSI NSGCT patients were analyzed for vascular invasion (VI), lymphatic invasion (LI), tunical invasion (TI) and quantitative determination of percentage of the primary tumor composed of embryonal carcinoma (%EMB), yolk sac carcinoma (%YS), teratoma (%TER) and seminoma (%SEM). These patients had undergone retroperitoneal lymphadenectomy or follow-up such that final stage included 55 pathologic stage I and 38 stage II or higher lesions. Two investigators were provided identical datasets for neural network analysis; one experienced researcher used custom Kohonen and back propagation programs and one less experienced researcher used a commercially available program. For each experiment, a subset of data was used for training, and subsets were blindly used to test the accuracy of the networks. In the custom back propagation network, 86 of 93 patients were correctly staged for an overall accuracy of 92% (sensitivity 88%, specificity 96%). Using Neural Ware commercial software 74 of 93 (79.6%) were accurately staged when all 7 input variables were used; however, accuracy improved from 84.9 to 87.1% when 2, 4 and 5 of the variables were used. Quantitative histologic assessment of the primary tumor and neural network processing of data may provide clinically useful information in the CSI NSGCT population; however, the expertise of the network researcher appears to be important, and commercial software in general use may not be superior to standard regression analysis. Prospective testing of expert methodology should be instituted to confirm its utility.

Germinoma↗
PubMed2002

Optimal neural network architecture selection: improvement in computerized detection of microcalcifications.

RATIONALE AND OBJECTIVES: The authors evaluated the effect of optimal neural network architecture selection on the performance of a computer-aided diagnostic system designed to detect microcalcification clusters on digitized mammograms. MATERIALS AND METHODS: The authors developed a computer program to detect microcalcification clusters automatically on digitized mammograms. Previously, they found that a properly selected and trained convolution neural network (CNN) could reduce false-positive (FP) findings and therefore improve the accuracy of microcalcification detection. In the current study, they evaluated the effectiveness of the CNN optimized with an automated optimization technique in improving the accuracy of the microcalcification detection program, comparing it with the manually selected CNN. An independent test data set was used, which included 472 mammograms selected from the University of South Florida public database and contained 253 biopsy-proved malignant clusters. RESULTS: At an FP rate of 0.7 cluster per image, the film-based sensitivity was 84.6% for the optimized CNN, compared with 77.2% for the manually selected CNN. For clusters imaged on both craniocaudal and mediolateral oblique views, a cluster could be considered detected when it was detected on one or both views. For this case-based approach, at an FP rate of 0.7 per image, the sensitivity was 93.3% for the optimized and 87.0% for the manually selected CNN. CONCLUSION: The classification of true and false signals is an important step in the microcalcification detection program. An optimized CNN can effectively reduce FP findings and improve the accuracy of the computer-aided detection system.

Breast Neoplasms↗
PubMed2006

[The applied research on neural network filtrated by rough-set in insect taxonomy].

This article provides demonstrations and calculations, using rough-set theory and method, of the math-morphological features (MMFs), such as form parameter, lobation and sphericity, etc. drawn from 28 species of insects of the Hemiptera, Lepidoptera and Coleoptera based on their images. The results are compared with statistical analysis made by Zhao Hanqing, and also with the traditional classifications through the pattern recognition of neural network on the basis of the rough-set disposal. The result of the experiments showed that when used in categorical taxonomy, the importance of MMF was ranked from high to low: (roundness-likelihood. eccentricity) > (hot-hole number, sphericity, circularity) > (lobation, form parameter). The results of pattern recognition by neural network were completely identical with those of traditional classifications. Accordingly, the conclusion was that this theory applied in insect taxonomy was more idealistic compared with statistical analysis method, and it had great significance when used with rough-set neural network.

Algorithms↗
PubMed2004

Robust stability of interval bidirectional associative memory neural network with time delays.

In this paper, the conventional bidirectional associative memory (BAM) neural network with signal transmission delay is intervalized in order to study the bounded effect of deviations in network parameters and external perturbations. The resultant model is referred to as a novel interval dynamic BAM (IDBAM) model. By combining a number of different Lyapunov functionals with the Razumikhin technique, some sufficient conditions for the existence of unique equilibrium and robust stability are derived. These results are fairly general and can be verified easily. To go further, we extend our investigation to the time-varying delay case. Some robust stability criteria for BAM with perturbations of time-varying delays are derived. Besides, our approach for the analysis allows us to consider several different types of activation functions, including piecewise linear sigmoids with bounded activations as well as the usual C1-smooth sigmoids. We believe that the results obtained have leading significance in the design and application of BAM neural networks.

Journal Article↗
PubMed2000

Neural network analysis of clinicopathological and molecular markers in bladder cancer.

PURPOSE: To evaluate retrospectively the ability of an artificial neural network (ANN) to predict bladder cancer recurrence within 6 months of diagnosis and stage progression in patients with Ta/T1 bladder cancer, and 12-month cancer-specific survival in patients with T2-T4 bladder cancer. MATERIALS AND METHODS: Data were analyzed using a NeuralWorks Professional II/Plus software package. The input neural data consisted of clinicopathological and molecular characteristics. Distinct patient groups were used for the prediction of stage progression and tumor recurrence in Ta/T1 bladder cancers, and 12-month cancer-specific survival for patients with T2-T4 tumors. ANN predictions were compared with those of four consultant urologists. RESULTS: The accuracy of the neural network in predicting stage progression and recurrence within 6 months for Ta/T1 tumors and 12-month cancer-specific survival for T2-T4 cancers was 80%, 75% and 82% respectively; with corresponding figures for clinicians being 74%, 79% and 65%. On restricting the validation subset to patients with T1G3 tumors in relation to stage progression, the sensitivity of the ANN analysis increased to 100% with a specificity of 78% and an overall accuracy of 82%. The performance of the ANN in predicting stage progression in T1G3 tumors was significantly higher than that of clinicians (p = 0.25 for the ANN and p = 0.008 for clinicians, McNemar test). CONCLUSIONS: Data analysis using an ANN has been shown to be a useful adjunct in predicting outcomes in patients with bladder cancer and out-performs clinicians' predictions of stage progression in the high risk group of patients with T1G3 disease.

Biomarkers, Tumor↗
PubMed2001

Noninvasive prediction of sudden death and sustained ventricular tachycardia after acute myocardial infarction using a neural network algorithm.

BACKGROUND: The early and accurate noninvasive identification of postinfarction patients at risk of sudden death and sustained ventricular tachycardia (arrhythmic events) still remains an unsolved problem. The aim of the present study was to identify the combination of clinical and laboratory noninvasive variables, easy to obtain in most patients, that best predicts the occurrence of arrhythmic events after an acute myocardial infarction. METHODS: Four hundred and four consecutive patients with acute myocardial infarction were enrolled and followed for a median period of 21.4 months. In each patient, 61 clinical and laboratory noninvasive variables were collected before hospital discharge and used for the prediction of arrhythmic events using an artificial neural network. RESULTS: During follow-up, 13 (3.2%) patients died suddenly and 11(2.5%) had sustained ventricular tachycardia. The neural network showed that the combination best predicting arrhythmic events included: left ventricular failure during coronary care stay, ventricular dyskinesis, late potentials, number of ventricular premature depolarizations/hour, nonsustained ventricular tachycardia, left ventricular ejection fraction, bundle branch block and digoxin therapy at discharge. The neural network algorithm allowed identification of a small high-risk patient subgroup (12% of the study population) with an arrhythmic event rate of 46%. The sensitivity and specificity of the test were 96 and 93% respectively. CONCLUSIONS: These results suggest that, in postinfarction patients, it is possible to predict early and accurately arrhythmic events by noninvasive variables easily obtainable in most patients. Patients identified as being at risk are candidates for prophylactic antiarrhythmic therapy.

Aged↗
PubMed2004

Notions of intuition and attention modeled by a hierarchically arranged generalized regression neural network.

In this paper, two psychological functions, intuition and attention, are modeled by a newly proposed hierarchically arranged generalized regression neural network (HA-GRNN). The main contribution of the paper is two-fold: 1) to provide an engineering basis for a macroscopic representation of psychology-oriented functions by means of artificial neural networks; 2) to propose a concrete model for the two functions, intuition and attention, in terms of the associated interactive and evolutionary processes within an HA-GRNN. In the simulation study, the effectiveness of an HA-GRNN is justified within the context of pattern classification tasks.

Attention↗
PubMed1992

Determination of eukaryotic protein coding regions using neural networks and information theory.

Our previous work applied neural network techniques to the problem of discriminating open reading frame (ORF) sequences taken from introns versus exons. The method counted the codon frequencies in an ORF of a specified length, and then used this codon frequency representation of DNA fragments to train a neural net (essentially a Perceptron with a sigmoidal, or "soft step function", output) to perform this discrimination. After training, the network was then applied to a disjoint "predict" set of data to assess accuracy. The resulting accuracy in our previous work was 98.4%, exceeding accuracies reported in the literature at that time for other algorithms. Here, we report even higher accuracies stemming from calculations of mutual information (a correlation measure) of spatially separated codons in exons, and in introns. Significant mutual information exists in exons, but not in introns, between adjacent codons. This suggests that dicodon frequencies of adjacent codons are important for intron/exon discrimination. We report that accuracies obtained using a neural net trained on the frequency of dicodons is significantly higher at smaller fragment lengths than even our original results using codon frequencies, which were already higher than simple statistical methods that also used codon frequencies. We also report accuracies obtained from including codon and dicodon statistics in all six reading frames, i.e. the three frames on the original and complement strand. Inclusion of six-frame statistics increases the accuracy still further. We also compare these neural net results to a Bayesian statistical prediction method that assumes independent codon frequencies in each position. The performance of the Bayesian scheme is poorer than any of the neural based schemes, however many methods reported in the literature either explicitly, or implicitly, use this method. Specifically, Bayesian prediction schemes based on codon frequencies achieve 90.9% accuracy on 90 codon ORFs, while our best neural net scheme reaches 99.4% accuracy on 60 codon ORFs. "Accuracy" is defined as the average of the exon and intron sensitivities. Achievement of sufficiently high accuracies on short fragment lengths can be useful in providing a computational means of finding coding regions in unannotated DNA sequences such as those arising from the mega-base sequencing efforts of the Human Genome Project. We caution that the high accuracies reported here do not represent a complete solution to the problem of identifying exons in "raw" base sequences. The accuracies are considerably lower from exons of small length, although still higher than accuracies reported in the literature for other methods. Short exon lengths are not uncommon.(ABSTRACT TRUNCATED AT 400 WORDS)

Base Sequence↗
PubMed2004

Non-linear survival analysis using neural networks.

We describe models for survival analysis which are based on a multi-layer perceptron, a type of neural network. These relax the assumptions of the traditional regression models, while including them as particular cases. They allow non-linear predictors to be fitted implicitly and the effect of the covariates to vary over time. The flexibility is included in the model only when it is beneficial, as judged by cross-validation. Such models can be used to guide a search for extra regressors, by comparing their predictive accuracy with that of linear models. Most also allow the estimation of the hazard function, of which a great variety can be modelled. In this paper we describe seven different neural network survival models and illustrate their use by comparing their performance in predicting the time to relapse for breast cancer patients.

Breast Neoplasms↗
PubMed2006

Comparison of neurofuzzy logic and neural networks in modelling experimental data of an immediate release tablet formulation.

This study compares the performance of neurofuzzy logic and neural networks using two software packages (INForm and FormRules) in generating predictive models for a published database for an immediate release tablet formulation. Both approaches were successful in developing good predictive models for tablet tensile strength and drug dissolution profiles. While neural networks demonstrated a slightly superior capability in predicting unseen data, neurofuzzy logic had the added advantage of generating rule sets representing the cause-effect relationships contained in the experimental data.

Fuzzy Logic↗
PubMed1993

Neural network classification of mutagens using structural fragment data.

A neural network was applied to a large, structurally heterogeneous data set of mutagens and non-mutagens to investigate structure-property relationships. Substructural data comprising a total of 1280 fragments were used as inputs. The training of the back-propagation networks was directed by an algorithm which selected an optimal subset of fragments in order to maximize their discriminating power, and a good predictive network. The system comprised three programs: the first used a keyfile of 100 fragments to generate training and test files, the second was the network itself and a procedure for ranking the effectiveness of these fragments and the third randomly replaced the lowest fragments. This cycle was then repeated. After running on a 386/33 PC several networks produced approximately 11% failures in the test set and 6% in the training set. By simplifying the output of the hidden layer it was possible to describe the hidden layer states in terms of clusters of mutagens and non-mutagens. Some of these clusters were structurally homogeneous and contained known mutagenic and non-mutagenic structural classes. This analysis provided a useful means of demonstrating how the network was classifying the data.

Cluster Analysis↗
PubMed2005

A neural network model of foraging decisions made under predation risk.

This article develops the cognitive-emotional forager (CEF) model, a novel application of a neural network to dynamical processes in foraging behavior. The CEF is based on a neural network known as the gated dipole, introduced by Grossberg, which is capable of representing short-term affective reactions in a manner similar to Solomon and Corbit's (1974) opponent process theory. The model incorporates a trade-off between approach toward food and avoidance of predation under varying levels of motivation induced by hunger. The results of simulations in a simple patch selection paradigm, using a lifetime fitness criterion for comparison, indicate that the CEF model is capable of nearly optimal foraging and outperforms a run-of-luck rule-of-thumb model. Models such as the one presented here can illuminate the underlying cognitive and motivational components of animal decision making.

Animals↗
PubMed1998

Automated segmentation of anatomic regions in chest radiographs using an adaptive-sized hybrid neural network.

The purposes of this research are to investigate the effectiveness of our novel image features for segmentation of anatomic regions such as the lungs and the mediastinum in chest radiographs and to develop an automatic computerized method for image processing. A total of 85 screening chest radiographs from Johns Hopkins University Hospital were digitized to 2 K by 2.5 K pixels with 12 bit gray scale. To reduce the amount of information, the images were smoothed and subsampled to 256 by 310 pixels with 8 bit. The determination approach consists of classifying each pixel into two anatomic classes (lung and others) on the basis of several image features: (1) relative pixel address (Rx, Ry) based on lung edges extracted through image processing using profile, (2) density normalized from lungs and mediastinum density, and (3) histogram equalized entropy. The combinations of image features were evaluated using an adaptive-sized hybrid neural network consisting of an input, a hidden, and an output layer. Fourteen images were used for the training of the neural network and the remaining 71 images for testing. Using four features of relative address (Rx, Ry), normalized density, and histogram equalized entropy, the neural networks classified lungs at 92% accuracy against test images following the same rules as for the training images.

Biophysical Phenomena↗
PubMed2000

On the exact solution of the Parity-N problem using ordered neural networks.

In this paper, an exact solution of the Parity-N function representation problem is provided using a fully connected feedforward (ordered) neural network. The solution requires neither training nor adaptation, and employs (N - 1) neurons. It is shown that any function sigma: R --> R with the property: sigma(- 1) = sigma(1) = 1 and sigma(0) = 0, can be used as the neural network activation function.

Models, Neurological↗
PubMed2005

Use of support vector machines and neural network in diagnosis of neuromuscular disorders.

In this study the performance of support vector machine (SVM)and back-propagation neural network were applied to analyze the classification of the electromyogram (EMG) signals obtained from normal, neuropathy and myopathy subjects. By using autoregressive (AR) modeling, AR coefficients were obtained from EMG signals. Moreover, the support vector machine and artificial neural network (ANN) were used as base classifiers. The AR coefficients were benefited as inputs for SVM and ANN. Besides, these coefficients were tested both in ANN and SVM. The results show that SVM has high anticipation level in the diagnosis of neuromuscular disorders. It is proved that its test performance is high compared with ANN.

Adolescent↗