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 1,135 records · Page 63Linked to original sources
PubMed1991

Segmentation of magnetic resonance images using an artificial neural network.

Signal intensities from intermediate and T2 weighted spin echo images of the brain were used as inputs into an artificial neural network (ANN). The signal intensities were used to train the network to recognize anatomically-important segments. The ANN was a self-organizing map (SOM) neural network which develops a continuous topographical map of the signal intensities within the two images. The neural network segmented images demonstrated good correlation with white matter, gray matter, and cerebral spinal fluid (CSF) spaces. This technique was rated better than manual thresholding of the intermediate images, but not as good as manual thresholding of the T2 weighted images.

Brain↗
PubMed1992

Classification of oxygen transport to tissue with neural networks.

In the current stage of the development of the EMPHO II it was shown that the transport of oxygen to tissue can be classified for all applications in real time by using neural networks. The resolution of the oxygen classification is 1%. The implementation of the neural network is possible on a commercial signal processor TMS 320C30 board which is compatible with an IBM PC. The implementation of the neural network is done fully in software and no special neurocomputer is necessary. The results are very promising so that the neural network design goals will be established in the future. In the future further efforts will be made in order to increase the data base and to train the network in more detail.

Animals↗
PubMed1994

Simulated neural networks to predict outcomes, costs, and length of stay among orthopedic rehabilitation patients.

Our purpose was to develop a set of simulated neural networks that would predict functional outcomes, length of stay, and costs among orthopedic patients admitted to an inpatient rehabilitation hospital. We used retrospective data for a sample of 387 patients between the ages of 60 and 89 who had been admitted to a single rehabilitation facility over a period of 12 months. Using age and data on functional capacity at admission from the Functional Independence Measure, we were successful in constructing networks that were 86%, 87%, and 91% accurate in predicting functional outcome, length of stay, and costs to within +/- 15% of the actual value. In each case the accuracy of the network exceeded that of a multiple regression equation using the same variables. Our results show the feasibility of using simulated neural networks to predict rehabilitation outcomes, and the advantages of neural networks over conventional linear models. Networks of this kind may be of significant value to administrators and clinicians in predicting outcomes and resource usage as rehabilitation hospitals are faced with capitation and prospective payment schemes.

Activities of Daily Living↗
PubMed1995

Comparison of multilayer neural network and Nearest Neighbor Classifiers for handwritten digit recognition.

The basic Nearest Neighbor Classifier (NNC) is often inefficient for classification in terms of memory space and computing time needed if all training samples are used as prototypes. These problems can be solved by reducing the number of prototypes using clustering algorithms and optimizing the prototypes using a special neural network model. In this paper, we compare the performance of the multilayer neural network and an Optimized Nearest Neighbor Classifier (ONNC) for handwritten digit recognition applications. We show that an ONNC can have the same recognition performance as an equivalent neural network classifier. The ONNC can be efficiently implemented using prototype and variable ranking, partial summation and distance triangular inequality based strategies. It requires the same memory space as, but less, training time and classification time than the neural network.

Computers↗
PubMed1995

Artificial neural network control of FES in paraplegics for patient responsive ambulation.

This paper describes an ART-1-based artificial neural network (ANN) adapted for controlling functional electrical stimulation (FES) to facilitate patient-responsive ambulation by paralyzed patients with spinal cord injuries. This network is to serve as a controller in an FES system developed by the first author which is presently in use by 300 patients worldwide (still without ANN control) and which was the first and the only FES system approved by the FDA. The proposed neural network discriminates above-lesion upper-trunk electromyographic (EMG) time series to activate standing and walking functions under FES and controls FES stimuli levels using response-EMG signals. For this particular application, we introduce several modifications of the binary adaptive resonance theory (ART-1) for pattern recognition and classification. First, a modified on-line learning rule is proposed. The new rule assures bidirectional modification of the stored patterns and prevents noise interference. Second, a new reset rule is proposed which prevents "exact matching" when the input is a subset of the chosen pattern. We show the applicability of a single ART-1-based structure to solving two problems, namely, 1) signal pattern recognition and classification, and 2) control. This also facilitates ambulation of paraplegics under FES, with adequate patient interaction in initial system training, retraining the network when needed, and in allowing patient's manual override in the case of error, where any manual override serves as a retraining input to the neural network. Thus, the practical control problems (arising in actual independent patient ambulation via FES) were all satisfied by a relatively simple ANN design.

Electric Stimulation Therapy↗
PubMed2004

Adaptive neural network control for a class of MIMO nonlinear systems with disturbances in discrete-time.

In this paper, adaptive neural network (NN) control is investigated for a class of multiinput and multioutput (MIMO) nonlinear systems with unknown bounded disturbances in discrete-time domain. The MIMO system under study consists of several subsystems with each subsystem in strict feedback form. The inputs of the MIMO system are in triangular form. First, through a coordinate transformation, the MIMO system is transformed into a sequential decrease cascade form (SDCF). Then, by using high-order neural networks (HONN) as emulators of the desired controls, an effective neural network control scheme with adaptation laws is developed. Through embedded backstepping, stability of the closed-loop system is proved based on Lyapunov synthesis. The output tracking errors are guaranteed to converge to a residue whose size is adjustable. Simulation results show the effectiveness of the proposed control scheme.

Algorithms↗
PubMed1999

Application of neural networks in structure-activity relationships.

Methodology and application of artificial neural networks in structure-activity relationships are reviewed focusing on the most frequently used three-layer feedforward back-propagation procedure. Two applications of neural networks are presented and a comparison of the performance with those of CoMFA and a classical QSAR analysis is also discussed.

Drug Design↗
PubMed1997

Classification of low back pain from dynamic motion characteristics using an artificial neural network.

STUDY DESIGN: Data were collected from 183 subjects who were randomly assigned to the training and test groups. During testing of the classification system, knowledge of the low back pain condition or motion characteristics of the patients in the test group was not made available to the system. OBJECTIVES: To determine specific characteristics of trunk motion associated with different categories of spinal disorders and to determine whether a neural network analysis system can be effective in distinguishing patterns. SUMMARY OF BACKGROUND DATA: Numerous studies have established the difficulty of evaluating lower back pain. Imaging techniques are expensive and ineffective in many cases. A technique for evaluation of lower back pain was developed on the basis of analysis of such dynamic motion features as shape, velocity, and symmetry of movements, using a neural network classification system. METHODS: Dynamic motion data were collected from 183 subjects using a triaxial goniometer. Features of the movement were extracted and provided as input to a two-stage neural network classifier governed by a radial basis function architecture. After training, the output of the classifier was compared with Québec Task Force pain classifications obtained for the patients. Linear and nonlinear classification techniques were compared. RESULTS: The system could determine low back pain classification from motion characteristics. The neural network classifier produced the best results with up to 85% accuracy on novel "validation" data. CONCLUSIONS: A neural network based on kinematic data is an excellent predictive model for classification of lower back pain. Such a system could markedly improve the management of lower back pain in the individual patient.

Adult↗
PubMed1993

Artificial neural networks for single photon emission computed tomography. A study of cold lesion detection and localization.

RATIONALE AND OBJECTIVES: An artificial neural network was developed for cold lesion detection and localization in single photon emission computed tomography (SPECT) images. METHODS: The network was trained for several noise levels and lesion sizes to identify lesions located in the center of small image neighborhoods. When scrolled across an image the trained network was able to identify cold abnormalities. The diagnostic performance of the technique was evaluated at two noise levels (50,000 and 100,000 counts/slice) and for two lesion sizes (radius: 1.0 cm and 1.5 cm) using the free-response operating characteristic (FROC) analysis. Furthermore, the same network was tested on a situation it was not trained on (80,000 counts/slice and a different reconstruction filter). RESULTS: The neural network showed high sensitivity and small false-positive rates per image for all test situations. These results suggest that neural networks are promising tools for computer-aided clinical diagnosis in SPECT:

Equipment Design↗
PubMed2006

Ion conducting polymer microelectrodes for interfacing with neural networks.

We have examined the stimulation and recording properties of conjugated polymer microelectrode arrays as interfaces with neural networks of dissociated cortical cells. In particular the stimulation properties were investigated as a means of supplying a neural network with information. The stimulation efficiency at low stimulation voltages was evaluated and referenced to bare indium tin oxide (ITO) electrodes. The polymer electrodes were electrochemically polymerized from a blend of poly(3,4-ethylenedioxythiophene)-poly(styrenesulfonate) (PEDOT-PSS) and ethylenedioxythiophene (EDOT) onto ITO microelectrodes. Dissociated cortical cells were then plated on the electrodes and cultivated to form neural networks. Polymer electrode stimulation evoked a much greater response from the network than stimulation from ITO electrodes. Neural interfaces using polymer electrodes could be maintained for several months.

Action Potentials↗
PubMed2004

[Application of artificial neural network based on the genetic algorithm in predicting the root distribution of winter wheat].

In this study, a controlled experiment of winter wheat under water stress at the seedling stage was conducted in soil columns in greenhouse. Based on the data gotten from the experiment, a model to estimate root length density distribution was developed through optimizing the weights of neural network by genetic algorithm. The neural network model was constructed by using forward neural network framework, by applying the strategy of the roulette wheel selection and reserving the most optimizing series of weights, which were composed by real codes. This model was applied to predict the root length density distribution of winter wheat, and the predicted root length density had good agreement with experiment data. The way could save a lot of manpower and material resources for determining the root length density distribution of winter wheat.

Algorithms↗
PubMed2001

Segmentation of medical ultrasonic image using hybrid neural network.

Objective. To solve one of the most difficult problems in multi-dimensional reconstruction of medical ultrasonic images: image segmentation. Method. A new segmental method based on hybrid neural network was presented in this paper. The hybrid neural network comprised two phases. The first phase was Kohonens self-organization neural network, which was used to segment and label the image coarsely. The feature vectors of those pixels within a specified distance from the cluster centers were employed to train the second phase--a three-layer perception network using back-propagation (BP) technique. Then the trained BP network was used to label every pixel of the image. In the end, a post-processing stage was used to remove the small isolated points and smooth out the contours of the segmented image. Result. The segmented image had smooth continuous edges, few noises or speckles, and the contour of ventricle was clear and accurate. Conclusion. Our method could segment the ultrasonic images accurately and effectively, and had a lot of advantages compared to traditional methods. The unsupervised segmentation problems could be solved using supervised methods.

Cardiovascular Physiological Phenomena↗
PubMed1989

On the ability of neural networks to perform generalization by induction.

The ability of neural networks to perform generalization by induction is the ability to learn an algorithm without the benefit of complete information about it. We consider the properties of networks and algorithms that determine the efficiency of generalization. These properties are described in quantitative terms. The most effective generalization is shown to be achieved by networks with the least admissible capacity. General conclusions are illustrated by computer simulations for a three-layered neural network. We draw a quantitative comparison between the general equations and specific results reported here and elsewhere.

Algorithms↗
PubMedDate not supplied

[Physical models of neural networks].

The state of art in computer modelling of neural networks with associative memory is reviewed. The available experimental data are considered on learning and memory of small neural systems, on isolated synapses and on molecular level. Computer simulations demonstrate that realistic models of neural ensembles exhibit properties which can be interpreted as image recognition, categorization, learning, prototype forming, etc. A bilayer model of associative neural network is proposed. One layer corresponds to the short-term memory, the other one to the long-term memory. Patterns are stored in terms of the synaptic strength matrix. We have studied the relaxational dynamics of neurons firing and suppression within the short-term memory layer under the influence of the long-term memory layer. The interaction among the layers has found to create a number of novel stable states which are not the learning patterns. These synthetic patterns may consist of elements belonging to different non-intersecting learning patterns. Within the framework of a hypothesis of selective and definite coding of images in brain one can interpret the observed effect as the "ideaś generating" process.

Animals↗
PubMed1996

Improving prediction of protein secondary structure using structured neural networks and multiple sequence alignments.

The prediction of protein secondary structure by use of carefully structured neural networks and multiple sequence alignments has been investigated. Separate networks are used for predicting the three secondary structures alpha-helix, beta-strand, and coil. The networks are designed using a priori knowledge of amino acid properties with respect to the secondary structure and the characteristic periodicity in alpha-helices. Since these single-structure networks all have less than 600 adjustable weights, overfitting is avoided. To obtain a three-state prediction of alpha-helix, beta-strand, or coil, ensembles of single-structure networks are combined with another neural network. This method gives an overall prediction accuracy of 66.3% when using 7-fold cross-validation on a database of 126 nonhomologous globular proteins. Applying the method to multiple sequence alignments of homologous proteins increases the prediction accuracy significantly to 71.3% with corresponding Matthew's correlation coefficients C alpha = 0.59, C beta = 0.52, and Cc = 0.50. More than 72% of the residues in the database are predicted with an accuracy of 80%. It is shown that the network outputs can be interpreted as estimated probabilities of correct prediction, and, therefore, these numbers indicate which residues are predicted with high confidence.

Amino Acid Sequence↗
PubMed1998

Effective learning in recurrent max-min neural networks.

Max and min operations have interesting properties that facilitate the exchange of information between the symbolic and real-valued domains. As such, neural networks that employ max-min activation functions have been a subject of interest in recent years. Since max-min functions are not strictly differentiable, we propose a mathematically sound learning method based on using Fourier convergence analysis of side-derivatives to derive a gradient descent technique for max-min error functions. We then propose a novel recurrent max-min neural network model that is trained to perform grammatical inference as an application example. Comparisons made between this model and recurrent sigmoidal neural networks show that our model not only performs better in terms of learning speed and generalization, but that its final weight configuration allows a deterministic finite automation (DFA) to be extracted in a straightforward manner. In essence, we are able to demonstrate that our proposed gradient descent technique does allow max-min neural networks to learn effectively.

Journal Article↗
PubMed1998

Gait restoration in a spinal cord injured subject via neuromuscular electrical stimulation controlled by an artificial neural network.

Attempts to restore gait in spinal cord injured subjects have stumbled on control difficulties associated to the neuromuscular system's non-linearity and time-variance. Thus, a simple autoadaptive artificial neural network has been devised to control gait swing generation by means of neuromuscular electrical stimulation. Both theoretical and experimental approaches were taken. The computer-based system consisted of a three-layer artificial neural network that read angular data from the hip, knee and ankle joints. The output signal consisted of variations on the applied stimulation pulse width. Surface electrical stimulation was applied to the femoral and peroneal nerves of one leg. Neural network training included off-line supervised learning schemes. The system was tested on a male subject with an incomplete C6-level lesion. Several tests were run to determine whether the off-line trained neural network could correctly control the motion. The effect of on-line learning upon the control performance was also evaluated. The system was found to control the motion with success only at times. Control performance was found to improve in response to the application of on-line learning. Learning stability following on-line learning was found to be satisfactory. In a final test, the artificial neural system had appropriate responses to an initial perturbation, which suggests that further research in this area should be pursued.

Algorithms↗
PubMed2005

Bayesian neural networks for bivariate binary data: an application to prostate cancer study.

Prostate cancer is one of the most common cancers in American men. The cancer could either be locally confined, or it could spread outside the organ. When locally confined, there are several options for treating and curing this disease. Otherwise, surgery is the only option, and in extreme cases of outside spread, it could very easily recur within a short time even after surgery and subsequent radiation therapy. Hence, it is important to know, based on pre-surgery biopsy results how likely the cancer is organ-confined or not. The paper considers a hierarchical Bayesian neural network approach for posterior prediction probabilities of certain features indicative of non-organ confined prostate cancer. In particular, we find such probabilities for margin positivity (MP) and seminal vesicle (SV) positivity jointly. The available training set consists of bivariate binary outcomes indicating the presence or absence of the two. In addition, we have certain covariates such as prostate specific antigen (PSA), gleason score and the indicator for the cancer to be unilateral or bilateral (i.e. spread on one or both sides) in one data set and gene expression microarrays in another data set. We take a hierarchical Bayesian neural network approach to find the posterior prediction probabilities for a test and validation set, and compare these with the actual outcomes for the first data set. In case of the microarray data we use leave one out cross-validation to access the accuracy of our method. We also demonstrate the superiority of our method to the other competing methods through a simulation study. The Bayesian procedure is implemented by an application of the Markov chain Monte Carlo numerical integration technique. For the problem at hand, our Bayesian bivariate neural network procedure is shown to be superior to the classical neural network, Radford Neal's Bayesian neural network as well as bivariate logistic models to predict jointly the MP and SV in a patient in both the data sets as well as in the simulation study.

Bayes Theorem↗