Registration of ultrasound liver images using mutual information technique

Abstract

Registration of medical images is done to investigate the disease process and understand normal development and aging. Image registration is the process of transforming different sets of medical image data into one coordinate system. Mutual information (MI) is a popular similarity measure for medical image registration. The preprocessing step for the registration process is done by means of two successive filters namely speckle reduction by anisotropic diffusion (SRAD) filter and median filter called S-mean filter. This work focuses on the registration of ultrasound liver images, and comparison is done by means of optimization techniques using MI, to bring utmost accuracy in computation time, and is very well suited for clinical applications.