Machine Learning for Classifying Abnormal Brain Tissue Progression based on Multi-parametric Magnetic Resonance Data

In vivo Magnetic Resonance Imaging (MRI) represents one of the major breakthroughs in medicine and biomedical sciences. In recent years, conventional MRI techniques have been complemented by several advanced MR modalities that target not only the morphology of an imaged organ, but also its functional properties: for instance, MR spectroscopy/MR spectroscopic imaging reveal metabolic information, perfusion MRI reveals microvascularisation, diffusion MRI reveals tissue architecture, etc. In neuroradiology, conventional MRI has a rather poor specificity in the diagnosis,prognosis and therapy follow-up of brain tumours. Several advanced MRI techniques have shown their own potential in brain tumour applications.However, there is a strong need for integrated data processing methods that can take advantage of the strengths and complementarity of all the available advanced MR techniques, in order to make a real impact in routine clinical practice. This doctoral project will develop classification methods that can cope with multi-modal MR data (in particular, anatomic MRI, spectroscopy, diffusion and perfusion MRI), and apply them to the diagnosis and follow-up of brain tumour and multiple sclerosis. One of the main goals is to solve some of the standing problems in brain tumour applications, e.g., non-invasive tumour subtyping and grading, prediction of tumour infiltration, or tumour recurrence after treatment. To this end, several data processing steps need to be designed and optimized: computation of relevant features from the data, matching the image resolutions among the different modalities, multi-modal data fusion and classification. Multi-modal MR data from brain tumour patients will be available from UZ Leuven. The data comes from patients with untreated brain tumours (low and high grade glioma; metastases), as well as longitudinal follow-up of glioma patients focusing on a new treatment approachcalled dendritic cell therapy. Data from multiple sclerosis patients will be available through a collaboration with Universite Claude Bernard Lyon I.nrpages: 170status: publishe