MICCAI 2019 Wednesday

MICCAI 2019 DAILY Théo Estienne 19 U-ReSNet: Ultimate coupling of Registration and Segmentation with deep Nets Théo Estienne is a PhD student at INRIA in Paris . His poster is about a network doing joint segmentation and registration. He has managed to make an end-to-end network without training two different networks separately, where you would have to freeze one, train the other one, and so on alternately. The main idea being that the segmentation will learn some brain structures which will help the registration to be done. This improves both segmentation and registration by training everything together.