Abstract

Magnetic resonance imaging (MRI) has become an essential tool in the diagnosis and managing of Multiple Sclerosis (MS).
Currently, the assessment of MS is based on a combination of clinical scores and subjective rating of lesion images by clinicians.
The purpose of this work is to provide a more objective classification of MS patients based on demographic features as well as
a range of quantitative geometric measures about the lesions which are obtained from MRI scans. Existing work on quantitative
modelling mainly focuses on lesion count and the total volume of all lesions in the brain. We show that, with the use of Minkowski
functionals to describe tumour geometry and image intensity features, we can build a reliable classification and prediction model
that achieves up to 64% total and 59% mean accuracy.