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 1,225 records · Page 68Linked to original sources
PubMed1992

Analysis of the clinical variables driving decision in an artificial neural network trained to identify the presence of myocardial infarction.

STUDY OBJECTIVE: To determine which clinical variables drive the output of an artificial neural network trained to identify the presence of myocardial infarction. DESIGN: Partial output analysis. SETTING: Tertiary university teaching center. PARTICIPANTS: Seven hundred six patients more than 18 years old presenting with anterior chest pain. MEASUREMENTS: Differential network output analysis. MAIN RESULTS: A methodology was developed as the first step in measuring the impact input clinical variables have on the output (diagnosis) of an artificial neural network trained to identify the presence of acute myocardial infarction. The methodology revealed that the network used the presence of ECG findings, as well as the presence of rales, syncope, jugular venous distension, response to trinitroglycerin, and nausea and vomiting, as major predictive sources. Although this first-step analysis studied individual variables, it must be stated that the network comes to clinical closure based on the settings of all variables in a pattern and that the impact of a single variable cannot be taken out of the context of a pattern. CONCLUSION: An artificial neural network trained to recognize the presence of myocardial infarction appears to place diagnostic importance on clinical variables that have not been shown previously to be highly predictive for infarction.

Adolescent↗
PubMed1996

Information processing in negative feedback neural networks.

Information theory suggests that extraction of the principal sub-space from data is useful when the input to a neural network is corrupted with additive noise. A number of neural network algorithms exist which can find this principal sub-space, many of which also extract the principal components of the input. However, when there is noise on both input and output of a network, simply extracting the principal sub-space (or components) is not sufficient to optimize information capacity. An approximate solution to maximizing information capacity would be to extract the principal sub-space of components with variances above a certain threshold, and then ensure that these are uncorrelated and that they have equal variance at the output. A neural network is described which uses negative feedback connections to achieve this uncorrelated, equal-variance solution.

Journal Article↗
PubMed2006

Combining neural network models for automated diagnostic systems.

This paper illustrates the use of combined neural network (CNN) models to guide model selection for diagnosis of internal carotid arterial (ICA) disorders. The ICA Doppler signals were decomposed into time-frequency representations using discrete wavelet transform and statistical features were calculated to depict their distribution. The first level networks were implemented for the diagnosis of ICA disorders using the statistical features as inputs. To improve diagnostic accuracy, the second level network was trained using the outputs of the first level networks as input data. The CNN models achieved accuracy rates which were higher than that of the stand-alone neural network models.

Diagnosis, Computer-Assisted↗
PubMed2000

Evaluating the frequency rate of hypomagnesemia in critically ill pediatric patients by using multiple regression analysis and a computer-based neural network.

OBJECTIVES: To determine the frequency rate of hypomagnesemia in patients admitted to the pediatric intensive care unit (ICU), and to identify subsets of patients (grouped by disease) who are at greatest risk of hypomagnesemia. We also compared a neural network model with multiple regression analysis to identify independent variables that would correlate with hypomagnesemia and to predict serum magnesium values in critically ill pediatric patients overall. DESIGN: Prospective, multicenter study. SETTING: Tertiary level medical/surgical pediatric ICUs. PATIENTS: Data were obtained at admission to the pediatric ICU for 463 patients from newborn to 18 yrs old who were admitted with a variety of surgical and nonsurgical conditions. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Total serum magnesium values were obtained within the first 24 hrs after admission in 463 pediatric patients admitted to four pediatric ICUs. Hypomagnesemia (defined as total serum magnesium <0.75 mmol/L) was found in 51 (11%) of the 463 patients, with the highest frequency rate (72%) and lowest mean serum magnesium level (0.66 +/- 0.17 mmol/L) in patients admitted after surgery with extensive osseous involvement (spinal fusion and craniofacial reconstruction). To determine whether hypomagnesemia could be predicted on the basis of other laboratory and clinical criteria, multiple regression analysis was performed and showed age, weight, and albumin levels weakly associated (r2 = .14, p < .001) with magnesium levels within the different diagnostic groups. These data were used to produce a mathematical model able to predict magnesium levels within 5% of the actual values in 23% of patients. A neural network was also created to compare its predictive capabilities to those of the multiple regression model. Once trained on a random subset (85%) of the patient population, the neural network was able to predict magnesium levels to within 5% of actual values for 88% of the remaining 15% of patients, comparing favorably with the predictions derived from the multiple regression model. CONCLUSIONS: Hypomagnesemia is not uncommon (11%) in critically ill pediatric patients, but is very common (72%) in patients admitted after surgery for spinal fusion or craniofacial reconstruction. Patients who undergo surgery for correction of scoliosis and craniofacial anomalies should have serum magnesium levels monitored closely after surgery. In other patients, a neural network or multiple regression model could help predict which patients would be at risk of developing hypomagnesemia, thereby focusing testing on patients likely to benefit from such testing.

Adolescent↗
PubMed1994

Reduction of false positives in computerized detection of lung nodules in chest radiographs using artificial neural networks, discriminant analysis, and a rule-based scheme.

A computer-aided diagnosis (CAD) scheme is being developed to identify image regions considered suspicious for lung nodules in chest radiographs to assist radiologists in making correct diagnoses. Automated classifiers--an artificial neural network, discriminant analysis, and a rule-based scheme--are used to reduce the number of false-positive detections of the CAD scheme. The CAD scheme first detects nodule candidates from chest radiographs based on a difference image technique. Nine image features characterizing nodules are extracted automatically for each of the nodule candidates. The extracted image features are then used as input data to the classifiers for distinguishing actual nodules from the false-positive detections. The performances of the classifiers are evaluated by receiver-operating characteristic analysis. On the basis of the database of 30 normal and 30 abnormal chest images, the neural network achieves an AZ value (area under the receiver-operating-characteristic curve) of 0.79 in detecting lung nodules, as tested by the round-robin method. The neural network, after being trained with a training database, is able to eliminate more than 83% of the false-positive detections reported by the CAD scheme. Moreover, the combination of the trained neural network and a rule-based scheme eliminates 96% of the false-positive detections of the CAD scheme.

Diagnosis, Computer-Assisted↗
PubMed2006

ON the influence of parameter theta- on performance of RBF neural networks trained with the dynamic decay adjustment algorithm.

The dynamic decay adjustment (DDA) algorithm is a fast constructive algorithm for training RBF neural networks (RBFNs) and probabilistic neural networks (PNNs). The algorithm has two parameters, namely, theta(+) and theta(-). The papers which introduced DDA argued that those parameters would not heavily influence classification performance and therefore they recommended using always the default values of these parameters. In contrast, this paper shows that smaller values of parameter theta(-) can, for a considerable number of datasets, result in strong improvement in generalization performance. The experiments described here were carried out using twenty benchmark classification datasets from both Proben1 and the UCI repositories. The results show that for eleven of the datasets, the parameter theta(-) strongly influenced classification performance. The influence of theta(-) was also noticeable, although much less, on six of the datasets considered. This paper also compares the performance of RBF-DDA with theta(-) selection with both AdaBoost and Support Vector Machines (SVMs).

Algorithms↗
PubMed2000

Modeling the toxicity of chemicals to Tetrahymena pyriformis using molecular fragment descriptors and probabilistic neural networks.

The results of an investigation into the use of a probabilistic neural network (PNN)-based methodology to model the 48-h ICG50 (inhibitory concentration for population growth) sublethal toxicity of 825 chemicals to the ciliate Tetrahymena pyriformis are presented. The information fed into the neural networks is solely based on simple molecular descriptors as can be derived from the chemical structure. In contrast to most other toxicological models, the octanol/water partition coefficient is not used as an input parameter, and no rules of thumb or other substance selection criteria are employed. The cross-validation and external validation experiments confirmed excellent recognitive and predictive capabilities of the resulting models and recommend their future use in evaluating the potential of most organic molecules to be toxic to Tetrahymena.

Animals↗
PubMed2003

Use of artificial neural networks to estimate production variables of broilers breeders in the production phase.

1. Although the poultry industry uses state-of-the-art equipment and up-to-date services, in Brazil it generally makes decisions involving all its production variables based on purely subjective criteria. This paper reports the use of artificial neural networks to estimate performance in production birds belonging to a South Brazilian poultry farm. 2. Recorded data from 22 broiler production breeder flocks were obtained, from April, 1998 to December, 1999, which corresponded to 689 data lines of weekly recordings. 3. These data were processed by artificial neural networks using the software NeuroShell 2 version 4.0 (Ward Systems Group). The artificial neural network models generated were compared and selected based on their largest determination coefficient (R2), lowest Mean Squared Error (MSE), as well as on a uniform scatter in the residual plots. The authors conclude that it is possible to explain the performance variables of production birds, with the use of artificial neural networks. 4. The method allows the decisions made by the technical staff to be based on objective, scientific criteria, allows simulations of the consequences related to these decisions, and reports the contribution of each variable to the variables under study.

Animal Feed↗
PubMed2005

Dynamic neural network models of the premotoneuronal circuitry controlling wrist movements in primates.

Dynamic recurrent neural networks were derived to simulate neuronal populations generating bidirectional wrist movements in the monkey. The models incorporate anatomical connections of cortical and rubral neurons, muscle afferents, segmental interneurons and motoneurons; they also incorporate the response profiles of four populations of neurons observed in behaving monkeys. The networks were derived by gradient descent algorithms to generate the eight characteristic patterns of motor unit activations observed during alternating flexion-extension wrist movements. The resulting model generated the appropriate input-output transforms and developed connection strengths resembling those in physiological pathways. We found that this network could be further trained to simulate additional tasks, such as experimentally observed reflex responses to limb perturbations that stretched or shortened the active muscles, and scaling of response amplitudes in proportion to inputs. In the final comprehensive network, motor units are driven by the combined activity of cortical, rubral, spinal and afferent units during step tracking and perturbations. The model displayed many emergent properties corresponding to physiological characteristics. The resulting neural network provides a working model of premotoneuronal circuitry and elucidates the neural mechanisms controlling motoneuron activity. It also predicts several features to be experimentally tested, for example the consequences of eliminating inhibitory connections in cortex and red nucleus. It also reveals that co-contraction can be achieved by simultaneous activation of the flexor and extensor circuits without invoking features specific to co-contraction.

Animals↗
PubMed2001

The complexity of linkage analysis with neural networks.

As the focus of genome-wide scans for disease loci have shifted from simple Mendelian traits to genetically complex traits, researchers have begun to consider new alternative ways to detect linkage that will consider more than the marginal effects of a single disease locus at a time. One interesting new method is to train a neural network on a genome-wide data set in order to search for the best non-linear relationship between identity-by-descent sharing among affected siblings at markers and their disease status. We investigate here the repeatability of the neural network results from run to run, and show that the results obtained by multiple runs of the neural network method may differ quite a bit. This is most likely due to the fact that training a neural network involves minimizing an error function with a multitude of local minima.

Diabetes Mellitus, Type 1↗
PubMed2002

Comparison between radiological and artificial neural network diagnosis in clinical screening.

The imaging protocol of the UK multicentre magnetic resonance imaging study for screening in women at genetic risk of breast cancer aims to assist in detecting and diagnosing malignant breast lesions. In this paper, we evaluate a three-layer, feed-forward, backpropagation neural network as an artificial radiological classifier using receiver operating characteristic (ROC) curve analysis and compare the results with those obtained using a proposed radiological scoring system for the study which currently supplements the radiologist's clinical opinion, in comparison with histological diagnosis. Based on the 76 symptomatic cases evaluated, descriptive features scored by radiologists showed considerable overlap between benign and malignant, although some features such as irregular contours and heterogeneous enhancement were more often associated with malignant pathology. In this preliminary evaluation, ROC analysis showed that the proposed scoring scheme did not perform well, indicating further refinement is required. When all 23 features were used in the neural network, its performance was poorer than that of the scoring scheme. When only ten features were used, limited to descriptors of enhancement characteristics, the neural network performed similar to the scoring scheme. This comparison shows that the neural network approach to clinical diagnosis has considerable potential and warrants further development.

Breast Neoplasms↗
PubMed1998

Automated neural network detection of wavelet preprocessed electrocardiogram late potentials.

The aim of the study is to investigate the potential of a feedforward neural network for detecting wavelet preprocessed late potentials. The terminal parts of a simulated QRS complex are processed with a continuous wavelet transform, which leads to a time-frequency representation of the QRS complex. Then, diagnostic feature vectors are obtained by subdividing the representations into several regions and by processing the sum of the decomposition coefficients belonging to each region. The neural network is trained with these feature vectors. Simulated ECGs with varying signal-to-noise ratios are used to train and test the classifier. Results show that correct classification ranges from 79% (high-level noise) to 99% (no noise). The study shows the potential of neural networks for the classification of late potentials that have been preprocessed by a wavelet transform. However, clinical use of this method still requires further investigation.

Arrhythmias, Cardiac↗
PubMed1993

Computer-simulated neural networks: an appropriate model for motor development?

The idea of an artificial neural network is introduced in a historical context, and the essential aspect of it, viz., the modifiable synapse, is compared to the aspect of plasticity in the natural nervous system. Based on such an artificial neural network, a model is presented for the way in which (the path along which) the connectivity in the spinal cord is modified during the period that a newborn 'learns' to control the movement of his forearm. In this way an automatic calibration of the receptors and the antagonists' recruitment of motor units is represented. The learning process is described in non-mathematical terminology. The model is then shown to be able after learning to reach target angles outside the training set of angles, and to be able to relearn when an important receptor has been made inoperative. In this way it is shown that the model is able to generalize, and that it is robust against at least some damage.

Computer Simulation↗
PubMed2001

A neural network for recognizing movement patterns during repetitive self-paced movements of the fingers in opposition to the thumb.

A data glove is a typical input device to control a virtual environment. At the same time it measures movements of wrist and fingers. The purposes of this investigation were to assess the ability of BrainMaker, a neural network, to recognize movement patterns during an opposition task that consisted of repetitive self-paced movements of the fingers in opposition to the thumb. The neural network contained 56 inputs, 3 hidden layers of 20 neurons, and one output. The 5th glove '95 (5DT), a commercial glove especially designed for virtual reality games, was used for finger motion capture. The training of the neural network was successful for recognizing the thumb, the index finger and the ring finger movements during the repetitive self-paced movements and neural network performed well during testing.

Adult↗
PubMed1996

Parametric signal restoration using artificial neural networks.

The problem of parametric signal restoration given its blurred/nonlinearly distorted version contaminated by additive noise is discussed. It is postulated that feedforward artificial neural networks can be used to find a solution to this problem. The proposed estimator does not require iterative calculations that are normally performed using numerical methods for signal parameter estimation. Thus high speed is the main advantage of this approach. A two-stage neural network-based estimator architecture is considered in which the vector of measurements is projected on the signal subspace and the resulting features form the input to a feedforward neural network. The effect of noise on the estimator performance is analyzed and compared to the least-squares technique. It is shown, for low and moderate noise levels, that the two estimators are similar to each other in terms of their noise performance, provided the neural network approximates the inverse mapping from the measurement space to the parameter space with a negligible error. However, if the neural network is trained on noisy signal observations, the proposed technique is superior to the least-squares estimate (LSE) model fitting. Numerical examples are presented to support the analytical results. Problems for future research are addressed.

Artifacts↗
PubMed2002

A review of evidence of health benefit from artificial neural networks in medical intervention.

The purpose of this review is to assess the evidence of healthcare benefits involving the application of artificial neural networks to the clinical functions of diagnosis, prognosis and survival analysis, in the medical domains of oncology, critical care and cardiovascular medicine. The primary source of publications is PUBMED listings under Randomised Controlled Trials and Clinical Trials. The rĵle of neural networks is introduced within the context of advances in medical decision support arising from parallel developments in statistics and artificial intelligence. This is followed by a survey of published Randomised Controlled Trials and Clinical Trials, leading to recommendations for good practice in the design and evaluation of neural networks for use in medical intervention.

Decision Support Techniques↗
PubMed1997

Using neural networks to identify patients unlikely to achieve a reduction in bodily pain after total hip replacement surgery.

OBJECTIVES: Fourteen patient-provided variables were chosen as potential predictors for improvement after total hip replacement surgery. These variables included patient demographic information, as well as preoperative physical function. METHODS: A neural network was trained to predict the relative success of total hip replacement surgery using this presurgical patient survey information. The outcome measure was improvement in the Medical Outcomes Study 36 Short Form Health Survey pain score between the preoperative assessment and the 1-year postoperative assessment. For the study sample, 221 patients were selected who had complete information for the composite outcome variable. A backpropagation feedforward neural network was trained to predict the output variable using the jackknife method. RESULTS: Performance of the neural network was assessed by calculating the area under the receiver operating characteristic curve for the network's ability to predict whether the pain score was improved after total hip replacement surgery. The observed area under the receiver operating characteristic curve was 0.79. For comparison, a linear regression model built using the same data had a receiver operating characteristic area of 0.74 (P = 0.23). CONCLUSIONS: This research therefore showed the ability of neural networks to predict the success of total hip replacement more accurately. Our results further indicate that it may be possible to predict which patients are at greatest risk of a poor outcome.

Activities of Daily Living↗
PubMed2003

[Artificial neural networks. Theory and applications in anesthesia, intensive care and emergency medicine].

Artificial neural networks (ANN) are constructed to simulate processes of the central nervous system of higher creatures. An ANN consists of a set of processing units (nodes) which simulate neurons and are interconnected via a set of "weights" (analogous to synaptic connections in the nervous system) in a way which allows signals to travel through the network in parallel. The nodes (neurons) are simple computing elements. They accumulate input from other neurons by means of a weighted sum. If a certain threshold is reached the neuron sends information to all other connected neurons otherwise it remains quiescent. One major difference compared with traditional statistical or rule-based systems is the learning aptitude of an ANN. At the very beginning of a training process an ANN contains no explicit information. Then a large number of cases with a known outcome are presented to the system and the weights of the inter-neuronal connections are changed by a training algorithm designed to minimise the total error of the system. A trained network has extracted rules that are represented by the matrix of the weights between the neurons. This feature is called generalisation and allows the ANN to predict cases that have never been presented to the system before. Artificial neural networks have shown to be useful predicting various events. Especially complex, non-linear, and time depending relationships can be modelled and forecasted. Furthermore an ANN can be used when the influencing variables on a certain event are not exactly known as it is the case in financial or weather forecasts. This article aims to give a short overview on the function of ANN and their previous use and possible future applications in anaesthesia, intensive care, and emergency medicine.

Anesthesiology↗