SEARCH · PubMed Health

Results for “predictive modelling”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 523 records · Page 29Linked to original sources

PubMed1990

A predictive model for vapor concentration in a nose-only inhalation chamber.

A unique nose-only inhalation chamber was designed and constructed to deliver uniform concentrations of gas, vapor, and aerosol contaminants to mice. This research investigated the fluid dynamics of a vaporous contaminant in the vertical flow chamber. The vapor was introduced by allowing the liquid phase of the contaminant to evaporate freely into the chamber interior. A contaminant mass transfer model was developed to predict concentrations generated by the system. The mathematical model of the system used clean airflow, liquid surface area, thickness of the stagnant air layer covering the liquid, system pressure, contaminant diffusion coefficient, and contaminant vapor pressure to compute the vapor concentration delivered to exposure ports. The equation was verified by placing various containers of methyl isobutyl ketone in the chamber and determining with a photospectrometer the resulting equilibrium concentrations. Vapor pressure, diffusion coefficient, and system pressure were held constant while airflow, surface area, and stagnant air layer thickness were varied systematically within the chamber. The resulting empirical data points were compared to the curves predicted by the theoretical model. Empirical concentrations fell within 0 to 48% of the theoretical values, showing that the equation can be used to choose values for airflow, surface area, and stagnant air layer thickness that will result in chamber concentrations in close proximity to the target concentration. If an exact concentration is essential, parameters may be individually adjusted to converge on the target concentration.

Animals↗

PubMed1999

Validation of predictive models describing the growth of Listeria monocytogenes.

In this study, predictions for growth rate of Listeria on food products were evaluated by both general applicable models and specific growth models. Literature values, obtained from a large number of publications, for growth rates in/on a variety of foods were compared by graphical and mathematical analysis with predictions given by various models. Apart for the great advantage of being generally applicable, the general models performed best. However, only small differences between the various models were observed. Model predictions were accurate within a factor of about two to four, depending on the type of product. The predictions should therefore not be considered as absolute; it is important to understand the limitations of the performance of models. All results and all assumptions should be criticised, but in many cases the accuracy will be sufficient to use these types of models as a tool in management decisions.

Animals↗

PubMed2003

Model predictive control with amplitude and rate actuator saturation.

In this work we show that the anti-wind-up-bumpless-transfer controller emerges from the structure of model predictive control (MPC) with quadratic objective and input constrains. The key to establish that relationship is the application of optimality conditions to the equivalent optimal control problem. The proposed framework employs a model of physical constraints as part of the controller architecture to ensure that the commands sent to the actuator do not exceed their specific limits and the internal states of the controller are well updated. Numerical examples are presented for illustrating the proposed control design methodology.

Journal Article↗

PubMed2006

Learning a predictive model for growth inhibition from the NCI DTP human tumor cell line screening data: does gene expression make a difference?

We address the problem of learning a predictive model for growth inhibition from the NCI DTP human tumor cell line screening data. Extending the classical Quantitative Structure Activity Relationship paradigm, we investigate whether including gene expression data leads to a statistically significant improvement of prediction quality. Our analysis shows that the straightforward approach of including individual gene expression as features does not necessarily improve, but on the contrary, may degrade performance significantly. When gene expression information is aggregated, for instance by features representing the correlation with reference cell lines, performance can be improved significantly. Further improvements may be expected if the learning task is structured by grouping features and instances.

Cell Line, Tumor↗

PubMed2006

Application of BRCA1 and BRCA2 mutation carrier prediction models in breast and/or ovarian cancer families of French Canadian descent.

The BRCAPRO, Couch, Myriad I and II, Ontario Family History Assessment Tool (FHAT), and Manchester models have been used to predict BRCA1 or BRCA2 mutation carrier status of women at high risk for developing the heritable form of breast and ovarian cancers. We have evaluated these models for their accuracy in classifying 224 French Canadian families with at least three cases of breast cancer (diagnosed before the age of 65 years), ovarian cancer, or male breast cancer where mutation status was known for an index affected case used to assess the model. This series includes 44 BRCA1 and 52 BRCA2 mutation-positive families. Using receiver operator characteristics analyses, the C-statistics were found to be 0.81, 0.80, 0.79, and 0.74 for the BRCAPRO, FHAT, Manchester, and Myriad II models, respectively, when incorporating both BRCA1 and BRCA2 mutation carrier predictions. For the BRCAPRO model, 75% scored greater than a 0.43 probability in the mutation-positive group and 75% scored less than 0.50 in the mutation-negative group. Only 38 of 128 (30%) mutation-negative group had a probability greater than 0.43 with the BRCAPRO model. While all models were highly predictive of carrier status, the BRCAPRO model was the most accurate where a cut-off of 10% would have eliminated 60 of 128 (47%) mutation-negative families for genetic testing and only miss 10 of 96 (10%) mutation-positive families. A review of the cancer phenotypes with high BRCAPRO probabilities showed that significantly more metachronous bilateral breast cancer cases occurred in BRCA1/2 mutation carrier families in comparison to mutation-negative families, a feature which is not discriminated in the BRCAPRO model.

BRCA1 Protein↗

PubMed2001

Cost prediction models for the comparison of two groups.

For trial-based economic evaluation where patient-specific cost data are not routinely available, cost prediction models are commonly used to estimate total cost for each patient. Typically, multiple regression techniques are used on data from diagnosis-matched, non-trial patients (where patient-level cost data are available) to model cost as a function of covariates that are observed on the trial subjects (e.g. length of hospital stay, procedures, etc.). The estimated beta coefficients provide a means of estimating the total cost for each patient in the trial. However, the variability of the beta coefficients due the measurement and sampling error is seldom included in the overall variance expression for mean costs by treatment group. In this paper we provide a method for estimating this variance and provide an example application

Angina, Unstable↗

PubMed1989

A psychosociomedical prediction model of response to treatment by chronically disabled workers with low-back pain.

There has been much interest in identifying variables that can predict which individuals are susceptible to developing chronic low-back pain. There currently are a number of studies that are evaluating primary predictors (which uninjured workers are likely to develop chronic low-back pain) and secondary predictors (which workers with acute episodes will develop chronic pain). The present study reports the first results from a large-scale investigation of tertiary predictors. Specifically, it addresses the issue of what psychosociomedical variables are predictive of success/failure in response to a comprehensive Functional Restoration treatment program by workers who are chronically disabled with low-back pain. Three stages were involved in the development of this prediction model. First, a group of treatment and research professionals who had extensive experience in the area of chronic low-back pain identified an array of 42 variables, from a larger pool of quantified physical, psychosocial, and medical parameters rated to be important with this patient population.(ABSTRACT TRUNCATED AT 250 WORDS)

Adult↗

PubMed2000

Application of model-predictive control based on artificial neural networks to optimize the fed-batch process for riboflavin production.

The fed-batch process for commercial production of riboflavin (vitamin B2) was optimized on-line using model-predictive control based on artificial neural networks (ANNs). The information required for process models was extracted from both historical data and heuristic rules. After each cultivation the process model was readapted off-line to include the most recent process data. The control signal (feed rate), however, was optimized on-line at each sampling interval. An optimizer simulated variations in the control signal and assessed the forecasted model outputs according to an objective function. The optimum feed profile for increasing the product yield (YB2/S) and the amount of riboflavin at the time of harvesting was adjusted continuously and applied to the process. In contrast to the control by set-point profiles, the novel ANN-control is able to react on-line to variations in the process and also to incorporate the new process information continuously. As a result, both the total amount of riboflavin produced and the product yield increased systematically by more than 10% and the reproducibility of seven subsequently optimized batches was enhanced.

Algorithms↗

PubMed2006

Applicability of the nipple-areola complex-sparing mastectomy: a prediction model using mammography to estimate risk of nipple-areola complex involvement in breast cancer patients.

The purpose of this study was to develop a prediction model that can be used to identify breast cancer patients at lowest risk for neoplastic nipple-areola complex (NAC) involvement to offer total NAC-sparing mastectomy with immediate reconstruction. Medical records, pathology slides, and mammograms were reviewed for all breast cancer patients treated with total mastectomy at Rhode Island Hospital between 2000 and 2004. The distance between the nipple and the closest tumor margin was measured using mammography. NAC involvement was identified in 42% of the 31 study patients. Mammographic distance, pathologic stage, and tumor size were identified as independent predictors of malignant NAC involvement by multivariate analysis (rho < 0.05). Based on these predictors, a linear discriminant score, the NAC Involvement Score (NACIS), was computed to distinguish between the presence and absence of NAC involvement. For individual patients, positive NACIS values (> or = -0.3665) were associated with NAC involvement with a sensitivity of 92%, specificity of 77%, and negative predictive value of 93%. These preliminary findings indicate that the NACIS formula may be a useful clinical tool for selecting low-risk patients for total NAC-sparing mastectomy with immediate reconstruction.

Adult↗

PubMed2006

Robust and nominal stability conditions for a simplified model predictive controller.

A new condition is derived that guarantees robust stability for a set of stable, linear time-invariant plants controlled by using a simplified model predictive control algorithm (SMPC). Discrete single-input-single-output control systems are considered in this paper. Uncertainty is treated in the time domain by considering the stabilization of a set of pulse response functions. The method presented is suitable for stabilizing a set of plants that are not necessarily related. Central to this method is a bounding function, which is a function of the model and controller parameters. The bounding function is designed to have a larger magnitude than all of the pulse response functions in the set of plants to be stabilized. Using this method, it was found that the bounding function is monotonically decreasing when a first-order plus dead-time model is used to design the controller. This allows the coincidence point used in SMPC to be employed directly as a tuning "knob" for robustness, and also simplifies the analysis for dead-time uncertainty. In addition, a comparison of two nominal stability conditions is provided.

Journal Article↗

PubMed2005

Electrostatic modeling predicts the activities of orthopoxvirus complement control proteins.

Regulation of complement activation by pathogens and the host are critical for survival. Using two highly related orthopoxvirus proteins, the vaccinia and variola (smallpox) virus complement control proteins, which differ by only 11 aa, but differ 1000-fold in their ability to regulate complement activation, we investigated the role of electrostatic potential in predicting functional activity. Electrostatic modeling of the two proteins predicted that altering the vaccinia virus protein to contain the amino acids present in the second short consensus repeat domain of the smallpox protein would result in a vaccinia virus protein with increased complement regulatory activity. Mutagenesis of the vaccinia virus protein confirmed that changing the electrostatic potential of specific regions of the molecule influences its activity and identifies critical residues that result in enhanced function as measured by binding to C3b, inhibition of the alternative pathway of complement activation, and cofactor activity. In addition, we also demonstrate that despite the enhanced activity of the variola virus protein, its cofactor activity in the factor I-mediated degradation of C3b does not result in the cleavage of the alpha' chain of C3b between residues 954-955. Our data have important implications in our understanding of how regulators of complement activation interact with complement, the regulation of the innate immune system, and the rational design of potent complement inhibitors that might be used as therapeutic agents.

Complement C3b Inactivator Proteins↗

PubMed2001

A new predictive model for adverse outcomes after elective thoracoabdominal aortic aneurysm repair.

BACKGROUND: Recent recommendations have emphasized individualized treatment based on balancing a patient's risk of thoracoabdominal aortic aneurysm rupture with the risk of an adverse outcome after surgical repair. The purpose of this study was to determine which preoperative risk factors currently predict an adverse outcome after elective thoracoabdominal aortic aneurysm repair. METHODS: A single, composite end point termed adverse outcome was defined as the occurrence of any of the following: death within 30 days, death before discharge from the hospital, paraplegia, paraparesis, stroke, or acute renal failure requiring dialysis. A risk factor analysis was performed using data from 1,108 consecutive elective thoracoabdominal aortic aneurysm repairs. RESULTS: The incidence of an adverse outcome was 13.0% (144 of 1,108 patients); predictors included preoperative renal insufficiency (p = 0.0001), increasing age (p = 0.0035), symptomatic aneurysms (p = 0.020), and extent II aneurysms (p = 0.0001). These risk factors were used to construct an equation that estimates the probability of an adverse outcome for an individual patient. CONCLUSIONS: This new predictive model may assist in decisions regarding elective thoracoabdominal aortic aneurysm operations. For patients who are acceptable candidates, contemporary surgical management provides favorable results.

Adolescent↗

PubMed2005

Predictive models of implicit and explicit attitudes.

Explicit attitudes have long been assumed to be central factors influencing behaviour. A recent stream of studies has shown that implicit attitudes, typically measured with the Implicit Association Test (IAT), can also predict a significant range of behaviours. This contribution is focused on testing different predictive models of implicit and explicit attitudes. In particular, three main models can be derived from the literature: (a) additive (the two types of attitudes explain different portion of variance in the criterion), (b) double dissociation (implicit attitudes predict spontaneous whereas explicit attitudes predict deliberative behaviour), and (c) multiplicative (implicit and explicit attitudes interact in influencing behaviour). This paper reports two studies testing these models. The first study (N = 48) is about smoking behaviour, whereas the second study (N = 109) is about preferences for snacks versus fruit. In the first study, the multiplicative model is supported, whereas the double dissociation model is supported in the second study. The results are discussed in light of the importance of focusing on different patterns of prediction when investigating the directive influence of implicit and explicit attitudes on behaviours.

Adult↗

PubMed1999

Properties of threshold model predictions.

Estimation of genetic parameters and accuracy of threshold model genetic predictions were investigated. Data were simulated for different population structures by using Monte Carlo techniques. Variance components were estimated by using threshold models and linear sire models applied to untransformed data, logarithmically transformed data, and transformation to Snell scores. Effects of number of categories (2, 5, and 10), incidence of categories (extreme, moderate, and normal), heritability in the underlying scale (.04, .20, and .50), and data structure (unbalanced and balanced) on accuracy of genetic prediction were investigated. The real importance of using a threshold model was to estimate genetic parameters. An expected heritability of .20 was estimated to be .22 and .10 by a threshold model and a linear model, respectively. Accuracy increased significantly with a larger number of categories, a more normal distribution of incidences, increased heritability, and more balanced data. Even threshold models were shown to be more efficient with more than two categories (e.g., binomial). Transformation of scale did not accomplish the purpose intended.

Animal Husbandry↗

PubMed1998

Evaluation of two outcome prediction models on an independent database.

OBJECTIVE: To evaluate the performance of the New Simplified Acute Physiology Score (SAPS II) and the admission Mortality Probability Model (MPM0) in a large independent database, using formal statistical assessment. DESIGN: Analysis of the database of a multicenter, multinational, prospective cohort study, EURICUS-I. SETTING: Eighty nine intensive care units (ICUs) from 13 European areas. PATIENTS: Data of 16,060 patients consecutively admitted to the participating ICUs were collected during a period of 4 months. Following the original SAPS II and MPM0 criteria, the analysis excluded: patients <18 ys of age; readmissions; patients admitted with acute myocardial infarction; burns; and patients in the postoperative period after coronary artery bypass surgery. All patients with a length of stay <8 hrs were excluded from the study to keep comparability between both systems. A total of 10,027 patients were analyzed. INTERVENTIONS: Collection of the first 24 hrs' admission data necessary for the calculation of SAPS II and MPM0 and basic demographic statistics. Vital status at discharge from the hospital was registered. MEASUREMENTS AND MAIN RESULTS: Despite having a good discriminative capability, as measured by the area under the receiver operating characteristic (ROC) curves (SAPS II: ROC = 0.822 +/- 0.005 SEM; MPM0: ROC = 0.785 +/- 0.006 SEM), both models presented poor calibration, with significant differences between observed and predicted mortality (Hosmer-Lemeshow goodness-of-fit tests H and C, p < .0001). Both SAPS II (predicted risk >40%) and MPM0 (predicted risk >30%) overestimated the risk of death. The evaluation of the uniformity of fit of SAPS II and MPM0 demonstrated large variations across the various subgroups of patients. CONCLUSIONS: The original SAPS II and MPM0 models did not accurately predict mortality on an independent large international multicenter ICU patient database. Results of studies utilizing general outcome prediction models without previous validation in the target population should be interpreted with prudence.

APACHE↗

PubMed2005

Predictive modeling of microorganisms: LAG and LIP in monotonic growth.

The variety of models that are currently being used in "Predictive Microbiology" or "Microbial Ecology" aiming at reproducing the growth curve of microorganisms motivates this study. It is widely agreed that no model can reproduce generically and consistently the "LAG Phase" of microorganism growth. To promote the objective of "predictive modeling", we present here a model that was derived from first biological and physical principles, which is shown to reproduce qualitatively as well as quantitatively all typical features captured experimentally in microorganism growth. In particular, this paper focuses on capturing and controlling of the "LAG Phase" a typical phase in microorganisms growth, at the initial growth stages, as well as the inflection point on the "ln curve" of the cell concentration, i.e. a Logarithmic Inflection Point referred here as "LIP". The proposed model also captures the Logistic Growth curve as a special case. Comparison of the solutions obtained from the proposed model with experimental data confirms its quantitative validity, as well as its ability to recover a wide range of qualitative features captured in experiments.

Bacteria↗

PubMed2005

Simple clinical variables predict liver histology in hepatitis C: prospective validation of a clinical prediction model.

OBJECTIVE: A recent single-center multivariate analysis of hepatitis C (HCV) patients showed that having any two criteria: 1) ferritin > or =200 microg/l and 2) spider nevi and/or albumin < or = 35 g/l predicted grade 2 or greater histological inflammation; the presence of any two of the following criteria: spider nevi, platelets < or =150 x 109/l, palpable splenomegaly and/or albumin < or =35 g/l predicted stage 2 or greater histological fibrosis. Absence of predictors also predicted a lack of inflammation and fibrosis. Our aim was prospectively to validate this clinical prediction model using an independent multicenter sample. MATERIAL AND METHODS: Eighty-one patients with previously untreated active chronic HCV underwent physical examination, laboratory investigation, and liver biopsy. Biopsies were read, in blinded fashion, by a single pathologist, using a modified Hytiroglou (1995) scale. The clinical scoring system was correlated with histology; likelihood ratios (LRs), Fisher's exact p-values, and receiver operating characteristics (ROCs) were calculated. RESULTS: Data recording was complete in 77 and 38 patients regarding fibrotic stage and inflammatory grade, respectively. For fibrosis, 3/3 patients with any three criteria (LR 17, positive predictive value (PPV) 100%), 4/5 patients with any two criteria (LR 5.1), and 15/47 with no criteria (LR 0.6, negative predictive value (NPV) 68%) had stage 2 or greater fibrosis on biopsy (p=0.01). For inflammation, 5/5 patients with both criteria (LR 15, PPV 100%), and 8/19 patients with no criteria (LR 0.5, NPV 58%) had moderate-severe inflammation on liver biopsy (p=0.036). When missing variables were assumed to be normal, recalculated LRs were almost identical. An alanine aminotransferase (ALAT) level <60 U/l may increase the NPVs. CONCLUSIONS: This independent multicenter data set has validated our published model which uses simple clinical variables accurately and significantly to predict hepatic fibrosis and inflammation in HCV patients.

Adult↗

PubMedDate not supplied

A survival predictive model in patients undergoing radical resection of ampullary adenocarcinoma.

BACKGROUND/AIMS: Radical resection with either pancreaticoduodenectomy or pylorus-preserving pancreaticoduodenectomy is considered to be the standard treatment for most ampullary carcinomas, but the prognostic predictive model has not yet been developed. METHODOLOGY: The pretreatment, treatment, and follow-up variables of data of 47 patients undergoing radical resection for the ampullary carcinoma were analyzed to determine the favorable prognostic variables. Employing the Kaplan-Meier method, the cumulative survival rates of the ampullary carcinoma were calculated. By Cox regression model, a stepwise multivariate analysis was performed to analyze the contributing factors of the survival rate, and a predictive survival equation was obtained. RESULTS: With the results of the univariate analysis, the variables significantly associated with favorable prognosis were younger age (<63 years), TNM stage (stage I or II or III), and the degree of tumor differentiation (well or moderately differentiated). When the above three variables were examined as covariates by Cox regression in multivariate analysis, the TNM stage and the degree of tumor differentiation were independently correlated with the survival. A predictive survival equation obtained with the beta-coefficients of the above three variables was as follows: S (t) = [So (t)] P, P = exp (0.0234 x age - 1.8744 x tumor differentiation + 1.1576 x TNM stage) CONCLUSIONS: This predictive survival equation can predict the survival and the favorable outcome of patients treated with radical resection of ampullary carcinoma.

Actuarial Analysis↗