PhD defence M.L.J. (Mathias) Polfliet

Dissertation in short:

This thesis deals with advances in groupwise image registration. Image registration remains an important task in medical image analysis. Whereas most methods are designed for the registration of two images (pairwise registration), there is an increasing interest in simultaneously aligning more than two images using groupwise registration given the increasing availability of medical imaging data, both at the individual and the population level. Groupwise image registration has shown promise in a number of applications dealing with large quantities of data, among others to increase registration accuracy and robustness, to improve the transformation smoothness and to reduce the methodological bias compared to pairwise registrations. However, directly comparing groupwise registrations to conventional repeated pairwise registrations is difficult due to several confounding factors impacting the algorithm. In this thesis, as a first contribution, we rigorously evaluate two registration methodologies in several experiments and investigate the di↵erences in performance. Secondly, we fill a gap in current literature on efficient (dis)similarity measures for multimodal groupwise image registration. These two contributions are distributed over four chapters.