Deformable registration methods for medical images: A review based on performance comparison

Deformable registration methods are widely used for the accurate registration of objects with large-scale deformation. in this paper,we present a detail review on performance analysis of deformable registration methods. we comprehensively review each registration method and describe its features,advantages,issues and challenges. deformable registration methods are further quantitatively compared and evaluated based on a set of criteria,which estimate the performance of each method. the performance of registration methods is estimated using root mean square error (rms),mutual information (mi),computational time complexity and memory requirement. it is found in our analysis that every registration method has its own strength to register deformable objects. however,due to large-scale variations in deformable objects most of the registration methods are not still a perfect choice in clinical applications. therefore,advanced and powerful registration methods are needed to develop in future,which can precisely,efficiently,and automatically register medical images with large-scale deformations. © pakistan academy of sciences.