Detection of internal displacement of tissues in ultrasound images using image registration technique

Changes in soft tissue elasticity are usually related to some abnormal pathological process. The difference between the elasticity of normal and abnormal tissues is significant in terms of Young´s shear modulus. This opens a new way of cancer diagnostic by ultrasonic imaging of tissues in which the tissue elasticity is measured and compared with elasticity of normal tissues. In this paper the results of displacement estimation of normal and abnormal breast tissues under external stress are calculated using image registration technique. Two ways are proposed. In the first way the intensity centers of gravity (ICOG) of tissue appeared in two ultrasound images taken with and without enforcing stress are calculated. The displacement is calculated by finding the local translation of ICOGs. The translation information is then used to calculate the elasticity of tissue for diagnostic purpose. in the second way, a more general model for displacement is considered. In this model we estimate not only the translation parameters but also the rotation and scaling parameters. A few control points are taken from two images and then are used to estimate displacement parameters in a minimum mean square sense. We calculate Young´s modulus from displacement values and then use it to identify tissue characterizations. We show that these procedures can successfully detect abnormal tissues.