Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Bayesian computational modeling”

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 73 records · Page 4Linked to original sources
PubMedDate not supplied

Designing for nonparametric Bayesian survival analysis using historical controls.

This paper gives a method for choosing the number of patients N0 out of N available patients to be randomized to current controls om a two-arm study when comparison of nonparametric survival curves is the anticipated method of data analysis. The criterion imposed is that of choosing N0 to minimize the posterior variance of the difference between the current control and experimental survival curves. A nonparametric Bayesian argument incorporating the survival curve of available historical controls establishes the criterion. Formulas and tables which facilitate this computation are presented.

Bayes Theorem↗
PubMed1985

A calculator program for clinical application of the Bayesian method of predicting plasma drug levels.

A pharmacokinetic program that allows individualization of drug dosage regimens through the Bayesian method is described. The program, which is designed for the Hewlett-Packard HP-41 CV calculator, is based upon the one-compartment open model with either instantaneous or zero-order absorption. Individualized estimation of the patient's kinetic parameters (clearance and volume of distribution) is performed by analyzing the plasma levels measured in the patient as well as considering the population data of the drug. After estimating the individual kinetic parameters by the Bayesian method, the program predicts the dosage regimen that will elicit the desired peak and trough plasma levels at steady state. For comparison purposes, the least-squares estimates for clearance and volume of distribution are calculated, and dosage prediction can also be made on the basis of the least-squares estimates. The least-squares estimates can be used to calculate population pharmacokinetic parameters according to the Standard Two-Stage method. Several examples of clinical use of the program are presented. The examples refer to patients with classic hemophilia who were treated with Factor VIII concentrates. In these patients, the Bayesian kinetic parameters of Factor VIII have been estimated through the calculator program. The Bayesian parameter estimates generated by the HP-41 have been compared with those determined by a Bayesian program (ADVISE) designed for microcomputers.

Computers↗
PubMed1980

Numerical evaluation of cytologic data. V. Bivariate distributions and the Bayesian decision Boundary.

The evaluation of cytologic data often involves the classification of observations into alternative categories (data sets). Plotting the elliptical contours of bivariate distributions provides immediate insight into the structure of data sets and their mutual relations. In this paper, the computation of tolerance ellipses and confidence ellipses for bivariate distribution is demonstrated, and the finding of a Bayesian decision boundary between two bivariate distributions is illustrated.

Bayes Theorem↗
PubMed1992

Predictive performance of Bayesian and nonlinear least-squares regression programs for lidocaine.

The predictive performance of two computer programs for lidocaine dosing were evaluated. Two-compartment Bayesian and nonlinear least-squares regression programs were used in two groups of patients (15 acute arrhythmia patients and 14 chronic arrhythmia patients). Lidocaine was given as a 1.5 mg/kg bolus and a 2.8 mg/min infusion for 48 h. A second bolus (0.5 mg/kg) was given 10 min after the first bolus over 2 min. Serum samples of the patients receiving lidocaine were drawn at 2, 15, 30 min and 1, 2, and 4 h and were used in forecasting the serum concentrations at 6, 8, 12, and 48 h. Predictive performance was assessed by mean error and mean-squared error. The results (mean +/- 95% confidence intervals) demonstrated the Bayesian program predicted a significant (p less than 0.05) difference at 12 h between the two arrhythmia groups (acute 0.52 [-0.95; -0.09] and chronic 0.28 [0.12; 0.44]). The results also demonstrated the Bayesian method was significantly more precise compared to the nonlinear least-squares regression program at 8, 12, and 48 h for the acute group. While caution is warranted, this study demonstrated that the predictive performance by a two-compartment Bayesian model is more accurate in predicting future lidocaine serum concentrations than that by nonlinear least-squares regression.

Acute Disease↗
PubMed1995

Bayesian estimation of p-aminohippurate clearance by a limited sampling strategy.

This study describes a methodology to calculated p-aminohippurate (PAH) clearance (CL) and volume of distribution (V) with both the population parameters and one or two samples taken during the disposition and the elimination phase after a single intravenous infusion. The computer program P-PHARM was used, and a log-normal distribution and a heteroscedastic residual error distribution were assumed. Ninety-six patients with and without renal insufficiency were available for analysis, and a two-compartment model was used for data modeling. Population parameters were evaluated for 70 patients (mean number of observed concentration per individual, 6) by a three-step approach. In step 1, the computer program was used to estimate the average pharmacokinetic parameters without taking into account the demographic and/or biological factors. In step 2, the relationship between the posterior individual estimates and the covariables was investigated with multiple linear stepwise algorithm. In step 3, the population parameters were re-estimated considering the relationship with the covariables. From the regression performed in step 2, the following covariables were included: serum creatinine, body surface area, and body weight. The population averages of CL and V were 30.7 +/- 2.36 L/h and 10.6 +/- 1.29 L, respectively. To evaluate the predictive performance of the population parameters, the remaining 26 patients were used. The population parameters combined with one or two individual PAH plasma concentrations led to a bayesian estimation of individual CL and V. This estimation was compared with the classical procedure of parameter estimation (individual fitting from multiple blood samples).(ABSTRACT TRUNCATED AT 250 WORDS)

Adult↗
PubMed1990

Classification of audiograms by sequential testing using a dynamic Bayesian procedure.

A new method for estimating audiograms using behavioral responses is presented. The method is based upon a modification of the Bayesian probability formula in which an outcome is predicted from a static set of events. In the new method, classification of audiograms by sequential testing (CAST), the probabilities of occurrence of audiogram patterns are dynamically updated according to the outcome of each test trial. Computer simulation using an infant response model suggests that the procedure is efficient, sensitive, and specific.

Algorithms↗
PubMed2025

The signed two-space proximity model for learning representations in protein-protein interaction networks.

MOTIVATION: Accurately predicting complex protein-protein interactions (PPIs) is crucial for decoding biological processes, from cellular functioning to disease mechanisms. However, experimental methods for determining PPIs are computationally expensive. Thus, attention has been recently drawn to machine learning approaches. Furthermore, insufficient effort has been made toward analyzing signed PPI networks, which capture both activating (positive) and inhibitory (negative) interactions. To accurately represent biological relationships, we present the Signed Two-Space Proximity Model (S2-SPM) for signed PPI networks, which explicitly incorporates both types of interactions, reflecting the complex regulatory mechanisms within biological systems. This is achieved by leveraging two independent latent spaces to differentiate between positive and negative interactions while representing protein similarity through proximity in these spaces. Our approach also enables the identification of archetypes representing extreme protein profiles. RESULTS: S2-SPM's superior performance in predicting the presence and sign of interactions in SPPI networks is demonstrated in link prediction tasks against relevant baseline methods. Additionally, the biological prevalence of the identified archetypes is confirmed by an enrichment analysis of Gene Ontology (GO) terms, which reveals that distinct biological tasks are associated with archetypal groups formed by both interactions. This study is also validated regarding statistical significance and sensitivity analysis, providing insights into the functional roles of different interaction types. Finally, the robustness and consistency of the extracted archetype structures are confirmed using the Bayesian Normalized Mutual Information (BNMI) metric, proving the model's reliability in capturing meaningful SPPI patterns. AVAILABILITY: S2-SPM is implemented and freely available under the MIT license at https://github.com/Nicknakis/S2SPM.

Protein Interaction Mapping↗
PubMed2026

A Bayesian framework for multivariate differential analysis.

Differential analysis is a routine procedure in the statistical analysis toolbox across many applied fields, including quantitative proteomics, the main illustration of the present paper. The state-of-the-art limma approach uses a hierarchical formulation with moderated-variance estimators for each analyte directly injected into the t-statistic. While standard hypothesis testing strategies are recognised for their low computational cost, allowing for quick extraction of the most differential among thousands of elements, they generally overlook key aspects such as handling missing values, inter-element correlations, and uncertainty quantification. The present paper proposes a fully Bayesian framework for differential analysis, leveraging a conjugate hierarchical formulation for both the mean and the variance. Inference is performed by computing the posterior distribution of compared experimental conditions and sampling from the distribution of differences. This approach provides well-calibrated uncertainty quantification at a similar computational cost as hypothesis testing by leveraging closed-form equations. Furthermore, a natural extension enables multivariate differential analysis that accounts for possible inter-element correlations. We also demonstrate that, in this Bayesian treatment, missing at random data should generally be ignored in univariate settings, and further derive a tailored approximation that handles multiple imputation for the multivariate setting. We argue that probabilistic statements in terms of effect size and associated uncertainty are better suited to practical decision-making. Therefore, we finally propose simple and intuitive inference criteria, such as the overlap coefficient, which express group similarity as a probability rather than traditional, and often misleading, p-values. The performance of this approach is evaluated through an extensive empirical study using both synthetic and controlled real-world proteomics datasets. Overall, we believe that this Bayesian framework for (multivariate) differential analysis provides a valuable and intuitive counterpart to standard methods at a comparable computational cost.

Bayes Theorem↗
PubMedDate not supplied

Qualitative probability versus quantitative probability in clinical diagnosis: a study using a computer simulation.

The use of Bayes' theorem as a diagnostic tool in clinical medicine normally requires an input of exact probability estimates. However, humans tend to think in categories ("likely," "unlikely," etc.) rather than in terms of exact probability. A computer simulation of the presenting features of a case of pelvic infection has been used to compare the effects of quantitative and qualitative probability estimates on the diagnostic accuracy of Bayes' theorem. For the commoner conditions (prior probability greater than or equal to 0.2) the use of a two- or three-category system is virtually equivalent to the use of exact probability. However, uncommon conditions (prior probability less than or equal to 0.03) are completely ignored by the qualitative system. It is concluded that the use of simple categories of probability is acceptable for a Bayesian diagnostic system provided that the target conditions have a relatively high prior probability.

Artificial Intelligence↗
PubMed1988

Evaluating diagnostic performance of clinical tests by spreadsheet modeling. Bayesian analysis using Ri/Cj ratio as a unifying concept.

We present a general spreadsheet model for evaluating diagnostic performance of clinical tests. Our model depicts test results as an r X c matrix, with r possible test results and c possible clinical states. Analysis of this matrix is based on the Ri/Cj ratio, calculated as a number of subjects having a specified result Ri within a given clinical state Cj, divided by total subjects within this clinical state. From this model, we can identify three special cases: (1) a 2 X c matrix, with two possible test results of T+ or T-, over c possible clinical states; (2) an r X 2 matrix, with r possible test results, over two possible clinical states of D+ or D-; and (3) a 2 X 2 matrix, with two possible test results over two possible clinical states. Application of the Ri/Cj ratio to the r X c matrix provides a useful approach to graphic analysis of multiple test results over multiple clinical states. The Ri/Cj ratio also provides a general approach to Bayesian analysis, in which likelihood ratio, relative operating characteristic analysis, sensitivity, and specificity represent special cases or special applications.

Bayes Theorem↗
PubMed1989

Computer-assisted optimization of aminophylline therapy in the emergency department.

The emergency department (ED) is a unique setting for pharmacokinetic-guided drug administration because of the need to rapidly optimize therapy. We compared outcomes in patients receiving intravenous aminophylline according to population-based ED guidelines (group 1) or Bayesian-derived pharmacokinetic estimates (group 2), we determined predictors for admission or discharge in our study group, and we assessed the ability of a Bayesian pharmacokinetic model to estimate theophylline requirements in the ED. The study population was composed of 82 patients (42 males, 40 females) with a mean age of 43 +/- 15.5 years. Fifteen patients were excluded because of protocol violations. Of the 67 cases studied, 30 were assigned to group 1, and 37 were assigned to group 2. Patient demographics, baseline theophylline concentration, and theophylline loading dose did not differ significantly between treatment groups. The aminophylline maintenance infusion was significantly (P less than .001) lower in group 1 (0.4 +/- 0.2 mg/kg/h) than in group 2 (0.6 +/- 0.2 mg/kg/h). Serum theophylline concentrations at one hour post-loading-dose did not differ significantly between treatment groups; however, significant differences were observed at two hours post-load (P less than .002) and four hours post-load (P less than .001). Baseline peak flow rate (PFR) was significantly (P less than .03) higher in group 1 (170 +/- 85 L/min) than in group 2 (132 +/- 62 L/min), but did not differ significantly at any other times throughout the study. The PFR one hour post-load (PFR-1) was the strongest (P less than .003) predictor of outcome.(ABSTRACT TRUNCATED AT 250 WORDS)

Adult↗
PubMedDate not supplied

Data-source effects on the sensitivities and specificities of clinical features in the diagnosis of rheumatoid arthritis: the relevance of multiple sources of knowledge for a decision-support system.

An experimental computer system was developed to support diagnosis of rheumatic disorders by computing diagnostic probabilities using modified likelihood ratios. The authors examined whether the performance of the model was affected by the settings in which the data used to derive the likelihood ratios were collected. The sensitivities and specificities of various clinical features for diagnosing rheumatoid arthritis (RA) were obtained from: 1) a study of 1,570 consecutive outpatients at a rheumatology clinic; 2) a review of the literature; 3) estimates by rheumatologists; and 4) a population study. Considerable variations in sensitivity and specificity but satisfactory agreement in likelihood ratios were found across the four data sets. The likelihood ratios were then used to compute the probabilities of RA in a test series of 570 of the rheumatology clinic outpatients. The model's diagnoses with likelihood ratios from the other sources were adequate. When the likelihood ratios from these sources were combined, discrimination came close to what could be achieved by using the likelihood ratios based on the data from the clinic. The method applied in the study, which makes use of variation of input data instead of variation of test series, and the results are relevant to assessing the external validity and transferability of Bayesian decision-support systems.

Adolescent↗
PubMed1987

Bayesian statistical theory in the preoperative diagnosis of pulmonary lesions.

We used a computerized Bayesian algorithm to assist in the preoperative diagnosis of pulmonary lesions. One hundred consecutive patients who were undergoing exploratory thoracotomy for newly discovered pulmonary lesions were prospectively evaluated. The Bayesian model used a total of 44 preoperative clinical and roentgenographic factors to categorize the lesions as benign or malignant. The Bayesian algorithm correctly categorized 96 of the 100 lesions, thereby providing an accuracy of 96 percent. The sensitivity of the model was 98 percent and the specificity was 87 percent. All but two of the 85 malignant lesions were correctly categorized and 13 of the 15 benign lesions were correctly analyzed by the model. These results indicate that computer-assisted diagnosis using the Theorem of Bayes may provide valuable preoperative information for the management of selected patients.

Adolescent↗
PubMedDate not supplied

Population pharmacokinetics. Theory and clinical application.

Good therapeutic practice should always be based on an understanding of pharmacokinetic variability. This ensures that dosage adjustments can be made to accommodate differences in pharmacokinetics due to genetic, environmental, physiological or pathological factors. The identification of the circumstances in which these factors play a significant role depends on the conduct of pharmacokinetic studies throughout all stages of drug development. Advances in pharmacokinetic data analysis in the last 10 years have opened up a more comprehensive approach to this subject: early traditional small group studies may now be complemented by later population-based studies. This change in emphasis has been largely brought about by the development of appropriate computer software (NONMEM: Nonlinear Mixed Effects Model) and its successful application to the retrospective analysis of clinical data of a number of commonly used drugs, e.g. digoxin, phenytoin, gentamicin, procainamide, mexiletine and lignocaine (lidocaine). Success has been measured in terms of the provision of information which leads to increased efficiency in dosage adjustment, usually based on a subsequent Bayesian feedback procedure. The application of NONMEM to new drugs, however, raises a number of interesting questions, e.g. 'what experimental design strategies should be employed?' and 'can kinetic parameter distributions other than those which are unimodal and normal be identified?' An answer to the later question may be provided by an alternative non-parametric maximum likelihood (NPML) approach. Population kinetic studies generate a considerable amount of demographic and concentration-time data; the effort involved may be wasted unless sufficient attention is paid to the organisation and storage of such information. This is greatly facilitated by the creation of specially designed clinical pharmacokinetic data bases, conveniently stored on microcomputers. A move towards the adoption of population pharmacokinetics as a routine procedure during drug development should now be encouraged. A number of studies have shown that it is possible to organise existing, routine data in such a way that valuable information on pharmacokinetic variability can be obtained. It should be relatively easy to organise similar studies prospectively during drug development and, where appropriate, proceed to the establishment of control systems based on Bayesian feedback.

Demography↗
PubMed1991

Bayesian diagnostic probabilities without assuming independence of symptoms.

The paper describes an application of Bayes' Theorem to the problem of estimating from past data the probabilities that patients have certain diseases, given their symptoms. The data consist of hospital records of patients who suffered acute abdominal pain. For each patient the records showed a large number of symptoms and the final diagnosis to one of nine diseases or diagnostic groups. Most current methods of computer diagnosis use the "Simple Bayes" model in which the symptoms are assumed to be independent, but the present paper does not make this assumption. Those symptoms (or lack of symptoms) which are most relevant to the diagnosis of each disease are identified by a sequence of chi-squared tests. The computer diagnoses obtained as a result of the implementation of this approach are compared with those given by the "Simple Bayes" method, by the method of classification trees (CART), and also with the preliminary and final diagnoses made by physicians.

Abdominal Pain↗