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 289 records · Page 16Linked to original sources

PubMed2006

Sensitivity analysis of an accident prediction model by the fractional factorial method.

Sensitivity analysis of a model can help us determine relative effects of model parameters on model results. In this study, the sensitivity of the accident prediction model proposed by Zegeer et al. [Zegeer, C.V., Reinfurt, D., Hummer, J., Herf, L., Hunter, W., 1987. Safety Effect of Cross-section Design for Two-lane Roads, vols. 1-2. Report FHWA-RD-87/008 and 009 Federal Highway Administration, Department of Transportation, USA] to its parameters was investigated by the fractional factorial analysis method. The reason for selecting this particular model is that it incorporates both traffic and road geometry parameters besides terrain characteristics. The evaluation of sensitivity analysis indicated that average daily traffic (ADT), lane width (W), width of paved shoulder (PA), median (H) and their interactions (i.e., ADT-W, ADT-PA and ADT-H) have significant effects on number of accidents. Based on the absolute value of parameter effects at the three- and two-standard deviation thresholds ADT was found to be of primary importance, while the remaining identified parameters seemed to be of secondary importance. This agrees with the fact that ADT is among the most effective parameters to determine road geometry and therefore, it is directly related to number of accidents. Overall, the fractional factorial method was found to be an efficient tool to examine the relative importance of the selected accident prediction model parameters.

Accidents, Traffic↗

PubMed2003

A preliminary study of the application of some predictive modeling techniques to assess atmospheric mercury emissions from terrestrial surfaces.

Predictive modeling techniques are applied to investigate their potential usefulness in providing first order estimates on atmospheric emission flux of gaseous soil mercury and in identifying those parameters most critical in controlling such emissions. Predicted data by simulation and statistical techniques are compared to previously published observational data. Results showed that simulation techniques using air/soil coupling may provide a plausible description of mercury flux trends with a RMSE of 24.4ngm(-2)h(-1) and a mean absolute error of 10.2ngm(-2)h(-1) or 11.9%. From the statistical models, two linear models showed the lowest predictive abilities (R2=0.76 and 0.84, respectively) while the Generalized Additive model showed the closest agreement between estimated and observational data (R2=0.93). Predicted values from a Neural Network model and the Locally Weighted Smoother model showed also very good agreement to measured values of mercury flux (R2=0.92). A Regression Tree model demonstrated also a satisfactory predictability with a value of R2=0.90. Sensitivities and statistical analyses showed that surface soil mercury concentrations, solar radiation and, to a lesser degree, temperature are important parameters in predicting airborne Hg flux from terrestrial soils. These findings are compatible with results from recent experimental studies. Considering the uncertainties associated with mercury cycling and natural emissions, it is concluded, that predictions based on simple modeling techniques seem quite appropriate at present; they can be useful tools in evaluating the role of terrestrial emission sources as part of mercury modeling in local and regional airsheds.

Air Pollutants↗

PubMed2026

Risk Factors and Predictive Model for Postoperative High Myopia in Children Undergoing Congenital Cataract Surgery With Intraocular Lens Implantation.

PURPOSE: To identify risk factors associated with the development of high myopia following congenital cataract surgery and to establish a robust predictive model. DESIGN: Retrospective clinical cohort study. SUBJECTS: This retrospective study included 106 pediatric patients who underwent congenital cataract surgery with primary IOL implantation (mean follow-up 8.19 years). The model was externally validated in an independent cohort of 72 patients with a mean follow-up of 7.83 years. METHODS: Preoperative and postoperative ocular biometric parameters were collected. Risk factors for postoperative high myopia were analyzed using Cox proportional hazards regression, which served as the basis for model construction. The predictive performance of the model was rigorously evaluated for discrimination and calibration. Discriminative ability was quantified using Harrell's C-index and the area under the receiver operating characteristic curve (AUC). Model calibration was assessed via calibration plots by comparing predicted probabilities with actual observed outcomes. Internal validation was performed using a bootstrapping method (500 iterations) to ensure model stability and adjust for potential overfitting. RESULTS: An initial postoperative refraction of <+0.75D, and a higher IOL Power to Axial length Ratio (IOL/AL ratio) were identified as significant risk factors for the development of postoperative high myopia. Shorter preoperative axial length was associated with a greater magnitude of postoperative myopic shift. The predictive model demonstrated robust performance, achieving a C-index of 0.711 (internal validation C-index: 0.713). The area under the receiver operating characteristic curve (AUC) values for predicting high myopia at 5 and 10 years were 0.858 and 0.745, respectively. Furthermore, calibration curves demonstrated excellent agreement between the predicted and observed outcomes throughout the follow-up period. In external validation, the model achieved a C-index of 0.825, 5-year AUC of 0.833, and 10-year AUC of 0.713. CONCLUSIONS: Our analysis established that initial postoperative refraction <+0.75D, and an elevated IOL/AL ratio are key determinants of high myopia risk following surgery. Shorter preoperative axial length was associated with a greater magnitude of postoperative myopic shift. This predictive framework provides clinicians with a practical tool to optimize preoperative IOL selection and identify high-risk infants who require vigilant myopia prevention and balanced amblyopia management.

Humans↗

PubMed1998

Tensions of the flexor digitorum superficialis are higher than a current model predicts.

Existing isometric force models can be used to predict tension in the finger flexor tendon, however, they assume a specific distribution of forces across the tendons of the fingers. These assumptions have not been validated or explored by experimental methods. To determine if the force distributions repeatably follow one pattern the in vivo tension of the flexor digitorum superficialis (FDS) tendon of the long finger was measured in nine patients undergoing open carpal tunnel release surgery. Following the release, a tendon force transducer (Dennerlein et al. 1997 J. Biomechanics 30(4), 395-397) was mounted onto the FDS of the long finger. Tension in the tendon, contact force at the fingertip, and finger posture were recorded while the patient gradually increased the force applied by the fingertip from 0 to 10 N and then monotonically reduced it to 0 N. The average ratio of the tendon tension to the fingertip contact force ranged from 1.7 to 5.8 (mean = 3.3, s.d. = 1.4) for the nine subjects. These ratios are larger than ratios predicted by current isometric tendon force models (mean = 1.2, s. d. = 0.4). Subjects who used a pulp pinch posture (hyper-extended distal interphalangeal joint (DIP)) showed a significantly (p = 0.02) larger ratio (mean = 4.4, s.d. = 1.5) than the five subjects who flexed the DIP joint in a tip pinch posture (mean = 2.4, s.d. = 0.6). A new DIP constraint model, which selects different force distribution based on DIP joint posture, predicts force ratios that correlate well with the measured ratios (r2 = 0.85).

Adult↗

PubMed1994

Approaches to predictive modeling.

A four-component clinical model for process improvement is presented: (1) patient-related risk factors, (2) clinical processes ordered by the attending physician, (3) the hospital's execution of the physician's plan, and (4) the patient's outcome, or outcomes, resulting from the first three factors. The goal of risk adjustment in the analysis of quality of care is to account for the contribution of patient-related risk factors, so that the patient's outcome can be used as an indicator of the care ordered by the physician and executed by the hospital. Risk adjustment is usually accomplished by comparing the patient's predicted outcome, based on the patient's risk factors, to the observed outcome. Historically, three approaches to the development of prediction models have been used: (1) selection and weighting of risk factors by expert opinion, (2) univariate analyses, and (3) multivariate analyses. Future prediction models will be based on neural network techniques or cluster analysis. As these prediction models have evolved, there has been a steady increase in their predictive power.

Cardiac Surgical Procedures↗

PubMed2005

Experimental verification of a portal dose prediction model.

Electronic portal imaging devices (EPIDs) can be used to measure a two-dimensional (2D) dose distribution behind a patient, thus allowing dosimetric treatment verification. For this purpose we experimentally assessed the accuracy of a 2D portal dose prediction model based on pencil beam scatter kernels. A straightforward derivation of these pencil beam scatter kernels for portal dose prediction models is presented based on phantom measurements. The model is able to predict the 2D portal dose image (PDI) behind a patient, based on a PDI without the patient in the beam in combination with the radiological thickness of the patient, which requires in addition a PDI with the patient in the beam. To assess the accuracy of portal dose and radiological thickness values obtained with our model, various types of homogeneous as well as inhomogeneous phantoms were irradiated with a 6 MV photon beam. With our model we are able to predict a PDI with an accuracy better than 2% (mean difference) if the radiological thickness of the object in the beam is symmetrically situated around the isocenter. For other situations deviations up to 3% are observed for a homogeneous phantom with a radiological thickness of 17 cm and a 9 cm shift of the midplane-to-detector distance. The model can extract the radiological thickness within 7 mm (maximum difference) of the actual radiological thickness if the object is symmetrically distributed around the isocenter plane. This difference in radiological thickness is related to a primary portal dose difference of 3%. It can be concluded that our model can be used as an easy and accurate tool for the 2D verification of patient treatments by comparing predicted and measured PDIs. The model is also able to extract the primary portal dose with a high accuracy, which can be used as the input for a 3D dose reconstruction method based on back-projection.

Algorithms↗

PubMed2006

Quantitative and predictive model of transcriptional control of the Drosophila melanogaster even skipped gene.

Here we present a quantitative and predictive model of the transcriptional readout of the proximal 1.7 kb of the control region of the Drosophila melanogaster gene even skipped (eve). The model is based on the positions and sequence of individual binding sites on the DNA and quantitative, time-resolved expression data at cellular resolution. These data demonstrated new expression features, first reported here. The model correctly predicts the expression patterns of mutations in trans, as well as point mutations, insertions and deletions in cis. It also shows that the nonclassical expression of stripe 7 driven by this fragment is activated by the protein Caudal (Cad), and repressed by the proteins Tailless (Tll) and Giant (Gt).

Animals↗

PubMed2003

Validation of a mathematical model predicting the response to growth hormone treatment in prepubertal children with idiopathic growth hormone deficiency.

OBJECTIVE: To validate a mathematical model developed by Ranke et al. (J Clin Endocrinol Metab 1999;84:1174-7783) to predict the GH response during the first years of GH replacement therapy. PATIENTS AND METHODS: 38 children with idiopathic GH deficiency (GHD) met all inclusion criteria for the prediction model, but the group differed in some characteristics from the cohort from which the model was derived. RESULTS: Using the model for the 1st year including maximum GH after stimulation and the equation for the 6th year, the predicted value corresponded well with actual height gain. Differences were found when the growth response of the 1st year excluding maximum GH and that of the 2nd-5th year were calculated, resulting in a significant underestimation of actual height gain (-0.63 to -1.07 cm/year). CONCLUSION: The mathematical prediction model tended to underpredict the growth response to GH treatment in our patients with pronounced GHD. The severity of GHD seems to be an important parameter for the 1st year prediction.

Body Height↗

PubMed2003

Lateral facial soft-tissue prediction model: analysis using Fourier shape descriptors and traditional cephalometric methods.

This study was designed to investigate the relationship between traditional skeletal cephalometric measurement and Fourier analysis of the lateral soft-tissue profile. A random sample of 121 untreated subjects of European descent, with wide ranges of malocclusions and underlying facial patterns, was selected in the Orthodontic Unit at the University of Melbourne. Lateral cephalograms were available for all subjects. Both traditional lateral cephalometric analysis and Fourier soft-tissue profile analysis were carried out. Multivariate statistical analysis among 11 hard-tissue cephalometric measurements and the first 50 Fourier harmonics was then performed. This analysis formed the basis for a subsequently proposed soft-tissue prediction model. From this model, 50 predicted x- and y-harmonics were generated for each subject in the total sample. Calculation of Pearson's correlation coefficients between the actual and predicted harmonics revealed strong relationships for many of the lower-order harmonics. To further test the model, the prediction-coefficients derived from all 121 subjects were then used to make predictions for the first 50 x- and y-harmonics for a subgroup of 10 independent test subjects. Once again, Pearson's correlations between the actual and predicted harmonics of the test model in the lower-order harmonics revealed strong associations. Superimposition of the actual and predicted soft-tissue outlines, however, revealed that much actual detail in the region between the nose and the chin was still lost using the predicted Fourier harmonics. This suggests that soft-tissue prediction based on this Fourier test model, while already useful in Forensic facial reconstruction, may not yet be appropriate for useful diagnosis and planning in clinical disciplines.

Adolescent↗

PubMed2006

Severe neutropenia in CHOP occurs most frequently in cycle 1: a predictive model.

Chemotherapy used to treat lymphoma can cause severe neutropenia. Risk models have identified factors that predict neutropenia across all chemotherapy cycles. We used clinical information obtained during pretreatment evaluation to develop a predictive model for severe neutropenia in the first cycle of cyclophosphamide, doxorubicin, vincristine and prednisone (CHOP) chemotherapy. This case series study included lymphoma patients receiving CHOP chemotherapy with or without rituximab who did not receive pre-emptive hematopoietic growth factor. Risk factors for neutropenia were identified from previously published models and included age >or=65 years, hypoalbuminemia, renal/cardiovascular disease, anemia, abnormal bone marrow and increased lactate dehydrogenase (LDH). A composite score equal to the number of pretreatment risk factors was used to predict severe neutropenia in cycle 1. Fifty-three percent of patients (47 of 89) had severe neutropenia, with 70% of first episodes occurring during cycle 1. Eighty-two percent of first-cycle, severe neutropenia events occurred in patients >or=65-years-old. In univariate analysis, age >or=65 years and increased baseline LDH were significantly associated with increased risk for severe neutropenia in cycle 1. In logistic regression modeling, the probability of severe neutropenia in cycle 1 increased as the number of pretreatment risk factors increased, with a one-unit increase in risk score resulting in a 2.3-fold increase in severe neutropenia. The study results suggest that data obtained before initiating CHOP-based chemotherapy can be used to identify those patients who are at risk for severe neutropenia in cycle 1. If validated, our model could be used to identify patients who would benefit from early use of growth factors.

Adolescent↗

PubMed1999

Shelf life of modified atmosphere packed cooked meat products: a predictive model.

The effect of temperature, concentration of dissolved CO2 and water activity on the growth of Lactobacillus sake was investigated by developing predictive models for the lag phase and the maximum specific growth rate of this specific spoilage organism for gas-packed cooked meat products. Two types of predictive model were compared: an extended Ratkowsky model and a response surface model. In general, response surface models showed a slightly better correlation, but the response surface model for the maximum specific growth rate showed illogical predictions at low water activities. The concentration of dissolved CO2 proved to be a significant independent variable for the maximum specific growth rate as well as for the lag phase of L. sake. Synergistic actions on the shelf life-extending effect were noticed between temperature and dissolved CO2, as well as between water activity and dissolved CO2. The developed models were validated by comparison with the existing model of Kant-Muermans et al. (1997) and by means of experiments in gas-packed cooked meat products. Both developed models proved to be useful in the prediction of the microbial shelf life of gas-packed cooked meat products.

Animals↗

PubMed2005

Clinical prediction model to characterize pulmonary nodules: validation and added value of 18F-fluorodeoxyglucose positron emission tomography.

BACKGROUND: The added value of 18F-fluorodeoxyglucose (FDG) positron emission tomography (PET) scanning as a function of pretest risk assessment in indeterminate pulmonary nodules is still unclear. OBJECTIVE: To obtain an external validation of the prediction model according to Swensen and colleagues, and to quantify the potential added value of FDG-PET scanning as a function of its operating characteristics in relation to this prediction model, in a population of patients with radiologically indeterminate pulmonary nodules. DESIGN, SETTING, AND PATIENTS: Between August 1997 and March 2001, all patients with an indeterminate solitary pulmonary nodule who had been referred for FDG-PET scanning were retrospectively identified from the database of the PET center at the VU University Medical Center. RESULTS: One hundred six patients were eligible for the study, and 61 patients (57%) proved to have malignant nodules. The goodness-of-fit statistic for the model (according to Swensen) indicated that the observed proportion of malignancies did not differ from the predicted proportion (p = 0.46). PET scan results, which were classified using the 4-point intensity scale reading, yielded an area under the evaluated receiver operating characteristic curve of 0.88 (95% confidence interval [CI], 0.77 to 0.91). The estimated difference of 0.095 (95% CI, -0.003 to 0.193) between the PET scan results classified using the 4-point intensity scale reading and the area under the curve (AUC) from the Swensen prediction was not significant (p = 0.058). The PET scan results, when added to the predicted probability calculated by the Swensen model, improves the AUC by 13.6% (95% CI, 6 to 21; p = 0.0003). CONCLUSION: The clinical prediction model of Swensen et al was proven to have external validity. However, especially in the lower range of its estimates, the model may underestimate the actual probability of malignancy. The combination of visually read FDG-PET scans and pretest factors appears to yield the best accuracy.

Aged↗

PubMed2006

Clinical features of high-risk older persons identified by predictive modeling.

The objective of this study was to describe the clinical features of older persons identified as high risk by a predictive modeling algorithm and to determine their suitability for clinical interventions like case management or disease management. A cross-sectional survey was undertaken at a community-based general internal medicine practice with 826 older patients enrolled in a Medicare-like health plan for military retirees and their dependents. Administrative claims data provided information about all 826 enrollees' chronic conditions, their use of health services, and the cost of those services during the past year. A survey mailed to 150 identified high-risk enrollees provided information about sociodemographic characteristics, general health, bed disability days, restricted activity days, activities of daily living (ADL) limitations, and instrumental activities of daily living (IADL) limitations. Compared to the 676 low-risk enrollees, the 150 high-risk enrollees had higher prevalence of eight individual chronic conditions, higher total chronic conditions (2.93 vs. 1.48, p < 0.001), higher annual rates of hospital admission (1.1 vs. 0.1, p < 0.001), more annual hospital days (7.3 vs. 0.5, p < 0.001), and higher total health insurance expenditures ($22,815 vs. $3,726, p < 0.001). The high-risk respondents to the survey (response rate = 80.0%) had suboptimal health (42.8% "fair or poor"), impaired functional ability (36.3% with 1+ ADL limitations, 58.1% with 1+ IADL limitations), and frequent health-related disruptions in their activities during the previous six months (38.7% with 1+ bed disability day, 52.3% with 1+ restricted activity day). A claims-based predictive modeling algorithm identifies older persons whose health, functional ability, and use of health services suggest they are good candidates for clinical interventions such as case management and disease management.

Aged↗

PubMed2004

Universal predictive models on octanol-air partition coefficients at different temperatures for persistent organic pollutants.

Owing to the importance of octanol-air partition coefficients (KOA) in describing the partition of organic pollutants from air to environmental organic phases, the paucity of KOA data at different environmental temperatures, and the difficulty or high expenditures involved in experimental determination, the development of predictive models for KOA is necessary. Approaches such as this are greatly needed to evaluate the environmental fate of the ever-increasing list of production chemicals. Partial least squares (PLS) regression with 18 molecular structural descriptors was used to develop predictive models based on directly measured KOA values of selected chlorobenzenes, polychlorinated biphenyls (PCBs), polychlorinated naphthalenes, polychlorinated dibenzo-p-dioxins/dibenzofurans, polybrominated diphenyl ethers, polycyclic aromatic hydrocarbons, and organochlorine pesticides (OPs). An optimization procedure resulted in two temperature-dependent universal predictive models that explained at least 91 % of the variance of log KOA. Model 1 was the more general of the two models that could be used for all the persistent organic pollutant (POP) classes investigated. Although model 1 performed poorly for select OPs, this was attributed to wide variability in structural types within this subset of POPs and their diversity compared to the other POP classes that were investigated. The exclusion of the structurally complex OP subset resulted in a more precise model, model 5. Intermolecular dispersive interactions (induced dipole-induced dipole forces) between octanol and solute molecules play a decisive role in governing KOA and its temperature dependence. Further investigations are needed to better characterize the steric structures of the POPs under study, especially of OPs.

Air↗

PubMed1998

Predictive modelling of the mechanical properties and failure processes in hydroxyapatite- polyethylene (Hapex) composite.

The development of a wide range of hydroxyapatite polyethylene composites for medical applications is increasing the need for accurate predictive modelling. The objective of this work was to elucidate the observed mechanical processes and failure processes in this material using the finite element analysis method. The need for full three-dimensional modelling of this material has been shown. The results from this predictive model lead to accurate predictions of measured mechanical properties, and allow deduction of possible routes to improved ductility at high volume fractions.

Journal Article↗

PubMed2006

Temporal transferability and updating of zonal level accident prediction models.

This paper examines the temporal transferability of the zonal accident prediction models by using appropriate evaluation measures of predictive performance to assess whether the relationship between the dependent and independent variables holds reasonably well across time. The two temporal contexts are the years 1996 and 2001, with updated 1996 models being used to predict 2001 accidents in each traffic zone of the City of Toronto. The paper examines alternative updating methods for temporal transfer by imagining that only a sample of 2001 data is available. The sensitivity of the performance of the updated models to the 2001 sample size is explored. The updating procedures examined include the Bayesian updating approach and the application of calibration factors to the 1996 models. Models calibrated for the 2001 samples were also explored, but were found to be inadequate. The results show that the models are not transferable in a strict statistical sense. However, relative measures of transferability indicate that the transferred models yield useful information in the application context. Also, it is concluded that the updated accident models using the calibration factors produce better results for predicting the number of accidents in the year 2001 than using the Bayesian approach.

Accidents, Traffic↗

PubMed2001

The potential of prediction models based on data from KIGS as tools to measure responsiveness to growth hormone. Pharmacia International Growth Database.

Various prediction models have been developed, based on data documented within KIGS (Pharmacia International Growth Database), for use in the growth hormone (GH) treatment of children with short stature resulting from GH deficiency (GHD) or other causes. In addition to the practical value of such models as part of a 'forward strategy' guiding GH treatment, we now propose that prediction models may also be useful for the identification of individual variance in responsiveness. In a comparison involving 1,800 children with idiopathic GHD (IGHD), 151 children who acquired GHD after treatment for medulloblastoma and 192 children with GHD accompanying craniopharyngioma, it was shown that the responsiveness to GH of patients with craniopharyngioma equalled that of IGHD patients, whereas patients with medulloblastoma were less responsive. These observations and the identification of 'good' and 'poor' responders to GH have practical clinical consequences (e.g. modification of treatment), and will, in the future, lead to the identification of those factors which determine the variability of sensitivity to GH. This will improve the efficacy and safety of GH treatment as well as reducing the costs involved.

Databases as Topic↗

PubMed2005

A new biodegradation prediction model specific to petroleum hydrocarbons.

A new predictive model for determining quantitative primary biodegradation half-lives of individual petroleum hydrocarbons has been developed. This model uses a fragment-based approach similar to that of several other biodegradation models, such as those within the Biodegradation Probability Program (BIOWIN) estimation program. In the present study, a half-life in days is estimated using multiple linear regression against counts of 31 distinct molecular fragments. The model was developed using a data set consisting of 175 compounds with environmentally relevant experimental data that was divided into training and validation sets. The original fragments from the Ministry of International Trade and Industry BIOWIN model were used initially as structural descriptors and additional fragments were then added to better describe the ring systems found in petroleum hydrocarbons and to adjust for nonlinearity within the experimental data. The training and validation sets had r2 values of 0.91 and 0.81, respectively.

Biodegradation, Environmental↗