My experiences are mainly founded in the field of prediction research, Artificial Intelligence and programming. During my education at Maastricht University, I have performed a wide series of simulations and developed a range of applications that involve machine learning. More specific, I have worked with classification & regression, evolutionary algorithms, signal processing, intelligent search techniques, multi-agent systems, but also on more technical issues such as multithreading, generics and remote procedure calls. My master thesis studies the problem of classification in class-imbalanced datasets, e.g. diagnosis of diseases with a low prevalence. Its main contribution is the introduction of a new approach which outperforms a number of well-known modern approaches including over- and undersampling, bootstrap aggregation, MetaCost and cost-sensitive learning. This thesis was performed in cooperation with CEBAM (KULeuven) and Department of General Practice (Maastricht University) under guidance of Prof. F. Buntinx and Dr. E. Smirnov. In my current thesis I investigate and develop novel methods to synthesize clinical prediction models (meta-analysis). These methods are based on the principles of evidence based medicine and aim to improve the generalization of future prediction models. Main emphasis is placed on a Bayesian perspective and the use of hierarchical models with implementation in R and WINBUGS. The project is guided by Prof. KGM Moons (Julius Center, UMC Utrecht) and Prof. EW Steyerberg (Department of Public Health, EramsusMC) and supervised by Dr. H. Koffijberg and Dr. Y. Vergouwe. Besides performing research, I enjoy photography and creating digital photomanipulations. Recently, I have succesfully exposed some of my works at the Royal Conservatory of Brussels.