Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Generative models”

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 865 records · Page 48Linked to original sources
PubMed1997

Determinants of postchemotherapy nausea and vomiting in patients with cancer. Quality of Life and Symptom Control Committees of the National Cancer Institute of Canada Clinical Trials Group.

PURPOSE: To assess whether prechemotherapy health-related quality-of-life (HQL) variables are associated with postchemotherapy nausea and vomiting (PCNV), and to determine their relationship to patient and treatment variables. PATIENTS AND METHODS: Eight hundred thirty-two chemotherapy-naive patients scheduled to receive antiemetic regimens containing a 5-hydroxytryptamine (5-HT3) antagonist with or without dexamethasone for moderately or highly emetogenic chemotherapy were enrolled. HQL was measured by the self-report European Organization for Research and Treatment of Cancer (EORTC) Care Quality of Life Questionnaire (QLQ-C30) within 7 days before chemotherapy. Prechemotherapy HQL scores, as well as other patient, disease, and treatment variables were compared in the groups of patients who had PCNV and those who did not have PCNV. All variables were assessed initially in a univariate analysis and then together in a multivariate analysis using step-wise logistic regression. The final model generated by the multivariate analyses was used in a risk factor analysis to predict PCNV. RESULTS: Univariate analyses identified 10 HQL variables and five patient and treatment characteristics that were associated with PCNV. In the multivariate analysis, the variables remaining in the final model included low social functioning, prechemotherapy nausea, female gender, highly emetogenic chemotherapy, and the lack of maintenance antiemetics (5-HT3 antagonists with or without dexamethasone) after chemotherapy. A history of low alcohol use was also associated with PCV, whereas increased fatigue and lower performance status were associated with PCN. In the risk factor analysis, the incidence of PCV increased from 20% in those having no risk factors to 76% in those having any four of the six risk factors. CONCLUSION: Several pretreatment HQL, patient, and treatment characteristics are associated with the occurrence of PCNV. Patients about to receive moderately or highly emetogenic chemotherapy should be screened for these factors and additional measures, such as behavior modification and modification of antiemetic therapy, should be considered in attempts to improve the control of PCNV.

Analysis of Variance↗
PubMed2000

Improving the local solution accuracy of large-scale digital image-based finite element analyses.

Digital image-based finite element modeling (DIBFEM) has become a widely utilized approach for efficiently meshing complex biological structures such as trabecular bone. While DIBFEM can provide accurate predictions of apparent mechanical properties, its application to simulate local phenomena such as tissue failure or adaptation has been limited by high local solution errors at digital model boundaries. Furthermore, refinement of digital meshes does not necessarily reduce local maximum errors. The purpose of this study was to evaluate the potential to reduce local mean and maximum solution errors in digital meshes using a post-processing filtration method. The effectiveness of a three-dimensional, boundary-specific filtering algorithm was found to be mesh size dependent. Mean absolute and maximum errors were reduced for meshes with more than five elements through the diameter of a cantilever beam considered representative of a single trabecula. Furthermore, mesh refinement consistently decreased errors for filtered solutions but not necessarily for non-filtered solutions. Models with more than five elements through the beam diameter yielded absolute mean errors of less than 15% for both Von Mises stress and maximum principal strain. When applied to a high-resolution model of trabecular bone microstructure, boundary filtering produced a more continuous solution distribution and reduced the predicted maximum stress by 30%. Boundary-specific filtering provides a simple means of improving local solution accuracy while retaining the model generation and numerical storage efficiency of the DIBFEM technique.

Algorithms↗
PubMed1987

Correlation of clinical stage, serum prostatic acid phosphatase and preoperative Gleason grade with final pathological stage in 275 patients with clinically localized adenocarcinoma of the prostate.

The usefulness of clinical stage, serum prostatic acid phosphatase and preoperative Gleason grade in predicting final pathological stage in patients with adenocarcinoma of the prostate remains controversial. To determine the predictive value of these 3 preoperative variables we reviewed 275 patients with clinically localized disease who were treated between April 1982 and February 1986. All patients were examined preoperatively and subsequently were operated upon by 1 urologist. Serum prostatic acid phosphatase was determined in all patients by the Roy method using thymolphthalein monophosphate as the substrate. The Gleason grade of each prostatic biopsy specimen was determined preoperatively by 1 pathologist, who also examined the final pathological specimen with respect to capsular penetration, and seminal vesicle and pelvic lymph node involvement. Using logistic regression analysis with the likelihood ratio chi-square test, clinical stage and Gleason grade had a direct correlation with capsular penetration (p less than 0.0001 and less than 0.0001, respectively), seminal vesicle involvement (p less than 0.0001 and less than 0.0001, respectively) and positive lymph nodes (p less than 0.0001 and less than 0.0002, respectively). Within the normal range of values (0.0 to 0.8 IU/l.) serum prostatic acid phosphatase correlated directly with capsular penetration (p less than 0.003) and seminal vesicle involvement (p less than 0.01) but not with lymph node involvement (p equals 0.08). Again with logistic regression analysis we determined that the best predictors of final pathological stage are not individual variables but models that use combinations of preoperative variables. The models generated are as follows: capsular penetration--serum prostatic acid phosphatase and Gleason grade (p less than 0.00001), seminal vesicle involvement--clinical stage and Gleason grade (p less than 0.00001), and lymph node involvement--clinical stage and Gleason grade (p less than 0.00001). With these models probability plots have been constructed so that the final pathological stage in patients with clinically localized prostatic cancer can be predicted preoperatively.

Acid Phosphatase↗
PubMed2002

An evaluation of the variance components approach: type I error, power and size of the estimated effect.

In order to evaluate the statistical properties of the variance components method implemented in SOLAR and GENEHUNTER2, the type I error rate, power and estimated size of modelled effects were determined using computer simulation. Results suggested that the type I error rate was quite conservative when the variance was completely due to random effects. However, when either a polygenic effect or an unlinked single locus component effect was included in the generating models the type I error rate was closer to the nominal level. The size of the polygenic or single locus effect appeared to influence the type I error rate. Results suggested that the variance components method underestimated the variance attributed to a single locus effect and overestimated the variance attributed to a polygenic effect in the models considered regardless of the source of variation.

Analysis of Variance↗
PubMed1994

An artificial neural network analogue of learning in autism.

An artificial neural network is simulated that shares formal qualitative similarities with the selective attention and generalization deficits seen in people with autism. The model is based on neuropathological studies which suggest that affected individuals have either too few or too many neuronal connections in various regions of the brain. In simulations where the model was taught to discriminate children with autism from children with mental retardation, having too few simulated neuronal connections led to relatively inferior discrimination of the two groups in a training set and, consequently, relatively inferior generalization of the discrimination to a novel test set. Too many connections produced excellent discrimination but inferior generalization because of overemphasis on details unique to the training set. It is concluded that, within the context of the current model, the neuropathological observations that have been described in the literature are sufficient to explain some of the unique pattern recognition and discrimination learning abilities seen in some people with autism as well as their problems with generalization and concept acquisition. The model generates testable hypotheses that have implications for understanding the pathogenesis, treatment, and phenomenology of autism.

Analysis of Variance↗
PubMed2005

Brugada syndrome and ischemia-induced ST-segment elevation. Similarities and differences.

INTRODUCTION: ST-Segment elevation is a common electrocardiogram (ECG) manifestation of acute transmural myocardial ischemia in leads facing the injury. Acute myocardial ischemia involving the right-ventricular (RV) outflow tract is known to induce a Brugada-like ECG. In this paper, we examined the electrophysiological bases for the similarities between the ECG characteristics of the Brugada syndrome model induced by terfenadine (5 mumol/L) and the ECG manifestations of the acute transmural no-flow ischemia model. METHODS: For both experimental simulations, we used isolated arterially perfused canine RV wedge preparations to record transmembrane action potentials (AP) from endocardium and epicardium together with a transmural pseudo-ECG (ECG); basic cycle length=400 to 2000 ms. RESULTS: In the presence of a prominent Ito-mediated AP notch, no-flow ischemia causes true ST-segment elevation because of selective depression and loss of the AP dome at some epicardial sites. In the absence of a prominent AP notch, ischemia ultimately produces an apparent ST-segment elevation, which is secondary to a prolongation of the R wave caused by marked transmural conduction delays. Similarly, in the Brugada syndrome model generated in preparations displaying a large epicardial Ito, ST-segment elevation was due to loss of the epicardial AP dome at some sites but not at others. Transmural conduction delay giving the appearance of ST-segment elevation is also observed in the Brugada model in preparations exhibiting smaller AP notch. In both models, propagation of the dome from the site at which it is maintained to a site at which it is lost may result in closely coupled phase 2 reentrant extrasystoles. CONCLUSION: Our results suggest that Ito can modulate the electrocardiographic manifestation of acute ischemia as well as that of the Brugada syndrome, and that both clinical entities are the result of a similar electrophysiological substrate.

Action Potentials↗
PubMed2003

Development and validation of k-nearest-neighbor QSPR models of metabolic stability of drug candidates.

Computational ADME (absorption, distribution, metabolism, and excretion) models may be used early in the drug discovery process in order to flag drug candidates with potentially problematic ADME profiles. We report the development, validation, and application of quantitative structure-property relationship (QSPR) models of metabolic turnover rate for compounds in human S9 homogenate. Biological data were obtained from uniform bioassays of 631 diverse chemicals proprietary to GlaxoSmithKline (GSK). The models were built with topological molecular descriptors such as molecular connectivity indices or atom pairs using the k-nearest neighbor variable selection optimization method developed at the University of North Carolina (Zheng, W.; Tropsha, A. A novel variable selection QSAR approach based on the k-nearest neighbor principle. J. Chem. Inf. Comput. Sci., 2000, 40, 185-194.). For the purpose of validation, the whole data set was divided into training and test sets. The training set QSPR models were characterized by high internal accuracy with leave-one-out cross-validated R(2) (q(2)) values ranging between 0.5 and 0.6. The test set compounds were correctly classified as stable or unstable in S9 assay with an accuracy above 85%. These models were additionally validated by in silico metabolic stability screening of 107 new chemicals under development in several drug discovery programs at GSK. One representative model generated with MolConnZ descriptors predicted 40 compounds to be metabolically stable (turnover rate less than 25%), and 33 of them were indeed found to be stable experimentally. This success (83% concordance) in correctly picking chemicals that are metabolically stable in the human S9 homogenate spells a rapid, computational screen for generating components of the ADME profile in a drug discovery process.

Algorithms↗
PubMedDate not supplied

An empirical model of hyperopic shift with corticosteroid modulation and refractive power prediction after photorefractive keratectomy.

Data from a selected sample of 158 patients who received excimer laser photorefractive keratectomy (PRK) in our center and obtained a desirable refractive outcome at 6-months were analyzed retrospectively to establish an empirical model of refractive power after PRK. This model, generated by a computer-assisted regression program, is presented as a multivariate non-linear quadratic equation. Since the coefficients of each variable (attempted correction and postoperative days) are statistically significant, the estimated influence of the variables on the postoperative refraction from this regression model is reliable and consistent. Therefore, this model has been used to direct the use of corticosteroids for modulating post-PRK refraction. The predictability of this model has also been corroborated by separate data from 1-year follow up of PRK recipients. The recipients whose refraction at 6 months agree with the calculation from this model have obtained an excellent outcome; while those whose refractive outcome deviates from this model have gotten undesirable results.

Computer Simulation↗
PubMed2000

Primary prevention, smoking, and smoking cessation: implications for future trends in lung cancer prevention.

BACKGROUND: Cigarette smoking is the major cause of lung cancer, and most lung cancers have historically occurred among current cigarette smokers or recent quitters. The relative risk of developing lung cancer declines in former smokers to approximately twice that of never smokers after 20 years of cessation, but it remains elevated indefinitely. As an increasing fraction of those who have ever smoked become former smokers, the absolute risk of lung cancer in the population will decline; and an increasing fraction of lung cancer cases will occur among former smokers. METHODS: Data from the 12-year follow-up of American Cancer Society Cancer Prevention Study I are used to estimate lung cancer death rates for smokers, never smokers, and former smokers by duration of cessation. These estimates are combined with population-based estimates of smoking status by birth cohort to generate models of lung cancer death rates for current and former smokers. RESULTS: Modeled estimates of lung cancer death rates suggest that, in the near future, the majority of lung cancer cases will be occurring among former smokers. CONCLUSIONS: Former smokers already have acted on the advice of their physicians to reduce their lung cancer risk; however, they will generate an increasing fraction of the cancers that are seen. Approaches must be developed for these former smokers to alter their risk if lung cancer death rates are going to continue to be affected. Additional primary preventive strategies, including chemoprevention, offer hope to these former smokers, as do secondary preventive efforts, such as screening.

Adolescent↗
PubMed1988

Control of feeding movements in the freshwater snail Planorbis corneus. III. Organization of the feeding rhythm generator.

(1) Neurons of different groups (for group classification, see Arshavsky et al. 1988a) have been polarized through an intracellular recording microelectrode in Planorbis corneus buccal ganglia during feeding rhythm generation. Group 1 neurons, active in the quiescence (Q) and in the protractor (P) phases of the cycle, and also group 2 and 4 neurons, active in the retractor (R) phase, have proved to be "influential", i.e., altering the rhythm generator operation. (2) Injection of a depolarizing current into group 1 neurons caused an increase of the rate of depolarization that neurons of this group exhibit in the Q- and P-phases of the feeding cycle. As a result, Q-phase shortened, the P-phase became longer, and the feeding rhythm accelerated. Opposite effects occurred when a hyperpolarizing current was injected into group 1 neurons. In some of the experiments, the hyperpolarization of group 1 neurons resulted in cessation of both their activity and the activity of all other protractor neurons. As a result, the P-phase of the cycle disappeared, i.e., the rhythm generator transited from A mode of operation to B mode. (3) With hyperpolarization of individual group 2 or 4 neurons, excitation of the R-phase neurons was delayed and the feeding rhythm phase shifted. This delay was accompanied by the enhanced activity of protractor neurons. (4) A generator model is considered in which two groups (1 and 2) of endogeneously active neurons are coordinated by the excitatory effect of group 1 on group 2 and the inhibitory action of group 2 on group 1. (5) Evidence is given that the different modes of rhythm generator operation (A, B and C, see Arshavsky et al. 1988a) are determined by different tonic inflow to group 1 neurons.

Action Potentials↗
PubMed1990

Simulating functional interactions in the brain: a model for examining correlations between regional cerebral metabolic rates.

A computer simulation model was developed to investigate the use of interregional correlations of cerebral metabolic rates to analyze functional interactions in the brain. The model generates simulated metabolic data for individual brain regions in a specified number of subjects, where there are defined functional couplings amongst the regions. Random numbers provide the variability seen in measured metabolic data. Correlational analysis is performed on these simulated data sets. The parameters of the model can be chosen so that simulated and actual metabolic data are very similar. The model demonstrates that the change in the correlation coefficient between normalized metabolic data in two brain regions is related to the change in the strength of the functional association between the two regions. The model also is used to explore the relations between patterns of correlations and the underlying sets of functional couplings. The results indicate that correlational analysis provides more information about regional involvement in neural systems than does region-by-region comparisons of absolute metabolic rates.

Adult↗
PubMed1996

An efficient tissue classifier for building patient-specific finite element models from X-ray CT images.

We developed an efficient semiautomatic tissue classifier for X-ray computed tomography (CT) images which can be used to build patient- or animal-specific finite element (FE) models for bioelectric studies. The classifier uses a gray scale histogram for each tissue type and three-dimensional (3-D) neighborhood information. A total of 537 CT images from four animals (pigs) were classified with an average accuracy of 96.5% compared to manual classification by a radiologist. The use of 3-D, as opposed to 2-D, information reduced the error rate by 78%. Models generated using minimal or full manual editing yielded substantially identical voltage profiles. For the purpose of calculating voltage gradients or current densities in specific tissues, such as the myocardium, the appropriate slices need to be fully edited, however. Our classifier offers an approach to building FE models from image information with a level of manual effort that can be adjusted to the need of the application.

Algorithms↗
PubMed2005

Pitch discrimination of patterned electric stimulation.

One reason for the poor pitch performance in current cochlear-implant users may be the highly synchronized neural firing in electric hearing that lacks stochastic properties of neural firing in normal acoustic hearing. This study used three different electric stimulation patterns, jittered, probabilistic, and auditory-model-generated pulses, to mimic some aspects of the normal neural firing pattern in acoustic hearing. Pitch discrimination was measured at standard frequencies of 100, 250, 500, and 1000 Hz on three Nucleus-24 cochlear-implant users. To test the utility of the autocorrelation pitch perception model in electric hearing, one, two, and four electrodes were stimulated independently with the same patterned electric stimulation. Results showed no improvement in performance with any experimental pattern compared to the fixed-rate control. Pitch discrimination was actually worsened with the jittered pattern at low frequencies (125 and 250 Hz) than that of the control, suggesting that externally introduced stochastic properties do not improve pitch perception in electric stimulation. The multiple-electrode stimulation did not improve performance but did not degrade performance either. The present results suggest that both "the right time and the right place" may be needed to restore normal pitch perception in cochlear-implant users.

Aged↗
PubMed1986

Use of biosynthetic human C-peptide in the measurement of insulin secretion rates in normal volunteers and type I diabetic patients.

We undertook this study to examine the accuracy of plasma C-peptide as a marker of insulin secretion. The peripheral kinetics of biosynthetic human C-peptide (BHCP) were studied in 10 normal volunteers and 7 insulin-dependent diabetic patients. Each subject received intravenous bolus injections of BHCP as well as constant and variable rate infusions. After intravenous bolus injections the metabolic clearance rate of BHCP (3.8 +/- 0.1 ml/kg per min, mean +/- SEM) was not significantly different from the value obtained during its constant intravenous infusion (3.9 +/- 0.1 ml/kg per min). The metabolic clearance rate of C-peptide measured during steady state intravenous infusions was constant over a wide concentration range. During experiments in which BHCP was infused at a variable rate, the peripheral concentration of C-peptide did not change in proportion to the infusion rate. Thus, the infusion rate of BHCP could not be calculated accurately as the product of the C-peptide concentration and metabolic clearance rate. However, the non-steady infusion rate of BHCP could be accurately calculated from peripheral C-peptide concentrations using a two-compartment mathematical model when model parameters were derived from the C-peptide decay curve in each subject. Application of this model to predict constant infusions of C-peptide from peripheral C-peptide concentrations resulted in model generated estimates of the C-peptide infusion rate that were 101.5 +/- 3.4% and 100.4 +/- 2.8% of low and high dose rates, respectively. Estimates of the total quantity of C-peptide infused at a variable rate over 240 min based on the two-compartment model represented 104.6 +/- 2.4% of the amount actually infused. Application of this approach to clinical studies will allow the secretion rate of insulin to be estimated with considerable accuracy. The insulin secretion rate in normal subjects after an overnight fast was 89.1 pmol/min, which corresponds with a basal 24-h secretion of 18.6 U.

Adult↗
PubMed1993

Model of lactose repressor core based on alignment with sugar-binding proteins is concordant with genetic and chemical data.

Using primary sequence similarity to arabinose-binding protein, D-glucose/D-galactose-binding protein, and ribose-binding protein (Vyas, N. K., Vyas, M. N., and Quiocho, F. A. (1991) J. Biol. Chem. 266, 5226-5237; Mowbray, S. L., and Cole, L. B. (1992) J. Mol. Biol. 225, 155-175), the core domain (residues 62-323) of the bacterial regulatory protein lac repressor has been aligned to these sugar-binding proteins of known structure. Although the sequence identity is not striking, there is strong overall homology based on two separate matrix scoring systems (minimum base change per codon (MBC/C) and amino acid homology per residue (AAH/R)) (mean score: MBC/C < 1.25, AAH/R > 5.50; random sequences: MBC/C = 1.45, AAH/R = 4.46). Similarly, the predicted secondary structure of the repressor exhibits excellent agreement with the known secondary structures of the sugar-binding proteins. Using this primary sequence alignment, the tertiary structure of the core domain of the lac repressor has been modeled based on the known structures of the sugar-binding proteins as templates. While the structure deduced for the repressor is hypothetical, the model generated allows a comparison between the predicted tertiary arrangement and the wealth of genetic and chemical data elucidated for the repressor. Important residues involved in operator and sugar binding and in protein assembly have been identified using genetic methods, and placement of these residues in the model is consistent with their known function. This approach, therefore, provides a means to visualize the core domain of the lac repressor that allows interpretation of genetic and chemical data for specific residues and rational design of future experiments.

Amino Acid Sequence↗
PubMed1987

Modeling: a method for program development.

Successful program development is crucial for nursing. The methodology of modeling systematically integrates program development strategies recommended in the nursing literature. In addition to guiding program planning and design, the series of diagrams which modeling generates can be used to guide program implementation, evaluation, refinement, and reformulation. The diagramming also facilitates communication with others about the program.

Health Services Needs and Demand↗
PubMed2005

Design of recombinant stem cell factor-macrophage colony stimulating factor fusion proteins and their biological activity in vitro.

Stem cell factor (SCF) and macrophage colony stimulating factor (M-CSF) can act in synergistic way to promote the growth of mononuclear phagocytes. SCF-M-CSF fusion proteins were designed on the computer using the Homology and Biopolymer modules of the software packages InsightII. Several existing crystal structures were used as templates to generate models of the complexes of receptor with fusion protein. The structure rationality of the fusion protein incorporated a series of flexible linker peptide was analyzed on InsightII system. Then, a suitable peptide GGGGSGGGGSGG was chosen for the fusion protein. Two recombinant SCF-M-CSF fusion proteins were generated by construction of a plasmid in which the coding regions of human SCF (1-165aa) and M-CSF (1-149aa) cDNA were connected by this linker peptide coding sequence followed by subsequent expression in insect cell. The results of Western blot and activity analysis showed that these two recombinant fusion proteins existed as a dimer with a molecular weight of approximately 84 KD under non-reducing conditions and a monomer of approximately 42 KD at reducing condition. The results of cell proliferation assays showed that each fusion protein induced a dose-dependent proliferative response. At equimolar concentration, SCF/M-CSF was about 20 times more potent than the standard monomeric SCF in stimulating TF-1 cell line growth, while M-CSF/SCF was 10 times of monomeric SCF. No activity difference of M-CSF/SCF or SCF/M-CSF to M-CSF (at same molar) was found in stimulating the HL-60 cell linear growth. The synergistic effect of SCF and M-CSF moieties in the fusion proteins was demonstrated by the result of clonogenic assay performed with human bone mononuclear, in which both SCF/M-CSF and M-CSF/SCF induced much higher number of CFU-M than equimolar amount of SCF or M-CSF or that of two cytokines mixture.

Animals↗
PubMed1991

A correlation between the distribution of biological apatite and amino acid sequence of type I collagen.

We have determined the localization of apatite within type I collagen fibrils of calcifying turkey leg tendons by both bright field and selected-area dark field (SADF) electron microscopy and have compared this to computer-modeled, chick type I collagen amino acid sequence data. Apatite crystals occur in both the gap and overlap zones at early stages of mineralization in an asymmetric pattern that corresponds to the polarity, N- to C- orientation, of the collagen molecule. Based on comparisons with computer-generated models of known amino acid sequence of collagen, it was determined for early stages of mineral deposition that apatite is restricted by areas of high hydrophobicity. The gap zone is less hydrophobic than the overlap zone on average but each of these zones had areas of high hydrophobicity that correlated with sites of low localization of mineral. Possible interactions between hydrophobic regions and the process of mineral deposition are discussed.

Amino Acid Sequence↗