Deformable Registration through Learning of Context-Specific Metric Aggregation

PUNEET K DOKANIA * (FRIST CO-AUTHOR); ENZO FERRANTE * (FIRST CO-AUTHOR); NIKOS PARAGIOS; RAFAEL MARINI SILVA

Quebec City

Workshop; 8th International Workshop on Machine Learning in Medical Imaging (MLMI 2017 - MICCAI 2017); 2017

MICCAI Society

We propose a novel weakly supervised discriminative algorithm for learning context specific registration metrics as a linear combination of conven- tional similarity measures. Conventional metrics have been extensively used over the past two decades and therefore both their strengths and limitations are known. The challenge is to find the optimal relative weighting (or parameters) of different metrics forming the similarity measure of the registration algorithm. Hand-tuning these parameters would result in sub optimal solutions and quickly become infea- sible as the number of metrics increases. Furthermore, such hand-crafted combi- nation can only happen at global scale (entire volume) and therefore will not be able to account for the different tissue properties. We propose a learning algo- rithm for estimating these parameters locally, conditioned to the data semantic classes. The objective function of our formulation is a special case of non-convex function, difference of convex function, which we optimize using the concave convex procedure. As a proof of concept, we show the impact of our approach on three challenging datasets for different anatomical structures and modalities.