Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “predictive modelling”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

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

At least 1,369 records · Page 76Linked to original sources
PubMedDate not supplied

Comparison of two dose prediction models for cisplatin.

Cisplatin toxicity could be decreased by adjusting its dosage to each patient. For this purpose, a limited sampling method was established and validated based on a Bayesian approach taken using the values of assays during a 5-day continuous infusion of cisplatin. Using this method, a dosing model to achieve a target plasma concentration of total platinum (Pt) was evaluated retrospectively; the calculated dose of cisplatin was 95.0 to 104.8% of the actual dose. This model was then studied prospectively and the actual plasma Pt concentration reached at the end of the infusion was 94.9% of the target concentration. A strong correlation was observed between the clearance of Pt and the calculated clearance of creatinine or Cockroft index (p = 1.7 x 10(-11), and this correlation was used to develop another cisplatin dosing model. With this model the actual concentration reached at the end of the infusion was 85.3% of the theoretical concentration. The Bayesian approach gave reliable results for most clinical uses, whereas the creatinine based model has to be improved.

Bayes Theorem↗
PubMed1993

[Surgical wound infection: the risk factors and a predictive model].

BACKGROUND: The identification of the factors associated with infection of the surgical wound and the groups of patients with greatest risk of developing the same may aid in the elaboration of prevention strategies. METHODS: A prospective follow up study of a group of 1,143 patients admitted to general and digestive surgical departments in the Ramón y Cajal Hospital over a period of 7 months was carried out to determine the accumulated incidence of infection of the surgical wound and quantify the associated risk factors. A mathematical model was developed by logistic multiple regression analysis allowing the identification of groups of patients with high risk of infection which were internally evaluated posteriorly. RESULTS: Surgically intervened patients (70% of those admitted) developed a mean of 11 wound infections out of 100 patients. Five independent factors (age, surgical classification, length of intervention, presurgical stay, and presence of a central route) were associated to increased risk of infection. CONCLUSIONS: The factors associated with surgical wound infection identified in this study are related to the degree of wound contamination, the intrinsic risk of the patients and quality of health care. The model obtained is more efficient than the traditional surgical classification for the identification of groups of patients with high risk of infection.

Chi-Square Distribution↗
PubMed1997

Radiotherapy-associated neutropenia and thrombocytopenia: analysis of risk factors and development of a predictive model.

Risk factors for unscheduled interruptions in radiotherapy courses completed between June 1989 and August 1995, lasting > or = 2 days, and associated with World Health Organization grade III-IV neutropenia or thrombocytopenia were studied retrospectively. A group of controls was randomly selected. Potential risk factors for myelosuppression were analyzed using univariate and multivariate analyses. The most important risk factors for treatment interruption with thrombocytopenia were concurrent chemotherapy (odds ratio [OR], 45.5; P < .001), increasing percentage of marrow irradiated (OR, 4.1 for each 20%; P < .001), and brain metastases (OR, 7.3; P = .01). Other significant (P < .05) factors were leukemia/lymphoma, bone or bone marrow metastases, and prior chemotherapy. The most important risk factors for treatment interruptions with neutropenia were concurrent chemotherapy (OR, 42.1; P < .001) and increasing percentage of marrow irradiated (OR, 3.3 for each 20%; P < .001). Similarly, the most important risk factors for treatment interruptions with both thrombocytopenia and neutropenia were concurrent chemotherapy (OR, 48.6; P < .001) and increasing percentage of marrow irradiated (OR, 3.9 for each 20%; P < .001). Other significant (P < .05) factors in these groups were bone marrow or brain metastases and previous chemotherapy. These data were used to create a model, assigning patients to groups at high, intermediate, or low risk for treatment interruption with thrombocytopenia. High-risk patients may be candidates for clinical trials of a platelet growth factor.

Adult↗
PubMedDate not supplied

[Use of an experimental space-time prediction model to design instruments for evaluation of higher nervous activity in man].

While examining the basic mechanisms responsible for the performance of the central nervous system (CNS), of interest are the situations in which the behaviour caused by these mechanisms and independent of education levels, habits, and occupation is shown to the highest extent. The paper presents an example of constructing and testing an experimental model which may assess different manifestations of CNS functions. The use of the model as textual examination procedures provides an insight into the higher nervous performance.

Central Nervous System↗
PubMed2005

Assessing the reliability of a QSAR model's predictions.

Quantitative structure activity relationships (QSAR) are one of the well-developed areas in computational chemistry. In this field, many successful predictive models have been developed for various property, activity or toxicity predictions. However, the predictive power of models for new query compounds is often not well characterized. The breadth of applicability of models is often not characterized. In other words, with a given QSAR model and a specific query compound to be predicted, can the model be used reliably for the desired prediction? In this study, we assessed the reliability of QSAR models' prediction on query compounds. Our approach, employing hierarchical clustering, was developed and tested using a test dataset containing 322 organic compounds with fathead minnow acute aquatic toxicity as the activity of interest. The hypothesis of the approach was that if a query compound is more similar to the compounds used to generate the QSAR model, it should be predicted more accurately. Thus, the core of the approach is to determine the relationship between the similarity of query compounds to the training set compounds of the QSAR model and the prediction accuracy given by that model. This relationship determination was achieved by comparing the results given by the two major components of the approach: objects clustering and activity prediction. With the resultant information from the two steps, a direct relationship was shown.

Models, Chemical↗
PubMed2004

A Skinfold Model to Predict Fat-Free Mass in Female Athletes.

OBJECTIVE: Despite widespread use of skinfolds to estimate body fatness, few prediction models have been validated on female athletes. Most skinfold models have been validated with hydrodensitometry, which does not account for the variability in bone density that may exist among female athletes. Our purpose was to develop a skinfold model that predicts fat-free mass (FFM) in female collegiate athletes. DESIGN AND SETTING: A skinfold model was developed using dual-energy x-ray absorptiometry (DEXA) as the criterion method. Four skinfold measures (abdominal, suprailiac, thigh, triceps), height, and weight were entered into a regression model. The best model was developed and validated by calculating the predicted error sum of squares statistic. SUBJECTS: Study participants included 101 National Collegiate Athletic Association Division I female athletes (age = 20.3 +/- 1.4 years, height = 166.7 +/- 7.8 cm, mass = 63.1 +/- 8.1 kg) from several sports. MEASUREMENTS: Each participant's FFM was measured via DEXA. Skinfold thicknesses were measured and entered into the regression model. RESULTS: The final regression model included mass and abdominal and thigh skinfolds: FFM = 8.51 + (0.809 x mass) - (0.178 x abdominal skinfold) - (0.225 x thigh skinfold). The model showed excellent predictive ability (R = 0.98, standard error of the estimate = 1.1 kg). Pairwise comparisons indicated that prediction error showed no overprediction or underprediction bias. CONCLUSIONS: In female collegiate athletes, FFM can be predicted accurately from body mass and abdominal and thigh skinfolds. This model is practical and can be used in most athletic settings.

Journal Article↗
PubMed2002

Measuring uncertainty in complex decision analysis models.

Prediction models used in support of clinical and health policy decision making often need to consider the course of a disease over an extended period of time, and draw evidence from a broad knowledge base, including epidemiologic cohort and case control studies, randomized clinical trials, expert opinions, and more. This paper is a brief introduction to these complex decision models, their relation to Bayesian decision theory, and the tools typically used to describe the uncertainties involved. Concepts are illustrated throughout via a simplified tutorial.

Bayes Theorem↗
PubMed1988

A clinical model for predicting survival following acute myocardial infarction in patients without cardiogenic shock: a multivariate (Cox) analysis.

A multivariate predictive model for early (six-month) survival based on Cox's proportional-hazards regression model was developed using data collected prospectively from 317 consecutive patients admitted with acute myocardial infarction to a coronary care unit (CCU). Of these, 63 (19.8%) died within the follow-up period. Patients with cardiogenic shock were excluded from the study. Variables associated with survival were sought from clinical, historical, electrocardiographic and radiographic variables recorded at the time of admission. On multivariate analysis, a stepwise selection procedure identified four variables which described the probability of survival for the six-month follow-up. These were: age, upper lung crepitations, marginal and also definite radiographic cardiomegaly on an anteroposterior radiograph. With this combination of clinical variables alone, using a survival probability partition value of 80%, the model had a sensitivity of 67% and a specificity of 75%. However, the model's predictive accuracy for death was 40%, compared to a predictive accuracy for survival of 90%. This clinical model is most useful for early discrimination of those patients at low risk of death within six months of CCU admission. Other predictive tests for premature death would need to exceed these discriminatory criteria to justify their cost and risks.

Aged↗
PubMed1996

Evaluation of artificial neural network modelling to predict torso muscle activity.

Owing to the complexities involved in obtaining direct measures of in vivo muscle forces, validation of predictive models of muscle activity has been difficult. An artificial neural network (ANN) model had been previously developed for the estimation of lumbar muscle activity during moderate levels of static exertion. The predictive ability of this model is evaluated in this study using several techniques, including comparison of response surfaces and composite statistical tests of values derived from model output, with multiple EMG experimental datasets. ANN-predicted activation levels were accurately modelled to within 3% across a range of experiments and levels of combined flexion/extension and lateroflexion loadings. The results indicate both a high degree of consistency in the averaged muscle activity measured in several different experiments, and substantiate the ability of the ANN model to predict generalized recruitment patterns. It also is suggested that the use of multiple comparison methods provides a better indication of model behaviour and prediction accuracy than a single evaluation criterion.

Databases, Factual↗
PubMed2006

Linear regression model for predicting patient-specific total skeletal spongiosa volume for use in molecular radiotherapy dosimetry.

UNLABELLED: The toxicity of red bone marrow is widely considered to be a key factor in restricting the activity administered in molecular radiotherapy to suboptimal levels. The assessment of marrow toxicity requires an assessment of the dose absorbed by red bone marrow which, in many cases, requires knowledge of the total red bone marrow mass in a given patient. Previous studies demonstrated, however, that a close surrogate-spongiosa volume (combined tissues of trabecular bone and marrow)-can be used to accurately scale reference patient red marrow dose estimates and that these dose estimates are predictive of marrow toxicity. Consequently, a predictive model of the total skeletal spongiosa volume (TSSV) would be a clinically useful tool for improving patient specificity in skeletal dosimetry. METHODS: In this study, 10 male and 10 female cadavers were subjected to whole-body CT scans. Manual image segmentation was used to estimate the TSSV in all 13 active marrow-containing skeletal sites within the adult skeleton. The age, total body height, and 14 CT-based skeletal measurements were obtained for each cadaver. Multiple regression was used with the dependent variables to develop a model to predict the TSSV. RESULTS: Os coxae height and width were the 2 skeletal measurements that proved to be the most important parameters for prediction of the TSSV. The multiple R(2) value for the statistical model with these 2 parameters was 0.87. The analysis revealed that these 2 parameters predicted the estimated the TSSV to within approximately +/-10% for 15 of the 20 cadavers and to within approximately +/-20% for all 20 cadavers in this study. CONCLUSION: Although the utility of spongiosa volume in estimating patient-specific active marrow mass has been shown, estimation of the TSSV in active marrow-containing skeletal sites via patient-specific image segmentation is not a simple endeavor. However, the alternate approach demonstrated in this study is fairly simple to implement in a clinical setting, as the 2 input measurements (os coxae height and width) can be made with either pelvic CT scanning or skeletal radiography.

Bone Marrow↗
PubMed2005

Comparison of artificial neural network and logistic regression models for prediction of mortality in head trauma based on initial clinical data.

BACKGROUND: In recent years, outcome prediction models using artificial neural network and multivariable logistic regression analysis have been developed in many areas of health care research. Both these methods have advantages and disadvantages. In this study we have compared the performance of artificial neural network and multivariable logistic regression models, in prediction of outcomes in head trauma and studied the reproducibility of the findings. METHODS: 1000 Logistic regression and ANN models based on initial clinical data related to the GCS, tracheal intubation status, age, systolic blood pressure, respiratory rate, pulse rate, injury severity score and the outcome of 1271 mainly head injured patients were compared in this study. For each of one thousand pairs of ANN and logistic models, the area under the receiver operating characteristic (ROC) curves, Hosmer-Lemeshow (HL) statistics and accuracy rate were calculated and compared using paired T-tests. RESULTS: ANN significantly outperformed logistic models in both fields of discrimination and calibration but under performed in accuracy. In 77.8% of cases the area under the ROC curves and in 56.4% of cases the HL statistics for the neural network model were superior to that for the logistic model. In 68% of cases the accuracy of the logistic model was superior to the neural network model. CONCLUSIONS: ANN significantly outperformed the logistic models in both fields of discrimination and calibration but lagged behind in accuracy. This study clearly showed that any single comparison between these two models might not reliably represent the true end results. External validation of the designed models, using larger databases with different rates of outcomes is necessary to get an accurate measure of performance outside the development population.

Adolescent↗
PubMed2006

A recessive Mendelian model to predict carrier probabilities of DFNB1 for nonsyndromic deafness.

Mutations in the DFNB1 locus, where two connexin genes are located (GJB2 and GJB6), account for half of congenital cases of nonsyndromic autosomal recessive deafness. Because of the high frequency of DFNB1 gene mutations and the availability of genetic diagnostic tests involving these genes, they are the best candidates to develop a risk prediction model of being hearing impaired. People undergoing genetic counseling are normally interested in knowing the probability of having a hearing impaired child given his/her family history. To address this, a Mendelian model that predicts the probability of being a carrier of DFNB1 mutations, using family history of deafness, has been developed. This probability will be useful as additional information to decide whether or not a genetic test should be performed. This model incorporates Mendelian mode of inheritance, the age of onset of the disease, and the current age of hearing family members. The carrier probabilities are obtained using Bayes' theorem, in which mutation prevalence is used as the prior distribution. We have validated our model by using information from 305 families affected with congenital or progressive nonsyndromic deafness, in which genetic analysis of GJB2 and GJB6 had already been performed. This model works well, especially in homozygous carriers, showing a high discriminative power. This indicates that our proposed model can be useful in the context of clinical counseling of autosomal recessive disorders.

Adult↗