In this paper, an automated approach to register CT and MR brain images based on mutual information and landmark features is described. Images are first rigidly registered by maximization of mutual information. A geometrical feature is then exploited by a differential geometrical operator. Landmarks are searched in the geometrical feature and registered elastically by thin-plate splines. Experimental results of 2D MR-TI to CT and MR-T2 to CT registration are presented.
Published in:
Intelligent Multimedia, Video and Speech Processing, 2004. Proceedings of 2004 International Symposium on
Date of Conference: 20-22 Oct. 2004