Publications
Overview | Projects | Publications | Presentations | Teaching
My research investigates approaches for deriving and validating prediction models when multiple sources of evidence are available. Particularly, we propose methods to aggregate previously published prediction models and associations with new data, methods to combine multiple datasets, methods to quantify heterogeneity and interpret external validity. Additional information is available on ResearchGate.
Janssen KJ, Siccama I, Vergouwe Y, Koffijberg H, Debray TP, Keijzer M, Grobbee DE, Moons KG. Development and validation of clinical prediction models: Marginal differences between logistic regression, penalized maximum likelihood estimation, and genetic programming. Journal of Clinical Epidemiology 2012; 65: 404-412. [Full Text]
Rietbergen C, Debray TPA, Janssen KJM, Klugkist I, Moons KGM. Bayesian methods in epidemiological and medical research: a systematic review. [Submitted].
Nieboer D, Vergouwe Y, Debray TPA, Koffijberg H, Moons KGM, Steyerberg EW. Modelling continuous predictor variables: a comparison between fractional polynomials and restricted cubic splines. [In progress]
Debray TPA, Koffijberg H, Vergouwe Y, Nieboer D, Steyerberg EW, Moons KGM. A framework for evaluating and distinguishing validity and generalization of prediction models. [In progress]
Debray TPA, Koffijberg H, Vergouwe Y, Moons KGM, Steyerberg EW. Aggregating published prediction models with individual patient data: a comparison of different approaches. [Submitted]
Debray TPA, Koffijberg H, Lu D, Vergouwe Y, Steyerberg EW, Moons KGM. Meta-analysis in prediction research: incorporating published univariable associations in diagnostic and prognostic modeling. [Submitted]
Lu D, Vergouwe Y, Eijkmans M, Debray TPA, Koffijberg H, Moons KGM, Steyerberg EW. Prognostic models based on literature and individual data: a Bayesian method with data augmentation. [In progress]
Debray TPA, Smirnov E, Nalbantov G, Vandenbruel A, Aertgeerts B, Buntinx F. Choice of methods for classification in Class Imbalanced Datasets is based on a trade-off between performance and clinical interpretability. [In progress]
Debray TPA. Classification in Imbalanced Datasets. MSc thesis (2009). Department of Knowledge Engineering. Maastricht University , Maastricht, The Netherlands. [Full Text]