Algorithms for Medical Image Analysis
In medical image analysis it is often desired to compare or fuse several images. Therefore, one of the fundamental problems is finding correspondences between anatomic structures. This project focuses on image registration - a method which computes correspondences by searching for a geometric transformation minimizing application specific similarity measures. Registration is performed on grey-value images (CT, MRI, confocal microscopy), as well as derived representations (label-/ signed distance field, triangulation). Further information is available in the detailed project description.