Malignant gliomas are aggressive brain tumours with poor prognosis. Standard magnetic resonance imaging (MRI) is the current imaging modality of choice in diagnosis and follow-up of these tumours, but it lacks specificity for assessing tumour aggressiveness and for predicting treatment outcome. Therefore it is beneficial to develop a multi-modal imaging protocol in which different advanced magnetic resonance techniques (magnetic resonance spectroscopic imaging (MRSI), diffusion tensor imaging (DTI), etc) revealing complementary information about metabolic properties and tissue structure, are measured within an examination time acceptable for patients. This PhD project fits within a larger collaboration with the departments of Radiology/Neurosurgery at the University Hospitals of Leuven and Ghent, where different data acquisition and different tumour therapies are currently tested. Considering the diversity of imaging and spectroscopic data being collected for each patient over time, fundamental computational problems arise, such as reliable extraction of meaningful features from low quality spectroscopic (MRSI) data, or optimal combination of all available data sources for robust classification. The mainresearch task of this PhD project is thus the design of computational methods for the analysis of MRSI signals and multi-modal classification techniques, in particular metabolic feature extraction with 3D spatial prior knowledge, and multi-modal approaches to 3D nosologic imaging. Thesenew methods are meant to assist clinicians in the following areas: (a) establish an accurate diagnosis, (b) make an early prognosis on success of therapy, (c) identify areas of microscopictumour infiltration, (d) identify mechanisms that contribute to success and failure of (new) therapeutic interventions. Additionally, this work will help provide a common baseline such that the data acquisition protocols in the two hospitals (Leuven and Ghent) will be harmonized in order to acquire prognostic information within the shortest measurement time.