PubMed Health⌕ Search

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 1,729 records · Page 96Linked to original sources
PubMed2002

New bioimpedance model accurately predicts lower limb muscle volume: validation by magnetic resonance imaging.

Conventional bioimpedance analysis (BIA) methods now simplify the representation of lower limb geometry and electrical properties for body composition estimation. In the present study, a three-dimensional model of the lower limb was assembled by segmentation of magnetic resonance cross-sectional images (MRI) for adipose tissue, skeletal muscle, and bone. An electrical network was then associated with this model. BIA and MRI measurements were made in six lean subjects (3 men and 3 women, age 32.2 +/- 6.9 yr). Assuming 0.85 S/m for the longitudinal conductivity of the muscle, the model predicted in the examined subjects an impedance profile that conformed well to the BIA impedance profile; predicted and measured resistances were similar (261.3 +/- 7.7 vs. 249 +/- 9 Omega; P = not significant). The resistance profile provided, through a simpler model, muscle area estimates along the lower limb and total leg muscle volume (mean 4,534 cm(3) for men and 4,071 cm(3) for women) with a mean of the absolute value of relative error with respect to MRI of 6.2 +/- 3.9. The new approach suggests that BIA can reasonably estimate the distribution and volume of muscles in the lower extremities of lean subjects.

Adipose Tissue↗
PubMed2001

Microbial growth modelling with artificial neural networks.

There is a growing interest in modelling microbial growth as an alternative to time-consuming, traditional, microbiological enumeration techniques. Several statistical models have been reported to describe the growth of different microorganisms, but there are accuracy problems. An alternate technique 'artificial neural networks' (ANN) for modelling microbial growth is explained and evaluated. Published data were used to build separate general regression neural network (GRNN) structures for modelling growth of Aeromonas hydrophila, Shigella flexneri, and Brochothrix thermosphacta. Both GRNN and published statistical model predictions were compared against the experimental data using six statistical indices. For training data sets, the GRNN predictions were far superior than the statistical model predictions, whereas the GRNN predictions were similar or slightly worse than statistical model predictions for test data sets for all the three data sets. GRNN predictions can be considered good, considering its performance for unseen data. Graphical plots, mean relative percentage residual, mean absolute relative residual, and root mean squared residual were identified as suitable indices for comparing competing models. ANN can now become a vehicle whereby predictive microbiology can be applied in food product development and food safety risk assessment.

Aeromonas hydrophila↗
PubMed2002

A prospective validated model for predicting axillary node metastases based on 2,000 sentinel node procedures: the role of tumour location [corrected].

AIMS: The purpose was to identify the independent predictive factors of axillary lymph-node metastases (ALNM) in infiltrating ductal carcinoma (IFDC) and to create a prospective, validated statistical model to predict the likelihood of ALNM in patients in the present era of sentinel lymph-node (SLN) biopsy and enhanced histopathology. METHODS: Univariate and multivariate analyses of 13 clinicopathological variables (including tumour location) were performed to determine predictors of ALNM in 1659 eligible SLN biopsy procedures. A logistic regression model was developed and then prospectively validated on a second population of 187 subsequent consecutive procedures. RESULTS: Age, pathological tumour size, palpability, lymphovascular invasion (LVI), histological grade, nuclear grade, ductal histological subtype, tumour location (quadrant) and multifocality were associated with ALNM in univariate analyses (P < 0.001). Of these, only palpability and histological grade were not statistically associated with ALNM in the multivariate analysis (P> 0.05). The frequency of ALNM in upper-inner-quadrant (UIQ) tumours was 20.6%, compared with 33.2% for all other quadrants (P<0.0005). There was no statistical difference between UIQ and other-quadrant tumours in any clinicopathological variables analysed. The logistic regression model, developed based on the population of 1659, had the same accuracy, sensitivity, specificity, positive predictive value and negative predictive value when applied prospectively to the second population. CONCLUSION: Tumour size, LVI, age, nuclear grade, histological subtype, multifocality and location in the breast were independent predictive factors for ALNM in IFDC. ALNM is less frequent in UIQ tumours than in other-quadrant tumours. Our prospectively validated predictive model could be valuable in pre-operative patient discussions, although staging of the axilla in the individual patient remains necessary.

Adult↗
PubMed2000

A simple model for predicting the free energy of binding between anthracycline antibiotics and DNA.

A theoretical model for predicting the free energy of binding between anthracycline antibiotics and DNA was developed using the electron density functional (DFT) and molecular mechanics (MM) methods. Partial DFT-ESP charges were used in calculating the MM binding energies for complexes formed between anthracycline antibiotics and oligodeoxynucleotides. These energies were then compared with experimental binding free energies. The good correlation between the experimental and theoretical energies allowed us to propose a model for predicting the binding free energy for derivatives of anthracycline antibiotics and for quickly screening new anthracycline derivatives.

Antibiotics, Antineoplastic↗
PubMed2004

Development of a novel approach for secondary modelling in predictive microbiology: incorporation of microbiological knowledge in black box polynomial modelling.

This research deals with the development of a novel secondary modelling procedure within the framework of predictive microbiology. The procedure consists of three steps: (i) careful formulation of the available microbiological information, both from literature and from the experimental case study at hand, (ii) translation of these requirements in mathematical terms under the form of partial derivatives throughout the complete interpolation region of the experimental design, and (iii) determination of parameter values with suitable optimisation techniques for a flexible black box modelling approach, e.g., a polynomial model or an artificial neural network model. As a vehicle for this procedure, the description of the maximum specific growth rate of Lactobacillus sakei in modified BHI-broth as influenced by suboptimal temperature, water activity, sodium lactate and dissolved carbon dioxide concentration is under study. The procedure results in a constrained polynomial model with excellent descriptive and interpolating features in comparison with an extended Ratkowsky-type model and classical polynomial model, by combining specific properties of both model types. The developed procedure is illustrated on the description of the lag phase as well. It is stressed how the confrontation with experimental data is very important to appreciate the descriptive and interpolating capacities of new or existing models, which is nowadays not always carefully performed. Alternatively, the first two steps of the novel procedure can be used as a tool to demonstrate clearly (possible) interpolative shortcomings of an existing model with straightforward spreadsheet calculations.

Carbon Dioxide↗
PubMed2003

A biokinetic model for predicting the retention of 3H in the human body after intakes of tritiated water.

Scarce published data on the long-term excretion of tritiated water from the human body have been re-evaluated in order to develop a biokinetic model describing the retention in the human body of 3H from tritiated water (HTO) that could be used for both prospective and retrospective radiation protection. A three-component exponential function is proposed to describe the elimination of 3H from HTO with biological half-times of 10 d (99.00%), 40 d (0.98%) and 350 d (0.02%) respectively. The model predicts a committed effective dose of 1.7 x 10(-11) Sv Bq(-1), comparable with that of the current ICRP Publication 56 and 72 models, and estimates the retention of 3H to within a factor of about 2 of the measured values up to 40 d after intake and about 5 at times longer than 100 d. The derivation of the model and the uncertainties associated with the various parameters are discussed.

Absorption↗
PubMed2005

Robust classification modeling on microarray data using misclassification penalized posterior.

MOTIVATION: Genome-wide microarray data are often used in challenging classification problems of clinically relevant subtypes of human diseases. However, the identification of a parsimonious robust prediction model that performs consistently well on future independent data has not been successful due to the biased model selection from an extremely large number of candidate models during the classification model search and construction. Furthermore, common criteria of prediction model performance, such as classification error rates, do not provide a sensitive measure for evaluating performance of such astronomic competing models. Also, even though several different classification approaches have been utilized to tackle such classification problems, no direct comparison on these methods have been made. RESULTS: We introduce a novel measure for assessing the performance of a prediction model, the misclassification-penalized posterior (MiPP), the sum of the posterior classification probabilities penalized by the number of incorrectly classified samples. Using MiPP, we implement a forward step-wise cross-validated procedure to find our optimal prediction models with different numbers of features on a training set. Our final robust classification model and its dimension are determined based on a completely independent test dataset. This MiPP-based classification modeling approach enables us to identify the most parsimonious robust prediction models only with two or three features on well-known microarray datasets. These models show superior performance to other models in the literature that often have more than 40-100 features in their model construction. AVAILABILITY: Our MiPP software program is available at the Bioconductor website (http://www.bioconductor.org).

Algorithms↗
PubMed2026

RP3Net: a deep learning model for predicting recombinant protein production in Escherichia coli.

MOTIVATION: Recombinant protein expression can be a limiting step in the production of protein reagents for drug discovery and other biotechnology applications. We introduce RP3Net (Recombinant Protein Production Prediction Network), an AI model of small-scale heterologous soluble protein expression in Escherichia coli. RP3Net utilizes the most recent protein and genomic foundational models. A curated dataset of internal experimental results from AstraZeneca and publicly available data from the Structural Genomics Consortium was used for training, validation and testing of RP3Net. RESULTS: RP3Net achieves an increase in area under the receiver operator curve (AUROC) of 0.15, compared to a baseline model. When experimentally validated on an independent, prospective, manually selected set of 97 constructs, RP3Net outperformed currently available models, with an AUROC of 0.83, delivering accurate predictions in 77% of the cases, and correctly identifying successfully expressing constructs in 92% of cases. AVAILABILITY AND IMPLEMENTATION: The model, along with installation and running instructions, is available under an MIT licence at https://github.com/RP3Net/RP3Net, DOI 10.5281/zenodo.17243498.

Escherichia coli↗
PubMed1997

Efficiency optimization of the selection period in therapeutic trials.

To determine eligibility for a (randomized) clinical trial, measuring the inclusion and exclusion criteria can be extended over a period of time. During this period, known as the selection period, a patient is repeatedly examined at certain time intervals. This study describes an approach for optimizing the efficiency of the selection period. Efficiency is defined as the costs of randomizing one patient. The objective is to construct prediction models based on data obtained early in the selection period to predict subsequent exclusions. A prediction model increases the efficiency if after its application the costs per randomization are lower. The approach is illustrated using data from the selection period of the Rotterdam Cardiovascular Risk Intervention (ROCARI) trial which was composed of five consecutive patient visits. At each visit, data to determine eligibility was obtained. We found that logistic regression models based on data of the first and second visit could predict exclusions during the third visit. Application of the prediction models suggested that in this particular trial the costs per randomization would decrease by $52. As the initial costs per randomization were $1444, there would be a 3.6% (52/1444) savings in recruitment costs under the prediction models, accounting for a savings of more than $450,000. We conclude that the use of data obtained early in a selection period can predict subsequent exclusions, and therefore could increase the efficiency of such a period. The approach could be applied to data obtained in a pilot study as well as data obtained in the beginning of a prolonged intake period.

Cost Savings↗
PubMed2006

Closed-loop control of mean arterial blood pressure during surgery with alfentanil: clinical evaluation of a novel model-based predictive controller.

BACKGROUND: In contrast to hypnosis, there is no surrogate parameter for analgesia in anesthetized patients. Opioids are titrated to suppress blood pressure response to noxious stimulation. The authors evaluated a novel model predictive controller for closed-loop administration of alfentanil using mean arterial blood pressure and predicted plasma alfentanil concentration (Cp Alf) as input parameters. METHODS: The authors studied 13 healthy patients scheduled to undergo minor lumbar and cervical spine surgery. After induction with propofol, alfentanil, and mivacurium and tracheal intubation, isoflurane was titrated to maintain the Bispectral Index at 55 (+/- 5), and the alfentanil administration was switched from manual to closed-loop control. The controller adjusted the alfentanil infusion rate to maintain the mean arterial blood pressure near the set-point (70 mmHg) while minimizing the Cp Alf toward the set-point plasma alfentanil concentration (Cp Alfref) (100 ng/ml). RESULTS: Two patients were excluded because of loss of arterial pressure signal and protocol violation. The alfentanil infusion was closed-loop controlled for a mean (SD) of 98.9 (1.5)% of presurgery time and 95.5 (4.3)% of surgery time. The mean (SD) end-tidal isoflurane concentrations were 0.78 (0.1) and 0.86 (0.1) vol%, the Cp Alf values were 122 (35) and 181 (58) ng/ml, and the Bispectral Index values were 51 (9) and 52 (4) before surgery and during surgery, respectively. The mean (SD) absolute deviations of mean arterial blood pressure were 7.6 (2.6) and 10.0 (4.2) mmHg (P = 0.262), and the median performance error, median absolute performance error, and wobble were 4.2 (6.2) and 8.8 (9.4)% (P = 0.002), 7.9 (3.8) and 11.8 (6.3)% (P = 0.129), and 14.5 (8.4) and 5.7 (1.2)% (P = 0.002) before surgery and during surgery, respectively. A post hoc simulation showed that the Cp Alfref decreased the predicted Cp Alf compared with mean arterial blood pressure alone. CONCLUSION: The authors' controller has a similar set-point precision as previous hypnotic controllers and provides adequate alfentanil dosing during surgery. It may help to standardize opioid dosing in research and may be a further step toward a multiple input-multiple output controller.

Adult↗
PubMed2005

[Modeling for prediction of arabinoxylans solubilization and endo-xylanase random attacking during mashing].

A model describing the solubilization of arabinoxylans and degradation by endo-xylanase random attacking during mashing was developed. The model was expected to predict the arabinoxylans concentration in wort at the settings of different initial value and mashing parameters for diminishing the negative effects of arabinoxylans on brewing. Results showed that the modeling errors range for the final concentration of arabinoxylans in wort was -9.5% to +13.6%. The model prediction accuracy for industrial scale mashing process was lower than that in laboratory scale. The errors were given 16.8% and 17.9%, respectively. The simulation results showed that arabinoxylans concentration was increased with the increase of mashing-in temperature, but it was decreased with prolonging the mashing-in time. The effect of initial arabinoxylans in malt on arabinoxylans concentration in wort was more remarkable than that of endo-xylanase activity in grist.

Endo-1,4-beta Xylanases↗
PubMed2006

Results of intensive chemotherapy in 998 patients age 65 years or older with acute myeloid leukemia or high-risk myelodysplastic syndrome: predictive prognostic models for outcome.

BACKGROUND: Elderly patients (age > or = 65 years) with acute myeloid leukemia (AML) generally have a poor prognosis. AML-type therapy results are often derived from studies in younger patients and may not apply to elderly AML. Many investigators and oncologists advocate, at times, only supportive care or frontline single agents, Phase I-II studies, low-intensity regimens, or 'targeted' therapies. However, baseline expectations for outcomes of elderly AML with 'standard' AML-type therapy are not well defined. The aim was to develop prognostic models for complete response (CR), induction (8-week) mortality, and survival rates in elderly AML, which would be used to advise oncologists and patients of expectations with standard AML type therapy, and to establish baseline therapy results against which novel strategies would be evaluated. METHODS: A total of 998 patients age > or = 65 years with AML or high-risk myelodysplastic syndrome (> 10% blasts) treated with intensive chemotherapy between 1980 and 2004 were analyzed. Univariate and multivariate analyses of prognostic factors associated with CR, induction (8-week) mortality, and survival used standard methods. RESULTS: The overall CR rate was 45% and induction mortality 29%. Multivariate analysis of prognostic factors identified consistent independent poor prognostic factors for CR, 8-week mortality, and survival. These included age > or = 75 years, unfavorable karyotypes (often complex), poor performance (3-4 ECOG [Eastern Cooperative Oncology Group]), longer duration of antecedent hematologic disorder, treatment outside the laminar airflow room, and abnormal organ functions. Patients could be divided into: 1) a favorable group (about 20% of patients) with expected CR rates above 60%, induction mortality rates of 10%, and 1-year survival rates above 50%; 2) an intermediate group (about 50-55% of patients) with expected CR rates of 50%, induction mortality rates of 30%, and 1-year survival rates of 30%; and 3) an unfavorable risk group (about 25-30% of patients) with expected CR rates of less than 20%, induction mortality rates above 50%, and 1-year survival rates of less than 10%. CONCLUSIONS: Prognostic models, based on standard readily available baseline characteristics, were developed for elderly patients with AML, which may assist in therapeutic and investigational decisions. These predictive models, based on a retrospective analysis, will require validation in independent study groups.

Age Factors↗
PubMed2003

Predicting structural models for silicon clusters.

This article introduces an efficient method to generate structural models for medium-sized silicon clusters. Geometrical information obtained from previous investigations of small clusters is initially sorted and then introduced into our predictor algorithm in order to generate structural models for large clusters. The method predicts geometries whose binding energies are close (95%) to the corresponding value for the ground-state with very low computational cost. These predictions can be used as a very good initial guess for any global optimization algorithm. As a test case, information from clusters up to 14 atoms was used to predict good models for silicon clusters up to 20 atoms. We believe that the new algorithm may enhance the performance of most optimization methods whenever some previous information is available.

Journal Article↗
PubMed2003

[Aortic valve replacement in patients over 70 years old: determinants of early death].

INTRODUCTION AND OBJECTIVES: There is controversy regarding the risk factors associated with early death in geriatric patients undergoing aortic valve replacement. We analyzed the risks in these patients and established an accurate model for predicting in-hospital mortality. PATIENTS AND METHOD: Univariate and multivariate analyses were made of the risk factors associated with early death in a group of 129 patients older than 70 years who underwent aortic valve replacement (May 1994-June 2001). The variables obtained by multivariate logistic regression were combined to produce an equation for the prediction of early death. The equation was tested using a receiver operating characteristic curve. RESULTS: Univariate analysis identified four factors related to early death: NYHA III-IV (p < 0.0001), ejection fraction < 40% (p < 0.05), aortic regurgitation (p < 0.05) and high left ventricular mass index (p < 0.05). Multivariate analysis revealed three independent risk factors: NYHA III-IV (p < 0.01), aortic regurgitation (p < 0.05), and small body surface area (p < 0.05). A lower mortality was observed in patients with a larger body surface area (0% for > 1.90 m2, 20% for < 1.40 m2). The estimated mortality with the predictive model was 7.06%, which was similar to the observed mortality of 7.80% (area under the ROC curve 0.87) and better than estimates obtained with the EuroSCORE (6.5%; area under the ROC curve 0.56). CONCLUSIONS: Risk factors associated with early death after aortic valve replacement in geriatric patients include functional status, aortic regurgitation, and small body surface area. Our model based on these factors accurately predicted operative mortality in our patients. Gender, prosthesis size, and pump time were not identified as risk factors.

Aged↗
PubMed1991

A model to predict canine pelvic limb musculoskeletal geometry.

We developed a model to predict the three-dimensional canine pelvic limb muscular geometry (i.e., all muscle moment arms during any instant in gait). Forty-one muscle origins and insertions, as well as external landmarks (to obtain anthropometric dimensions) were marked on both pelvic limbs of five dogs and digitized on biplanar radiographs. Reference frames in the pelvis, femur, and tibia established the three-dimensional coordinates of each origin, insertion, and landmark. A set of dimensionless 'scaled coordinates' was created by dividing the actual origin and insertion coordinates by selected anthropometric dimensions of each animal. Combining scaled coordinates from all ten limbs produced an averaged 'template' of scaled coordinates. To provide limited validation of the scaling procedure, we measured the anthropometric dimensions between externally palpable landmarks of two additional pelvic limbs. The anthropometric dimensions were multiplied by the averaged template coordinates to calculate two new sets of hindlimb muscle coordinates within the three bony reference frames. The two limbs then were dissected, muscle endpoints were marked, and biplanar radiographs of each of the limb segments were digitized. The actual coordinates so obtained were similar to those predicted by the template and anthropometric measures.

Animals↗
PubMed1997

Assessment of a remote sensing-based model for predicting malaria transmission risk in villages of Chiapas, Mexico.

A blind test of two remote sensing-based models for predicting adult populations of Anopheles albimanus in villages, an indicator of malaria transmission risk, was conducted in southern Chiapas, Mexico. One model was developed using a discriminant analysis approach, while the other was based on regression analysis. The models were developed in 1992 for an area around Tapachula, Chiapas, using Landsat Thematic Mapper (TM) satellite data and geographic information system functions. Using two remotely sensed landscape elements, the discriminant model was able to successfully distinguish between villages with high and low An. albimanus abundance with an overall accuracy of 90%. To test the predictive capability of the models, multitemporal TM data were used to generate a landscape map of the Huixtla area, northwest of Tapachula, where the models were used to predict risk for 40 villages. The resulting predictions were not disclosed until the end of the test. Independently, An. albimanus abundance data were collected in the 40 randomly selected villages for which the predictions had been made. These data were subsequently used to assess the models' accuracies. The discriminant model accurately predicted 79% of the high-abundance villages and 50% of the low-abundance villages, for an overall accuracy of 70%. The regression model correctly identified seven of the 10 villages with the highest mosquito abundance. This test demonstrated that remote sensing-based models generated for one area can be used successfully in another, comparable area.

Animals↗
PubMed2000

A predictive risk model for outcomes of ischemic stroke.

BACKGROUND AND PURPOSE: The great variability of outcome seen in stroke patients has led to an interest in identifying predictors of outcome. The combination of clinical and imaging variables as predictors of stroke outcome in a multivariable risk adjustment model may be more powerful than either alone. The purpose of this study was to determine the multivariable relationship between infarct volume, 6 clinical variables, and 3-month outcomes in ischemic stroke patients. METHODS: Included in the study were 256 eligible patients from the Randomized Trial of Tirilazad Mesylate in Acute Stroke (RANTTAS). Six clinical variables and 1-week infarct volume were the prespecified predictor variables. The National Institutes of Health Stroke Scale, Barthel Index, and Glasgow Outcome Scale were the outcomes. Multivariable logistic regression techniques were used to develop the model equations, and bootstrap techniques were used for internal validation. Predictive performance of the models was assessed for discrimination with receiver operator characteristic (ROC) curves and for calibration with calibration curves. RESULTS: The predictive models had areas under the ROC curve of 0.79 to 0.88 and demonstrated nearly ideal calibration curves. The areas under the ROC curves were statistically greater (P<0.001) with both clinical and imaging information combined than with either alone for predicting excellent recovery and death or severe disability. CONCLUSIONS: Combined clinical and imaging variables are predictive of 3-month outcome in ischemic stroke patients. Demonstration of this relationship with acute clinical variables and 1-week infarct information supports future attempts to predict 3-month outcome with all acute variables.

Aged↗