A Parameters Selection Scheme for Medical Image Registration

This paper presents a novel approach for automating the process of medical image registration. Considering rigid image registration, we propose a methodology in which the image registration process is divided in two phases. Phase 1 is used to obtain an accurate estimate of the rotations and a rough estimate of the translations. Phase 2 is used to finetune the translation estimates. In each phase, a fuzzy logic controller is used to adjust the registration parameters to obtain accurate transformation estimates. The proposed methodology produces highly accurate medical image registrations while reducing the required user interaction and maintaining excellent computational efficiency