BraTS is an international competition typically held every year focused on the evaluation of state-of-the-art methods for the segmentation of brain tumors in magnetic resonance imaging (MRI) scans. BraTS 2017 utilized multi-institutional pre-operative MRI scans and focuses on the segmentation of intrinsically heterogeneous (in appearance, shape, and histology) brain tumors, namely gliomas. In order to pinpoint the clinical relevance of this segmentation task, BraTS’17 also focused on the prediction of patient overall survival, via integrative analyses of radiomic features and machine learning algorithms.
The competition on prediction attracted a large number of groups from around 30 different leading universities and research centres. A team of MôLAB researchers took part in this competition in a team together with the University of Bern and Bern Inselspital. The predictive algorithms developed using machine learning approaches ranked second place.
The results of the competition were made public during the MICCAI 2017 conference in Montreal and the details can be found here