Glioblastoma is the most common primary malignant brain tumor. It is infiltratively growing and associated with poor prognosis. Treatment is planned based on multi-modal anatomic and functional imaging and always includes a multi-disciplinary approach. So far, image structures are interpreted with basic volumetric measures at best, and the rich patho-physiological information of advanced clinical imaging sequences is not made use of, as the analysis of complex multi-parametric, multimodal and even multi-temporal image data sets is still a major challenge. In the proposed project, we will pursue a consolidated effort to analyse clinical imaging data together with genetic and histological information to generate new imaging, image processing, and modelling tools. We will estimate local tumor burden with advanced radiomics techniques, and integrate this with quantitative magnetic resonance imaging sequences. We will estimate tumor proliferation and infiltration patterns using biophysical growth models, and make an effort to develop new techniques for the probabilistic alignment of pre- and postoperative images using biomechanical priors. There will be a strong emphasis on machine learning in all four projects. Imaging and modelling tools will be implemented and validated on clinical data, to enable the translational use of our technologies and to maximize their impact on the personalized treatment of glioma patients.