Probabilistic Machine Learning for Lesion and Tumour Detection, Segmentation and Disease Prediction in Patient Brain Images

Abstract

Tal Arbel's research focuses on developing probabilistic machine learning techniques in computer vision and medical image analysis, with applications to neurology and neurosurgery. She will describe recent work developing probabilistic graphical models for brain tumour/lesion detection and segmentation, which were successfully applied to the MICCAI BRaTs Challenge public datasets and to large-scale, multi-scanner, multi-center clinical trial datasets of patients with Multiple Sclerosis.  Additional graphical models were developed for accurate detection and segmentation of lesions in contrast-enhanced images, as well as in longitudinal MRI, both important markers of new disease activity and for assessing treatment effects in clinical trials. She will  also describe recent work for the prediction of future new lesion activity based on baseline MRI, and for automatically identifying potential responders to treatment, leading to the possibility of personalized medicine. 

Biography

Prof. Tal Arbel, PhD, is the Director of the Probabilistic Vision Group and Medical Imaging Lab, Centre for Intelligent Machines, and a full professor in the Department of Electrical and Computer Engineering, McGill University. Her research program focuses on the development of probabilistic and machine learning techniques in computer vision and medical image analysis, particularly in neurology and neurosurgery. She has extensive expertise in developing probabilistic graphical models for brain tumour/lesion detection and segmentation. Tools developed in her lab are currently being used in the software analysis pipeline of her industrial partner for the analysis of almost all new MS treatments throughout the world. She is also working on computational neuroanatomy, with the objectives of generating automatic discoveries of the variability of the cortex from brain images, as well as the automatic identification of biomarkers of disease progression in Multiple Sclerosis and in cancer. Finally, her team has been focused on developing fast and accurate multi-modal image registration frameworks in order to improve image-guidance for brain tumour resections. Prof. Arbel has co-organized a number of major international conferences in two fields, including serving as co-organizer and satellite events chair for MICCAI 2017, area chair/program committee member for CVPR and MICCAI, and General Chair for a joint national conference (AI/GI/CRV/IS).  She currently serves on the editorial boards of IEEE Transactions on Pattern Analysis and Machine Intelligent (TPAMI) and the Journal of Computer Vision and Image Understanding (CVIU).