Image registration is an important task in medicine, especially when images have been acquired by different scanner/sensor types, since they provid e information on different body structures (bones, muscles, vessels…). Severa l techniques have been proposed in th e past, and among those, Normalized Mutual Information has been proven as successful in many cases. Recently, Normalized Compression Distance, that makes use of real-world compresso rs, has been proposed as a simple yet effective technique for image registration. It is especially suitable for the case of CT-MRI registration, and may improve timings over Normalized Mutual Information. However, other image modalities such as PET pose some problems and do not achieve accurate registration. In this paper we analyze and propose a valid approach for image registration using compressi on that works properly for different combinations of CT, MRI and PET images.