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 1,297 records · Page 72Linked to original sources
PubMed2005

Artificial neural networks in prediction of bone density among post-menopausal women.

Artificial neural networks (ANN) are promising tools in learning complex interplay of factors on a particular outcome. We performed this study to compare the predictive power of ANN and conventional methods in prediction of bone mineral density (BMD) in Iranian post-menopausal women. A database of 10 input variables from 2158 participants was randomly divided into training (1400), validation (150) and test (608) groups. Multivariate linear regression and ANN models were developed and validated on the training, and validation sets and outcomes (femoral neck and lumbar T-scores) were predicted and compared on the test group using different numbers of input variables. Results were evaluated by comparing the mean square of differences between predicted and reference values (non-central chi-square test) and by measuring area under the receiver operating characteristic curve (AUROC) around cut-off value of -2.5 for T-scores. For models with less than 3 input variables in femoral neck and 4 variables in spinal column, performance of regression and ANN models was almost the same. As more variables imported into models, ANN outperformed linear regression models. AUROC varied in 2 to 10 variable models as follows: for ANN in spine, from 0.709 to 0.774; linear models in spine, from 0.709 to 0.744; ANN in femoral neck, from 0.801 to 0.867; linear models in femoral neck, from 0.799 to 0.834. The ANN model performed better than five established patient selection tools in the test group. Superior performance of neural networks than linear models demonstrate their advantage especially in mass screening applications, when even a slight enhancement in performance results in significant decrease in number of misclassifications.

Aged↗
PubMed2026

engGNN: a dual-graph neural network for omics-based disease classification and feature selection.

Omics data, such as transcriptomics, proteomics, and metabolomics, provide critical insights into disease mechanisms and clinical outcomes. However, their high dimensionality, small sample sizes, and intricate biological networks pose major challenges for reliable prediction and meaningful interpretation. Graph neural networks offer a promising way to integrate prior knowledge by encoding feature relationships as graphs. Yet, existing methods typically rely solely on either an externally curated feature graph or a data-driven generated graph, which limits their ability to capture complementary information. To address this, we propose the external and generated Graph Neural Network (engGNN), a dual-graph framework that jointly leverages both external biological networks and data-driven generated graphs. Specifically, engGNN constructs a biologically informed undirected feature graph from established network databases and complements it with a directed feature graph derived from tree-ensemble models. This dual-graph design produces more comprehensive representations, thereby improving predictive performance and interpretability. Through extensive simulation studies and real-world applications to three independent gene expression datasets, engGNN consistently demonstrates strong classification performance compared with competitive baselines. Beyond classification, engGNN provides feature- and source-level interpretability, enabling biologically meaningful analyses such as pathway enrichment analysis. Taken together, these results highlight engGNN as a robust, flexible, and interpretable framework for disease classification and biomarker discovery in high-dimensional omics contexts.

Graph Neural Networks↗
PubMed2004

Criteria for exponential [corrected] stability of Cohen-Grossberg neural networks.

In this paper, the Cohen-Grossberg neural network models without and with time delays are considered. By constructing several novel Lyapunov functionals, some sufficient criteria for the existence of a unique equilibrium and global exponential stability of the network are derived. These results are fairly general and can be easily verified. Besides, the approach of the analysis allows one to consider different types of activation functions, including piecewise linear, sigmoids with bounded activations as well as C1-smooth sigmoids. In the meantime, our approach does not require any symmetric assumption of the connection matrix. It is believed that these results are significant and useful for the design and applications of the Cohen-Grossberg model.

Neural Networks, Computer↗
PubMed1998

Peptide design aided by neural networks: biological activity of artificial signal peptidase I cleavage sites.

De novo designed signal peptidase I cleavage sites were tested for their biological activity in vivo in an Escherichia coli expression and secretion system. The artificial cleavage site sequences were generated by two different computer-based design techniques, a simple statistical method, and a neural network approach. In previous experiments, a neural network was used for feature extraction from a set of known signal peptidase I cleavage sites and served as the fitness function in an evolutionary design cycle leading to idealized cleavage site sequences. The cleavage sites proposed by the two algorithms were active in vivo as predicted. There seems to be an interdependence between several cleavage site features for the constitution of sequences recognized by signal peptidase. It is concluded that neural networks are useful tools for sequence-oriented peptide design.

Amino Acid Sequence↗
PubMed2000

Ranitidine hydrochloride X-ray assay using a neural network.

A simple X-ray powder diffractometric (XRD) method with artificial neural networks (ANNs) for data modelling was developed to recognize and quantify two crystal modifications of ranitidine HCl in mixtures and thus, provide information about the solid state of the bulk drug. The method was also used to quantify ranitidine HCl from tablets in the presence of other components. An ANN consisting of three layers of neurons was trained by using a back-propagation learning rule. A sigmoid output function was used in the hidden layer to facilitate non-linear fitting. Unlike other techniques the ANN method described here employed pattern recognition on the entire XRD pattern. Correct classification was mainly influenced by the XRD pattern resolution. It was shown that data transformations improved the quantitative performance when the XRD patterns were not contaminated by other components. Only smoothed X-ray diffractograms were required to distinguish between the two crystalline forms in a mixture. In the case of ranitidine-HCl quantification from tablets, where significant interference with tablet excipients was present, better results were obtained without data transformations. The trained ANN perfectly quantified ranitidine HCI polymorphic forms from mixtures (mean sum of squared error was less than 0.02%) and ranitidine HCl form 1 from tablets (recovery = 98.65). Excellent quantification performance of the ANN analysis. demonstrated in this study, serves as an indication of the broad potential of neural networks in pattern analysis. While the system described has been developed to interpret XRD patterns, peak detection has implications in every chemical application where the recognition of peak-shaped signals in analytical data is important.

Anti-Ulcer Agents↗
PubMed2006

Optimized quality of histologic images allows the use of neural network scanning in diagnosis of fungal infection of abnormal nails.

OBJECTIVE: The neural network scanning (NNS) system, formerly known as Papnet, is capable of selecting fungi in cervical smears. The objective of this study was to investigate whether the optimized quality of histologic images created using a combination of coagulant fixation and microwave histoprocessing allows the application of this computer-assisted microscopy in the diagnostic process. STUDY DESIGN: In a prospective study, 117 abnormal nails clinically suspect for fungal disease werefixed in a coagulant fixative, BoonFix, processed in a microwave histoprocessor to obtain optimal paraffin sections and stained with the periodic acid-Schiff (PAS) method. The stained paraffin sections were randomly numbered and screened by two independent pathologists for diagnosis of fungal hyphae and spores. The same sections were subsequently scanned by a neural network, and a maximum of 128 digital images produced by the system were screened and diagnosed by pathologists. Using light microscopy as the gold standard for diagnosis of fungi, the usefulness of NNS was then assessed. RESULTS: The fungi and spores were clearly demonstrated in the paraffin sections, and the NNS system detected and recorded them efficiently. The hyphae and spores could be identified in these pixilated images with relative ease. Of the 117 examined cases, 50 were positive and 47 negative by both methods. In the 20 remaining cases, NNS did not present images of fungi that were present in the histologic sections. In practice, this implies that only 67 out of 117 cases (57%) must be screened by light microscopy. NNS recorded not only fungi and spores in the 128 digital images but also artifacts, such as round, deeply PAS-positive granules of talcum powder, which by light microscopy might be mistaken for fungal spores. CONCLUSION: NNS proves applicable for the selection of spores and fungi if the histologic sections are of sufficiently high quality. As a result, the number of slides to be screened by light microscopy can be reduced substantially. In a throughput diagnostic laboratory handling a large number of such cases this technology can be highly valuable.

Adolescent↗
PubMed1996

Dipole source localization of MEG by BP neural networks.

The purpose of this study was to examine the usefulness of BP neural network for source localization of MEG. Since the performance of this method does not depend on the complexity of brain parameters and source models, a homogeneous brain model and a single current dipole source are assumed for convenience. Localization accuracy was examined in relation to the configuration and scale of the network. As a result, average error for position and moment estimations was within 2% while the maximum error was about 5%. It was therefore concluded that the neural network method, was useful for MEG source localization, though some improvements are still necessary.

Brain↗
PubMed2006

[Research on EEG classification with evolving cascade neural networks].

To correctly classify EEG with different mental tasks, a new learning algorithm for Evolving Cascade Neural Networks (ECNNs) is described to avoid over-fitting of a neural network due to noise and redundant features. The learning algorithm calculates the value of a fitness function on validate set and accordingly updates the connection weights on training set. The learning algorithm uses the regularity criterion for selecting the neurons with relevant connection. If the value Cr calculated for the rth neuron is less than the value Cr-1 calculated for the previous (r-1) neuron, the features that feed the rth neuron are relevant, else they are irrelevant. An ECNN starts to learn with one input node and then, adding new inputs as well as new hidden neurons, evolves it. The trained ECNN has a nearly minimal number of input and hidden neurons as well as connections. The algorithm is applied to classify EEG with two mental tasks. The trained ECNN has correctly classified 83.1% of the testing segments. It shows a better result, compared with a standard BP network.

Algorithms↗
PubMed1999

[Neural network approach to medical grading of rheumatoid arthritis].

A grading system based on neural network has been developed and used int he medical grading diagnosis of a case group of Rheumatoid Arthritis presenting Cold syndrome intermingled with Heat syndrome. The possibility of using neural network in TCM diagnosis is preliminarily explored in this paper.

Algorithms↗
PubMed1997

Classification and staging of dementia of the Alzheimer type: a comparison between neural networks and linear discriminant analysis.

OBJECTIVE: To examine the utility of artificial neural networks (ANNs) for differentiating patients with Alzheimer disease from healthy control subjects and for staging the degree of dementia. DESIGN: Comparison of the classification abilities of ANNs with the statistical technique of linear discriminant analysis (LDA) using the results of 11 neuropsychological tests as predictors. PARTICIPANTS: Ninety-two patients with a diagnosis of probable Alzheimer disease (referred from a geriatric clinic) and 43 elderly control subjects (independently solicited). The patients met National Institute of Neurological and Communicative Disorders and Stroke-Alzheimer's Disease and Related Disorders Association criteria for probable dementia, with clinical ratings of dementia severity derived from the Cambridge Examination for Mental Disorders of the Elderly (CAMDEX). MAIN OUTCOME MEASURES: Classifications between and within groups were determined by using LDA and ANNs, and more detailed comparisons of the 2 methods were performed by using chi2 analyses and unweighted and weighted kappa statistics. RESULTS: Linear discriminant analysis correctly identified 71.9% of cases. Artificial neural networks, trained to classify the subjects using the same data, correctly classified 91.1% of the cases. Subsidiary analyses showed that although both techniques effectively discriminated between the control subjects and patients with dementia, the ANNs were more powerful in discriminating severity levels within the dementia population. The analyses for goodness of fit revealed that the ANN classification produced a better fit to the actual data. A comparison of the weighted proportion of agreement between the criterion and predictor variables also showed that the ANNs clearly outperformed LDA in classification accuracy for the full data set and patients-only data set. CONCLUSION: The results demonstrate the utility of ANNs for group classification of patients with Alzheimer disease and elderly controls and for staging dementia severity using neuropsychological data.

Aged↗
PubMed1996

A neural network model for stroboscopic alternative motion.

A neural network which models multistable perception is presented. The network consists of sensor and inner neurons. The dynamics is established by a stochastic neuronal dynamics, a formal Hebb-type coupling dynamics and a resource mechanism that corresponds to saturation effects in perception. From this a system of coupled differential equations is derived and analyzed Single stimuli are bound to exactly one percept, even in ambiguous situations where multistability occurs. The network exhibits discontinuous as well as continuous phase transitions and models various empirical findings, including the percepts of succession, alternative motion and simultaneity, the percept of oscillation is explained by oscillating percepts of a continuous phase transition.

Algorithms↗
PubMed1999

Predicting cytomegalovirus disease after renal transplantation: an artificial neural network approach.

Outcome prediction is becoming increasingly important in medicine, but when a resource is scarce the need for accurate prediction becomes acute. Prediction based on biostatistical models has been in use for many years, but in areas such as renal transplantation their results have been disappointing. Recently however, there has been growing interest in the use of artificial neural networks for prediction. The creation of a large database containing high quality data on renal transplantation patients in Wales offers an ideal opportunity to research a new area viz., the ability of these techniques to accurately predict outcomes such as the appearance of disease in transplant recipients or the time to graft failure. This paper describes the use of neural networks to identify patients who risk the development of cytomegalovirus disease--a significant cause of mortality and morbidity in these patients. The neural networks we examined produced overall correct classifications well in excess of 80% in each of the two groups involved, diseased and non-diseased. These predictions are a considerable improvement on current methods and encourage the belief that renal transplantation data may respond well to analysis by neural networks.

Adolescent↗
PubMed2003

Neural networks in the prediction of survival in patients with colorectal cancer.

It is important to predict outcome for colorectal cancer patients following surgery, as almost 50% of patients undergoing a potentially curative resection will experience relapse. It is clear that present prognostic categories such as Dukes or TNM staging are too broad, and further refining is required to prognosticate for high-risk subgroups. One approach is to determine a phenotype associated with recurrence. We compared 2 methods of analyzing such data. Pathologic data from a large clinical trial was analyzed for 403 patients. The outcome modeled was disease recurrence. The results from logistic regression analysis and a neural network approach are compared with respect to receiver operator characteristic plots, which estimate the fit of the model. The best logistic regression model gives a result of 66%, and the neural network approach 78%. The conclusion from this study is that the neural network approach is superior to regression analysis. Further analyses are in progress using a larger patient sample size (n > 1000), improved statistical models, and a more refined neural network.

Colorectal Neoplasms↗
PubMed2001

Neural network modelling of Cryptosporidium and Giardia concentrations in the Delaware River, USA.

Artificial neural networks are brain-like structures used in mathematical modelling that excel in pattern recognition. In this research, a simple feed-forward artificial neural network, trained by error back-propagation algorithm, was used as a tool to relate peak Cryptosporidium and Giardia concentrations with other biological, chemical and physical parameters in surface water. Multiple water quality parameters at a water treatment plant intake on the Delaware River, New Jersey, USA, collected in 1996, were provided to the authors for recognition analysis. Water samples were classified as "background" and "above background" based on the concentration of full and empty oocysts and cysts of Cryptosporidium and Giardia. The results of this preliminary effort were encouraging. Parameters significant to the identification of each protozoa were identified, eight for Cryptosporidium and seven for Giardia by a stepwise elimination technique. Data withheld from the model training was used to validate the trained models and evaluate the most effective internal architecture. In both cases, the best prediction performance was found when the number of internal nodes was twice that of the input parameters in single hidden-layer architecture. Predictions for the classification of the verification data set resulted in no false-negatives (mis-prediction of above background protozoa concentrations) when the models were optimally trained.

Animals↗
PubMed1999

A neural network based predictor of residue contacts in proteins.

We describe a method based on neural networks for predicting contact maps of proteins using as input chemicophysical and evolutionary information. Neural networks are trained on a data set comprising the contact maps of 200 non-homologous proteins of well resolved three-dimensional structures. The systems learn the association rules between the covalent structure of each protein and its correspondent contact map by means of a standard back propagation algorithm. Validation of the predictor on the training set and on 408 proteins of known structure which are not homologous to those contained in the training set indicate that this method scores higher than statistical approaches previously described and based on correlated mutations and sequence information.

Animals↗
PubMed2005

Attractor neural networks and spatial maps in hippocampus.

Attractor neural network theory has been proposed as a theory for long-term memory. Recent studies of hippocampal place cells, including a study by Leutgeb et al. in this issue of Neuron, address the potential role of attractor dynamics in the formation of hippocampal representations of spatial maps.

Animals↗
PubMed1984

Collective properties of neural networks: a statistical physics approach.

Among the various models proposed so far to account for the properties of neural networks, the one devised by Little and the one derived by Hopfield prove to be the most interesting because they allow the use of statistical mechanics techniques. The link between the Hopfield model and the statistical mechanics is provided by the existence of an extensive quantity. When the synaptic plasticity behaves according to a Hebbian procedure, the analogy with the classical spin glass models studied by Van Hemmen is complete. In particular exact solutions describing the steady states of noisy systems are found. On the other hand, the Little model introduces a Markovian dynamics. One shows that the evolution equation obeys the microreversibility principle if the synaptic efficiencies are symmetrical. Therefore, assuming that such a symmetry materializes, the Little model has to obey a Gibbs statistics. The corresponding Hamiltonian is derived accordingly. At last, using these results, both models are shown to display associative memory properties. In particular the storage capacity of neural networks working along with the Little dynamics is similar to the capacity of Hopfield neural networks. The conclusion drawn from the study of the Hopfield model can be extended to the Little model, which is certainly a more realistic description of the biological situation.

Action Potentials↗
PubMed2003

Global stability of neural networks with distributed delays.

In this paper, a model describing the dynamics of recurrent neural networks with distributed delays is considered. Some sufficient criteria are derived ensuring the global asymptotic stability of distributed-delay recurrent neural networks with more general signal propagation functions by introducing real parameters p>1, q(ij)>0, and r(jj)>0, i,j=1, em leader,n, and applying the properties of the M matrix and inequality techniques. We do not assume that the signal propagation functions satisfy the Lipschitz condition and do not require them to be bounded, differentiable, or strictly increasing. Moreover, the symmetry of the connection matrix is also not necessary. These criteria are independent of the delays and possess infinitely adjustable real parameters, which is important in signal processing, especially in moving image treatment and the design of networks.

Biophysical Phenomena↗