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 667 records · Page 37Linked to original sources
PubMed2005

[A primary study on simulation of fracture strength based on BP neural networks].

OBJECTIVE: To model the relationship between stimulating stress and fracture strength using BP neural networks, and to provide a theoretical basis for accurate prediction of the rate of fracture healing. METHODS: The bilateral tibiae in New Zealand rabbits were osteotomized and fixed by stress-relaxation plate(SRP) and rigid plate(RP), respectively. The stress shielding rate and bending strength of the healing fractures were measured at 2 to 48 weeks postoperatively. A BP neural network was constructed and trained using the experimental data of the stress-relaxation group. Then the trained network was used for simulation to predict fracture strength of the two groups from the stress at the fracture site. RESULTS: With the input of the data that has been used to train the network, fracture strength similar to those measured in experiment was calculated from the BP neural network. However, poor results were obtained with the input of new data. CONCLUSION: BP neural network can be used to investigate the influence of various factors on fracture healing quantitatively, and to predict the rate of healing. However, the model still needs to be perfected. More experimental or clinical data are needed to train the network

Animals↗
PubMed2006

The use of artificial neural networks in decision support in cancer: a systematic review.

Artificial neural networks have featured in a wide range of medical journals, often with promising results. This paper reports on a systematic review that was conducted to assess the benefit of artificial neural networks (ANNs) as decision making tools in the field of cancer. The number of clinical trials (CTs) and randomised controlled trials (RCTs) involving the use of ANNs in diagnosis and prognosis increased from 1 to 38 in the last decade. However, out of 396 studies involving the use of ANNs in cancer, only 27 were either CTs or RCTs. Out of these trials, 21 showed an increase in benefit to healthcare provision and 6 did not. None of these studies however showed a decrease in benefit. This paper reviews the clinical fields where neural network methods figure most prominently, the main algorithms featured, methodologies for model selection and the need for rigorous evaluation of results.

Animals↗
PubMed2004

Abciximab pharmacodynamic model with neural networks used to integrate sources of patient variability.

OBJECTIVE: Our objective was to develop a computational model for predicting abciximab-induced inhibition of ex vivo platelet aggregation from the administered dose and readily available patient clinical characteristics by use of a neural network approach. METHODS: A back-propagation neural network was designed to establish the relationship between abciximab dosing, patient clinical history, and effect (inhibition of 20 micromol/L adenosine diphosphate-induced ex vivo platelet aggregation). The neural network was trained by use of data from 8 (out of 47) patients undergoing coronary angioplasty and 30 healthy individuals. Final neuron connection weights were used to evaluate significant patient covariates. The final neural network was validated via (1) predicting the effects of the validation database (remaining 39 patients) and (2) predicting the individual patient doses to achieve 20% of baseline platelet aggregation. RESULTS: The trained neural network successfully captured the complex pharmacodynamic profiles of abciximab without specifying a structural model and identified several patient covariates that significantly contribute to establishing the abciximab dose-effect relationship, including stable angina with nitrate treatment, previous myocardial infarction, and smoking. A wide distribution of individual bolus doses of abciximab was predicted, suggesting the potential for dosing individualization while improving the risk of adverse drug events. The mean predicted dose (16.9 mg) was in agreement with the results from a previously published concentration-effect relationship for abciximab (18.9 +/- 2.0 mg). CONCLUSIONS: These findings suggest the usefulness of neural network methods to individualize dosing for drugs with a narrow therapeutic index when real-time measures of drug concentration and effect are unavailable, but future clinical studies are required for prospective validation.

Abciximab↗
PubMed2005

A delayed neural network for solving linear projection equations and its analysis.

In this paper, we present a delayed neural network approach to solve linear projection equations. The Lyapunov-Krasovskii theory for functional differential equations and the linear matrix inequality (LMI) approach are employed to analyze the global asymptotic stability and global exponential stability of the delayed neural network. Compared with the existing linear projection neural network, theoretical results and illustrative examples show that the delayed neural network can effectively solve a class of linear projection equations and some quadratic programming problems.

Algorithms↗
PubMed2003

Neural network approach to classify infective keratitis.

PURPOSE: Infective keratitis is a major sight-threatening condition in developing countries like India. An early diagnosis of infective keratitis is critical to its treatment. Epidemiological trends, morphological features of corneal ulceration and presence of other risk factors often dictate choice of initial treatment. This work assesses the usefulness of classification of infective keratitis by artificial neural network (ANN). METHODS: Forty input variables from each of the sixty-three known bacterial or fungal ulcers provided the basis for training a three layer feed-forward neural network. The trained neural network classified another set of forty-three corneal ulcers. RESULTS: Trained artificial neural network could classify correctly all sixty-three cornea ulcers in the training set. In the test set, the artificial neural network correctly classified 39 out of 43 cornea ulcers. Specificity for bacterial and fungal categories was 76.47% and 100% respectively. Accuracy of classification by neural network was 90.7% and compared significantly better than clinicians' prediction of 62.8% (p < 0.01). CONCLUSION: ANN has the potential to help clinicians classify corneal ulcers more accurately.

Corneal Ulcer↗
PubMed1996

Neural networks in ventilation-perfusion imaging.

PURPOSE: To optimize the performance of artificial neural networks in the prediction of pulmonary embolism from ventilation-perfusion (V-P) scans. MATERIALS AND METHODS: Neural networks were constructed with a set of V-P scan criteria that included sharpness and completeness of perfusion defects and involved quantification of abnormalities by using a continuous numeric scale. Several network parameters were systematically varied. Networks were trained with 150 cases and tested with 30 different cases. Findings were compared with those of pulmonary angiography. RESULTS: Networks capable of performing as well as experienced nuclear medicine physicians could be constructed with few V-P scan features. A brief training period was optimal (50-100 iterations). Further training diminished network performance. CONCLUSION: Effective neural networks can be constructed by using a limited number of unconventional V-P scan features. Several parameters can be adjusted to optimize performance.

Adolescent↗
PubMed1995

Application of neural networks to the real-time diagnosis of acute toxoplasmic infection in immunocompetent patients.

Neural networks constitute a relatively new, radically different approach to the interpretation and recognition of subtle diagnostic patterns in multivariate data. In this study the use of neural networks with a single serum sample for rapid real-time recognition of recent toxoplasmic infection was investigated. A neural-network model was implemented on the basis of data obtained by four serological methods--dye test, indirect fluorescence assay, indirect hemagglutination assay, and IgM immunosorbent agglutination assay--and was "trained" to extract features of acute infection by application to an analysis of 65 immunocompetent patients, 10 of whom were in fact acutely infected. The trained model correctly classified all 10 cases of acute infection. On its application to 61 additional infected patients, this method correctly identified seven cases as potentially acute. Our study shows that neural networks can discern diagnostic patterns from variables that individually have limited utility in the diagnosis of acute toxoplasmosis.

Acute Disease↗
PubMed2006

Global exponential stability of multitime scale competitive neural networks with nonsmooth functions.

In this paper, we study the global exponential stability of a multitime scale competitive neural network model with nonsmooth functions, which models a literally inhibited neural network with unsupervised Hebbian learning. The network has two types of state variables, one corresponds to the fast neural activity and another to the slow unsupervised modification of connection weights. Based on the nonsmooth analysis techniques, we prove the existence and uniqueness of equilibrium for the system and establish some new theoretical conditions ensuring global exponential stability of the unique equilibrium of the neural network. Numerical simulations are conducted to illustrate the effectiveness of the derived conditions in characterizing stability regions of the neural network.

Algorithms↗
PubMed1995

Extraction of fuzzy rules using neural networks with structure level adaptation and its application to diagnosis for hepatobiliary disorders.

First, this paper presents the reasoning and the learning method for fuzzy rules using structure level adaptation of neural networks. In a usual neural network's mechanism, during learning process of rules, we can observe the following two behaviors: Case 1: If a neural network does not have enough neurons to be satisfied to infer, then the input weight vector will have a tendency to fluctuate greatly, even after a certain long period the learning process. In this case, the network needs to generate a new neuron as its parent's attribute is inherited. Case 2: If a neural network has enough neurons to infer, and even if the input weight vector of each neuron will converge to a certain value, then we shall be able to turn out unnecessary neurons from the network in the calculation. In this case, because it is necessary to delete a redundant neuron to the calculation, the neuron is annihilated without affecting the performance of the network. By observing such behaviors, we can generate or annihilate the specified neuron respectively to achieve an overall good system. In the proposed method, we described a procedure to derive the neuron generation/annihilation automatically and applied the procedure to learning system. Next, we apply such procedure to the learning system in which the experimental data related to hepatobiliary disorders is used. We use a real medical database containing the results of ten biochemical terms test for four hepatobiliary disorders. We have 536 case data, including some errors. After the learning, by using 179 data chosen randomly from database, the proposed system converged to a certain small value and this constructed network has the optimal structure for these teaching data. In addition, we get that the fuzzy rules have some meanings related to the degree of the input weight vector, and the fuzzy rules for hepatobiliary disorders are extracted from the learned network with respect to the degree of input weight vector. Moreover, to verify the validity of the diagnosis of the proposed method, the feed-forward calculation was implemented using extracted fuzzy rules for all databases. As a result, the proposed system correctly diagnosed more than 70%.

Biliary Tract Diseases↗
PubMed2000

Interacting neural networks.

Several scenarios of interacting neural networks which are trained either in an identical or in a competitive way are solved analytically. In the case of identical training each perceptron receives the output of its neighbor. The symmetry of the stationary state as well as the sensitivity to the used training algorithm are investigated. Two competitive perceptrons trained on mutually exclusive learning aims and a perceptron which is trained on the opposite of its own output are examined analytically. An ensemble of competitive perceptrons is used as decision-making algorithms in a model of a closed market (El Farol Bar problem or the Minority Game. In this game, a set of agents who have to make a binary decision is considered.); each network is trained on the history of minority decisions. This ensemble of perceptrons relaxes to a stationary state whose performance can be better than random.

Algorithms↗
PubMed2004

[Research on the application of neural network to diagnosis of cardiopathy].

Neural networks can fit any nonlinear function. After drawing out several characteristic parameters from the three-dimension spectrum for high frequency QRS waves, we input them into the network and trained the network. In this way, we can get a m-dimension curved surface in the m-dimension space which is constructed by those parameters, and this curved surface divides the space into two parts: the unhealthiness and the health. Now, the network can automatically distinguish between the healthiness and the unhealthiness according to their three-dimension spectrum for high frequency QRS waves.

Algorithms↗
PubMed2006

[Artificial neural network in the prediction of nosocomial infection risk].

OBJECTIVE: To establish a model based on artificial neural network in the prediction of nosocomial infection risk. METHODS: Clinical data of 27,352 inpatients extracted from hospital information system were cleaned and coded, and the model of prediction in nosocomial infection risk was developed based on artificial neural network. RESULTS: The structure of artificial neural network is {16-6-1}-BP, and the fit rate of prediction was 0.9891. The area under ROC curve was 0.986. CONCLUSION: Artificial neural network model can be used as a tool for nosocomial infection forecasting, which can provide supplementary information for the diagnosis and control of nosocomial infection.

Cross Infection↗
PubMed1995

Use of neural networks in predicting the risk of coronary artery disease.

Artificial neural networks were created to predict the occurrence of coronary artery disease based on information from the serum lipid profile. The development of the networks involved a strategy which permitted learning from censored observations. The networks were developed with data from the Cholesterol Lowering Atherosclerosis Study, which followed serum lipoprotein levels and clinical events in 162 patients over a period of up to 10 years. Inputs consisted of seven different mean lipid values, and the desired output was the time period during which a complication of coronary artery disease was predicted to occur. Cross-validation was performed by splitting the data into separate training and testing sets, scoring the performance of the neural network strategy on the testing sets, and comparing scores with those obtained from Cox regression models developed on the same training data. Performance of the neural network strategy exceeded that of Cox regression in predicting clinical outcomes (66% vs 56%, McNemar's test P = 0.005). The network design provided an effective approach to predicting outcomes from a clinical trial with variable follow-up times.

Adult↗
PubMed2006

Learning in human neural networks on microelectrode arrays.

This paper describes experiments involving the growth of human neural networks of stem cells on a MEA (microelectrode array) support. The microelectrode arrays (MEAs) are constituted by a glass support in which a set of tungsten electrodes are inserted. The artificial neural network (ANN) paradigm was used by stimulating the neurons in parallel with digital patterns distributed on eight channels, then by analyzing a parallel multichannel output. In particular, the microelectrodes were connected following two different architectures, one inspired by the Kohonen's SOM, the other by the Hopfield network. The output signals have been analyzed in order to evaluate the possibility of organized reactions by the natural neurons.f The results show that the network of human neurons reacts selectively to the subministered digital signals, i.e., it produces similar output signals referred to identical or similar patterns, and clearly differentiates the outputs coming from different stimulations. Analyses performed with a special artificial neural network called ITSOM show the possibility to codify the neural responses to different patterns, thus to interpret the signals coming from the network of biological neurons, assigning a code to each output. It is straightforward to verify that identical codes are generated by the neural reactions to similar patterns. Further experiments are to be designed that improve the hybrid neural networks' capabilities and to test the possibility of utilizing the organized answers of the neurons in several ways.

Embryonic Stem Cells↗
PubMed1998

Neural networks predict response biases of female túngara frogs.

Artificial neural networks have become useful tools for probing the origins of perceptual biases in the absence of explicit information on underlying neuronal substrates. Preceding studies have shown that neural networks selected to recognize or discriminate simple patterns may possess emergent biases toward pattern size of symmetry--preferences often exhibited by real females--and have investigated how these biases shape signal evolution. We asked whether simple neural networks could evolve to respond to an actual mate recognition signal, the call of the túngara frog, Physalaemus pustulosus. We found that not only were networks capable of recognizing the call of the túngara frog, but that they made remarkably accurate quantitative predictions about how well females generalized to many novel calls, and that these predictions were stable over several architectures. The data suggest that the degree to which P. pustulosus females respond to a call may often be an incidental by-product of a sensory system selected simply for species recognition.

Animal Communication↗
PubMed2006

Real-time learning capability of neural networks.

In some practical applications of neural networks, fast response to external events within an extremely short time is highly demanded and expected. However, the extensively used gradient-descent-based learning algorithms obviously cannot satisfy the real-time learning needs in many applications, especially for large-scale applications and/or when higher generalization performance is required. Based on Huang's constructive network model, this paper proposes a simple learning algorithm capable of real-time learning which can automatically select appropriate values of neural quantizers and analytically determine the parameters (weights and bias) of the network at one time only. The performance of the proposed algorithm has been systematically investigated on a large batch of benchmark real-world regression and classification problems. The experimental results demonstrate that our algorithm can not only produce good generalization performance but also have real-time learning and prediction capability. Thus, it may provide an alternative approach for the practical applications of neural networks where real-time learning and prediction implementation is required.

Computer Systems↗
PubMed2004

Analysis of speedup as function of block size and cluster size for parallel feed-forward neural networks on a Beowulf cluster.

The performance of feed-forward neural networks trained with the backpropagation algorithm on a dedicated Beowulf cluster is analyzed. The concept of training set parallelism is applied. A new model for run time and speedup prediction is developed. With the model the speedup and efficiency of one iteration of the neural networks can be estimated as a function of block size and cluster size. The model is applied to three example problems representing different applications and network architectures. The estimation of the model has a higher accuracy than traditional methods for run time estimation and can be efficiently calculated. Experiments show that speedup of one iteration does not necessarily translate to a shorter training time toward a given error level. To overcome this problem a heuristic extension to training set parallelism called weight averaging is developed. The results show that training in parallel should only be done on clusters with high performance network connections or a multiprocessor machine. A rule of thumb is given for how much network performance of the cluster is needed to achieve speedup of the training time for a neural network.

Cluster Analysis↗
PubMed1996

Singular perturbation analysis of competitive neural networks with different time scales.

The dynamics of complex neural networks must include the aspects of long- and short-term memory. The behavior of the network is characterized by an equation of neural activity as a fast phenomenon and an equation of synaptic modification as a slow part of the neural system. The main idea of this paper is to apply a stability analysis method of fixed points of the combined activity and weight dynamics for a special class of competitive neural networks. We present a quadratic-type Lyapunov function for the flow of a competitive neural system with fast and slow dynamic variables as a global stability method and a modality of detecting the local stability behavior around individual equilibrium points.

Algorithms↗