SEARCH · PubMed Health

Results for “Predictive modeling”

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 271 records · Page 15Linked to original sources

PubMed2005

Meta-analysis of food safety information based on a combination of a relational database and a predictive modeling tool.

The management of microbial risk in food products requires the ability to predict growth kinetics of pathogenic microorganisms in the event of contamination and growth initiation. Useful data for assessing these issues may be found in the literature or from experimental results. However, the large number and variety of data make further development difficult. Statistical techniques, such as meta-analysis, are then useful to realize synthesis of a set of distinct but similar experiences. Moreover, predictive modeling tools can be employed to complete the analysis and help the food safety manager to interpret the data. In this article, a protocol to perform a meta-analysis of the outcome of a relational database, associated with quantitative microbiology models, is presented. The methodology is illustrated with the effect of temperature on pathogenic Escherichia coli and Listeria monocytogenes, growing in culture medium, beef meat, and milk products. Using a database and predictive models, simulations of growth in a given product subjected to various temperature scenarios can be produced. It is then possible to compare food products for a given microorganism, according to its growth ability in these products, and to compare the behavior of bacteria in a given foodstuff. These results can assist decisions for a variety of questions on food safety.

Databases, Factual↗

PubMed2006

[Developing the predictive model for the group at high risk for colon cancer].

OBJECTIVES: We developed the predictive model for the incidence of colon cancer by utilizing the health screening data of the National Health Insurance in Korea. We also explored the characteristics of the high risk group for colon cancer. METHODS: The predictive model was used to determine those people who have a high risk for colon cancer within 2 years of their NHI health screening, and we excluded the people who had already been treated for cancer or who were cancer patient. The study population is the insured of the NHI, aged 40 or over and they had undergone health screening from the year 2000 to 2004, according to NHI health screening formula. We performed logistic regression analysis and used SAS Enterprise Miner 4.1. RESULTS: This study shows that there exists a higher rate of colon cancer in males than females. Also, for the population in their 60s, the incidence rate of colon cancer is much higher by 5.36 times than that for those people in their 40s. Amongst the behavioral factors, heavy drinking is the most important determinant of the colon cancer incidence (7.39 times in males and 21.51 times in females). CONCLUSIONS: Our study confirms that the major influencing factors for the incidence of colon cancer are drinking, lack of exercise, a medical history of colon polypus and a family history of colon cancer. As a result, we can choose the group that is at a high risk for colon cancer and provide customized medical information and selective management services according to their characteristics.

Adult↗

PubMed2000

The cardiovascular event reduction tool (CERT)--a simplified cardiac risk prediction model developed from the West of Scotland Coronary Prevention Study (WOSCOPS).

The clinical decision to treat hypercholesterolemia is premised on an awareness of patient risk, and cardiac risk prediction models offer a practical means of determining such risk. However, these models are based on observational cohorts where estimates of the treatment benefit are largely inferred. The West of Scotland Coronary Prevention Study (WOSCOPS) provides an opportunity to develop a risk-benefit prediction model from the actual observed primary event reduction seen in the trial. Five-year Cox model risk estimates were derived from all WOSCOPS subjects (n = 6,595 men, aged 45 to 64 years old at baseline) using factors previously shown to be predictive of definite fatal coronary heart disease or nonfatal myocardial infarction. Model risk factors included age, diastolic blood pressure, total cholesterol/ high-density lipoprotein ratio (TC/HDL), current smoking, diabetes, family history of fatal coronary heart disease, nitrate use or angina, and treatment (placebo/ 40-mg pravastatin). All risk factors were expressed as categorical variables to facilitate risk assessment. Risk estimates were incorporated into a simple, hand-held slide rule or risk tool. Risk estimates were identified for 5-year age bands (45 to 65 years), 4 categories of TC/HDL ratio (<5.5, 5.5 to <6.5, 6.5 to <7.5, > or = 7.5), 2 levels of diastolic blood pressure (<90, > or = 90 mm Hg), from 0 to 3 additional risk factors (current smoking, diabetes, family history of premature fatal coronary heart disease, nitrate use or angina), and pravastatin treatment. Five-year risk estimates ranged from 2% in very low-risk subjects to 61% in the very high-risk subjects. Risk reduction due to pravastatin treatment averaged 31%. Thus, the Cardiovascular Event Reduction Tool (CERT) is a risk prediction model derived from the WOSCOPS trial. Its use will help physicians identify patients who will benefit from cholesterol reduction.

Age Factors↗

PubMedDate not supplied

Decision curve analysis: a novel method for evaluating prediction models.

BACKGROUND: Diagnostic and prognostic models are typically evaluated with measures of accuracy that do not address clinical consequences. Decision-analytic techniques allow assessment of clinical outcomes but often require collection of additional information and may be cumbersome to apply to models that yield a continuous result. The authors sought a method for evaluating and comparing prediction models that incorporates clinical consequences,requires only the data set on which the models are tested,and can be applied to models that have either continuous or dichotomous results. METHOD: The authors describe decision curve analysis, a simple, novel method of evaluating predictive models. They start by assuming that the threshold probability of a disease or event at which a patient would opt for treatment is informative of how the patient weighs the relative harms of a false-positive and a false-negative prediction. This theoretical relationship is then used to derive the net benefit of the model across different threshold probabilities. Plotting net benefit against threshold probability yields the "decision curve." The authors apply the method to models for the prediction of seminal vesicle invasion in prostate cancer patients. Decision curve analysis identified the range of threshold probabilities in which a model was of value, the magnitude of benefit, and which of several models was optimal. CONCLUSION: Decision curve analysis is a suitable method for evaluating alternative diagnostic and prognostic strategies that has advantages over other commonly used measures and techniques.

Decision Support Techniques↗

PubMed2005

Applicability of clinical prediction models in acute myocardial infarction: a comparison of traditional and empirical Bayes adjustment methods.

INTRODUCTION: Several clinical prediction models have been developed to predict outcome after acute myocardial infarction. Updating to local circumstances may be required to make such models better applicable. We aimed to compare traditional and empirical Bayes (EB) methods to perform such updating. METHODS: We focused on 16 geographical regions within the GUSTO-I trial, which included 40,830 patients with acute myocardial infarction; of whom, 2851 (7.0%) had died by 30 days. Differences in mortality between regions were studied with traditional adjustment for case mix in logistic regression models and with EB methods. These methods updated predictions for new patients while accounting for the uncertainty in the traditionally estimated mortality differences. RESULTS: The case mix in the regions differed with respect to important predictive characteristics such as age, presence of shock, and anterior infarct location (all P < .001). These differences did not explain regional differences in 30-day mortality, which varied between 80% and 120% with traditional analyses (P < .01). The EB estimates for regional differences were much smaller (between 93% and 107%). CONCLUSIONS: Statistically significant differences in case mix and 30-day mortality were noted between geographical regions. The practical implications of this heterogeneity were, however, limited when model predictions were updated with EB methods.

Bayes Theorem↗

PubMed2004

Predictive modelling of growth and measurement of enzymatic synthesis and activity by a cocktail of Brochothrix thermosphacta.

The possibility was examined of developing a predictive model that combined microbial growth (increase in cellular number) and extracellular enzyme activity of a cocktail of three strains of Brochothrix thermosphacta. Estimations of growth and enzyme activity were made within a three-dimensional matrix of conditions: temperature 2-20 degrees C, pH value 4.0-7.5 and water activity (a(w)) 0.95-0.995. A model which predicted growth based on increases in cell number was constructed. No extracellular lipases were detected, but slight proteolytic reactions were observed. Although it was not possible to model protease activity, the growth model and information relating to enzyme activity will be made freely available in a database on the Internet.

Colony Count, Microbial↗

PubMed1991

Exploration of prediction models for caries risk assessment of the geriatric population.

The purpose of this study was to demonstrate a method for establishing a model designed to predict the caries risk of elderly individuals. Twenty-three patients over the age of 65 and living in a semi-independent retirement center were examined; several variables were collected and analyzed for their ability to predict the development of new carious lesions. The analysis was performed using logistic regression where the proportion of new decay was used as the dependent variable. The model for prediction of combined coronal and root caries included the variables flow rate, gender, and root caries index. The proposed method has the advantages of easily collected data, individualized criteria, and the ability to order patients as to the relative risk of developing decay.

Aged↗

PubMed1996

Clinical variables associated with colorectal cancer on colonoscopy: a prediction model.

OBJECTIVES: We prospectively evaluated the ability of various indications for colonoscopy to predict colorectal cancer. METHODS: Indications and findings were prospectively recorded for 1223 consecutive colonoscopies performed during an 18-month period at a University Hospital. Colonoscopies performed on 981 patients for indications which included colorectal cancer in the differential diagnosis were included in the study. A group of 653 patients was randomly selected to derive a model predictive of colorectal cancer at colonoscopy; the remaining 328 patients were used to validate the model. RESULTS: Colorectal cancer was found in 44/981 patients (4.5%). Univariate analysis of the derivation set showed that age > 55, occult bleed, anemia, iron deficiency, weight loss, and abnormal CT scan were associated with finding colorectal cancer at colonoscopy in the derivation set, p < 0.05. History of polyps was negatively associated with finding colorectal cancer. Stepwise selection was used to develop a predictive model by using three independent variables-age > 55, iron deficiency, and weight loss. Assigning a score of 1 to each variable, cancer was present in 0% [95% confidence intervals (CI), 0-0.6%], 5.3% (95% CI, 2.3-8.3%), and 17.9% (95% CI, 4.8-31%) of patients in the validation sample with a score of 0, 1, and > or = 2, respectively (p < 0.001 by chi2 for trend). The model was predictive of finding colorectal cancer at colonoscopy, independent of the location or stage of the cancer. CONCLUSIONS: Age, iron deficiency, and weight loss are important independent predictors of colorectal cancer in patients referred for colonoscopy.

Aged↗

PubMed1999

Multivariate predictive models for group A beta-hemolytic streptococcal pharyngitis in children.

OBJECTIVES: To create predictive models for the clinical diagnosis of group A beta-hemolytic streptococcal (GABHS) pharyngitis in children. METHODS: Patients aged 6 months to 18 years presenting to a pediatric ED with suspected GABHS pharyngitis were prospectively enrolled in the study. Clinicians recorded pertinent clinical information using a standardized form and obtained a throat swab to culture GABHS using a reference standard method. Twelve demographic and clinical features of patients with positive throat cultures were compared with the features of patients with negative throat cultures. Significantly different features were entered in a stepwise logistic regression analysis to create predictive models for the diagnosis. RESULTS: Eighty-five patients (29%) were culture-positive and 212 (71%) were culture-negative for GABHS. Respective mean ages were 6.2 years and 6.1 years in the two groups. Univariate chi-square analysis of the 12 features identified six variables that were significantly associated with GABHS. All significant features were initially included in a stepwise logistic regression analysis. In model I, four independent variables were identified: moderate to severe presentation of tonsillar swelling, moderate to severe tenderness and enlargement of cervical lymph nodes, the presence of scarlatiniform rash, and the absence of moderate to severe coryza, yielding a 95% probability for GABHS. Excluding the rare scarlatiniform rash, the remaining variables were used in the second regression analysis. In model II, three independent variables were identified: moderate to severe tonsillar swelling, moderate to severe tenderness and enlargement of cervical lymph nodes, and absence of moderate to severe coryza, yielding a probability of 65% for the diagnosis. A probability of <15% was observed in the absence of scarlatiniform rash, the absence of moderate to severe tenderness and enlargement of cervical lymph nodes, and the presence of moderate to severe coryza. CONCLUSIONS: In children with moderate to severe presentation of tonsillar swelling, tenderness and enlargement of cervical lymph nodes, and the absence of coryza, the probability of a positive throat culture is >65%. Conversely, in the absence of a moderate to severe presentation of tonsillar swelling, enlargement of cervical nodes, and the presence of coryza, the probability of a positive throat culture is <15%. If prospectively validated, these models could be integrated into a consistent treat, test, and no treatment/no testing approach to the clinical management of childhood pharyngitis.

Adolescent↗

PubMed2001

A predictive model of well-being and self-care for rural elderly women in Taiwan.

A predictive model of well-being and self-care was tested with 284 women aged 60-88 living in rural communities in Taiwan. The variables studied were age, marital status, social class, social support, perceived health, self-care agency, self-care behavior, and perceived well-being. The model was tested by path analysis. The parameters of the model were estimated with the maximum likelihood method. In the refined model, chi2 (12, N = 284) = 15.18, p =.23), 33% of the total variance in well-being was explained by self-care behavior, social support, and perceived health; 66% of the total variance in self-care behavior was explained by self-care agency and social support; 49% of the total variance in self-care agency was explained by social support, perceived health, social class, and age; 14% of the total variance in social support was explained by marital status, social class, and age; and 8% of the total variance in perceived health was explained by social support. The goodness-of-fit index was.99, indicating that the refined model fit the data well. The findings of this study contribute to a greater understanding of this predictive model for application with older women in rural Taiwan.

Adaptation, Psychological↗

PubMed1998

Derivation of a clinical prediction model for the emergency department diagnosis of ectopic pregnancy.

OBJECTIVE: To derive a clinical prediction model for estimating the pretest probability of ectopic pregnancy in ED patients with first-trimester abdominal pain or vaginal bleeding. METHODS: All hemodynamically stable first-trimester patients presenting to the ED of a tertiary care military teaching hospital over a 14-month period with a chief complaint of abdominal pain and/or vaginal bleeding had clinical data coded prior to determining outcome. They were then followed longitudinally until a criterion standard pregnancy outcome was established. RESULTS: Of the 486 patients enrolled, 280 (58%) had viable intrauterine pregnancies, 167 (34%) had nonviable intrauterine pregnancies, and 39 (8%) had ectopic pregnancies. Using a recursive partitioning model, a high-risk group was derived (that was separated from intermediate and low-risk groups), consisting of patients with abdominal peritoneal signs or definite cervical motion tenderness, with a sensitivity of 31% (95% CI: 17-48%), a specificity of 93% (95% CI: 90-95%), a positive likelihood ratio of 4.3, and a negative likelihood ratio of 0.74. A low-risk group, consisting of patients with either fetal heart tones or tissue at the cervical os, or the absence of pain other than midline menstrual-like cramping and lacking any pelvic tenderness, was differentiated from an intermediate-risk group, with a sensitivity of 96% (95% CI: 81-100%), a specificity of 22% (95% CI: 18-26%), a positive likelihood ratio of 1.2, and a negative likelihood ratio of 0.17. CONCLUSION: A clinical prediction model for estimating the probability of ectopic pregnancy in ED patients has been derived. It may prove to have practical clinical application for estimating pretest probability of ectopic pregnancy as well as assisting in medical decision making when laboratory and ultrasonographic findings are nondiagnostic. Clinical application should await prospective validation in an independent sample.

Abdominal Pain↗

PubMed2000

Accurate predictive modeling of response variables under dynamic condition without the use of past response data

One promising attribute of the dynamic predictive modeling method introduced by Rollins et al. [D.K. Rollins, J. Liang, P. Smith, Accurate simplistic predictive modeling of nonlinear dynamic processes, ISA Transactions 37(4) (1998) 193-203] is its ability to accurately predict output response without the use of online output data. The proposed method only needs online input data to accurately predict output behavior once the semi-empirical model has been identified using offline data. This ability is critical to chemical processes because many output variables (such as chemical composition) are often measured infrequently, inaccurately, or not at all. In addition, in the presence of extremely high measurement noise of the output variable, this work will demonstrate very accurate predictive performance. Finally, this article will show that the method of Rollins et al. can predict better without the use of output data than with the use of output data in the case of large measurement variance. Thus, the proposed method is being recommended for its accuracy, especially in situations where online output response data is limited or inaccurate.

Journal Article↗

PubMedDate not supplied

Optimal stopover decisions of migrating birds under variable stopover quality: model predictions and the field data.

Dataset on departure fuel loads, stopover length and fuel deposition rate of the European robins Erithacus rubecula during their migration in the Baltic area is presented. We test these empirical data against the predictions of an optimal migration model assuming that robins minimize time spent on migration, and that fuel deposition rate varies stochastically. The latter assumption sets this model apart from the alternative ones and makes it more realistic. In particular, it is applicable in frequently observed situations when fuel deposition rate is negative. Our model assumes stochastic variation of the fuel deposition rate at sites along the migratory rout and thus is applicable when negative values of fuel deposition rate are recorded. The model predicts the relationship between fuel deposition rate and departure fuel load rather well. The agreement between the observed and the predicted values of optimal stopover duration is much poorer. Predictions of optimal migration theory are known to be dependent on the form of flight equation chosen. Our model fits the data best when the costs of transport are low. This supports the idea that transport costs of fuel stores may be low, especially when fuel stores are modest.

Animal Migration↗

PubMed1999

Validating and comparing predictive models.

The bias and accuracy factors introduced by Ross [Ross, T., 1996. Indices for performance evaluation of predictive models in food microbiology. J. Appl Bacteriol. 81, 501-508] for the evaluation of the performance of models in 'predictive food microbiology' are refined by basing the calculation of those measures on the mean square differences between predictions and observations. The use of the indices is extended by presenting formulae and methods which enable evaluation of the difference between alternative models for growth of an organism of interest over a domain of environmental factors. This is done by calculating the integral mean of the square differences between the models under investigation over the domain of the environmental variables common to those models, or a sub-region of it. The use of the techniques is exemplified by evaluating the difference between four published models for the growth rate of psychrotrophic pseudomonads.

Bias↗

PubMed2002

[Improvement of prediction model of retention values of anions with sodium carbonate-sodium bicarbonate as eluents].

A prediction model of retention values of anions with sodium carbonate-sodium bicarbonate as eluents has been improved. The improved model is suitable for a wider range of concentrations of sodium carbonate-sodium bicarbonate eluents than the old one set up several years ago. An extensive set of experimental retention data obtained for 7 anions (nitrate, sulfate, oxalate, selenate, iodide, phosphate and arsenate ) using the eluents of varied concentrations was dealt by binary linear regression process. The improved model gave satisfactory performance with correlation coefficients over 0.999 for nitrate, sulfate, oxalate, selenate and iodide, of 0.9971 for phosphate, and of 0.995 7 for arsenate. A computer program has been designed with the prediction model to simulate the separation of anions and used to optimize the routine analysis with sodium carbonate-sodium bicarbonate as the eluent.

English Abstract↗

PubMed2002

Model-predictive control of hyperthermia treatments.

A model-predictive controller (MPC) of the thermal dose in hyperthermia cancer treatments has been developed and evaluated using simulations with one-point and one-dimensional models of a tumor. The developed controller is the first effort in: 1) the application of feedback control to pulsed, high-temperature hyperthermia treatments; 2) the direct control of the treatment thermal dose rather than the treatment temperatures; and 3) the application of MPC to hyperthermia treatments. Simulations were performed with different blood flow rates in the tumor and constraints on temperatures in normal tissues. The results demonstrate that 1) thermal dose can be controlled in the presence of plant-model mismatch and 2) constraints on the maximum allowable temperatures in normal tissue and/or the pulsed power magnitude can be directly incorporated into MPC and met while delivering the desired thermal dose to the tumor. For relatively high blood flow rates and low transducer surface intensities--factors that limit the range of temperature variations in the tumor, the linear MPC, obtained by piece-wise linearization of the dose-temperature relationship, provides an adequate performance. For large temperature variations, the development of nonlinear MPC is necessary.

Computer Simulation↗

PubMed2000

Total-body skeletal muscle mass: development and cross-validation of anthropometric prediction models.

BACKGROUND: Skeletal muscle (SM) is a large body compartment of biological importance, but it remains difficult to quantify SM with affordable and practical methods that can be applied in clinical and field settings. OBJECTIVE: The objective of this study was to develop and cross-validate anthropometric SM mass prediction models in healthy adults. DESIGN: SM mass, measured by using whole-body multislice magnetic resonance imaging, was set as the dependent variable in prediction models. Independent variables were organized into 2 separate formulas. One formula included mainly limb circumferences and skinfold thicknesses [model 1: height (in m) and skinfold-corrected upperarm, thigh, and calf girths (CAG, CTG, and CCG, respectively; in cm)]. The other formula included mainly body weight (in kg) and height (model 2). The models were developed and cross-validated in nonobese adults [body mass index (in kg/m(2)) < 30]. RESULTS: Two SM (in kg) models for nonobese subjects (n = 244) were developed as follows: SM = Ht x (0.00744 x CAG(2) + 0.00088 x CTG(2) + 0.00441 x CCG(2)) + 2.4 x sex - 0.048 x age + race + 7.8, where R:(2) = 0.91, P: < 0.0001, and SEE = 2.2 kg; sex = 0 for female and 1 for male, race = -2.0 for Asian, 1.1 for African American, and 0 for white and Hispanic, and SM = 0.244 x BW + 7.80 x Ht + 6.6 x sex - 0.098 x age + race - 3.3, where R:(2) = 0.86, P: < 0.0001, and SEE = 2.8 kg; sex = 0 for female and 1 for male, race = -1.2 for Asian, 1.4 for African American, and 0 for white and Hispanic. CONCLUSION: These 2 anthropometric prediction models, the first developed in vivo by using state-of-the-art body-composition methods, are likely to prove useful in clinical evaluations and field studies of SM mass in nonobese adults.

Adult↗

PubMed2006

A predictive model for postoperative intraocular pressure among patients undergoing laser in situ keratomileusis (LASIK).

PURPOSE: The aim of this study was to develop a predictive model based on preoperative variables for estimating postoperative intraocular pressure (IOP) of those eyes undergoing LASIK surgery, to predict the amount of underestimated IOP after LASIK for myopia and myopic astigmatism. DESIGN: Pretest-post-test longitudinal study. METHODS: Both eyes of 193 eligible subjects who underwent LASIK procedures at the Department of Ophthalmology, National Taiwan University Hospital, from July 2000 to December 2002 for myopia and myopic astigmatism were identified to build up the predictive models. IOPs were measured with noncontact air-puff tonometry. Information on age, gender, preoperative central corneal thickness (CCT), preoperative central corneal curvature (CCK), preoperative spherical equivalent refractive error, and ablation depth was collected and applied for predicting postoperative IOP after LASIK based on linear mixed model. RESULTS: Significant predictors for postoperative IOP after myopic LASIK procedures included age, gender, preoperative IOP, ablation depth, preoperative CCT, and preoperative spherical equivalent refractive errors. The linear mixed model, taking into account these significant preoperative correlates and the correlation of IOPs between both eyes of the same patient, explained 91% of the variation of postoperative IOP. CONCLUSIONS: A statistical model was developed for predicting the amount of underestimated IOP after LASIK for myopia and myopic astigmatism, which is of clinical importance to uncover ocular hypertension among patients whose information on postoperative IOP immediately after LASIK is not available.

Adult↗